Citation
Carbonic anhydrases and bicarbonate transport in larval mosquitoes

Material Information

Title:
Carbonic anhydrases and bicarbonate transport in larval mosquitoes
Creator:
Seron, Theresa J
Publication Date:
Language:
English
Physical Description:
xiii, 147 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Antibodies ( jstor )
Bicarbonates ( jstor )
Complementary DNA ( jstor )
Enzymes ( jstor )
Epithelial cells ( jstor )
Larvae ( jstor )
Messenger RNA ( jstor )
Midgut ( jstor )
pH ( jstor )
Protein isoforms ( jstor )
Dissertations, Academic -- Fisheries and Aquatic Sciences -- UF
Fisheries and Aquatic Sciences thesis, Ph. D
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 2004.
Bibliography:
Includes bibliographical references.
General Note:
Printout.
General Note:
Vita.
Statement of Responsibility:
by Theresa J. Seron.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
022441638 ( ALEPH )
880637440 ( OCLC )
880438533 ( OCLC )

Downloads

This item has the following downloads:


Full Text











CARBONIC ANHYDRASES AND BICARBONATE TRANSPORT
IN LARVAL MOSQUITOES















By

THERESA J. SERON













A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY



































Copyright 2004

by

Theresa J. Seron














DEDICATION

I wish to dedicate this dissertation to my incredible family, Mom, Dad,

Grandparents, Tracey, and George. My family has witnessed my struggles and triumphs and has helped me through it all. I am so proud to call them my family. I cannot thank them enough for all of their support. This achievement is really a reflection of all of us.

I also want to dedicate this milestone to my soon to be husband, Dr. Peter Lovell. Coming home to his love, humor, and music, has given me true joy. My family would be incomplete ifI did not mention our furry companions, Frodo and Princess, who remind us that a nap can solve most problems.














ACKNOWLEDGEMENTS

I would like to acknowledge my dissertation committee, Dr. Paul J. Linser, Dr. Edward J. Phlips, Dr. Leonid Moroz, Dr. Robert Greenberg, and Dr. Shirley Baker, for their suggestions and comments on the final rewriting of this document. I want to thank my project supervisor, Dr. Paul J. Linser, for allowing me to form my own project goals and the space to tackle them.

There are a number of people at The Whitney Laboratory who I would like to thank for their assistance with this dissertation project as well as their friendship. Dr. Judith Ochrietor devoted her time and energy to improving every aspect of this dissertation. Judy assisted me with experimental designs, introduced me to real time PCR, provided a wealth of knowledge about molecular biology, and was a great person with which to share a laboratory and office. Dr. Andrea Kohn provided molecular biology teaching and advice along with being a fantastic person to work with and be inspired by. Leslie vanEkeris taught me how to do mosquito dissections and provided many of the mosquito guts that I photographed for this document. Dr. Bill Harvey provided insight into the ionic transport mechanisms of the mosquito and the editing of this manuscript. Dr. Dmitri Boudko was instrumental in the expression of the anion exchanger and the production of amplified cDNA libraries. Jessica Roberts-Misterly and














TABLE OF CONTENTS



DEDICATION ...... ........................................... .. .. ........... ........... ....... ...........................
ACKNOWLEDGEMi ENTS .. I...i..iti................................................................................. i

LIST OF TABLES................................ .......................................................................

ELIST OF FIGUREiS .............................................................. .....................................1 Ix

AB STRACT .....d......F..... .............................................................................................1 i

CHAPTER

1 INTRODUCTION ........................................1

Alkaline Gut ................. ................... ...................................................................16
m O cCarbonic Anhydrase. .Activiy......................... 17...4
Mosquito Developent and Control.. ......... ......................................................
Carbonic Anhydrase Inhibition ......t*.............................. .........................................6
Bicarbonate Transport .................. .... .............. .................. ........................
Gut Ao Malizazion Model .............. .... ..................................... ................
Specific Ais............ ............................................................................................18





2 MATERIALS AND MEg o , ir e DS... ci ........ir l . tic....it.................................................... 19
Experimental Insects.......... .... ..... ..............................................14
Preparation and FEixation of Tissue............ ............................................. 15
Bromnothymol Blue Qualitative Assay ................ ..... ................... ..............16
]Effect of Methazolamide on the Alkcalization of the Midgut of Live Larvae. .... ... 16 180 Exchange Method to Measure Carbonic Anhydrase Activity...................,...... 17
Isolation of RNA and Synthesis of eDNA ............... .... ............. ....... ................... .. 18
Bioinformatics ................... ................... ................... .............................1
Cloning of CA fi'om Aedes aegpti Larval Midgut ............................. .. ............ 19










Immu1 niso Prchmstcrn ........ .... .. . .. . . . ....... . s..... ...... ...... ... ....... . .... . ..... s.. 30
Immunohistochemistry ..... ..... ........ .......... ..... ..... ... ... ... .. ........... ...3
CA Protein ressin ..... . ... . .. . .... . . .. . . ..... ....... . . ... ... ....... ...... .. ....... ... a . .. .32
Anion Exchanger Oocyte Expression ...... ..... ... ..... ..................... .... ........... ...... .. .. . 34
Anion Exchanger Physiology .............. ....... . .... . ..............a*. ..... ..... ...... ....34

3 CARBONIC ANHYDRASE IN THE MIDGUT OF LARVAL AEDES
AEGYPTI: CLONING, LOCALIZATION, AND INHIBITION ................................39
Introduction ................... ........... ..... ... ...................................................3
Results...... ............. ..... ... . ... a. .......tAss ........ .. ....... ... . . . . ... . . .... . ... . .. .. .. . . . . . . . ... 4 0
Bromothymol Blue Qualitative Assay ..... ..... ......... ............ .... ... .......... ...... ....40
Carbonic Anhydrase Activity and Alkalization ... . . . . ... . . . . . .. . .... .. . ... . . . . . . . . . . . . . .. 4 2
'80 Isotope-Exchange Experiments ................... ................... ................... ........43
Cloning of Carbonic Anhydrase from Aedes aegypti Larvae ..........................43
Localization of the Enzyme in the Midgut Epithelium: Carbonic
Anhydrase Enzyme Histochemistry ................... ...................... ........ ....... ... ..45
In Situ H ybridization ..................... .... ................ .... ................. ........4 6
Discus sion................... ................... ................... .................. .........4

4 A GPI-LINKED CARBONIC ANHYDRASE EXPRESSED IN THE
LARVAL MOSQUITO MIDGUT .................. . ............. ............... ................ .......... 61
Introduction ., ........ ................... ................................................6
R esults.......................................................... .........................................................62
Bioinformatics ofAedes aegypti CA ................. ...... . .... .................. .......... . ...62
Sequence Comparisons of CA IV - like Isoforms .. . . . ... .... . . . ........ ... . . . . . ... ......... 62
Localization of CA IV-like Isoform in the Mosquito Midgut .................... .....65
Real Time PCR Analysis ofAedes aegypti CA IV-like Transcripts ...............65
Immunolocalization of CA IV-like Protein in the Mosquito Gut....................66
Antibody Cross-Reactivity with Other Mosquito Species ....... ..... .....................67
Phospholipase C Treatm ent . .......... . ... .. .......... .... . ... ................... ....... . .........68
Discussion ................... ...... ....... ...... ................... .,......... ..................6

5 ANION EXCHANGER EXPRESSED WITHIN THE LARVAL
ANOPHEL ES GAMBIE MO SQUITO. .... . ... ................... .... .. .... .....8

Introduction .... ... .. ........ . ..a.. . ............ a.e....... .... . .a..... .. t...... ..83
R esults... ... ......... ... ..... ... . . .. ...... ...... .. . . . .... . ..... .. ......... ... ..... ...... ...... ... ... ..84
An. gambiae AE Sequence Analysis ...... .... . .... ..... ... ....... ....... 8'4
BT Sequence~~r Coprsn..................... .......................8










6 CYTOSOLIC CA EXPRESSION IN LARVAL ANOPHELES GAMBIAE............. 115

Introduction ................... .............. ............... .. ......................................... ......... 1 15
R results ............................................. ............ ... .................. ............ ................1 16
A nopheles gambiae (1CA Sequence Anralysis ........................... .................. ..1 1 6
Localization of CA Activity in Anopheles gambiae Larvae ...................... ...118
Localization of Cytosolic CA mRNA in Anopheles gambiae Larvae ........... 118
Antibody Localization of CA Protein ....................*........... ......................... .1 19
Bacterial Expression and Purification of Anopheles gambiae
C y2)tosolic C A . . ........................ ....................... ................ ..........................1 19
D iscussion .......................................................................... ..................................120

7 CONCLUSIONS AND FUTURE DIRECTIONS........................ . ......... .... .... ... 131

C onclusions.......................... ........ ....... .... ...... ....... .......... ....... ....... ............ .......1 3 1
New Model ........................................ ............................................................. 134
Future Directions ..................... ........... ........................ete...... .....................137

REFEENCES ......................... .. ...................... .. ......... ................................................140

BIOGRAPHICAL SKFETCH .................. .....................*......*........................ ............147















LIST OF TABLES

Table .............................................................................................................................
2 - 1***. PC**primerseq** u* ences** .**.*.*.* . *. .*.. . . . . . . .. .. . . . . .. . . .. .. ... . . .. .. *.. ... . .. . ....... ... . .. .. .... *.*..
2-1. PCR prim er sequences ....b.....*........................................ ...........

2-2. Composition of all solutions used in Xenopus oocyte expression of
An. gambiae AE ................. ...................................... .........















LIST OF FIGURES

figure .. .... ............................ ................... ........................... ''..**'***''***"'*'*''*'''***'*** "' .

1-1. Illustration showing the regions of the larval mosquito gut .................................. 11

1-2. Illustration of the mosquito life cycle .............. ..... ....................... ............ ............. 12

1-3. Preliminary mosquito anterior midgut model based on M Sexta...........................13

2-1. Efficiency plots for real-time PCR primers. . .....................s .......... ........... ... ..37

2-2. Three-dimensional (Cn3D) depiction of human CA IV (1ZNC)..........................38

3-1. Effect of CA inhibition on culture medium pH with fourth-instar Ae.
aegyp ti. larvae ...................................... ...........................................O. ..o..O......... ..5 3

3-2. Effect ofmethazolamide on the alkalization of the midgut using
Bromothymol Blue (BTB) assay ofpH within living, but isolated, gut tissue......54 3-3. Relative activity of CA in different pooled segments of the midgut of larval
Ae. aegpti... ..................... .............. ................... .............. . ............ ....5

3-4. Carbonic anhydrase from the midgut of larval Ae. aegypti ........................... .. .....56

3-5. Comparison of the extrapolated amino acid sequences of A-CA with
six putative dipteran CA genes identified in the D. melanogaster
gene databases........... ......... . . 0 ........... .............. . .. .......... .... ......................... ...... 57

3-6. Polymerase chain reaction (PCR) analysis ofAe. aegypti amplified cDNA
frm different gut regions ............ . ............ ....... ......... ... .... .......................... 58

3-7. Hansson' s histochemistry of whole mount Ae. aegypti gut ................... ...............59

3-8. Localization of CA mRNA expression in larval Ae. aegypti ...................... .... ..... ... 60









4-3. Localization of CA mRNA in a whole mount preparation of early 4t
instar Ae. ae yp i ................... ................... ... .................... ....................... .......74
iegyjpti..a..m...........a *..... ..*...a ....... .

4-4. Expression of CA mRNA in Ae. aegypti anterior midgut ............................... ...... ..75

4-5. Localization of CA V-like message within Ae. aegypti CNS tissue ....................76

4-6. Relative quantification of CA IV-like message in Ae. aegypti larvae
using real tim e PCR ................. ...... .m..... ..................................... ..... ... .a ... ..... 77

4-7. Ae. aegypti and An. gambiae CA protein labeling . . . . . . . . . . .. . . . . ...... .......... . ...... ...78

4-8. The Ae. aegypti CNS ganglia express the CA IV-like isoform ............................79

4-9. Immunolocalization of mosquito CA IV-like enzyme in Aedes albopictus ..........80

4-10. High magnification ofimmunoreactive muscle fibers within the Aedes
albopictus m idgut. ..................... ......... ..................................................... .... ...........81

4-11. Immunoreactivity ofAe. aegypti guts for the CA IV-like isozyme ................... ...82

5- 1 . Structural prediction of the An. gambiae AE 1 . . . . . . . ... ... . . . ... . . .... ... . .. ...... . .... .....96

5-2. Putative amino terminus CA II binding motif ............ ...... ................. ......... ........97

5-3. Homology tree depicting the amino acid identity between several BTs ...............98

5-4. Alignment of carboxy terminus amino acids ofAn. gambiae and D.
m elanogaster A Es. ..... ......... .... .. . . .... ..... .... ........ ...........................a............ ..99

5-5. Alignment of An. gambiae and human AEs ..... .....................0............ 1 00

5-6. Localization of AgAE1 mRNA within whole mount An. gambiae
larvae................... ............. .... ..................... ., ..........a.... .101

5-7. Localization of AgAE1 mRNA in muscle, nerve, and trachea in
An. gambiae ...............li..e.m...............ease......m em.102

5-8. In situ hybridization of AgAE1 in whole mount An. gambiae consistently
shows positive labeling of tracheal fibers along the midgut..... ......... ..... ...... 103









5-11. Larval An. gambiae displays strong AgAE1 expression in the hindgut,
the pylorus .....(). . .... . .... .. ...... . .... . ..6. . . . . . ............ .... 106

5-12. Localization of AE mRNA in An. gambiae shows abundant labeling of
the M alpighian tubules ........... ..... . ..... .... .. . ..... ...... .... ....... . .... ..... ......... ............ . 107

5-13. Expression of AE mRNA was found throughout the ventral midgut
g a n g lia ................... ..... .. . ...... .. .. .... - . � . � .. . � . . . . - -............- � ... " .. . - - - * - - - - *1 0 8

5-14. Sense AE probes display no specific hybridization .... ........... ............. .... ..... .... . ...109

5-15. Antibody localization of AgAE1 protein to the gastric caeca in An.
gam biae larvae ...... ....... ................... ...... ... .... .. ...... .. ..... ...... . -. .... -. ..... .. ........... ......110
5-16. Localization of AgAE1 protein within the PMG ofAn. gambiae larvae.............111

5-17. Neuronal cells within the AMG display immunoreactivity for our
An. gambiae AE specific antibody...... ..... .. ..... .. ...... . ...... ..... . ... ..... ..... . ....... .. ... ..112
5-18. Current-voltage (I-V) plots depicting ion transport by the AgAE1 expressing
oocytes in contrast to the water injected control oocytes .................................... 113

5-19. Inhibition of AgAE1 mediated chloride transport by DIDS.................................1 14

6-1. Clustal alignment of active sites within An. gambiae, D. melanogaster,
and hum an CA proteins .... . ....... .. ...... .... . .................. .. ...... ..... .... . ............ ... ........ .. ..... . 122
6-2. Phylogenetic analysis between mammalian (human and mouse) and
dipteran (An. gambiae and D. melanogaster) CAs ...................................... ........123

6-3. Localization of An. gambiae CA activity ...... . . . ..... ....... .... ...... ... . ...... ..... ......1 24

6-4. Localization of CA mRNA expression within An. gambiae whole mounts........125

6-5. Localization of CA mRNA expression within the posterior midgut of An.
gam biae .... ............. . .......... .. ....... ............... ... ........... ........................ 2

6-6. Localization of CA mRNA expression within the hindgut.............................. ....127

6-7. Localization of CA protein within gastric caeca of An. gambiae larvae .............128














Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy


CARBONIC ANHYDRASES AND BICARBONATE TRANSPORT IN LARVAL MOSQUITOES


By

Theresa J. Seron

May 2004

Chair: Edward J. Phlips
Cochair: Paul J. Linser
Major Department: Fisheries and Aquatic Sciences

Carbonic anhydrase (CA) is an important enzyme due to its involvement in many pH-dependent, physiological processes. CA reversibly converts CO2 and -I20 into bicarbonate and a proton. The anterior midgut lumen of the larval mosquito has an extremely alkaline pH; therefore, we hypothesized that an active CA within the epithelial cells surrounding this region would rapidly produce bicarbonate to buffer the high pH. A cDNA cloning strategy, followed by in situ hybridization, was employed to isolate and localize CA and anion exchanger (AE) transcripts within the mosquito gut. Localization of CA enzymatic activity was assessed via histochemical analyses. Enzymatic and electrophysiological analyses of recombinant CAs and AE were also performed. In this









basal side of the anterior midgut. This CA resembles the mammalian CA IV isozyme in that a glycosylphosphatidylinositol (GPI)-link tethers the enzyme to the extracellular membrane. The other CA isoform, an active cytosolic enzyme, was localized to the gastric caeca and posterior midgut regions. It was also determined that the AE transports chloride and is expressed in the gastric caeca, posterior midgut, and Malpighian tubules. We were unable to detect CA within the anterior midgut epithelial cells using a variety of assays. My studies have therefore led to an alternative hypothesis that one or more CAs within the mosquito gut, but located outside of the anterior midgut epithelial cells, contribute to buffering the alkaline pH of 11 within the anterior midgut lumen. The localization of two CA isoforms, one extracellular and the other cytosolic, and an AE possessing a putative CA binding sequence, to the regions flanking the anterior midgut, supports the prediction of a bicarbonate transport metabolon within the gastric caeca and posterior midgut regions. Such a metabolon has only been studied in mammals, however, the colocalization of CA and AE within the mosquito gut suggests a similar network of bicarbonate production and transport.













CHAPTER 1
INTRODUCTION

Insects represent one of the most numerous and diverse groups of animals on the

planet. One particularly successful group of insects is the well-studied Pterogota (winged insect) group. This grouping includes the Lepidoptera (butterflies and moths) as well as the Diptera (flies). These insects have been extensively studied due to their huge impact on the lives of humans. For example, the Lepidopteran, Manduca sexta, is a great pest to tobacco companies that rely on abundant and healthy tobacco crops, on which M sexta feeds. Also, mosquitoes (Dipterans) are responsible for transmitting a host of diseases to humans as well as other mammals by injecting pathogens, along with the anti-coagulants from their salivary glands, to aid in bloodletting. The pathogens that cause these diseases can be viruses or various parasites (eg. protozoans).

Mosquitoes belong to the order Diptera, family Culicidae. According to the American Mosquito Control Association, there are more than 2500 different species throughout the world, with 150 species in the United States (Darsie and Morris, 2000; Spielman and D'Antonio, 2001). Mosquitoes act as vectors for a wide variety of diseases such as malaria, yellow fever, west nile virus, and dengue fever. Recent reports estimate that fifty to one hundred million cases of dengue fever occur annually, along with several hundred thousand cases of the life-threatening form of the disease, dengue hemorrhagic






2


(Gubler, 1997). Another mosquito example, Anopheles gambiae kills millions of people each year in Afirica by infecting them with the deadly Plasmodium parasite that causes malaria. Many studies have therefore been undertaken to understand the life cycle and physiology of these insects that exert such a large socio-economic impact.

The mosquito's ability to acquire, harbor and transmit deadly pathogens has

spurred research into the workings of the mosquito gut. Specific cells of the midgut, which express a proton pumping V-ATPase, have been found to be preferentially invaded by pathogens (Shahabuddin and Pimenta, 1998). Studies have also shown that the mosquito gut is not a static organ but is comprised of several different regions. Each region displays different characteristics and is made up of different cell types.

Alkaline Gut

Larval mosquitoes, as well as some caterpillars, are known to possess a highly alkaline digestive system (Dadd, 1975). The tobacco hornworm, M sexta, has a gut lumen pH that can exceed 11, while the larval mosquito, Aedes aegypti, displays a pH greater than 10 in its anterior midgut region (Zhuang et al., 1999). These insects are not only unharmed by this caustic pH, but are able to generate this property while maintaining cellular homeostasis.

The larval midgut is involved in ionic and osmotic regulation as well as digestion, absorption, and excretion (Clements, 1992). It is subdivided into four structurally distinguishable regions: cardia, gastric caeca, anterior stomach, and posterior stomach (Fig. 1-1). Each of these regions consists of one cell laver of eithelial cells, composed





3


cells are capable of maintaining physiological homeostasis while facing a pH range of 711 along the length of the mosquito gut lumen (Dadd, 1975). This range in pH, along the length of the mosquito gut, is presumed to support digestive and assimilation functions (Clements, 1992). The epithelial cells of the anterior midgut (AMG) surround a highly alkaline lumen (pH 11) while those of the gastric caeca (GC) and posterior midgut (PMG) surround a neutral to mildly alkaline lumen (pH 7-8; Clements, 1992; Zhuang et al., 1999). The different pH values found along the midgut may support the various metabolic functions that are active in each gut region. The gastric caeca perform ion and water transport, the anterior midgut performs alkaline digestion, the posterior midgut performs nutrient absorption, and the Malpighian tubules (part of the hindgut) actively transport potassium and fluid (Clements, 1992).

The role of the alkaline pH in the anterior midgut is a point of some controversy.

It has been suggested that the high pH contributes to the digestion of plant detritus and, in particular, to the dissociation of tannin-protein complexes (Martin et al., 1980). The high pH restricts the conglomeration of proteins within the anterior midgut that could interfere with the insects' nonnrmal physiology. These complexes could also interfere with insect digestion by blocking the active sites of many different digestive enzymes. Therefore, the alkaline gut serves as a proposed benefit to the insects by allowing ingested food to remain soluble. The alkalinity therefore keeps the gut free from attachable tannin-protein complexes and enhances the assimilation of proteins. Berenbaum's review (1980) of Lepidopteran insects correlated gut pH (range from 7.0-10.3) with diet. Caterpillars





4


Although this alkaline digestive strategy is well documented in insects, the molecular processes involved have not been clearly defined.

Carbonic Anhydrase

Carbonic anhydrase (CA), a blood enzyme, first described by Meldrum and Roughton in 1933, catalyzes the reversible hydration of carbon dioxide to form bicarbonate and a proton (CO2 + H20 ++ HCO3" + H+; Meldrum and Roughton, 1933). Carbonic anhydrase was first characterized in erythrocytes as the result of a search for a catalytic factor that would enhance the transfer of bicarbonate from the erythrocyte to the pulmonary capillaries (Meldrm and Roughton, 1933). Since it was first described, CA has been shown to play an important role in most acid/base transporting epithelia. Fourteen different CA isoforms have been characterized to date in mammals (HewettEmmett and Tashian, 1996). These enzymes have been determined to function in pH regulation and ion balance, thereby performing a crucial role in many biological processes such as respiration, bone resorption, renal acidification, gluconeogenesis, aqueous humor production, gastric acid production, cerebrospinal fluid formation, and signal processing (Dodgson, 1991; Sly and Hu, 1995; Hewett-Emmett and Tashian, 1996; Lindskog, 1997; Sun and Alkon, 2002).

Various types of epithelial cells, such as those described in the mammalian

kidney, contain CAs that can provide large quantities of bicarbonate for buffering cells and their microenvironment. Polarized epithelia play an important role in partitioning chvsioloically distinct compartments. and in maintaining cell and tissue homeostasis.






5


differential roles in homeostasis and function. Elucidating the distribution of CAs along the mosquito midgut epithelium may uncover the mechanisms responsible for the unique alkaline physiology of the mosquito gut.

Mosquito Development and Control

Part of the success of insects can be attributed to the structural adaptation of their integument, which functions as skin, skeleton, sensory and respiratory organ, and food reserve (Rockstein, 1964). The advantage of having an extremely strong integument is offset by the disadvantage of not being able to grow significantly in size. Insects have overcome this growth-limiting problem by shedding their integument and rebuilding a new larger one. This process of ecdysis (molting) is used as a tool for marking the different stages of development in many insect species. While mosquito control can target different stages of mosquito development, this project focuses on the larval enzymes, specifically early fourth instar, which begins immediately after the third molt. Careful attention was paid to the stage of insect development in all experiments due to a previous study that showed insect enzymes to decrease or completely arrest prior to molting (Jungreis et al., 1981).

The mosquito life cycle begins at hatching from the egg (Fig.1-2). At this point the fully independent mosquito is called a first instar larva. Successive molts mark the transition to the next larval instar, four larval instars in all. In each instar, the larvae possess a series of morphological characteristics, some particular to that stage. However, there are only slight changes in internal organs such as the midgut. Within a day or two





6


species require a bloodmeal in order to nourish their developing eggs. However, the males do not ingest blood but instead feed on fruit or do not feed at all (Clements, 1992).

Mosquito control tactics use different methods for controlling mosquito larvae as compared to the flying adults. Mosquito larvae are confined to the water in which they develop, whereas the adults are free-flying and therefore highly mobile. Pesticide sprays are employed against the flying adult mosquitoes, but dragonflies and butterflies are also ill-affected. An arguably better strategy for mosquito control is to target the larvae before they are capable of biting and transmitting disease. Mosquito larvae are voracious eaters, incessantly consuming particulates in the water around them, taking in almost anything. Because of this non-discretional eating behavior, the wriggling larvae can potentially consume a larvacidal agent if placed in the water. Determining the physiological roles of larval mosquito gut enzymes and metabolic transporters may provide a lead for constructing mosquito larvacides.

Carbonic Anhydrase Inhibition

The focus of this project is to examine the distribution and expression of CAs

within the fourth instar of larval development of two species of mosquito, Ae. aegypti and An. gambiae. A tangential result of characterizing mosquito CAs may be in the development of mosquito-specific inhibitors. If a CA is discovered to be essential for mosquito development or homeostasis, a specific inhibitor of precisely this mosquito CA isoform could be developed. Since virtually all organisms contain CA enzymes, an inhibitor that would compromise this mosquito CA while not affecting any other





7


characterization of mammalian CA isoforms. For example, the acidic sulfonamide benzolamide has been used for the preferential inhibition of extracellular CA while not compromising any intracellular CA activity (Tong et al., 2000). This occurs due to the inability of benzolamide to readily penetrate cell membranes (Tong et al., 2000). The wealth of information pertaining to mammalian CA isoforms and their specific inhibitors provides a basis for comparisons with CAs that are discovered in the mosquito midgut. Sulfonamide CA inhibitors are widely used to treat a number of conditions including glaucoma, gastro-duodenal ulcers, and cancer, by lowering the production of fluids and acids. Parkkila et al. (2000) showed that the invasion of renal cancer cells in vitro could be inhibited with CA inhibitors. If larval mosquito physiology is dependent upon the generation or maintenance of the alkaline gut, and CA is a necessary component, then the possibility exists for the use of CA inhibitors as mosquito larvacides.

Bicarbonate Transport

The site(s) of bicarbonate production by CA may not be as important as the

translocation of the bicarbonate that is produced. Transporters can facilitate the passage of bicarbonate and other ions through otherwise impermeable cell membranes. Bicarbonate transporters compose a large family of membrane proteins that includes the anion exchangers (AEs), sodium bicarbonate cotransporters (NBCs), and members of the sulfate transporter group that can also transport bicarbonate (Alper et al., 2001). Most of the BT proteins consist of a cytosolic anchoring domain as well as a 10-14 membranesnanning transnorter domain (Alper et al.. 2001). Also. evidence exists that some AEs






8


intracellular carboxy terminus of AE1 was found to contain a cytosolic CA II binding site (Vince and Reithmeier, 2000; Sterling et al., 2002a). A metabolon, a complex of membrane proteins involved in regulation of bicarbonate metabolism and transport, defines the relationship between the CA and AE proteins (Sterling et al., 2001a). This bicarbonate transport metabolon, is thus capable of transporting bicarbonate as soon as it is available from the CA enzyme. Transport can be in either direction, into or out of the cell, and is therefore predictively capable of maintaining a tight hold on pH. The occurrence of such a tight bicarbonate control mechanism could be very advantageous to the mosquito. With such a large pH gradient across the membrane, a bicarbonate transport metabolon could ensure that the pH on either side of the membrane is strictly monitored. This bicarbonate transport metabolon has only been identified in a mammalian system. Despite this fact, an insect gut model that employs such a bicarbonate transport metabolon is easy to envision. Because of the strong pH gradient that is maintained in the mosquito gut, it is reasonable to propose that a bicarbonate transport metabolon could exist in this system as well.

Gut Alkalization Model

My first physiological model of the larval mosquito midgut was derived from the tobacco hornworm, M sexta, which also uses an alkaline digestive strategy. In this model, several proteins contribute to the high alkalinity (Fig. 1-3). These are the CA, the I1 V-ATPase, and the cation and anion exchangers. The H' V-ATPase is thought to be the energizer of the system by using ATP, and pumping protons out into the lumen of the





9


combines carbon dioxide and water to produce bicarbonate and a proton ion. The bicarbonate is pushed from the epithelial cell, across to the lumen side by the anion exchanger, in trade for a chloride ion. The proton then gets stripped off of the bicarbonate and, along with the proton pumped across by the V-ATPase, is brought back into the cell in exchange for a potassium ion (Wieczorek et al., 2000). This potassium ion combines with the carbonate to produce potassium carbonate, which is hypothesized to be responsible for the high alkaline pH of the anterior gut region. This hypothesis stems from the fact that potassium ions are actively produced by the Malpighian tubules and are circulated throughout the gut via the hemolymph (Clements, 1992). Potassium carbonate also has a pKA greater than 10 and can therefore contribute to the gut alkalization.

The goal of this project was to expand and adapt this model to the larval mosquito by completing several clear objectives. These objectives are outlined within the following specific aims.

Specific Aims

1. Determine whether CA is involved in buffering the high alkalization of the larval
mosquito gut.

A. Determine if a CA enzyme is present within the mosquito gut. Determine which
regions of the larval Ae. aegypti gut display CA activity using CA histochemistry
and 'O80 isotope exchange.

B. Determine if CA-specific inhibitors, such as acetazolamide, can influence larval
midgut alkalization.

2. Determine whether CA is expressed in the larval mosquito gut.






10


B. Use experimental and bioinformatical approaches to determine if CA expressed in
the mosquito gut is similar to a characterized mammalian CA isoform.
Furthermore, determine the subcellular location of mosquito CA isoforms as
cytosolic, membrane-bound, mitochondrial, or GPI-linked.

C. Determine which regions of the mosquito gut express CA mRNA and protein
using in situ hybridization, real time PCR, and immuno-localization.

3. Determine whether anion exchangers (AE) are involved in the pH regulation of the
larval mosquito gut.

A. Clone and characterize an AE from the larval mosquito gut that uses the
bicarbonate produced by CA as a substrate.

B. Determine whether the mosquito AE transports chloride using a Xenopus oocyte
expression assay.

C. Determine if AE is expressed in the same regions of the mosquito gut as the CA.
Co-localization of CA and AE would support the existence of a bicarbonate
transport metabolon.

D. Determine if the AE contains the amino acid sequence predicted to be necessary
for binding CA. If indeed the AE protein is predicted to bind CA, a bicarbonate transport metabolon within the larval mosquito gut could maximize bicarbonate
production and transport.

4. Present a new larval mosquito model that reflects the studies in this dissertation.
Bring together all localized components of mosquito gut physiology into one model.






11









Midgut Hindgut
II
GC AMG PMG MT
cardia rectum











Figure 1-1. Illustration showing the regions of the larval mosquito gut. The midgut is
compsed of the cardia, gastric caeca (GC), anterior midgut (AMG), and the posterior midgut (PMG). The hindgut is composed of the Malpighian
tubules (MT) and the rectum.





12




2nd
1t instar 2nd instar eggs larva larva hatch Ist molt 2nd mofl

-- rd,
3rd instar . larva

3rd
molt adult

4th instar larva emerge
4th
moff



pupa

Figure 1-2. Illustration of the mosquito life cycle. The four life stages are egg, larva,
uanna and adult The larval stane consists of four different instars. Early





13


Blood ClIpllhr gut ml Lumt () (ea) (Ahe) pin 6.5 pIs 7.0 pH= 11.0




ItlPj.
='-l1) ) =3
rATP















AYb = 0mV AY =240 mV I I
* AYT 210 mV +


HV-TPm B Aunocarrdng K puy H,-A Ada= Calon i iii: g 0 Caanlcansvdms

m Amino acid: K+ etrn.1oror
Channel :

Figure 1-3. Preliminary mosquito anterior midgut model based on M sexta. This
theoretical model places a CA II-like isoform within the cell cytosol where
it combines carbon dioxide and water to form bicarbonate and a proton.
Alkalization is driven by a proton pumping V-ATPase that resides in the
apical membrane and pumps protons into the lumen. A chloride/













CHAPTER 2
MATERIALS AND METHODS

Experimental Insects

Ae. aegypti eggs were obtained from a colony maintained by the United States Department of Agriculture (USDA) laboratory in Gainesville, Florida. The eggs were allowed to hatch in 20 ml of 2% artificial seawater (ASW; 8.4 mM NaC1, 1.7 mM KCI,

0.1 mM CaCl2, 0.46 mM MgC12, 0.51 mM MgSO4, and 0.04 mM NaHCO3). The mosquito larvae were reared in 2% ASW at room temperature. The Ae. aegypti larvae were fed a mixture of yeast and liver powder (1:1.5 g respective dry weight; ICN Biomedicals Inc., Aurora, Ohio). Eight to ten days were required for this species to reach the early fourth instar.

An. gambiae eggs were obtained from the Centers for Disease Control and

Prevention (CDC) in Atlanta, Georgia. Strict handling guidelines were followed with this particular species, which does not currently inhabit Florida, due to its inherent ability to acquire and transmit the Plasmodium protozoan, which causes malaria. This Anopheles species was therefore reared in deionized water inside of a locked incubator set at 300C. A mesh screen served as a second barrier within the incubator while the sealed (but not airtight) containers harboring the An. gambiae larvae served as the third barrier against escane The An. gambiae larvae were fed a Wardlev tronil fish flkce fnnd (The Trt7





15


reach the early fourth instar. Late fourth instar larvae that went unused were sacrificed to prevent any chance of emerging adults.

Preparation and Fixation of Tissue

To dissect out the midgut, the heads of the cold-immobilized larvae were pinned

down using fine stainless-steel pins to a Sylgard layer at the bottom of a Petri dish containing hemolymph substitute solution consisting of 42.5 mM NaC1, 3.0 mM KC1, 0.6 mM MgSO4, 5.0 mM CaC12, 5.0 mM NaHCO3, 5.0 mM L-succinic acid, 5.0 mM L-malic acid, 5.0 mM L-proline, 9.1 mM L-glutamine, 8.7 mM L-histidine, 3.3 mM L-arginine, 10.0 mM dextrose, 25 mM Hepes and adjusted to pH 7.0 with NaOH (Clark et al., 1999). The anal segment and the saddle papillae were removed using ultra-fine scissors and forceps, and an incision was made longitudinally along the thorax. The cuticle was gently pulled apart and the midgut and gastric caeca were removed. In some cases, the gut contents enclosed in the peritrophic membrane slid out, leaving behind the empty midgut. In other cases, it was necessary to remove the peritrophic membrane and its contents manually. For enzyme histochemistry, fixation was in 3% glutaraldehyde in 0.1 M phosphate buffer, pH 7.3, overnight at 40C (Ridgway and Moffet, 1986). For in situ hybridization and immunohistochemistry, dissected tissues were fixed overnight in 4% paraformaldehyde in 0.1 M cacodylate buffer, pH 7.2 (0.1 M phosphate buffer, pH 7.2 was used for Ch 4 and 5 in situ). In some cases, the dissected larval midguts were photographed using a Nikon FX-35DX photographic camera mounted on a Nikon SMZ10 dissecting microscope. In other cases, digital images were acquired using a Leica






16


Bromothymol Blue Qualitative Assay

A qualitative test to detect carbonic anhydrase activity in mosquito larval midgut homogenate was adapted from the test described by Tashian (1969). The procedure included immersing a piece of Whatman no.1 paper in a solution made with 0.15% Bromothymol Blue (BTB) in ice-cold 25 mM Tris HC1, 0.1 M Na2SO4, pH 8.0. The paper was allowed to soak completely in this blue solution and was placed on ice for 30 minutes. The colored filter was then transferred to a Petri dish with a hole in the lid. Samples of mosquito larval midgut homogenate were prepared by sonicating midguts of early fourth instar larvae in ice-cold 25 mM Tris HC1, 0.1 M Na2SO4 pH 8.0, with protease inhibitor cocktail (Sigma-Aldrich, St. Louis, MO; diluted 1:1000). An autopipette was used to spot exactly 4 pl samples on the paper. Controls were also spotted. The controls included a buffer with protease inhibitor and controls for the liver/yeast food added to the medium in which the mosquito larvae were reared. These food controls included a range of concentration from 1 to 100 gg/ml liver powder and yeast. Carbonic anhydrase from bovine erythrocytes (Sigma-Aldrich) dissolved in the same buffer described above was used as a positive control.

A steady stream of CO2 at 34.5 KPa was blown for 3 seconds through the opening on the lid of the Petri dish, and the dish was sealed and kept on ice. The formation of yellow spots in a few seconds was indicative of carbonic anhydrase activity.

Effect of Methazolamide on the Alkalization of the Midgut of Live Larvae
The effect of a CA inhibitor. methazolamide. on gut alkalization and the capacity
The effect of a CA inhibitor. methazolamide. on Qut alkalbation and the canacitv





17


mM Tris HC1, 0.1 M Na2SO4 buffer, pH 8.5. BTB solution was added to each well until a 0.003% solution was achieved. Five live early fourth-instar larvae that had been placed in BTB indicator solution for 2 hours were added to each of the wells, and the larvae were allowed to adjust to their new environment for 30 minutes. Methazolamide dissolved in Dimethyl Sulfoxide (DMSO; Sigma-Aldrich) at concentrations ranging from 10 M to 8x 10-3 M was added to the wells. Controls included wells containing DMSO with BTB indicator but no inhibitor and wells containing BTB indicator but no DMSO. The plates were scanned using a Hewlett Packard ScanJet 6100C scanner before addition of the inhibitor and 5 hours later. In addition, the midguts were dissected and photographed to record the pH within the gut lumen as revealed by the color of ingested BTB.

18o Exchange Method to Measure Carbonic Anhydrase Activity

Tissue homogenate carbonic anhydrase activity was measured using the 180O

exchange method (Silverman and Tu, 1986). Midguts were dissected, and the peritrophic membrane was removed together with its contents. Individual measurements of CA activity were performed with pooled samples of gastric caeca, anterior midgut, posterior midgut and Malpighian tubules. The method involved adding '80-labeled NaHCO3 to

0.1 M Hepes buffer, pH 7.6, at 9.50C. The disappearance of O8 isotopes from CO2 and/or HCO3" upon addition of the enzyme preparations was monitored. Measurements of O8 in CO2 were accomplished with a mass spectrometer, using a CO2-permeable inlet that allowed very rapid, continuous measurement of the isotopic content of CO2 in





18


methazolamide to a final concentration of 10-6 M. Recombinantly expressed and purified mosquito CAs were also tested for activity using this assay.

Isolation of RNA and Synthesis of cDNA

Total RNA was isolated from freshly dissected fourth instar mosquito larval midguts using TRI Reagent (Molecular Research Center Inc., Cincinnati, Ohio) according to the manufacturer's instructions. Briefly, 100 Ae. aegypti gut epithelial organs, including fore-, mid-, and hindgut (approximately 20 mg) were dissected in HSS and transferred to a sterile microcentrifuge tube containing TRI Reagent (600 Ipl). The tissue was homogenized and incubated for 5 min at room temperature. The homogenate was then extracted with chloroform (40 jiL) and precipitated with isopropanol (100 tL). The RNA pellet was washed with 75% ethanol (200 pL), air-dried and resuspended in 50 pL diethylpyrocarbonate (DEPC; Sigma-Aldrich)-treated H20. RNA concentrations were calculated from the absorbance at 260 nm. Total RNA (10 pig) was reversetranscribed for 2 hours at 420C in a 20 jl reaction mixture using 5 pmol of oligo(dT)1218, RNasin (1:40 dilution), 1X first strand buffer, 1 mM dNTPs, and 200 units (U) of Superscript II reverse transcriptase (Invitrogen Inc., Carlsbad, California). This cDNA was used to clone the first fragment ofAe. aegypti CA.

Bioinformatics

The National Center for Biotechnology Information (NCBI) website

(www.ncbi.nlm.nih.gov) was used for the majority of the bioinformatical data presented in this study. The first mosquito genome, An. gambiae, was released in 2002 (Holt et al.,





19


as analyzing PCR products. The NCBI Blast Flies database

(www.ncbi.nlm.nih.gov/BLAST/Genome/FlyBlast.html), together with the Ensembl database (www.ensembl.org/Anopheles gambiae/) were used to predict the number of CA genes in the Drosophila melanogaster and An. gambiae genomes by inputting the Ae. aegypti CA as the search sequence. These partial sequence results were then annotated to reflect the 2 full-length CA sequences that we have cloned from An. gambiae and presented within this manuscript.

Ensemble is a joint project between the European Bioinformatics Institute and the Sanger Institute to bring together genome sequences with annotated structural and functional information. The NCBI protein database (pdb) and the BLAST were used in

conjunction with the 3-dimensional structure viewer (Cn3D; Hogue, 1997) for the prediction of antibody accessible peptide regions in mosquito proteins. BLAST analyses

also confirmed that the chosen antigenic peptides were unique. The conserved domain database (CDD; Marchler-Bauer et al., 2002) and the conserved domain architecture retrieval tool (CDART; Geer et al., 2002) were used to predict the function of our newly cloned mosquito proteins. Alignments were produced using Clustal W (Thompson et al., 1994), as implemented in DNAman software (Lynnon Biosoft, Vaudreuil, Quebec, Canada).

Cloning of CA from Aedes aegypti Larval Midgut Degenerate oligonucleotides were designed against the regions of conserved aminn nidc amnng CA nroteins as determined hv the BLAST analysis of several





20


The primer sequences used initially for Ae. aegypti CA were CA5F and CA3R (see Table 2-1). PCRs were performed in a total volume of 20 pl, and the reaction mixture contained 0.1 gg of cDNA as template, 0.2 pM of each primer, 200 jM each of dNTPs, 1X PCR buffer and 1 U of Taq polymerase (Promega; Madison, Wisconsin). The PCR cycling profile was: 94C for 5 min, 550C for 2 min and 720C for 3 min, followed by six cycles of 94�C for 0.5 min, 530C (in increments of 20C/cycle) for 1 min and 720C for 1 min and 35 cycles of 94oC for 0.5 min, 450C for 1 min and 720C for 2 mmin followed by a final extension at 720C for 15 min. The PCR products were visualized on 1% agarose gels and specific products were isolated using a QIAquick gel extraction kit (Qiagen, Inc, Valencia, California), diluted 1:100 in water, and used as template for a second, identical PCR. The resulting 297 base-pair (bp) product was gel-purified, ligated into pGem-T (Promega) and transformed into JM109 Escherichia coli (Promega) for subcloning. This partial Ae. aegypti CA cDNA was completed using amplified cDNA pools from gastric caeca and posterior midgut.

Construction of Amplified cDNA Pools

Adapter-ligated, amplified cDNA pools ("libraries") were constructed from

different regions of the fourth instar larval gut of both Ae. aegypti and An. gambiae using a technique optimized for invertebrate tissues (Matz et al., 1999). The gastric caeca, anterior midgut, posterior midgut, rectal salt gland, Malpighian tubules, and anal papillae often larvae were dissected in HSS and collected separately, resulting in six discreet tissue pools. The tissue was dissolved in Buffer D (500 uL; 4 M guanidine thiocyanate,
tissue nools. The tissue was dissolved in Buffer D (500 giL: 4 M ~uanidine thiocyanate,





21


was vortexed and centrifuged at 14,000 g for 30 seconds at 40C. The upper, aqueous phase was transferred to a clean tube and 5 pL glycogen solution (Pharmacia Quick Prep Micro RNA purification kit, Piscataway, New Jersey). The RNA was precipitated by the addition of 100% ice-cold ethanol (550 gL) followed by centrifugation at 14,000 g for 6 minutes at room temperature. The supernatant was removed and 1 mL of ice-cold ethanol (80%) was added. The mixture was centrifuged at 14,000 g for 10 minutes at room temperature, the supernatant was removed, and the pellet was air-dried.

For first strand synthesis, the pellet was resuspended in DEPC-treated water (5 jiL) and combined with the TRsa primer (1 pJM; Table 2-1). This mixture was incubated at 50oC for 3 minutes and immediately placed on ice. Then 1X ligation buffer (Marathon cDNA Amplification kit, BD Biosciences, Palo Alto, California), 0.01 M DDT, 1 U Superscript II (Life Technologies; Rockville, Maryland), and 0.5 piL dNTP mix (10 mM each dNTP, Marathon cDNA Amplification kit) were added to a total volume of 10.5 pL. This reaction mixture was incubated at 420C for 1 hour and immediately put on ice.

For second strand synthesis, DEPC-treated water (49 pL) was added to the first strand reaction mix. The mixture was then combined with 1.6 jiL dNTP mix (10 mM each, Marathon cDNA Amplification kit), 1X reaction buffer (Marathon cDNA Amplification kit), and 4 jL second strand synthesis enzyme mix (Marathon cDNA Amplification kit) in 80 iL total volume. The reaction mix was then incubated at 160C for 1.5 hours. T4 DNA polymerase (1 U; Marathon cDNA Amplification kit) was added
to the reaStin mixture and the entire mixture was inchated at 16oC for an additional 0.5
tn the renctinn mirtiire 2nd the entire mixture wa~ incubated at 1 60C for an additional 0 5





22

The reaction mix (80 jiL) was combined with 40 pL phenol and 40 pL

chloroform and centrifuged at 14,000 g for 10 minutes. The upper, aqueous phase was removed and transferred to a clean tube. The cDNA was precipitated by the addition of 3 M sodium acetate (8 tL, pH 5.0) and 100% ethanol (160 jiL). The mixture was centrifuged at 14,000 g for 15 minutes at room temperature. The supernatant was removed and the pellet was air-dried.

For adaptor ligation, the cDNA pellet was resuspended in DEPC-treated water (6 pL) and combined with 1 jiM adaptor, 1X ligase buffer, and 1 U T4 ligase (Marathon cDNA Amplification kit) in 10 pL total volume. This mixture was stored overnight at 16oC. For cDNA amplification, the ligation mixture (10 jiL) was combined with 40 pL DEPC-treated water. PCR amplification was then performed using the Advantage kit (BD Biosciences). The diluted cDNA (1 kL) was combined with 1X advantage buffer,

0.4 gL dNTP mixture (10 mM each), 0.1 pjM DAP and TRsa primers (Table 2-1), and 0.4 L advantage enzyme mix in 20 tL total volume. The cycling profile consisted of 94oC for 30 seconds, 660C for 1 minute, and 720C for 2.5 minutes. The reaction was analyzed on a 1% agarose gel after 12, 16, and 20 cycles. A final chase step was then performed to ensure that all cDNAs were completely double-stranded. Both 5' and 3' adaptor primers were added to the PCR reactions and two cycles of 770C for 1 min, 650C for 1 min, and 720C for 2.5 min were performed. The resulting collections of amplified cDNA were then diluted 1:50 and used as template for subsequent PCR experiments.

Amplified cDNA pools from An. gambiae were used to clone two CA cDNAs and






23


determined by BLAST analysis using characterized proteins against the An. gambiae genome. See table 2-1 for all initial primer sequences.

3' and 5' Rapid Amplification of cDNA Ends and Sequencing

Full-length cDNAs were obtained by rapid amplification of cDNA ends (RACE), (Zhang and Frohman 1997, modified by Matz et al., 1999). Exact primers were defined according to the 5' adaptor (DAP primer) along with a reverse primer specific to the cloned fragment, and 3' TRsa adaptor (TRsa primer) along with a forward primer specific to the cloned fragment (see Table 2-1 for adaptor primer sequences). These ends, which included the 5' and 3' UTR sequences, were then used to design PCR primers to produce a single product with consensus start and stop codons.

Plasmid DNA from individual colonies was purified using a Qiaprep

Plasmid Mini kit (Qiagen). The plasmid DNA (50 ng) was then sequenced using the ABI Prism Big Dye Terminator Cycle Sequencing Kit (PE Biosystems, Foster City, California) and the reaction products were analyzed on an ABI Prism 310 Genetic Analyzer (PE Biosystems).

Construction of In Situ Hybridization Probes Sense and antisense digoxygenin (DIG)-labeled cRNA probes were generated by in vitro transcription using a DIG RNA labeling kit (Roche Molecular Biochemicals, Indianapolis, Indiana). The initial in situ hybridization experiment, presented in chapter 3, used a cRNA probe derived from the original 297 bp Ae. aegypti CA sequence. The in situ experiments presented in chapters 4 and 5 utilized the fuIll-leneth CA and AE






24


(New England Biolabs (NEB); Beverly, Massachusetts) and 1X buffer 3 (NEB) at a total volume of 20 pL for 1 hour at 370C. For the sense probe, the pGEM-T vector containing the 297 bp CA sequence was linearized by incubating 2 tg ofplasmid with Not I restriction enzyme and 1X buffer 3 (NEB) in a total volume of 20 iL for 1 hour at 370C. After digestion, the volume was brought to 100 i�L with the addition of 80 pL water. A phenol/ chloroform extraction was performed such that 100 piL of phenol/ chloroformisoamyl alcohol was added to the linearized plasmid and the solution was centrifuged at 14,000 g for 1 minute. The upper aqueous phase was transferred to a new tube and 100 kL chloroform was added. After centrifugation at 14,000 g for 1 minute, the upper aqueous phase was transferred to a new tube and the chloroform step was repeated. The linearized plasmid DNA was precipitated by the addition of 10 pL sodium acetate (3 M, pH 2.5) and 200 tL cold ethanol (100%). The DNA was incubated at -80oC for 15 minutes and then centrifuged at 14,000 g for 10 minutes at 40C. The supernatant was removed and the DNA pellet was washed by the addition of 500 jtL ethanol (70%) followed by centrifugation at 14,000 g for 5 minutes at 40C. The supernatant was removed and the pellet was air-dried and then resuspended in 13 p.L DEPC-treated water.

The full-length Ae. aegypti CA was subcloned into pCR 4-TOPO plasmid using a PCR manufactured 5' Sal I restriction site and a 3' Xho I site. Therefore, the pCR 4TOPO plasmid was linearized by incubating 2 jig ofplasmid with either Sal I, 1X Sal I buffer, and BSA, or Xho I, 1X buffer 2, and BSA (NEB). The pCR 4-TOPO plasmid was also used for the generation of the An. gambiae CA and AE robes. For these nroes. the





25


These mixtures were all incubated at 370C for 2 hours to ensure complete linearization of the plasmids. After digestion the uncut pCR 4-TOPO plasmids were compared to the cut plasmids on a 1% agarose gel to confirm linearization. The cut plasmids (10 gL) were cleansed using a Qiaquick PCR Purification kit (Qiagen Inc, Valencia, California).

For in vitro translation, the resuspended pellet or purified plasmids were

combined with 1X transcription buffer, 1X NTP labeling mixture, RNase inhibitor (20 U), and 40 U T3 RNA polymerase (or SP6 for pGEM-T plasmids) or 40 U T7 RNA polymerase. For the Ae. aegypti CA probes, T7 polymerase was used with the Sal I cut plasmid to produce the antisense probe, while T3 polymerase was used with the Xho I cut plasmid to produce the sense (control) probe. The pCR 4-TOPO plasmid used for the generation of the An. gambiae CA and AE probes contained the CA and AE sequences in the reverse configuration. Therefore, for these An. gambiae probes, T3 was used with the Not I linearized plasmids to produce the antisense probes, while T7 was used with the Pme I cut plasmids to produce the sense (control) probes. The mixtures were incubated at 370C for 2 hours followed by the addition of 20 U DNase I and incubation at 370C for 15 minutes. The DNase I reaction was stopped by the addition of 0.5 tL of EDTA (500 mM). The DIG-labeled cRNA was then precipitated by the addition of 2.5 LL of LiC1 (4 M) and 75 pL cold ethanol (100%). The mixture was incubated overnight at -200C and centrifuged at 14,000 g for 10 minutes at 40C. The supernatant was removed and the pellet was washed with 50 pL cold ethanol (75%). The centrifugation step was repeated and the pellet was air-dried and resuspended in 100 gL DEPC-treated water. The probes





26


In Situ Hybridization

The in situ hybridization experiments presented in chapters 4 and 5 added an

additional fixation step due to a recommendation by Dr. Dmitri Boudko to increase the clarity of the in situ labeling. A glass electrode fitted to a micromanipulator was used to inject 4% paraformaldehyde into the thoracic cavity, just behind the head. Successful perfusion was easily identified by the cessation of the otherwise constant muscle twitching along the length of the body. This injection of fixative served to preserve the cellular integrity and protect against the many proteases that exist within the mosquito gut. For in situ hybridization, methods were adapted from Westerfield (1994). The midguts were washed with PBS at room temperature and then incubated in 100% methanol at -200C for 30 minutes to ensure permeabilization of the gut tissue. The tissue was washed (5 min each wash) in 50% methanol in PBST (Dulbecco's phosphate buffered saline [Sigma-Aldrich] plus 0.1% Tween-20), followed by 30% methanol in PBST and then PBST alone. The tissue was fixed in 4% paraformaldehyde in 0.1 M sodium cacodylate buffer (or 0.1 M phosphate buffer) for 20 min. at room temperature and washed with PBST. The larval midguts were digested with proteinase K (10 gg/ml in PBST) at room temperature for 10 min, washed briefly with PBST and fixed again, as described previously.

Prehybridization of the tissue was accomplished by incubation in HYB solution (50% formamide, 5X SSC [1X SSC equals 0.15 M NaC1, 0.015 M Na-citrate buffer pH

7.0], 0.1% Tween-20) for 24 hours at 550C. The larval midguts were transferred to





27

55C with 50% formamide in 2X SSCT for 30 min (twice), 2X SSCT for 15 min and

0.2X SSCT for 30 min (twice). For detection, the tissue was incubated in PBST containing 1% blocking solution (Roche Molecular Biochemicals) for 1 h at room temperature. The tissue was incubated with anti-DIG-alkaline phosphatase (Roche Molecular Biochemicals) diluted 1:5000 in blocking solution for 4 hours at room temperature. The tissue was washed with PBST and incubated in alkaline phosphatase substrate solution (Bio Rad Laboratories, Hercules, CA, USA) until the desired intensity of staining was achieved (2-3 hours).

CA Histochemistry

Carbonic anhydrase activity was detected in isolated Ae. aegypti midguts using Hansson's method (Hansson, 1967), as modified by Ridgway and Moffet (1986). The procedure involved the incubation of isolated, 3% glutaraldehyde-fixed midguts in 1.75 mM CoSO4, 53 mM H2SO4, 11.7 mM KH2PO4, and 15.7 mM NaHCO3 (pH 6.8). The incubation medium contains a high concentration of bicarbonate, which stimulates the production of CO2 and hence a decrease in pH in the presence of CA. The acidic pH then stimulates the formation of insoluble black cobalt salts which were visualized using 0.5% (NH4)2S in distilled water. Therefore, micro-sites of active CA liberation of CO2 from bicarbonate dehydration become apparent with this assay. Removal of the bicarbonate substrate (NaHCO3) eliminated staining.

Real Time PCR

Region-specific cDNA was produced from dissected mosquito tissue using the





28


minutes at 750C. The lysed tissues were treated with 2 U of DNase I for 30 minutes at 370C. The DNase I was then inactivated by heating to 750C for 5 minutes. For the reverse transcription reaction, 10 1L of cell lysate was combined with 4 itL dNTP mix (contains 2.5 mM each dNTP) and 5 jM random decamer first strand primer in 16 IL total volume. The mixture was incubated at 700C for 3 minutes and then chilled on ice for 1 minute. This mixture was then combined with 1X RT buffer, 1 U M-MLV reverse transcriptase, and 10 U RNase inhibitor, and incubated at 420C for 1 hour. The reverse transcriptase was then inactivated by incubation at 950C for 10 minutes. Primers (Table

2.1) were designed using Primer Express software (Applied Biosystems; Foster City, California). The SYBR Green PCR Master mix, which includes SYBR Green I dye, Amplitaq Gold DNA Polymerase, dNTPs, and buffer, was used for all real time PCR investigations. Each cycle of PCR was detected by measuring the increase in fluorescence caused by the binding of the SYBR Green dye to double-stranded DNA using an ABI Prism 7000 Sequence Detection System (Applied Biosystems). Initially, each primer set, including the control 18s ribosomal RNA (Genbank accession M95126), was assessed to determine the optimal concentration of primer to be used. All real time experiments used the same 2-step cycling profile: 500C for 2 minutes followed by 95oC for 10 minutes and 40 cycles of 950C for 15 seconds and 600C for 1 minute. Whole gut cDNA (100 nM) was used as template with 500 nM, 300 nM, 100 nM, or 50 nM of each primer set and 1X SYBR green I master mix in 25 piL total volume. Each reaction was done in triplicate. The optimal concentration was then chosen based on the amplification





29


whole gut cDNA were used as template with the appropriate concentration of primers and 1X SYBR green I master mix in 25 gL total volume. The threshold cycle number (Ct) was plotted versus the log of the template concentration and the slope (m) and intercept

(b) were determined (Figure 2-1). These pre-determinations were then used in the standardized comparison of the amount of 18s transcript and CA transcript in each of the cDNA samples tested. For each analysis, a control containing all of the necessary PCR components except the cDNA template was run. To determine the relative expression level for each transcript analyzed, the following equation was used: (Ct-b)/m. The average log ng for each transcript was then compared to the average log ng of 18s RNA transcript to normalize the values. Then the expression levels were determined relative to the transcript with the greatest normalized log ng value and expressed in a bar graph using Microsoft Excel software.

Antibody Production

An antigenic peptide consisting of eighteen amino acids was chosen from the Ae. aegypti CA sequence for antibody production. In order to increase the probability that this antibody would be specific for this particular CA sequence (in the event that other CA isoforms were isolated from the mosquito gut), attempts were made to synthesize an antigenic peptide that would be specific to this isoform. The well-characterized mammalian CA isoforms served as a model in trying to choose a unique CA peptide sequence. The comparison of the mosquito CA with the mammalian isoforms yielded a pectide sequence from the amino (NT terminus where CA isoforms showed the mot
nentide sequence from the amino (N~ tenniniis where ('A jqnfnrmq QI1AXVrII the mncf





30


more accessible to antibody probing. Furthermore, three-dimensional analyses (Cn3D v4.1 NCBI) of predicted CA IV structures (human 1ZNC and mouse 2ZNC) predicted that the N terminus is exposed and accessible (Figure 2-2). An antigenic peptide was therefore chosen from the N terminus of the Ae. aegypti CA sequence. This peptide sequence (GVINEPERWGGQCETGRR) was sent to Sigma-Genosys (Woodlands, Texas), where it was synthesized and conjugated to bovine serum albumin (BSA). The synthetic peptide-BSA construct and Freund's incomplete adjuvant were injected into two rabbits to elicit an immune response. Prior to injection, a blood sample from each rabbit was collected to serve as the control pre-immune serum. Every two weeks a blood sample was collected from the rabbits, the fraction of immunoglobulin G (IgG) pooled, and another dose of the peptide-BSA construct administered. Three months after the initial injections, the final bleeds were collected and used for all immunohistochemical analyses.

We also raised antibodies against an An. gambiae cytosolic CA peptide and an

anion exchanger (AE) peptide. These antibodies were produced by the Aves Labs, Inc., (Tigard, Oregon), using a similar strategy to that described above. However, these antibodies were produced in hens. The synthesized peptides were conjugated to BSA and injected into two hens each. The immunoglobulin Y (IgY) antibodies were collected from the hens' eggs, pooled, and purified.

Immunohistochemistry

The specificity of the antibodies in the resultant antisera was determined. The






31


pre-incubation medium (pre-inc) for a minimum of 1 hour, and then incubated in primary antibody (1:1000) overnight at 4oC. The guts were then washed in pre-inc and incubated in FITC-conjugated goat anti-rabbit (GAR) or Alexa-GAR secondary antibody (Jackson ImmunoResearch, West Grove, Pennsylvania, 1:250 dilution) overnight at 40C. The whole mount preparations were rinsed in pre-inc and mounted onto slides using pphenylenediamine (PPD, Sigma-Aldrich) in 60% glycerol. In some cases Draq 5 (Jackson ImmunoResearch, 1:1000 dilution) was applied before mounting to visualize nuclear DNA. The samples were examined and images captured using the Leica scanning confocal microscope.

Live preparations were examined, following a similar procedure, to ensure that antibodies were capable of localizing extracellular proteins only. In this case, the primary antibody was applied to the live guts for four hours. The samples were washed in TBS, and fixed as described previously, before the secondary antibody was applied. The samples were mounted on slides as described above.

The live gut assays were also performed to determine whether this specific CA is tethered to the cell membrane via a GPI linkage. Ten live gut preparations were incubated with phosphoinositol-specific phospholipase C (PI-PLC, 1:100 in HSS; SigmaAldrich) for 90 minutes at 370C. PI-PLC was used as a tool in determining the presence of a GPI link. Controls in which the guts were incubated in HSS alone were also performed. The guts were then washed in HSS, fixed, and treated with primary and secondary antibodies as described above.





32


CA Protein Expression

Recombinant Ae. aegypti and An. gambiae CAs were produced using the pET1i00 vector (Invitrogen). Specific primers were designed to amplify each cDNA. The 3' primers included the sequence 5' to and including the native stop codon. The 5' primers contain the sequence CACC preceeding the native start codon for correct frame insertion (See Table 2-1 for primer sequences). PCRs were performed using 1 U of Platinum P4 polymerase (Invitrogen), the gastric caeca cDNA collections as template (200 ng), 1X Pfx amplification buffer, 1.2 mM dNTP mixture, 1 mM MgSO4, and 0.3 ptM of each primer in a total volume of 50 jiL. A three-step PCR protocol was used consisting of 940C for 2 minutes followed by 30 cycles of 940C for 30 seconds, 550C for 30 seconds, and 680C for 1 minute.

The resultant blunt-ended cDNAs (4 gL from PCR mix) were ligated with the

pET 100 directional Topo vector (1 jL and 1 pL salt solution; Invitrogen) for 10 minutes at room temperature. Top 10 chemically competent E. coli (50 pL; Invitrogen) were transformed by incubating 3 pL of ligation mix with the cells for 30 minutes on ice, followed by a heat shock of 420C for 30 seconds. SOC (250 gL) was added to the cells and they were then incubated at 370C for 30 minutes with shaking. The transformation mix (100 jiL) was then plated on a LB-carbenicillin plate (50 jg/mL) and incubated overnight at 370C. Colonies were sequenced using Big Dye version 1.1 as described previously. The purified plasmids (10 ng each) were transformed into BL21 Star (DE3) cells (Invitrogen) for CA expression as described above. However, after SOC addition





33


mL of fresh LB-carb and was grown at 370C with shaking. Optimization experiments were performed in order to facilitate the production of the greatest quantity of CA protein. For production of CA protein, isopropylthio-P-galactoside (IPTG, 1 mM final concentration; Stratagene, La Jolla, California) was added when the culture had attained an optical density of 0.5 at a 600 nm wavelength. Achieving this density took about 1.5 hours of growth at 370C and 200 rpm. Zinc, in the form of zinc sulfate (0.5 mM final concentration), was added along with the IPTG to facilitate the proper conformation of an active CA protein. In order to optimize the duration of the induced growth phase, samples were collected every hour for six hours. These samples were analyzed on an SDS-Page 4-12% Bis-Tris gel to compare CA protein content. Four hours of growth was determined to be ideal for the production of the truncated Ae. aegypti CA IV-like and full-length An. gambiae CA II-like proteins.

Total protein was collected using the Probond Purification System according to

the manufacturers instructions for soluble proteins (Invitrogen). The cells were harvested by centrifugation, sonicated in native buffer (250 mM NaPO4, 2.5 M NaCi; Invitrogen) with lysozyme (1 mg/mL; Sigma-Aldrich), and centrifuged again to collect a crude protein extract. The supernatant was applied to a Probond nickel column (Invitrogen) and washed free of non-specific binding contaminants. The nickel column binds the CA protein due to the added histidine tag, a repeat of six histidine residues within the pET100 expression vector that is inserted after the carboxy-terminus of the CA protein. CA was eluted by adding imidazole (250 mM; Invitrogen) to the column, which competes with





34


Anion Exchanger Oocyte Expression

The full-length anion exchanger (AE) sequence was subcloned into the pXOOM vector, which is optimized for both oocyte and mammalian expression (Jespersen et al., 2002; a generous gift from Dr. T. Jespersen). In addition to a T7 RNA polymerase promoter, this vector contains Xenopus-specific 5' and 3' UTR sequences flanking the insert in both directions. cRNA synthesis was performed using the T7 mMessage mMachine kit (Ambion, Austin, Texas), after the cDNA was linearized using PMEI.

One day after surgical removal of the eggs from the frog, the eggs were injected

with either AE cRNA or water (control). After injection the eggs were incubated at 160C for 4 days, long enough for measurable protein production and expression. The oocytes were maintained in ND96 (96 mM NaC1, 2 mM KC1, 1 mM MgC12, 10 mM HEPES, pH

7.4 with NaOH). The medium was changed daily and dead oocytes were removed.

Anion Exchanger Physiology

Expression of the An. gambiae AE was examined using 2-electrode voltage clamp electrodes. The voltage electrodes were pulled using 1.2 mm glass (M1B120F-3, World Precision Instruments), and showed resistances between 1-2 Mf. Oocytes were clamped to -50 mV and stepped from -90 mV to +70 mV in 10 mV increments. The water injected eggs served as the control in evaluating any activity exerted by endogenous proteins found in the Xenopus oocytes. Several different solutions were used to determine the exchanger's functional activities (refer to table 2-2). The transporter blockers, 4,4'-diisothiocyanodihydrostilbene-2,2'-disulfonate (DIDS, Calbiochem, La





35


A do nrate CAImers: -' "-- "
CA5F:5' GAR CAR TTY CAY TKY CAY TGG GG
CA3R: 5' GTI ARI SWN CCY TCR TA
N=G,A,T,C; K=G,T; S=G,C; W=A,T; Y=C,T; RA,G
Amplifed cDNA ada _________ ___ r p_DAP:5' CGA CGT GGA CTA TCC ATG AAC GCA
TRsa:5' CGC AGT CGG TAC TTT TT TT TTT T Anopheles exact CAprmers:
1CA2F:5' CAG TCA CCT ATC GAC CTA AC
AgICA4R:5' CTC GCG TGT TCA ATG GT G
A4CA1F:5' GGA GGC GTC CTT GGC AAC
CAI2R: 5' CTG CAC TGA CCG GAA GTT G Anopes exact AE rmers: __ .___AE1F:5' CCT GGA AG AAA CGG CAC G E4R:5' CCT CGAGCT GGT GCA C .T C Aedn CA Real time PCRpmers:
5SPCAFI:5' GCA ACA CTG CTT CCG TCT ACA A
5SPCARI:5' CCG GTT CGT TAA TAA CTC CAT TG
18s RIBF:5' CGC TAC TAC CGA TGG ATT ATT TAG TG 18s RIBR:5' GTC AAC TTC AGC GAT TCA AAT GTA A Aed CA session rime
ExCAshortF:5' CACC ATG _GAC .GAA __TGG CAC T
ExCAshortR: 5' TTA GTA .ATC CAT .ATC IGGT GTG GT
nh/e CAe mers:
Anh-o-jg~m__________ ____ExCA4F:5 CACC ATG GCA TCA AAA ACA ACA AAG CA4end:5' TTA CAG CTT CGA AAG CAC AAC GG

Table 2-1. PCR primer sequences.






36

















Salt for 98M value mW #1 #2 #3 #4
98N 98K MN-CI 9K-CL mM mM mM ml Solution: Ix 4x lx 4x lx 4x lx 4x
NaC 58.44 5.73 22.91 2 0.12 0.47 0 0 0 0 KCI 74.55 2 0.15 0.60 7.31 29.22 0 0 0 0
- -
Na Gluconate 218.1 0 0 0 0 21.37 85.50 2 0.44 1.74 K Gluconate 234.2 0 0 0 0 2 0.47 1.87 22.95 91.81 Choline Cl 139.6 0 0 0 0 0 0 0 0 MgSO47H20(120.36) 246.5 0 0 0 0 OS 0.12 0.49 05 0.12 0.49 MgCI26H20(59.7) 203 s 0.10 0.41 s 0.10 0.41 0 0 0 0 CaCi22H1120(110.98) 147.02 s 0.07 0.29 s 0.07 0.29 0 0 0 0 Ca Glucate 430.38 0 0 0 0 US 0.22 0.86 U 0.22 0.86
-- -53
HEPES (fee be 238.3 10 2.38 9.53 10 2.38 9.53 10 2.38 9.53 10 2.38 9.53 EGTAforlnside0Oout 380.4 0 0 0 0 0 0 0 0 pH 7.2 (4M NaOH) 7.2 (4M KOH) 7.2 (4M NaOH) 7.2 (4M KOH)



Table 2-2. Composition of all solutions used inXenopus oocyte expression of An.
gambiae AE. Total molarity and pH were kept constant in all solutions.
Expression profiles were recorded in high sodium (#1), high sodium minus chloride (#3), high potassium (#2), and high potassium minus chloride (#4).






37














Ad CA Primer Unerization
35

BrCA equations:
30 y= -4.265x + 39.8
125.R2 = 0.9776
25


2 18s equations: 20
5y 3.1312x + 27.781 SR' = 0.9931
15
*BrCA 18s primers WG
io ::- ::1: : :I *l8s primrs WG
0 1 2 LogR n3 4 5

Figure 2-1. Efficiency plots for real-time PCR primers. Serially diluted cDNA samples
were tested with each primer set to determine the efficiency of
amplification. A linear regression was performed to determine the slope and
intercept for each primer set. These values were then used in an algorithm
to compare cDNA concentrations within the samples.






38









































Figure 2-2. Three-dimensional (Cn3D) depiction of human CA IV (1ZNC). The green
barrel represents an alpha helix structure, the tan arrows represent beta sheets, and the colored strings represent extended loop structures. The yellow coloring represents the accessible, extended loop peptide region
agais .t. ch t. homologous4. aegpt C'A antibod asi r fl-asedl













CHAPTER 3
CARBONIC ANHYDRASE IN THE MIDGUT OF LARVAL AEDES AEGYPTI:
CLONING, LOCALIZATION, AND INHIBITION' Introduction

Bicarbonate (and ultimately carbonate) ions are produced in vivo primarily by the enzymatic action of carbonic anhydrase (CA). Its activity contributes to the transfer and accumulation of H+ or HCO3- in bacteria, plants, vertebrates and invertebrates. Although there are innumerable reports related to the isolation of CA from vertebrates, studies involving CA from invertebrates are very rare and there are no reports of the isolation of CA from adult or larval mosquitoes.

There is strong immunohistochemical (Zhuang et al., 1999) and physiological (Clark et al., 1999; Boudko et al., 2001b) evidence that an electrogenic, basal Hl VATPase energizes luminal alkalinization in the anterior midgut of the larval mosquito by producing a net extrusion of protons out of the lumen and a hyperpolarization of the basal membrane. In contrast, HIt V-ATPase appears to be localized in the apical membrane of the posterior midgut and gastric caeca providing a reversed Ht- pumping capacity relative to the anterior midgut (Zhuang et al., 1999). A system capable of generating a high luminal pH is likely to be buffered by carbonate (CO3-2), which has a pKa of approximately 10.5.





40


The purpose of this study was to determine the presence and location of CA in the midgut of larval Ae. aegypti and to clone and characterize the enzyme. To investigate the role of CA in the alkalization of the larval midgut, the effects of CA inhibitors were tested. Here, we report the cloning and localization of the first CA from mosquito larvae and, in particular, from the midgut epithelium of larval Ae. aegypti. A cDNA clone isolated from fourth-instar Ae. aegypti midgut (termed A-CA) revealed sequence homology to the a-carbonic anhydrases (Hewett-Emmett, 2000). Histochemistry and in situ hybridization showed that the enzyme appears to be localized throughout the midgut, although preferentially in the gastric caeca and posterior regions. In addition, classic carbonic anhydrase inhibitors such as acetazolamide and methazolamide inhibit the mosquito enzyme in the midgut.

Results

Bromothymol Blue Qualitative Assay

This assay allowed the identification of samples of solubilized midgut tissue

containing CA activity by spotting them onto a filter paper soaked in a basic buffered solution containing a pH indicator, bromothymol blue (BTB). As stated previously, BTB changes color from yellow (at pH<7.6) to blue when the pH increases above this value. The principle behind the assay is based on the fact that CA catalyzes the conversion of CO2 into bicarbonate with the concomitant release of protons (Donaldson and Quinn, 1974). The presence of protons lowers the pH in those regions of the paper where the spotted samples contain the enzyme. As the DH falls below 7.6, these soots rapidly





41


enzymatic reaction takes only a few seconds, and it can be delayed if the solutions, the paper and the samples are kept cold on ice. However, a few seconds is usually sufficient to discriminate the samples that contain CA from those lacking enzymatic activity. The assay must be performed quickly since, after approximately one minute the entire filter paper turns yellow, probably as a result of the uncatalyzed hydration of carbon dioxide absorbed by the solution at this basic pH.

The test has proved useful in determining the presence of small amounts of CA in homogenates of mosquito larvae. The assay was also used to detect CA activity qualitatively, in fractions obtained from affinity chromatography (Osborne and Tashian, 1975) of larval homogenates. The affinity chromatographic procedure, which employs a bound CA inhibitor (p-aminomethyl benzyl sulfonamide (p-AMBS); Sigma), produced two peaks of CA activity upon exposure to the standard elution buffers. The amount of protein that we were able to produce by this technique was, however, very small and resisted several efforts at direct microsequencing. This change in color was inhibited by acetazolamide and methazolamide when these inhibitors (105 M) were added to the samples prior to spotting on the dye-impregnated filter papers. Inhibition of the reaction resulted in blue spots that did not change color upon addition of CO2. The positive control containing commercial CA turned yellow when carbon dioxide was added, and this color change was also inhibited by acetazolamide and methazolamide. This finding confirmed that the yellow color of the spots was due to the action of CA and that the mosquito larva contains active CA.





42


Carbonic Anhydrase Activity and Alkalization

A classic CA inhibitor methazolamide, was tested in live fourth instar larvae to examine the influence of CA on the maintenance of the pH extremes inside the midgut, and the effect of the enzyme on the net alkalinization of the growth medium by the intact animals. Previous investigations have shown that living mosquito larvae excrete bicarbonate, which results in the net alkalization of their surrounding aqueous medium (Stobbart, 1971). Equal numbers of living larvae of equivalent age and size were placed in culture plate wells containing lightly buffered medium and the pH indicator BTB. The tissue culture plates used in this assay were scanned before and after addition of various concentrations of methazolamide. In the absence of methazolamide, the blue color of the medium, indicating a pH of at least 7.6, was maintained (Stobbart, 1971). Actual measurement of the pH in each well showed a slow increase over time (data not shown). Upon addition of methazolamide, the culture medium slowly became acidic, with a resulting change in color to yellow as the pH dropped below 7.6 (Figure 3-1). All of the controls that did not contain methazolamide remained blue. Addition of methazolamide, at various concentrations, to wells containing only medium with BTB (no mosquito larva control) remained blue. These data show that CA activity is present in the living larvae and that it plays some role in acid/base excretion.

Moreover, fourth instar larvae cultured in BTB-containing medium ingest the dye, which can then be used as a visible indicator of the pH in the gut lumen. Treatment of the cultured larvae with methazolamide showed a direct impact of inhibited CA activity






43


alkalinization of the midgut was inhibited by methazolamide as shown by the color change of the BTB indicator. Interestingly, the effect was most pronounced in the anterior midgut, where the pH indicator changed from blue in the midgut of larvae reared in the absence of inhibitor to yellow in as little as 30 minutes when methazolamide (106 M) was added to the culture. The indicator also changed color progressively from blue through green to yellow in the gastric caeca (Figure 3-2). No apparent change was observed in the posterior midgut. The color of the midgut in this region was yellow both in the untreated larvae and in the larvae treated with methazolamide. Since the pH of the posterior midgut has been associated with values close to 7.6, no change in color was evident using this qualitative method.

O80 Isotope-Exchange Experiments

The relative activity of CA, normalized to total protein content, was calculated as described by Silverman and Tu (1986). The relative activity of CA was highest in the gastric caeca, followed by the posterior midgut and Malpighian tubules (Figure 3-3). The relative activity of CA in the anterior midgut was either extremely low or non-existent, falling at or below that of the buffer blank. The specificity of the reaction was confinnrmed by complete inhibition with the addition of 10-6 M methazolamide (results not shown). Cloning of Carbonic Anhydrase from Aedes Aegypti Larvae

We utilized a cDNA cloning strategy to obtain a specific carbonic anhydrase

cDNA from the midgut epithelial cells of the larval Ae. aegypti. A comparison of twelve CA sequences, including two putative CA sequences that had been annotated but not





44


bp partial sequence was used to derive exact PCR primers for a modified 3'- and 5'RACE (Frohman and Zhang, 1997, modified by Matz, 1999). Amplified cDNA pools

from each region of the isolated gut, facilitated the eventual cloning of a single contiguous cDNA (Matz, 1999). The final contiguous region spanned both start and stop codons, and encoded a polypeptide of 298 residues (GenBank accession number AF395662). Figure 3-4A shows an alignment of the Ae. aegypti carbonic anhydrase (ACA) amino acid sequence with several other, previously characterized members of this extensive a gene family. Figure 3-4B shows a homology tree depicting the percentage of identical amino acids between sequences, generated using DNAman software. Figure 35A shows the alignment between A-CA and six putative CA gene sequences from the D. melanogaster genome that our homology search (BLAST) revealed. Four of the D. melanogaster genes (AAF54494, AAF56666, AAF57140, AAF57141) had not previously been annotated. Figure 3-5B shows the homology tree generated with these sequences. A-CA has a putative molecular mass of 32.7 kDa. The translated A-CA protein sequence possesses a characteristic eukaryotic-type CA signature sequence within the polypeptide (amino acid residues 99-115; Fernley, 1988).

To examine the possibility of regionalized expression of the A-CA, PCR using

exact primers was performed on amplified cDNA pools from the various sections of the gut. Figure 3-6 shows an ethidium-bromide-stained agarose gel. PCR products of the expected, 894 nucleotide length, are readily seen in the gastric caeca and the posterior midgut regions. Anal papillae (not shown), anterior midgut, Malpighian tubules and





45


product was also discernible in the anterior midgut. This PCR analysis also revealed higher molecular mass products in the anterior midgut and Malpighian tubules that may represent additional carbonic anhydrases specific to larval Ae. aegypti (Figure 3-6). This result is shown only to display the gut regions in which the A-CA clone was derived. The lack of an 894 bp product in the other gut regions may simply be due to poor quality cDNA pools from those regions. However, the cloning of A-CA from both the gastric caeca and posterior midgut regions is consistent with the location of enzyme activities described above.

Localization of the Enzyme in the Midgut Epithelium: Carbonic Anhydrase Enzyme
Histochemistry

To further analyze the regional and cellular expression of CA in the midgut epithelium of larval mosquitoes, a modified Hansson's histochemical reaction was performed on whole mount preparations of the gut (Hansson, 1967). Figure 3-7 summarizes the results of this analysis. Carbonic anhydrase activity was detected in a non-uniform pattern along the length of the gut. The most intense staining was evident in the gastric caeca and the posterior midgut. Staining was less intense in the anterior midgut. At higher magnification, it was obvious that cellular heterogeneity with regard to CA activity also exists. This is particularly evident in the posterior midgut, where very large and regularly spaced cells appear nearly white on a background of dark CA reaction product. The larger cells have been characterized as "columnar" or ion-transporting cells (Volkman and Peters, 1989b). Surrounding these large cells are more numerous smaller





46


cuboidal cells. In addition, the distal cells of each lobe of the gastric caeca, termed Cap cells, show little or no histochemical staining, suggesting further cellular heterogeneity with respect to CA distribution in the gut (Figure 3-7). In Situ Hybridization

To further characterize the localization of A-CA expression, in situ hybridization was performed using a portion (approximately 300 bp) of the central coding region of the cDNA. Figure 3-8 shows typical results of this type of analysis. A strong hybridization signal was evident in the gastric caeca and the posterior midgut. Lower levels of hybridization were evident in other gut regions. As with the CA histochemical stain, higher magnification revealed that the relatively small cuboidal cells exhibit more intense labeling than do the large columnar cells (Figure 3-8B).

Discussion

The search for the enzyme in the midgut of the larval mosquito was triggered by the observations of a pH value around 11 in the anterior midgut lumen and a high bicarbonate concentration (Zhuang et al., 1999; Boudko et al., 2001b). The presence of CA in the midgut of the larval mosquito has been suggested before by investigations of the epithelium of larval lepidopteran midgut. Carbonic anhydrase has been studied in Manduca sexta, where the enzyme has been associated with the fat body, midgut and integumentary epithelium (Jungreis et al., 1981). The enzyme has also been localized in the goblet cells of the epithelium of Hyalophora cecropia using Hansson's histochemical stain. The same procedure showed that the columnar cells were devoid of activity





47


Even though a number of genes and their products have been isolated from the midgut of Ae. aegypti, and the role of CA in the alkalization of the midgut has been suggested (Turbeck and Foder, 1970; Haskell et al., 1965; Ridgway and Moffett, 1986; Boudko et al., 2001b), there have been no reports of the isolation or cloning of CA or of the localization of the enzyme within the midgut of larval mosquitoes. This is the first recorded cloning of a CA from a mosquito, and is also the first to be cloned from any arthropod. Our results show that at least one (and perhaps more) CA is present in the midgut of larval Ae. aegypti. The CA of larval Ae. aegypti (A-CA) is inhibited by classical carbonic anhydrase inhibitors such as methazolamide and acetazolamide. Methazolamide has the most potent effect on A-CA. Direct physiological measurements of ion fluxes from living larval mosquito midgut epithelial cells also show methazolamide to be a very potent inhibitor of ion movements and balance (Boudko et al., 2001a).

To investigate the distribution of CA in the midgut of the larval mosquito, we

employed both in situ hybridization and enzyme histochemistry. Our results indicate that enzymatic activity is greatest in the gastric caeca and the posterior midgut, as demonstrated by the intense staining obtained using Hansson's method and by in situ hybridization using cRNA probes. Measurements of activity using the '"O exchange method in pools of dissected regions of the gut corroborate these findings. In addition, the enzyme seems to be preferentially associated with the small cuboidal cells in the midgut epithelium, as determined both by enzyme histochemistry and by in situ





48


As reviewed in Clements (1992), two major cell types have been defined in the gastric caeca by inferring functional states from cytological findings. These two major cell types have been called ion-transporting cells and resorbing/secreting cells (Volkman and Peters, 1989a,b) and they correspond to the columnar and cuboidal cells mentioned above with the ion-transporting cells being equivalent to the columnar cells and the resorbing/secreting cells being the cuboidal cells (Zhuang et al., 1999). Neither of these cell types, as characterized in the larval mosquito gut, parallels the structurally unique qualities of the lepidopteran goblet cell. Nonetheless, our results indicate that, as in lepidopterans, CA activity is preferentially associated with one of two distinct cell types whose functional complementation must produce the alkalization and ionic balances regulated by the gut. These results are consistent with the observations of lepidopteran midgut by Turbeck and Foder (1970). In the larval lepidopteran midgut, two morphologically distinct cell types have been long recognized: goblet cells and columnar cells. Goblet cells posses both the proton-pumping V-ATPase and CA activity (Harvey, 1992; Ridgway and Moffet, 1986; Wieczorek et al., 1999). One of the enigmas of using the pioneering analyses of insect model systems such as M sexta to produce testable hypotheses for gut alkalinization in mosquito larvae has been the apparent absence of goblet cells from mosquitoes. Previous investigations have inferred different functional cell types in the larval mosquito gut epithelium. We are currently developing antibody probes for A-CA. Immunocytochemical analyses of A-CA distribution in comparison with other key components of gut function, such as V-ATPase (Zhuang et al., 1999).





49


It is interesting to note that the lowest concentration of CA in the midgut

epithelium occurs in the region that surrounds and probably regulates the region of highest luminal pH, the anterior midgut. The pKa of CO3-2 is approximately 10.5 and, hence, this anion is likely to be the primary buffer of the pH 10.5-11 gut contents within the anterior midgut. Our results therefore suggest that the major buffering anion in this area of the midgut is probably not produced by local CA but instead either upstream, in the gastric caeca, or downstream, in the posterior midgut, where CA levels are very high. This result, and results presented elsewhere (Boudko et al., 2001a), are consistent with a model in which a major function of the anterior midgut is to pump protons out of this region of the gut lumen, promoting the conversion of HCO3 to CO32. A comprehensive model of the regulation of ion homeostasis and gut alkalization in the larval mosquito awaits the characterization and localization of other major components of the system in addition to CA. It will also be very important to resolve the question of whether multiple CAs are expressed in the midgut and how each is distributed in this dynamic tissue.

Quantitative evidence corroborating the distribution of CA within the midgut and supporting the histochemical and in situ observations was obtained using the 80exchange mass spectrometric method. The results obtained with this method indicate that the gastric caeca exhibit the highest level of carbonic anhydrase, relative to total protein content, followed by the posterior midgut and the Malpighian tubules. The anterior midgut showed levels of activity so low that two possibilities could be considered: either the method could not detect the enzyme or it is absent from the anterior midgut. The





50


levels of activity in the anterior midgut might be too low to be detected using the 180 method, but that the enzyme is present throughout the entire length of the midgut.

In summary, our evidence demonstrates the existence of CA in Ae. aegypti larvae and it also suggests that the gastric caeca and posterior midgut exhibit the highest levels of CA activity. In addition, the enzyme seems to be associated with the small cuboidal cells of the midgut epithelium. Furthermore, enzyme activity has also been detected in membrane preparations isolated from whole midguts and could be due to the presence of more than one isoenzyme. Carbonic anhydrase activity has previously been demonstrated in the epithelium of the larval midgut of six species of lepidopterans, in which it has been associated with the particulate fractions of the homogenate (Turbeck and Foder, 1970). This is consistent with our hypothesis that there might be more than one CA and that one of these enzymes may be associated with the plasma membrane.

What is the role of CA in the alkalization mechanism? BTB proved useful in

monitoring the impact of CA inhibition on the maintenance of gut luminal pH and the excretion of acid/base. As mentioned earlier, Ae. aegypti larvae typically alkalize the medium in which they are reared by secreting bicarbonate ions (Stobbart, 1971). The ingestion of CA inhibitors altered the metabolism of the larvae to the point that the metabolic products secreted into the medium change the pH of the environment, shifting it towards more acidic values than those observed in the absence of inhibitors. The lowering of the pH of the medium might be related to a decrease in the rate of secretion of HCO3. The effect of the ingestion of CA inhibitors on the secretion of bicarbonate





51


maintenance of an alkaline pH within the midgut lumen (Boudko et al., 2001a). It is plausible that a decrease in the rate of secretion of bicarbonate is elicited by inhibiting the CA enzyme.

A simple model of bicarbonate transport fails to explain how the high pH is

achieved within the anterior midgut of the larval mosquito. At a pH of approximately 11, similar to that observed within the anterior midgut, the majority of bicarbonate is present as carbonate. In fact, measurements of lepidopteran midgut fluid have shown that it contains 37 mM carbonate and 17 mM bicarbonate (Turbeck and Foder, 1970). Since the pH of a 0.1 M solution of sodium bicarbonate is only approximately 8.3, secretion of bicarbonate alone cannot be responsible for the high pH observed in the anterior midgut (Dow, 1984). It could, however, explain the pH values at the gastric caeca and posterior midgut. The mechanism for maintenance of an alkaline pH within the anterior midgut must be more complex than just a simple buffering of a physiological solution with bicarbonate. Although this mechanism has been investigated (Wieczorek et al., 1999; Zhuang et al., 1999; Boudko et al., 2001a), its details remain unclear. However, the evidence suggests that a basal, electrogenic H+ V-ATPase energizes luminal alkalization in the midgut of larval mosquitoes (Zhuang et al., 1999; Boudko et al., 2001b). Although the electrogenic transport of K drives the pH gradient, there must also be flux of one or more weak anions in the opposite direction to maintain homeostasis. Several transporters are thought to participate in this mechanism.

Another line of evidence suggests that the levels of carbon dioxide in the





52


mM in the midgut lumen in larval Hyalophora cecropia (Turbeck and Foder, 1970). Recent measurements using capillary zone electrophoresis of larval Ae. aegypti fluids have revealed a bicarbonate/carbonate level as high as 50.8+4.21 mM in the midgut lumen compared with 3.96-2.89 mM in the hemolymph (Boudko et al., 2001a). These values correlate with those observed by Turbeck and Foder (1970). This combined evidence suggests that the CO2 that reaches the midgut lumen in the larvae of lepidopterans is rapidly converted to a mixture of bicarbonate and carbonate. The role of CA in the alkalization process would be of great significance. The generation of antibodies against A-CA will facilitate a detailed analysis of the cellular and subcellular distribution of this key enzyme in this system.





53














A



B


Figure 3-1. Effect of CA inhibition on culture medium pH with fourth-instar Ae.
aegypti larvae. Mosquito larvae typically alkalize the medium in which
they are reared (Stobbart, 1971). (A) Six culture wells each containing five
fourth-instar larvae incubated for 5 hours in medium containing 0.003%
Bromothymol blue (BTB). The blue color is retained, indicating a pH
greater than 7.6. (B) The same as A, except that each well also contains a
different concentration of the CA inhibitor methazolamide ranging from
10 to 10"3 M from left to right. A yellow color indicates a pH below 7.6.





54




























Figure 3-2. Effect of methazolamide on the alkalization of the midgut using
Bromothymol Blue (BTB) assay of pH within living, but isolated, gut tissue. Gut tubes were dissected after pre-loading with BTB and then
incubated for 5 hours in hemolymph substitute (Clark et al., 1999)
in the absence (A) or presence (B) of 10.6 M methazolamide. The loss of
blue coloration in B shows that the internal pH of the gut lumen has
dropped below 7.6. Scale bar represents 300 rm.





55










12

10

8

C6




2

0
GC AM PM MT


Figure 3-3. Relative activity of CA in different pooled segments of the midgut of larval
Ae. aegypti. Midguts were dissected from early fourth-instar larvae and
separated into gastric caeca (GC), anterior midgut (AM), posterior midgut
(PM) and Malpighian tubules (MT). The relative activity of CA was
measured using the 'O mass spectrometry method (Silverman and Tu, 1986), normalized to total protein content. The activity of the anterior
midgut was lower than that of the water blank and, thus, is set as "zero"
activity.








56





A A * s aeqyp t .... LA LF VAT.LP ST I.,ADWHX PrPArIav, w
A MorCA1Poo917 . .:. . : . : : :: - : . : : : :-u
Zebratih-Q925* 28
an A -PO7451 .....................A 28
MoueA14-NPO35927----.-.-. .MAI.I.TWfz.AADGC 45
C.elegans-T16575 . . ....... . . . . .........................TGNWATCDDD EC 24
RatCrV-P0204 i7 .T.rT VAPSTDSHWCTR IQAxaacamfl 50

Ades megypti NeV 94
HoraeeOA-POOC17 SE. 7
h r nCAS-P074 8 . 77
bMsoCUlA4-PC35927 av I WAUDPRGYDQLTBPZ.DECgmaGAITYQLBI.P . P 92 C.elegaae-t14575 74 RfatCAIV-NPO2047 s98

ass. aegypti ..1. .a 137 RormeCA-POO917 *s26
ebrar tia.-O 2051 a . 125 umanCAS-P07451 - . I. 125 oue CA14-PO15927 A so 142 C.elegmn-T16S75 121 RlatCAXV-NPOS2047? SE 146

Aedes aegypti a86
NorseCA -00917 *
ebratish-Q92051 174 HumanCAS-074S1 174
Moarse3CA 4 -P1o35927 192 C.elegans-T16575. 162 RatCATV--1O62047 TG.DVSDP, i .IPIT 195

Aedes egypti 234
HBorsc -300917 N. .SDtT. 222 i bratss h- 92051 rAN .YeT.. 1l 221 HumanCAS-1307451 TTK. . xag NT. 221 ato' A4-MPO35927 VIP.. Q . . 2 iP 240 C. legans-T16T575 YIl.. .VRLOB ]m 210 atCAZ -P062047 VSSCLE.OEB ,fitz.BAl ZEX 245

ade segypti. . . . T PSAIREDSGOR 282 MoreCAl-P91 . ......... 20
ebrat ish--Q92051 TPCK K . . . ........ 260
HuaaEnC U.3-P07451 P . ........... 260
Moue.CA1-O35927 pan E GRP.TTE 29D C..legans-T1575 Z. . . .gtQzag 2Z vtZVAN .......... 246
RatCAXV-PO62047 Exa glCLN 294
Aneda agyfpt� I. SLTLI VIZA.Al.. . . . . . . . . . . . . .. . . . . . . . . . . .. 298
orsm A2-P 0917 ...................................... ..... -- ...... 2 60
Zefria-h -Q92051 ............................................... 260
HWuman - ]PO 745 1 ............................................... 260
nonmea1 4-NPO 35927 LGLGVGrZAC LACL.YF ZAQK nnI samaRATTEA 337 C.elegan- 15,, 7 ..... . ... . ........................... ..... 246
RMatAX V-PO620 47 1.LVPTTLVASBF. . - . . . . . . - . . . . - . . . . - - . - . - . - . . .. 309





B Homology (%)
100 80 60 40 20 II I I

Aedes aegvpti


HorseCAl-P0917
61%
Zebrafish-Q9205 31% 56%

HumanCA3-P07451 35%

27%o
MouseCA14-NP035927 33%

Caenorhabditis-T 16575


RatCAIV-NP062047








57





A A...a aey o . -...M AFVATLLSTI .A=UYVT =WAwE s R.av i Pt 43*
Dros AAES7140 . . . . . -- . . . .. . ...**...........*........ O
Dr.om AA 57141 ... ........... ... .. .. O
Dros AA 56666 MS EIAvTKScr.Vfa a1MVt QRmAURNCHCAGRTXOS.PZAT 49 Dros AJtgd54494 ..- . .x.m.MPRMvGQTPDnEu..WaDsNNCD GPPK .w 43 Dro AAIP4 9948 . . R)C RmqprAxvIAP L-XCAB I.V'.DPGnOMGPwXEDY to Drow Abr44817 . . . . - - - . .. - - .. . - . . ... . k4BI14ManInuarrankGIPAJng6axY 25

Aads aegypti IDLTYaVxsDrAP S PIRM.G.G . sQrWeS 90 Dros AAPS7140 ... - . . OP- 37 Dros AAPS7141 . . .%DLCSSGaOIIQSP-LLRTVs:BISD.... * IPVN 43 Dro APrS666 6 TTAZ AvarjZGNLLPLEM INNGHsTVITIPKVN. ... . VTEVGE 94 Dros &Afl4494 90

Dros P.Af417 HZlPSS KM.Gll Z,4V dlrrl*V PG-.y-W RVD N73

Ades egyp. t .... 130 Cros AAr57140 83
Dros RS77t1 l A . . .
Droe AAP56666 * * 140 Dros AAPS4494 .. 130 Dres A A49948 148PLOY P 4t8w 2 Drom AA444817 EL : F * * L~][ 120

ndes aegypti 8 11
Drom aUP 7140P 130
Drow AA57141 138 Dropm A r5666 t. XF- * 2 Dros aar54494 - . THY8N. . TPMEAZ ZI5B177
Drow AP49948 .94 8 ta. .rG
Dro AAlie44817 . . H V166

Anda. *pM 226 Drom 45 7 T40 .180.
Dros ARA57141 R...... ....V...EL....D 188.. 5
Droe AAP546666 T.ArN L .236
Dros AAW54494 VXcaD VVDDL4 227
Dros AAN49948 ZDRKOE -r V 243 Dros aAA448e7 mn1.HrGDR LP . 214

Ade. aegpt 275
Dros AA57140 A 229
Drom 7rs1 i 7 235
Drom rAAS 6666 PP 5 B ~PAv A285
DroM APr4494 205 * VVEVBraga 274
Dram h r A49948 T * , .3z291 Dro AAr ALF7 R264

Andes mgypti eaSn CALeLiVeAAZAALLAKi . .th ... eao.a.s..e 29 Dres aPar7140 GAot.................. .......... -.. 235
Droe RP S7141.. I.. ( II. ..I I. II. . * ** * ** , ,, .. . .. . ..... 235
Dros AAES6666 TzaAmzz.KYDWFT. ... . . . . . . . . . . . . . . . . . . . . . . 304
Dros AM5449,4 SGSAGLTASSLGIf AG LI. . . . . . . . . . .. . . . . .-. 302
Dros an49948 HanSGBXPL17DAEWAA nhRAVLMpLvvsAALnterrP 335
Dros r4 d4817 InGG. .. . .. .. .. . . . . * . . - . . - * . . . . . . . . . . . . ...... 270


B Homology (%)

100 80 60 40 20 11 i I i

Aedes aegypti
38%
Drars AAF4494


Dros AAF57140 28%
49%
Drns AAFS71417
3% 27%1

Dros AAFS6666


Dros AAF49948
3r~o
Drs AAF44817




Figure 3-5. Comparison of the extrapolated amino acid sequences of A-CA with six
putative dipteran CA genes identified in the D. melanogaster gene





58











1 2 3 4 5 6 7 bp




1500>

500




Figure 3-6. Polymerase chain reaction (PCR) analysis ofAe. aegypti amplified cDNA
from different gut regions. PCR was performed using exact primers for the
cloned A-CA. Anterior midgut (lane 2), gastric caeca (lane 3), posterior
midgut (lane 4), whole gut RNA control (lane 5), Malpighian tubules (lane
6) and a water template control (lane 7) are shown. Note the primary
product in gastric caeca and posterior midgut samples at the expected size
of approximately 894 nucleotides. Also note the absence of this band from
other gut regions but the appearance of bands of higher molecular masses.
Lane 1 is a 100 bp molecular mass ladder (Promega).





59














A




















Figure 3-7. Hansson's histochemistry of whole mount Ae. aegypti gut. A. Intense dark
staining is observed in the cardia, gastric caeca (GC) and posterior midgut
(PMG), indicating CA activity. B. Higher magnification of the gastric
caeca. The distal lobes of the gastric caeca (Cap cells) exhibit relatively
low levels of reaction product, indicating lower levels of enzyme activity in
these cells relative to other cells of the gastric caeca. C. Higher
magnification of the PMG shows large, relatively unstained columnar cells
(*) contrasted with the smaller stained cuboidal cells (arrow). Scale bars






60






A. cardia




GC




AMG PMG



I
MT

B. C.




I .,







Figure 3-8. Localization of CA mRNA expression in larval Ae. aegypti. A. An isolated
whole mount gut probed with DIG-labeled cRNA for A-CA. Abundant hybridization is observed in the cardia, gastric caeca (GC), and posterior
midgut (PMG). B. The smaller cuboidal cells (arrow) display stronger
hybridization than the larger columnar cells (*). C. Isolated midgut reacted
wiP eses cnto*cN fr . cale... rs repesnt30 *ilfl m













CHAPTER 4
A GPI-LINKED CARBONIC ANHYDRASE EXPRESSED IN THE LARVAL MOSQUITO MIDGUT

Introduction

The CA enzyme expressed in the midgut of larval mosquitoes shares some

characteristics with the mammalian CA IV isozyme, including a glycosyl-phosphatidylinositol (GPI) link to the plasma membrane. Mammalian CA IV enzymes have been found in dynamic organs such as kidney, lung, gut, brain, eye, and capillary endothelium (Chegwidden and Carter, 2000). The human CA IV isoform was found to be as active as the CA II isoform in carbon dioxide hydration and even more active in bicarbonate dehydration (Baird et al., 1997). Studies of larval mosquito CAs are being pursued to better understand the alkaline gut system. As the anterior midgut of the larval mosquito lacks a highly active cytosolic CA II-like isozyme (previous chapter; Corena et al., 2002), the presence of a highly active CA IV-like isozyme within the mosquito gut may be able to provide the buffering capacity that is needed within the highly alkaline anterior midgut. A more detailed characterization of larval Aedes aegypti CA is presented in this study as well as the sequence of a homologous CA isoform from Anopheles gambiae. New tools and techniques, such as the generation of a mosquito-specific CA antibody and real time PCR, as well as improved methodology for in situ hybridization, have enabled





62


Results

Bioinformatics ofAedes Aegypti CA

We have previously cloned a CA cDNA from the Ae. aegypti midgut (accession number AF395662; Corena et al., 2002). Our initial structure prediction indicated that the protein was cytosolic. However, further characterization has indicated that this CA is actually membrane associated via a GPI-link. We have determined that the CA propeptide sequence encodes an extracellular protein with a hydrophobic tail region. The first 17 amino acids of the propeptide are predicted by the Simple Modular Architecture Research Tool (SMART) program to be the signal sequence (Letunic et al., 2002). This sequence "flags" the message for transport to the endoplasmic reticulum (ER). Using the PSORT II server, the prediction of membrane topology (MTOP) indicates that the Ae. aegypti CA sequence is GPI anchored. Amino acid G-276, is predicted by the GPI prediction server (Eisenhaber et al., 1999) to be the site for GPI attachment. The hydrophobic tail (L278-A289) allows translocation of the transcript through the ER plasma membrane and is also predicted to stabilize the protein with the membrane until the pre-formed GPI anchor is transferred to the protein. The hydrophobic tail is then cleaved to produce a completely extracellular protein that is tethered to the cell by the GPI link (for a review of GPI-linked proteins see Brown and Waneck, 1992). Sequence Comparisons of CA IV-like Isoforms

I also cloned a CA IV-like cDNA from the gut ofAn. gambiae, an important vector in the spread of malaria. This CA isoform (Ensembl gene ID:





63


gambiae genome. These cloned mosquito cDNAs from Ae. aegypti and An. gambiae are 61% identical to each other in amino acid residues and show similarities to the mammalian CA IV isozyme. However, in contrast to the mammalian CA IV, which is encoded by 7 exons (Sly and Hu, 1995), only 3 exons make up the An. gambiae CA isoform. Alignment of the mosquito CA IV-like isoforms from Ae. aegypti and An. gambiae with various mammalian CA IV isozymes reveals conserved features within this CA isoform (Fig. 4-1). For example, the multiple leucine (L) residues within the amino terminus of the mammalian CA IV propeptides that comprise the signal sequence are found in the Ae. aegypti and An. gambiae CA IV-like isoforms. One important feature of the mosquito CA IV-like sequences is the conserved alignment of G-69 (human CA IV numbering) with the human, bovine, and rabbit CA IV sequences. This particular amino acid residue has been changed to glutamine (Q) in rat and mouse CA IV, which results in reduced enzyme activity (Tamai et al., 1996b). Additionally, all of the CA IV sequences, including the mosquito isoforms, display a hydrophobic tail region. In addition to the conserved CA IV-like features of GPI-linked proteins, there are also conserved cysteine residues (C28 and C211, human CA IV numbering) between all of these CAs (Fig. 4-1). It has been determined via cysteine labeling, proteolytic cleavage and sequencing that these two cysteine residues, in the human CA IV, form a disulfide bond (Waheed et al., 1996). A second disulfide bond is present in the mammalian CA IVs between residues C6 and C18 (human CA IV numbering; Waheed et al., 1996). This second pair of cysteine residues, and hence the resultant disulfide bond, is not present in either of the





64


Although the mosquito CA isoforms display features similar to mammalian CA

IVs, such as a 5' signal sequence, a hydrophobic 3' tail, and extracellular GPI expression, there is one striking difference in the amino acid composition of the mosquito CAs' active sites. The active site within all of the 14 characterized mammalian CA isoforms is tightly conserved. Three histidine residues (His-94, His-96, and His-119) are essential for CA activity through their coordinated binding of a required zinc molecule. The absence of one or more of these histidine residues results in inactive proteins called CArelated proteins (CARPs), as found in mammalian CA isoforms VIII, X, and XI (Tashian et al., 2000). The mosquito CA IV-like isoforms contain all three of the required histidine residues along with all of the other 13 highly conserved residues found in most other CAs (refer to Fig 4-1; Tashian, 1992; Sly and Hu, 1995; Tamai et al., 1996a). However, as the alignment shows in Figure 4-1, there is a conserved gap within the active site of the mosquito CAs that is not present in any of the mammalian active sites. Because this shortened active site was found in mosquitoes but was not found in any mammalian CA isoform, I searched the Drosophila melanogaster genome for potential CA homologs. The D. melanogaster genome was found to contain 14 putative CA genes (ENSF00000000228), the same number found in An. gambiae. One out of the fourteen D. melanogaster CA isoforms was found to contain the identical number of deleted amino acids as the mosquito forms within the active site region. This D. melanogaster CA sequence (accession number CG3940-PA) may also be a GPI-linked isoformn due to the presence ofa lysine-rich 5' signal sequence and hydrophobic tail region. Figure 4-2





65


of the human CAs, which show several additional amino acids within the conserved active site region.

Localization of CA IV-like Isoform in the Mosquito Midgut

In situ hybridization analyses indicate that the Ae. aegypti CA message is

expressed most heavily within the epithelial cells of the gastric caeca and posterior midgut (Fig. 4-3). An antisense cRNA probe corresponding to the entire cDNA sequence generated strong cytoplasmic staining of the proximal gastric caeca, while the distal Cap cells showed no detectable hybridization (Fig. 4-3B). Rostral to the gastric caeca, a strong localization was evident in a small subset of cardia cells that encircle the tissue, forming a collar (Fig 4-3B). These "collar cells" are clearly different from the surrounding cells in this same area. This technique also highlighted a set of specific epithelial cells that are found only in a subset of the posterior midgut. These CA-positive cells form a ring of about 5 cells in width that circumscribe the lower-posterior gut region (Fig. 4-3C). The CA message was also localized to longitudinal and circular muscle fibers of the anterior and posterior midgut (Fig. 4-4). Following the longitudinal muscle fibers, in close association, are distinct nerve fibers that also show strong CA mRNA expression (Fig. 4-4). Epithelial cells of the anterior midgut were clearly void of signal beneath the labeled muscle and nerve cells. Specific staining was also evident however within the abdominal ganglia central nervous system (CNS) and peripheral nerve tissue (Fig. 4-5). No labeling was seen in the Malpighian tubules. Real Time PCR Analysis ofAedes aegvpti CA IV-like Transcripts






66


dissected and the head, gastric caeca (GC), anterior midgut (AMG), posterior midgut (PMG), and Malpighian tubules (MT) were pooled. RNA was isolated from each tissue sample for subsequent real time PCR analysis. Ae. aegypti ribosomal RNA (Genbank accession number M95126) was used to normalize the quantity of transcript from each sample. The results are presented in graph format in figure 4-6. Gastric caeca contain the greatest quantity of CA message within the gut sections (Fig. 4-6). The head tissue contained roughly half as much message as the gastric caeca (Fig. 4-6). The localization of CA IV-like message within the larval head is consistent with the localization of CA message to CNS tissue by in situ hybridization. The anterior midgut, posterior midgut, and Malpighian tubule collections showed CA message only marginally greater than zero (Fig. 4-6).

Immunolocalization of CA IV-like Protein in the Mosquito Gut

The amino terminal peptide sequence (GVINEPERWGGQCETGRR) was chosen from the Ae. aegypti CA sequence as an antigen for antibody production. The resultant antiserum was used to immunolocalize the CA IV-like isoform within the mosquito gut. The pre-immune serum was used as a control for all experiments. Immunoreactivity was found within the gastric caeca region of the gut as well as on muscle fibers along the anterior midgut (Fig. 4-7A). A subset of anterior muscle fibers displays the strongest and most striking labeling on their extracellular surface, while other muscle fibers show little or none. Immunoreactivity was also found within the CNS ganglia and immunoreactive nerve fibers that traverse the out (Fig. 4-8R. There ws no imminnoractivity deterted in





67


Antibody Cross-Reactivity with Other Mosquito Species

The 18 amino acid sequence from the Ae. aegypti CA, used to elicit the antibody response, shares 14 identical residues with the homologous An. gambiae CA protein (refer to Fig. 4-1), and therefore, the antiserum recognizes the CA IV-like isoform present in An. gambiae as well. The immunoreactivity within the An. gambiae gut displays a strikingly similar, yet species-distinctive pattern of CA IV-like protein expression (Fig. 47B). Similar to Ae. aegypti (Fig. 4-7A), not all of the muscle fibers were localized in the An. gambiae gut. However, in An. gambiae the immunolabeled muscle fiber network runs down the lateral sides of the midgut, while the dorsal and ventral muscle fibers are not immunoreactive (compare Fig. 4-7A to 4-7B).

The high sequence conservation between the chosen antigenic peptide from Ae. aegypti and the An. gambiae CA, prompted us to check other mosquito species for immunoreactivity. The other species tested also displayed the same strong labeling on a subset of anterior gut muscle fibers that included both circular and longitudinal muscle fibers. Figures 4-9 and 4-10 display the immunoreactive results obtained from Aedes albopictus and are representative of the results from the other mosquito species including Ae. aegypti and An. gambiae. The CA IV-like immunolabeling is clearly confined to only a subset of actin-containing muscle fibers (Fig. 4-9D). Labeled phalloidin, which binds to actin, labeled all of the muscle fibers within the gut, while the antibody for CA IV-like mosquito CA only recognized a subset of the anterior muscle fibers. The CA antibody is also specific for the extracellular plasma membrane of these muscle cells.






68


discovery. This immunoreactive subset of CA-containing muscle fibers traverses the cells that surround the highly alkaline anterior gut lumen. Determining the role of these CA-specific muscle fibers holds promise for deciphering the necessary CA component of mosquito gut alkalization.

Phospholipase C Treatment

In order to validate the CA IV-like isoform cloned from Ae. aegypti is indeed GPI linked to the membrane, live fourth instar Ae. aegypti larvae were subjected to phosphoinositol-specific phospholipase C (PI-PLC) treatment and subsequent immunohistochemistry. This compound is capable of breaking the GPI-anchor and therefore severs GPI-linked proteins from the plasma membrane. Larvae subjected to PIPLC treatment showed a decrease in CA immunoreactivity along the midgut muscle fibers, as compared to the non PI-PLC treated controls (Fig. 4-11). This evidence supports the prediction that the mosquito CA IV-like isoform is in fact GPI-linked to the outer plasma membrane.

Discussion

In this study, we show that two GPI-linked CAs are expressed in the midguts of two different mosquito species that rely on an alkaline digestive strategy. These mosquito CAs share characteristics with the mammalian CA IV isozyme, including the GPI link to the membrane. In situ hybridization localized CA message predominantly to the gastric caeca and posterior midgut epithelial cells, as well as muscle and nerve fibers along the anterior midgut, and CNS ventral ganglia. Real time PCR analyses confirmed





69


while the head sample contained roughly half of the gastric caeca concentration. CA immunoreactivity was most striking on specific muscle fibers of the anterior midgut, along with labeling of the gastric caeca and CNS ganglia. The localization of CA protein in a distinct subset of muscle fibers was found in several different mosquito species. This finding suggests that many mosquito species express a similar CA IV-like protein as well as confirms the immunoreactivity by only a subset of muscles. Immnunolocalization of the CA V-like isozyme within the mosquito gut and CNS also demonstrates that the CA message is being translated into protein and agrees with the localization of CA mRNA in the gastric caeca and muscle fibers. Interestingly, not all of the muscle fibers that show CA mRNA expression also show CA protein. Therefore, only a fraction of the muscle fibers that contain the CA mRNA are translating the message into protein. This may represent a form of regulation, in which the muscle fibers not expressing the CA protein could be "turned on" to translate the CA message if needed. This ability would be very advantageous if indeed this particular CA isoform is involved in buffering the alkaline gut.

The posterior midgut was also found by in situ hybridization to express the CA mRNA. However, both the real time PCR analysis of CA mRNA expression and the immunolocalization of CA protein failed to determine the presence of this particular CA within the posterior midgut. One explanation for this may be the inability of our CA antibody to permeate the posterior midgut tissue. However, the real time analysis, which is very specific for a region ofmRNA, also did not find this particular CA message in





70


region. Evidence for this is supported by genome data showing 14 different CA isoforms, all with regions of high nucleotide identity. The posterior midgut region does display CA activity, but apparently not as a result of the GPI-linked CA isoform presented here. The specific isoform or number of CAs contributing to the activity of the posterior midgut is still unknown.

We have previously shown that the application of CA-specific inhibitors

dramatically decreases the alkaline gut pH, and in fact is lethal to the larval mosquitoes (Corena et al., 2002). We now present evidence that a CA found in the mosquito gut is most similar to the mammalian CA IV isozyme but contains a novel active site motif unlike any of the mammalian CA IV isoforms (Fig. 4-1). The finding of a novel CA active site within the mosquito may facilitate the construction of a mosquito-specific CA inhibitor for use in larval mosquito control. We are hopeful that the ongoing mosquito CA crystallization project will yield further significant structural differences from the mammalian CA IV structure. These differences may be useful in the design of a mosquito-specific CA inhibitor.

Out of the 14 mammalian CAs identified thus far as cytosolic, membrane-bound, secreted, and mitochondrial, only CA IV has a GPI link to the cell membrane. The localization of this highly active mammalian isozyme to dynamic tissues such as the gut, brain, kidney, and lung supports the important catalyst role of CA. It should not be surprising that the gut of a mosquito, a highly alkaline and fluctuating system, has been found to contain a presumably active CA IV-like isoform as well. The single amino acid






71


CA IV enzyme (Tamai et al., 1996b). Mutating glutamine-63 to glycine within the rodent sequence resulted in almost three times greater CA activity (Tamai et al., 1996b). Unlike the rodent sequences, both of the mosquito CA IV-like sequences display the high activity glycine residue adjacent to histidine-69 (Human CA IV numbering, refer to Fig. 4-1).

The task ahead is to decipher ifa GPI-linked CA is better equipped to function in a highly dynamic system than other CA isoforms. Perhaps the GPI link affords the mosquito CA enzyme a characteristic advantage in buffering such an alkaline pH through its exclusively extracellular expression. Residing at the plasma membrane intrinsically affords this isozyme the best location for monitoring CO2 and HCO3 flux. Indeed, mammalian CA Vs are expressed on membrane surfaces where large fluxes of CO2 and /or HCO3" are expected (Sly, 2000). The most compelling ability of GPI-linked proteins is that they are known to elicit second messengers for signal transduction (Brown and Waneck, 1992). The alkaline pH of the larval mosquito gut was found to drop within two to three minutes after being narcotized or just simply handled (Dadd, 1975). This "handling effect" lends itself to our prediction that larval mosquitoes may exert neuronal control over the generation of the gut lumen's pH. Since a GPI-linked CA was localized within the mosquito gut and CNS tissue we propose that a GPI-linked CA may regulate the pH of the mosquito gut by severing the GPI-link and starting a signal cascade.

Further studies are being pursued within the mosquito gut to encompass the

localization and characteristics of other CA isnfnrmsa well a hirhnnate Pverhanaerc







72


Antienic tide
Ades CA MIALVATLL----PSTIRADE PTPA-Anoph CA NKSFTLLLCYALFVLRAARGDE PTPG--TNGVMSEPERWGGQCDNGRR PIDLTI Human CA IV MRMLLALLALSAARP SHW QAESSNYPCLVPVENGGNCQKD-R PINIVT Bovine CA IV RLLLALLVLAAAP H IQVKPSNYTCLEPDWEGSCQNN-R PVNIVT Rabbit CA IV MQLLLL LALGALRPLA ELHEE IQA--SNYSCLGPDKEQBDCQKS- PINIVT Marine CA IV HLLLLALLALAYVAPST DSGW IQTKDPRSSCLGPEEPGACEN-Q PINIVT Rat CA IV NQLLLALLALAYVAPST DSHC IQAKEPNSHCSGPEQWTGDCE- PINIVT
* 1 * * * * * Aedes CA QAAVKEGDFAPFLF-SNYNPIRNAQL GHSIQIDSTDPSVTLYGGGLG Anoph CA A~AVRGQFAPLFF-SNYMLPLKQPR GSlQIRSAI, Human CA IV TKAKVDKKLGRFFFSGYDK-KQTWTV GH-ASISGGGLPAPYQ Bovine CA IV AKTQLDPNLGRFSSGYNM-KW GH-PSIAGGGLSTRYQA Rabbit CA IV TKAEVDRSLGRFHFSGYDQ-REARL GHSVMVSLGD-EISISGGGLPARYRA Marine CA IV ARTKVRPRLTPFILVGYDQ-KQQPI QBTVEMTLGG-GACIIGGDLPARYEAl Rat CA IV SKTKLNPSLTPFTFVGYDQ- QHSVESLGE-DIYIFGGDLT
* * * ** * * * Aedes CA HWG IAGVRYG SRYNS--LTEAGAVKNVAVVGVLFRVSNQD Anoph CA HWG LDTRYG VERDTRnYAS--EDALgQANASQP Human CA IV HWSDLPYKG SLDGERF HGTSPEDEIAVLAFLVEAGTQV Bovine CA IV HNSRAtRG FDGERYA IV E GASONQFAEDEIAVLAFMVEDG-SE Rabbit CA IV HWSQCLDRI LDGRSIVQKETGTSGWEVQD--DDSIAVLAFLVEAGPTM Mutine CA IV NWSNZDNG IDERFIVR 'LTS----.--SEDSEDEFAV Rat CAIV HNSEESMEG. IDGKHP VV E G--DVQDSDSEDKIAVV
** *** * * ** ** Aedes CA NTENVVLETSQDIRDAAGKSAPLK-GKLSPHNPLPENRTSY S CAE Anoph CA !WHIDTILDTATEIQNEVGEALLR-GKLSPYNLLPSNRTS GS CAE RHuman CA IV NEGGFQPLVEALSNIPEMSTTMAE-SSLLDLLPKEEKLRHY CD Bovine CA IV NVNFQPLVEALSDIPRE TGVSLFDLLPEESLRY S CD Rabbit CA IV EGFQPLVTAASIPGTNTTMAP-SSLDLLPAELRY S CSE Marine CA IV NKGFQPLVEALPSISKPHSTSTVRE-SSLQDMLPPSGSCDE Rat CA IV NEGFQPLVEALSRLSKPFTNSTVSE-SCLQDPE LSAYS E
* S. a* e ****** * **e Aedes CA TVFTELSLPVSLDQVELfT- - - IRDPSGHELV LPLNARALVTDYSG Anoph CA TVFTESISVS---IDQTGREL PLRALVATEQGFA Human CA IV TVFREPIQLRREQILAFSQKLYYDELGTV Bovine CA IV TVFKPIQLHRDQILAFSQKLFDDQQSLGQRQL QPL Rabbit CA IV TVKEPIRLERDQILEFSSKLYYDQRS LLPLPL Murine CA IV TVYKQPIKIHNQFLEFSKNLYDQPLGRVF LLSLPL Rat CA IV TVFEEPIKIRKDQFLEFSKKLYYDQEQ IPLGRLLSLPL
** * * * ** a*** *
Att tA A A
Aedes CA PCLSLTLI-..... ------.-...----- - ---- -------a-------Anoph CA TD SNVVFLGAIVLLVITSRLSY----------------------------e --a
RiSuman CA IV PALLGPMLACLLAGFLR -----. -----------------------------------Bovine CA IV PTLLAPVLACLTVGFLR---- -. - -.--.- ..--. --------------Rabbit CA IV PTLL LLR--------------------------- ----------- ----Maurine CA IV PTLLVPTLTCLVANFLQ------- --------------- ------------------Rat CA IV PTLLVPTLTCLVASFL .----- ---------------------------- ---------Figure 4-1. Alignment of several mammalian CA IV enzymes with two mosquito CA
isoforms. The leucine-rich signal sequences are found in all aligned isoforms, along with the 3 essential zinc-binding histidines (red), and
-v a in4 r-ide (.r~n tha n~r -i.las * a~d Th -,mo ,-v






73





AAL72625 Aedes CA FVLDQMHFHWG--- -----SEHTIAGVRYGQELHMVHHDS AAQ21365 Anoph CA FVLDQMHFHWG-- ------ SEHTLDDTRYGLELHLVHHDT CG3940-PA Dros CA FVVEQIHMWW-------- SEHTINDIRYPLEVHIVHRNT P00915 Human CA I YRLFQFHFHWG--STNEHGSEHTVDGVKYSAELHVAHWNS P00918 Human CA II YRLIQFHFHWG--SLDGQGSEHTVDKKKYAAELHLVHWNT P07451 Human CA III YRLRQFHLHWG--SSDDHGSETVDGVEYAAELHLVHWNP P22748 Human CA IV YQAKQLHLHWS--DLPYKGSEHSLDGEHFAMEMHIVHEKE AAB47048 Human CA V YRLKQFHFHWG--AVNEGGSEHTVDGRAYPAELHLVHWNS CAC42429 Human CA VI YIAQQMHFHWGGASSE I SGSEHTVDGIRHVIEIHIVHYNS P43166 Human CA VII YRLKQFHFHWG--KKHDVGSEETVDGKSFPSELHLVRWNA JNO576 Human CA VIII FELYEVRFHWG--RENQRGSEHTVNFKAFfMLHLIWNS AAB14950 Human CA IX YRALQLHLHWG--AAGRPGSEHTVEGHRFPAEIHVVWHLST Q9NS85 Human CA X HRLEEIRLHFG--SEDSQGSELLNGQAFSGEVQLIYNH AABO2662 Human CA XI HRLSELRLLFG--ARDGAGSEHQINQGFSAEVQLIHFNQ AAN23981 Human CA XII YSATQLHLHWG-NPNDPHGSEHTVSGQHFAAELHIVHYNS BAA85002 Human CA XIV YVAAQLHLHWG-QKGSPGGSEHQINSEATFAELHIVHYDS
*** *


Figure 4-2. Clustal alignment of CA protein sequences. All characterized human CA
isoforms are presented along with putative GPI-linked isoforms from Ae.
aegypti, An gambiae, and D. melanogaster. Three histidine residues that are
required for the essential binding of zinc are shaded in blue. Note that I or more of these histidine residues are missing from the inactive human CArelated proteins VIII, X, and XI while all three histidines are present within the Dipteran sequences. The three CAs from Dipterans contain a shortened
active site region (marked by red dashes) when compared to any of the human
or other mammalian CA sequences. This difference may provide a potential
target for mosquito-specific CA inhibitors, for use as larvacides.





74





A B











C













Figure 4-3. Localization of CA mRNA in a wholemount preparation of early 4 instar
Ae. aegypti. A. The wholemount gut preparation localizes CA message to specific cells of the gastric caeca (GC) and posterior midgut (PMG). B. A
subset of cardia (arrows) and gastric caeca cells display the CA message.
The distal lobes of the caeca, called Cap cells, display no staining (*). C.
There is a distinctive labeling pattern of CA message within a specific band
of posterior epithelial cells. In addition, numerous trachea (arrows) are
heavily labeled along the length of the midgut. Scale bar represents 300 pm
in A, 150 pm in B, 75 pm in C.





75


A
AMG

PMG




BZ












C








AMG

Figure 4-4. Expression of CA mRNA in Ae. aegypti anterior midgut. While the cardia,
gastric caeca and posterior midgut display heavy epithelial hybridization,
there is also specific CA mRNA expression seen in muscle and nerve cells.
A. A representative whole mount Ae. aegypti larvae displaying the strong
CA expression in epithelia, along with muscle fiber staining that can be
overlooked at low magnification. B. The beginning of the anterior midgut
shows hybridization to both muscle (arrowheads) and nerve fibers (arrows).
The labeled fibers reveal striated muscle running longitudinally down the





76





A.















B.













Figure 4-5. Localization of CA IV-like message within Ae. aegypti CNS tissue. A. In
situ hybridization localized the CA IV-like mRNA within all ventral ganglia
CNS clusters (arrows) as well as hair sensory cells (*) and longitudinal
nerve fibers (arrowheads). B. The sense control probe displayed no specific
hybridization. Scale bar represents 300 pm.





77












Aedes aegypti Carbonic Anhydrase
1.5






*g
�n i~6C i
0J






Ae.aegpi Tisue Sctions

Figure 4-6. Relative quantification of CA IV-like message in Ae. aegypti larvae using
real time PCR The gastric caeca tissue displays the greatest amount of CA
IV-like message. Data was normalized to the gastric caeca (GC) sample. The anterior and posterior midgut along with the Malpighian tubules display very
little CA message. The head section displays roughly half the amount of
message found in the gastric caeca.
U


a

MO Mr

0


Figure 4-6. Relative quantification of CA IV-like message in Ae. aegypti larvae using
real time PCRL The gastric caeca tissue displays the greatest amount of CA
NV-like message. Data was normalized to the gastric caeca (GIC) sample. The anterior and posterior midgut along with the Malpighian tubules display very
little CA message. The head section displays roughly half the amount of
message found in the gastric caeca.





78















A B















Figure 4-7. Ae. aegypti and An. gambiae CA protein labeling. The antibody generated
against the Ae. aegypti CA can also be used to localize the homologous CA
isoform within An gambiae. The larvae were incubated with phalloidin
(red) and the CA-specific antiserum (green). Colocalization of the red and
green signals appears yellow. A. The antibody localization shows the
strongest labeling in Ae. aegypti for a subset of muscle fibers in the anterior
midgut and the proximal portions of the gastric caeca. B. Antibody
localization ofAn. gambiae CA is depicted by the yellow muscle fibers,
while the red muscle fibers are not recognized by the antibody. The scale
bar represents 300 pm in A, 150 pm in B.





79














AB














Fig 4-8. The Ae. aegypti CNS ganglia express the CA IV-like isoform. A. Pre-immune
serum does not show any detectable labeling of the CNS tissue. B. Strong
immunolabeling for the mosquito CA IV-like isoform is displayed in the
ventral ganglion clusters, as displayed by the fluorescent green coloring as
compared to the yellow control (pre-immune) ganglia. The scale bars
represent 100 pm.





80






























Figure 4-9. Immunolocalization of mosquito CA IV-like enzyme in Aedes albopictus.
Muscle and nerve fibers within the anterior midgut region are heavily labeled.
A. Selective labeling of particular muscle fibers (green). B. Labeled phalloidin (red) was used to localize actin and labels all muscle fibers,
including those that were not recognized by the antibody against mosquito
CA. C. A nuclear label (blue) was used to distinguish cell numbers present.
D. Overlay of all three signals. Colocalization of the green and red signals
appear yellow. The CA antibody recognizes only a subset of anterior muscle
fibers, and is seen in several different mosquito species. The scale bar
represents 50 pm.





81















A B















Figure 4-10. High magnification ofimmunoreactive muscle fibers within the Aedes
albopictus midgut. A. Labeling of muscle fibers appears to be
extracellular. B. Cross section of the same fiber demonstrates that the
localization pattern is confined to the extracellular plasma membrane of
the midgut muscle fibers. The scale bars represent 25 pm.





82



A B













D













Figure 4-11. Immunoreactivity ofAe. aegypti guts for the CA IV-like isozyme. The red
labeling is specific for muscle fibers. The green labeling shows localization
of the mosquito CA IV-like protein. The yellow labeling shows colocalization of the mosquito CA IV-like isoform and actin. Prior to
immunolabeling, the guts were treated with PI-PLC to determine if the CA IV-like isoforms are GPI-linked to the cell membrane. A. Immunolabeling
of the gastric caeca, without the PI-PLC treatment, displays heavy yellow
labeling of the GPI-linked CA isoform. B. After PI-PLC treatment there is
no CA IV-like immunolabeling of the gastric caeca. C. The anterior midgut displays the immunolocalization of the CA V-like isoform along a subset of
muscle fibers. D. After PI-PLC treatment the yellow immunolabeling for
the GPI-linked mnrnito CA is orently reduced The dereSae














CHAPTER 5
ANION EXCHANGER EXPRESSED WITHIN THE LARVAL ANOPHELES GAMBIAE MOSQUITO Introduction

The larval mosquito gut provides an ideal model for studying epithelial transport

due to its cellular simplicity, being only one cell layer thick. The transport of bicarbonate within the larval mosquito gut was prompted by several studies (Boudko et al., 2001a,b; Corena et al., 2002). Bicarbonate is the main pH buffer in most complex organisms (Sterling and Casey, 2002) so it was predicted that de-protonated bicarbonate (ie. carbonate) is necessary for attaining the highly alkaline lumen of the larval mosquito midgut. However, the anterior midgut was found to apparently lack an active CA enzyme (Corena et al., 2002). An alternative to bicarbonate being rapidly produced within the anterior midgut, is the transport of bicarbonate into the anterior midgut. This alternative was supported by a previous study which implicated a chloride/ bicarbonate anion exchanger within the larval mosquito gut through the use of self-referencing ionselective (SERIS) microelectrodes (Boudko et al., 2001b). However, the molecular identity of the chloride/ bicarbonate anion exchanger (AE) was not determined. We now present the first anion exchanger (AE) to be cloned and characterized from the An. gambiae mosquito, in an attempt to unravel the physiology of an extremely alkaline






84


Baltz, 1999). AE localization will also distinguish whether the AE is co-expressed within the same epithelial cells as the HIt V-ATPase and if the polarity of basal or apical expression is also the same. The localization of a If V-ATPase within the larval mosquito gut was found to be apical in the gastric caeca, basal in the AMG, and then apical again in the PMG (Zhuang et al., 1999).

The highly dynamic system of alkaline digestion in the larval mosquito gut does not exist in any known mammalian system. However, mammalian organs such as the kidney are able to perform many parallel functions of the mosquito gut, such as water regulation, filtration, and ionic homeostasis. The CA, AE, and I+ V-ATPase proteins, in particular, have been extensively studied and localized within the mammalian kidney due to their dynamic roles in acid-base balance (Huber et al., 1999; Schwartz, 2002). The colocalization and polar expression of an AE with a I+ V-ATPase will define the epithelial cells of the mosquito gut as resembling the mammalian kidney A-intercalated cell type, the B-intercalated cell type, or the non-A non-B intercalated cell type (types as defined by Brown and Breton, 1996 and Kim et al., 1999).

Results

An. gambiae AE Sequence Analysis

The full length An. gambiae AE 1 (AgAE1) cDNA was cloned from midgut tissue and contains 3309 bases (accession number AY280611) with a molecular weight of 123 kDa for the predicted protein. The NCBI conserved domain search tool (CDD)
a4te* A .Lr ned that - , l t s e wa pr o . a a ily of b aic na bate + a + transporters






85


This complex of BTs contains both the solute carrier 4A (SLC4) and solute carrier 26A (SLC26A) proteins. More specifically, the EnsEMBL database places this particular AE cDNA sequence within the anion exchange/ band3 protein family (ENSF00000000189) as 1 of the 8 putative anion exchange band3 transcripts (ENSANGP000000101 12) encoded in the An. gambiae genome. These 8 transcripts arise from 3 different genes (ENSANGG00000007623, ENSANGG00000004501, and ENSANG00000012483). The gene that gives rise to the cloned AE that we are presenting (ENSANGG00000007623) is located on chromosome 3R. The other 2 genes (ENSANGG00000004501 and ENSANG00000012483) are located on chromosome 2L (Hubbard et al., 2002; Clamp et al., 2003).

The 1102 amino acids comprising the An. gambiae AE form a cytosolic

framework at the amino terminus while the carboxy terminus is composed of 12 transmembrane spanning domains also with an intracellular cytosolic terminus (hmmtop v.2; Tusnady and Simon 1998; Tusnady and Simon 2001). This hmmtop prediction was generated based on two assumptions: 1), that the CA binding site is within the carboxy terminus; and 2), that the C-terminus is intracellular, as is found for all known AEs. This structure is consistent with the predicted structure of the Drosophila melanogaster sodium dependent anion exchanger (NDAE1), which consists of a 12 membranespanning pattern with intracellular carboxy and amino termini (Romero et al., 2000). The highly conserved sequence identity of the An. gambiae AE1 with respect to the D.
melanoastor NTAE1 allows the nredictive 17 transmembrane-nanning dnmnin of the
molnnnanetorMThAPl 2hIAUYC the nreelirtive 19 frnnQmemhrsne-cn2nnlno dnmnin~ Af the





86


The amino terminus of the AgAE protein contains 523 cytoplasmic residues

enriched with multiple binding sites for cytoskeletal proteins (Bairoch et al., 1997). The carboxy terminus contains the membrane-spanning domains that are responsible for ion transport. According to the PROSITE motif search and the hmmtop server, sites of potential post-translational modification include 2 cAMP and cGMP-dependent protein kinase phosphorylation sites, 16 protein kinase C phosphorylation sites, 11 casein kinase II phosphorylation sites, 16 N-myristoylation sites, 1 prokaryotic membrane lipoprotein lipid attachment site, and 1 leucine zipper pattern (Bairoch et al., 1997; Gupta et al., 2002). The last 82 amino acids of the An. gambiae AE carboxy terminus are predicted to project into the cytosol, the correct orientation that is expected for binding of a cytosolic CA. The CA II binding site comprises a hydrophobic amino acid residue followed by at least two acidic residues within the next four residues (Vince and Reithmeier, 2000). Figure 5-2 displays the CA II binding sites found in several AE proteins along with the putative CA II binding site (LDDIM) in the An. gambiae AE. BT Sequence Comparisons

Following the sequence prediction that this An. gambiae BT is an AE, the closest characterized protein sequence is the NDAE1 (accession number AAF98636) from D. melanogaster. The An. gambiae AEl shares 72% identity with NDAEI (Fig. 5-3). However, the greatest similarity is with an uncharacterized splice form ofNDAEI (AAF52497) that contains an inserted sequence that is also found in the An. gambiae AE sequence (Fig. 5-4). Other BTs such as the sodium bicarbonate cotransoorters (NBCs)






87


that are also capable of transporting sulfate (SLC26A group) show only 11% amino acid identity (Fig. 5-3).

Anion exchangers, specifically AE2 and AE3, were determined to be pH

sensitive. AE3 is stimulated by intracellular alkalization whereas AE1 is not. More specifically, a region of amino acids (WRETARWIKFEE) within the carboxy terminus is responsible for the pH sensitivity seen in AE2 (Vince et al., 2000). The An. gambiae AE sequence in this same region contains 14 of the 16 residues found within AE2 while the other two amino acids are conserved (Fig. 5-5). AEl shares only 8 of the 16 amino acids in this region.

Localization of Anion Exchanger mRNA in An. gambiae Larvae

A DIG-labeled antisense cRNA probe comprising the full length AE cDNA was employed to localize the AgAE1 mRNA. A DIG-labeled sense probe was used as a control. The AgAEl mnRNA was found in every region of the larval gut including gastric caeca, anterior midgut, posterior midgut, Malpighian tubules, and rectum (Fig. 5-6). In the gastric caeca and posterior midgut regions, the probe was localized to epithelial cells. Within the gastric caeca the labeling is most intense in the area where the lobes face the lumen. The gastric caeca labeling was confined to the proximal cells, whereas the Cap cells displayed no label (Fig. 5-6A,B). The rectum displayed staining in a small subset of epithelial cells along with tracheoles (Fig. 5-6D).

Muscle, nerve, and trachea cells that traverse the outer plasma membrane of the

anterior gut epithelial cells were labeled with the AE antisense nrobe (Fig. 5-7. Laheled




Full Text
The mammalian anion exchanger (AE1) has been shown to physically bind a
cytosolic CA isoform (Sterling et al., 2001a). An AE has been cloned from the An.
gambiae midgut (refer to chapter 5) and protein expression has been localized to the
gastric caeca and posterior midgut regions. This AE also has a putative CA binding site
within its intracellular carboxy terminus. The membrane localization of this predicted
cytosolic CA may be caused by its interaction with a membrane protein such as an AE.
Future experiments will reveal if this cytosolic CA is indeed coupled to an AE within the
gastric caeca and posterior midgut regions of the mosquito gut, forming a bicarbonate
transport metabolon. The existence of such a metabolon within gut regions flanking the
highly alkaline anterior midgut may be the mechanism through which cellular
homeostasis is maintained.


51
maintenance of an alkaline pH within the midgut lumen (Boudko et al., 2001a). It is
plausible that a decrease in the rate of secretion of bicarbonate is elicited by inhibiting the
CA enzyme.
A simple model of bicarbonate transport fails to explain how the high pH is
achieved within the anterior midgut of the larval mosquito. At a pH of approximately 11,
similar to that observed within the anterior midgut, the majority of bicarbonate is present
as carbonate. In fact, measurements of lepidopteran midgut fluid have shown that it
contains 37 mM carbonate and 17 mM bicarbonate (Turbeck and Foder, 1970). Since the
pH of a 0.1 M solution of sodium bicarbonate is only approximately 8.3, secretion of
bicarbonate alone cannot be responsible for the high pH observed in the anterior midgut
(Dow, 1984). It could, however, explain the pH values at the gastric caeca and posterior
midgut. The mechanism for maintenance of an alkaline pH within the anterior midgut
must be more complex than just a simple buffering of a physiological solution with
bicarbonate. Although this mechanism has been investigated (Wieczorek et al., 1999;
Zhuang et al., 1999; Boudko et al., 2001a), its details remain unclear. However, the
evidence suggests that a basal, electrogenic FT V-ATPase energizes luminal alkalization
in the midgut of larval mosquitoes (Zhuang et al., 1999; Boudko et al., 2001b). Although
the electrogenic transport of K+ drives the pH gradient, there must also be flux of one or
more weak anions in the opposite direction to maintain homeostasis. Several transporters
are thought to participate in this mechanism.
Another line of evidence suggests that the levels of carbon dioxide in the
hemolymph of lepidopterans are lower than those within the midgut lumen. The
concentration of CCb has been determined to be near 5 mM in the hemolymph and 50


Several acid-base controlling proteins have now been identified within the
mosquito gut (Zhuang et al., 1999; Corena et al., 2002). The distribution of these
93
proteins along the length of the mosquito gut is both heterogeneous and discontinuous.
Along the length of the gut, non-adjacent regions such as the gastric caeca and posterior
midgut display similar protein expression and CA activity, while the region that separates
them, the anterior midgut, displays a different pattern of protein expression. This is not
surprising as the highly alkaline pH of the anterior midgut also contrasts with the nearly
neutral pH of the flanking gut regions. The most intriguing part of the novel expression
profile of these mosquito proteins is the parallel expression profile for the same proteins
within the mosquito midgut and the well-characterized mammalian kidney. As
characteristic of most epithelia, the epithelial cells found in both the mammalian kidney
and the mosquito midgut share several specific morphological features. Both populations
are mitochondria-rich, display apical microvilli, contain active cytosolic CA activity, and
express a proton translocating FC V-ATPase on specific domains of their plasma
membranes (Clements, 1992; Sterling et al., 2001a). Similarly, cell polarity proteins can
also be compared between the A (alpha) intercalated cells of the collecting tubules of the
mammalian kidney and the mosquito epithelial cells of the gastric caeca and posterior
midgut. The A cell subpopulation of kidney intercalated cells expresses, and is defined
by, an apical H+ V-ATPase, and a basolateral AE (Matsumoto et al., 1994). The
epithelial cells of the gastric caeca and posterior midgut also express an apical FT V-
ATPase (Zhuang et al., 1999), and a basolateral AE. Furthermore, B (beta) intercalated
cells of the mammalian kidney are defined by a Ff V-ATPase expressed within the
basolateral plasma membrane and an apical AE. This apical AE is different from AE1,


CHAPTER 5
ANION EXCHANGER EXPRESSED WITHIN
THE LARVAL ANOPHELES GAMBIAE MOSQUITO
Introduction
The larval mosquito gut provides an ideal model for studying epithelial transport
due to its cellular simplicity, being only one cell layer thick. The transport of bicarbonate
within the larval mosquito gut was prompted by several studies (Boudko et al., 2001a,b;
Corena et al., 2002). Bicarbonate is the main pH buffer in most complex organisms
(Sterling and Casey, 2002) so it was predicted that de-protonated bicarbonate (ie.
carbonate) is necessary for attaining the highly alkaline lumen of the larval mosquito
midgut. However, the anterior midgut was found to apparently lack an active CA
enzyme (Corena et al., 2002). An alternative to bicarbonate being rapidly produced
within the anterior midgut, is the transport of bicarbonate into the anterior midgut. This
alternative was supported by a previous study which implicated a chloride/ bicarbonate
anion exchanger within the larval mosquito gut through the use of self-referencing ion-
selective (SERIS) microelectrodes (Boudko et al., 2001b). However, the molecular
identity of the chloride/ bicarbonate anion exchanger (AE) was not determined. We now
present the first anion exchanger (AE) to be cloned and characterized from the An.
gambiae mosquito, in an attempt to unravel the physiology of an extremely alkaline
digestive system.
Localizing an AE within the larval mosquito gut is important because intracellular
pH is known to be regulated by exchangers of bicarbonate and chloride (Phillips and
83


8
intracellular carboxy terminus of AE1 was found to contain a cytosolic CAII binding site
(Vince and Reithmeier, 2000; Sterling et al., 2002a). A metabolon, a complex of
membrane proteins involved in regulation of bicarbonate metabolism and transport,
defines the relationship between the CA and AE proteins (Sterling et al., 2001a). This
bicarbonate transport metabolon, is thus capable of transporting bicarbonate as soon as it
is available from the CA enzyme. Transport can be in either direction, into or out of the
cell, and is therefore predictively capable of maintaining a tight hold on pH. The
occurrence of such a tight bicarbonate control mechanism could be very advantageous to
the mosquito. With such a large pH gradient across the membrane, a bicarbonate
transport metabolon could ensure that the pH on either side of the membrane is strictly
monitored. This bicarbonate transport metabolon has only been identified in a
mammalian system. Despite this fact, an insect gut model that employs such a
bicarbonate transport metabolon is easy to envision. Because of the strong pH gradient
that is maintained in the mosquito gut, it is reasonable to propose that a bicarbonate
transport metabolon could exist in this system as well.
Gut Alkalization Model
My first physiological model of the larval mosquito midgut was derived from the
tobacco homworm, M. sexta, which also uses an alkaline digestive strategy. In this
model, several proteins contribute to the high alkalinity (Fig. 1-3). These are the CA, the
H+ V-ATPase, and the cation and anion exchangers. The FT V-ATPase is thought to be
the energizer of the system by using ATP, and pumping protons out into the lumen of the
anterior midgut. This sets up a potential difference across the membrane of about 210 mv
(Harvey, 1992). In this model, the predicted cytosolic CA within the anterior midgut


44
bp partial sequence was used to derive exact PCR primers for a modified 3'- and 5'-
RACE (Frohman and Zhang, 1997, modified by Matz, 1999). Amplified cDNA pools
from each region of the isolated gut, facilitated the eventual cloning of a single
contiguous cDNA (Matz, 1999). The final contiguous region spanned both start and stop
codons, and encoded a polypeptide of 298 residues (GenBank accession number
AF395662). Figure 3-4A shows an alignment of the Ae. aegypti carbonic anhydrase (A-
CA) amino acid sequence with several other, previously characterized members of this
extensive a gene family. Figure 3-4B shows a homology tree depicting the percentage of
identical amino acids between sequences, generated using DNAman software. Figure 3-
5A shows the alignment between A-CA and six putative CA gene sequences from the D.
melanogaster genome that our homology search (BLAST) revealed. Four of the D.
melanogaster genes (AAF54494, AAF56666, AAF57140, AAF57141) had not
previously been annotated. Figure 3-5B shows the homology tree generated with these
sequences. A-CA has a putative molecular mass of 32.7 kDa. The translated A-CA
protein sequence possesses a characteristic eukaryotic-type CA signature sequence within
the polypeptide (amino acid residues 99-115; Femley, 1988).
To examine the possibility of regionalized expression of the A-CA, PCR using
exact primers was performed on amplified cDNA pools from the various sections of the
gut. Figure 3-6 shows an ethidium-bromide-stained agarose gel. PCR products of the
expected, 894 nucleotide length, are readily seen in the gastric caeca and the posterior
midgut regions. Anal papillae (not shown), anterior midgut, Malpighian tubules and
rectal salt gland showed little or no PCR product. When the PCR products were
subjected to a second round of PCR using the same primers, an appropriately sized


72
Aedes CA
Anoph CA
Human CA IV
Bovine CA TV
Rabbit CA IV
Murine CA TV
Rat CA IV
Aedes CA
Anoph CA
Human CA XV
Bovine CA IV
Rabbit CA IV
Murine CA IV
Rat CA IV
Aedes CA
Anoph CA
Human CA XV
Bovine CA IV
Rabbit CA IV
Murine CA XV
Rat CAIV
Aedes CA
Anoph CA
Human CA XV
Bovine CA IV
Rabbit CA IV
Murine CA XV
Rat CA IV
Aedes CA
Anoph CA
Human CA IV
Bovine CA IV
Rabbit CA IV
Murine CA XV
Rat CA IV
Aedes CA
Anoph CA
Human CA IV
Bovine CA IV
Rabbit CA XV
Murine CA IV
Rat CA IV
MIALFVATLL PSTIRADE1
MKSFTLLLCYALFVLHAARGDE1
MRMLLALLALSAARPSASAESHW
MRLLLALLVLAAAPPQAF&ASHW
MQLLFALLALGALRPLAG EELHW
MQLLLALLALAYVAPST-EDSGWi
MQLLLALLALAYVAPST-JEDSHWi
* 1
Antigenic peptide
PTPAPNfGVINEPERWGGQCETGRRSPIDLTY
¡PTPG- -TNGVMSEPERWGGQCDNGRRC£ PIDLTI
QAESSNYPCLVPVKWGGNCQKD-RC S PINIVT
IQVKPSNYTCLEPDEWEGSCQNN-RC S PVNIVT
IQA--SNYSCLGPDKWQEDCQKS-RC S PINIVT
IQTKDPRSSCLGPEKWPGACKEN-QQ S PINIVT
IQAKEPNSHCSGPEQWTGDCKKN-QQsjpiNIVT
k k k kirk t
QAAVKGDFAPFLF-SNYMNPIRNAQLTHTGHSIQIDSTDPSVTLYGGGLPGKFVLDttiHF
AAAVRGQFAPLFF-SNYMLPLKQPRVTNTGHSIQINNRDSAITMQGGGLGGRFVLD QMHF
TKAKVDKKLGRFFFSGYDK-KQTWTVQjNNGHSVMMLLEN-KASISGGGLPAPYQAKQ LHL
AKTQLDPNLGRFSFSGYNM-KHQWWQNNGHTVMVLLEN-KPSIAGGGLSTRYQAT Q LHL
TKAEVDHSLGRFHFSGYDQ-REARLVEN NGHSVMVSLGD-EISISGGGLPARYRAlj QLHL
ARTKVNPRLTPFILVGYDQ-KQQWPIK NNQHTVEMTLGG-GACIIGGDLPARYEAVQLHL
SKTKLNPSLTPFTFVGYDQ-KKKWEVKHNQHSVEMSLGE-DIYIFGGDLPTQYKAigLHL
HWG Sfi
HWG Si
HWSDLPYKGS
HWSRAMDRGS
HWSQELDRGS
HWSNGNDNGS
HHSEESNKGS
kk kkk
TIAGVRYGQBLHMVHHDSRYNS--LTEAGAVKNAVAVIGVLFHVSNQD
TLDDTRYGi]e LHLVHHDTRYAS--LEDAVQARNGVAVLGVLFHVGSQP
sldgehfaMe mhivhekekgtsrnvkeaqdpedeiavlaflveagtqv
SFDGERFAME3MHIVHEKEKGLSGNASQNQFAEDEIAVLAFMVEDG-SK
IVHQKETGTSGNEVQD--SDDSIAVLAFLVEAGPTM
SIDGRHFAMElMHIVHKKLTS SKEDSKDKFAVLAFMIEVGDKV
S IDGKHFAiMHWHKKMTTGDKVQDSDSKDKXAVLAFMVEVGNEV
CAESVIW
lACAESVlW
CDEB
CDEB
lACSEI
CDETVIW
iBTVljw
** ****** * *
NTHMDWLETSQDIRDAAGKSAPLK-GKLSPHNPLPKNRTSYFPSEGS
NMHIDTILDTATEIQNEVGKEALLR-GKLSPYNLLPSNRTSFYRyegs:
NEGFQPLVEALSNIPKPEMSTTMAE-SSLLDLLPKEEKLRHYFPiLGS:
NVNFQPLVEALSDIPRPNMNTTMKEGVSLFDLLPEEESLRHYFF V LGS:
NEGFQPLVTALSAISIPGTNTTMAP-SSLWDLLPAEEELRHYFFYMGS:
nkgfqplvealpsiskphststvre-sslqdmlppstkmytyfryngs:
NEGFQPLVEALSRLSKPBTNSTVSE-SCLQDMLPEKKKLSAYFpjYQGS:
*
TVFTESLPVSLDQVELFKT IHDPSGHELVIKfKSLQPLNARALVYHTDMDYSGs}gAI
TVFTESISVSLEQVERFKA ihdqtgrelvn netr SVQPLNTRALVYATEWDQHGNNFA
TVFREPIQLHREQILAFSQKLYYDKEQTVSMKDNVR PLQQLGQRTVIKSjGAPGRPLPHAL
TVFQKPIQLHRDQILAFSQKLFYDDQQKVNMTDN vjRPVQSLGQRQVFRS GAPGLLLAQPL
TVFQEPIRLHRDQILEFSSKLYYDQERKMNMKD N V R PLQRLGDRSVFKS QAAGQLLPLPL
TVYKQPIKIHKNQFLEFSKNLYYDEDQKLNMKDKV R PLQPLGKRQVFKSHAPGQLLSLPL
tvfeepikihkdqflefskklyydqeqklnmkdmvbplqplgnrqvfrsHASGRLLSLPL
. * .**.**.*.
PKLSLTLIVAAIAALLAK
TKMTSNWFLGAIVLLVITSRLSYH
PALLGPMLACLLAGFLR
PTLLAPVLACLTVGFLR
PTLLVPTLACVMAGLLR
PTLLVPTLTCLVANFLQ
PTLLVPTLTCLVASFLH
Figure 4-1. Alignment of several mammalian CA IV enzymes with two mosquito CA
isoforms. The leucine-rich signal sequences are found in all aligned
isoforms, along with the 3 essential zinc-binding histidines (red), and
cysteine residues (green) that form disulfide bonds. The reduced activity in
rodent CA IVs is caused by the glycine-69 mutation to glutamine (orange;
Tamai et al., 1996b), which the mosquito CAs do not display. Important
conserved residues are boxed. The position of mammalian signal sequence
cleavage is shown and the following amino acid is residue #1 in the
functional protein. Antigenic peptide sequence is also displayed.


125
A
B
Figure 6-4. Localization of CA mRNA expression within An. gambiae whole mounts.
A. A view of the entire gut region hybridized with the CA cRNA probe.
The most prominent hybridization was found in the gastric caeca and
posterior midgut. B. A higher magnification view of the gastric caeca. Both
the proximal cells and distal cap cells of the gastric caeca show CA mRNA
expression. Rostral to the gastric caeca the cardia region also displayed
intense label (arrow). C. Top view of the cardia displays consistent labeling
around the circumference of the cardia region (arrows). Scale bars represent
300 pm in A, 75 pm in B and C.


135
Although CA is a reversible enzyme, the mammalian CAIV is even faster than
CA II at bicarbonate dehydration. This ability, along with prior studies that have found a
very low concentration of bicarbonate in the hemolymph surrounding the mosquito
anterior midgut, have led to a prediction of the role of the CA IV-like enzyme in the
mosquito midgut. The high alkalinity of the AMG lumen is energized by a H+ V-
ATPase, which pumps protons from the epithelial cells into the hemolymph surrounding
the anterior midgut. Only the AMG contains a basally-oriented H+ V-ATPase, whereas
the other regions of the midgut (gastric caeca and PMG) express an apically-oriented H+
V-ATPase (Zhuang et al., 1999). The CA IV-like enzymes that are suspended in the
hemolymph are in the perfect position to provide a sink for these protons. Through the
action of the CA IV-like enzyme in the hemolymph, protons are combined with
bicarbonate to form carbon dioxide and water. Due to this coordinated effort, the H+ V-
ATPase can keep pumping protons out into the hemolymph without creating a
concentration gradient. The CA activity on the hemolymph side of the AMG therefore
allows the H+ V-ATPase to function more efficiently.
In contrast to the benefit that the CA IV-like enzyme provides to the FT V-
ATPase, the existence of CA within the epithelial cells of the AMG would actually
hinder the H+ V-ATPase. The benefits of not having a CA in the AMG are twofold, due
to the reversibility of CA. First, because there is no CA to dehydrate bicarbonate, there is
no competition with the H+ V-ATPase for protons. Secondly but most importantly,
because there is no CA to hydrate carbon dioxide to provide protons to the H+ V-ATPase,
the protons must come into the cell from the lumen to replace the protons that are being
pumped out to the hemolymph. This need for proton replacement may be the driving


LIST OF TABLES
Table page
2-1. PCR primer sequences 35
2-2. Composition of all solutions used in Xenopus oocyte expression of
An. gambiae AE 36
vm


142
Halstead, S. B. (1997). in: Dengue and dengue hemorrhagic fever, (ed. Gubler, D. J. &
Kuno, G.), CAB International, Wallingford, U.K. pp. 23-44.
Hansson, H. P. J. (1967). Histochemical demonstration of carbonic anhydrase activity.
Histochemie 11, 112-128.
Harvey, W. R. (1992). Physiology of V-ATPases. J. Exp. Biol. 172,1-17.
Haskell, J, A., Clemons, R. D. and Harvey, W. R, (1965). Active transport by the
Cecropia midgut. I. Inhibitors, stimulants and potassium-transport. J. Cell. Comp.
Physiol. 65,45-56.
Hewett-Emmett, D. (2000). Evolution and distribution of the carbonic anhydrase gene
families. Exs. 90, 29-76.
Hewett-Emmett, D. and Tashian, R. E. (1996). Functional diversity, conservation, and
convergence in the evolution of the alpha-, beta-, and gamma-carbonic anhydrase gene
families. Mol. Phylogenet. Evol. 5, 50-77.
Holt, R. A. Subramanian, G. M. Halpern, A. Sutton, G. G. Charlab, R. Nusskern, D.
R. Wincker, P. Clark, A. G. Ribeiro, J. M. Wides, R. et al. (2002). The genome
sequence of the malaria mosquito Anopheles gambiae. Science 298, 129-149.
Hogue, C. W. (1997). Cn3D: a new generation of three-dimensional molecular structure
viewer. Trends Biochem. Sci. 22, 314-316.
Hubbard, T., Barker, D., Birney, E., Cameron, G., Chen, Y., Clark, L., Cox, T.,
Cuff, J., Curwen, V., Down, T. et al. (2002). The ensembl genome database project.
Nucleic Acids Res. 30, 38-41.
Huber, S., Asan, E., Jons, T., Kerscher, C., Puschel, B. and Drenckhahn, D. (1999).
Expression of rat kidney anion exchanger 1 in type A intercalated cells in metabolic
acidosis and alkalosis. Am. J. Physiol. 277, F841-849.
Jespersen, T., Grunnet, M., Angelo, K., Klaerke, D. A. & Olesen, S. P. (2002). Dual
function vector for protein expression in both mammaliancells and Xenopus laevis
oocytes. Biotechniques 32, 536-538, 540.
Jungreis, A. M., Barron, D. N. and Johnston, J. W. (1981). Comparative properties of
tobacco homworm Manduca sexta, carbonic anhydrases. Am. J. Physiol. 241, R92-99.
Kim, J., Kim, Y. H., Cha, J. H., Tisher, C. C., and Madsen, K M. (1999). Intercalated
cell subtypes in collecting tubule and cortical collecting duct of rat and mouse. J. Amer.
Soc. Nephrol. 10, 1-12.


118
appear to be CA homologs. Further studies must be performed to determine whether any
differences in the activity or inhibitory profile of these dipteran CA proteins exists due to
this active site difference. This conserved amino acid change found in An. gambiae and
D. melanogaster, but not humans, may suggest an evolutionary importance that could be
exploited in future mosquito larvacide production. Ongoing efforts aimed at
crystallization and x-ray analyses with collaborators at the University of Florida will also
reveal the structural identity of this An. gambiae cytosolic CA.
Localization of CA Activity in Anopheles gambiae Larvae
A modified version of Hanssons CA histochemistry method (Hansson, 1967) was
used to localize CA activity within the An. gambiae larvae. Precipitated cobalt salts
marked the regions of CA activity within the larvae. The regions of CA activity include
the cardia, gastric caeca and posterior midgut (Fig. 6-3). A small subset of specific cells
within the rectum also stained positively for CA activity (Fig. 6-3E).
Localization of Cytosolic CA mRNA in Anopheles gambiae Larv ae
A DIG-labeled antisense RNA probe was utilized to localize the CA message.
The most intense labeling was consistently viewed within epithelial cells of the gastric
caeca and posterior midgut (Fig. 6-4). The localization of cytosolic CA mRNA to the
gastric caeca and posterior midgut correlates with the location of CA activity within the
An. gambiae gut, as determined by CA histochemistry (refer to Fig. 6-3). The cardia was
also labeled, as well as a subset of nerve cells and fibers that traverse the AMG
longitudinally (Fig. 6-4B,C). Within the posterior midgut, the labeling was confined to
the outer edges of a subset of large columnar cells and the small cuboidal cells (Fig.6-5).


12
st
1 instar
larva
2nd instar
larva
adult
emerge
3rd instar
larva
3rd
molt y
th
4 instar
larva
pupa
Figure 1-2. Illustration of the mosquito life cycle. The four life stages are egg, larva,
pupa, and adult. The larval stage consists of four different instars. Early
fourth instar larvae, following the third molt, were chosen for all
experiments. The female mosquito continues the cycle by laying eggs,
usually after a required blood meal.


132
functional identities. Two different CA isoforms from larval mosquitoes were partially
characterized within this study. One GPI-linked CA isoform was found in both An.
gambiae andAe. aegypti, and the other cytosolic CA isoform was found in An. gambiae.
The additional characterization of more mosquito CAs in the future, will present a great
opportunity for studying the evolution of CAs within the same a CA family.
A GPI-linked CA isoform, similar to the mammalian CA IV isoform was
localized to the gastric caeca and a specific subset of muscle fibers in the anterior midgut
region of both Ae. aegypti and An. gambiae. This isoform is different from the other a
CA isoforms (characterized in mammals) in that the entire CA protein is located
extracellularly, with only the GPI-link maintaining an association to the plasma
membrane. These GPI-linked CA isoforms from two different mosquito species also
have a direct homolog in the D. melanogaster genome database. These three dipteran CA
IV-like isoforms all display a shortened active site region. How this novel active site
affects the activity of the CA protein is unknown. However, it is known that mammalian
CA IV proteins are oriented by the GPI attachment so that the active site is directed away
from the membrane, thereby affording the greatest accessibility of substrate to the active
site. The GPI-linked CA of the anterior midgut muscle fibers would be in the prime
location and conformation for taking up substrate from the hemolymph.
The other full-length CA cDNA isolated from An. gambiae gut tissue is a
cytosolic CA isoform. The expression of mRNA was localized to the cardia, gastric
caeca and posterior midgut epithelial cells. These regions were also shown to contain an
active CA enzyme through CA histochemistry. Recombinant expression of this An.
gambiae CA protein in bacteria, produced a purified CA-active eluate, as measured by


7
characterization of mammalian CA isoforms. For example, the acidic sulfonamide
benzolamide has been used for the preferential inhibition of extracellular CA while not
compromising any intracellular CA activity (Tong et al., 2000). This occurs due to the
inability of benzolamide to readily penetrate cell membranes (Tong et al., 2000). The
wealth of information pertaining to mammalian CA isoforms and their specific inhibitors
provides a basis for comparisons with CAs that are discovered in the mosquito midgut.
Sulfonamide CA inhibitors are widely used to treat a number of conditions including
glaucoma, gastro-duodenal ulcers, and cancer, by lowering the production of fluids and
acids. Parkkila et al. (2000) showed that the invasion of renal cancer cells in vitro could
be inhibited with CA inhibitors. If larval mosquito physiology is dependent upon the
generation or maintenance of the alkaline gut, and CA is a necessary component, then the
possibility exists for the use of CA inhibitors as mosquito larvacides.
Bicarbonate Transport
The site(s) of bicarbonate production by CA may not be as important as the
translocation of the bicarbonate that is produced. Transporters can facilitate the passage
of bicarbonate and other ions through otherwise impermeable cell membranes.
Bicarbonate transporters compose a large family of membrane proteins that includes the
anion exchangers (AEs), sodium bicarbonate cotransporters (NBCs), and members of the
sulfate transporter group that can also transport bicarbonate (Alper et al., 2001). Most of
the BT proteins consist of a cytosolic anchoring domain as well as a 10-14 membrane-
spanning transporter domain (Alper et al., 2001). Also, evidence exists that some AEs
are capable of physically binding CA enzymes. Thus, the fourth extracellular loop of
AE1 contains a glycosyl-phosphatidyl-inositol (GPI)-linked CA IV binding site and the


130
0 1234 56 24
Figure 6-9. Protein gels and western blots of recombinantly expressed CA protein. A.
Brilliant blue staining of CA protein induction (33 kDa) from 0 to 24 hours.
B. Fast green staining of recombinant CA protein expression (33 kDa). C.
The XPRESS (XP) antibody detected a 33 kDa band. Pre-immune sera
from two chickens inoculated with a conjugated CA peptide (PI) did not
detect a band at 33 kDa. The antisera collected from one of the inoculated
chickens, detects a visible protein band at the expected 33 kDa (94; red
arrow). The antisera collected from the other chicken (93), does not
recognize the 33 kDa protein band on the western blot. The lane containing
the molecular weight standard is marked with an M.


90
The AgAEl was determined to be a functional protein with the capacity to transport
chloride. No sodium ion or potassium ion dependence was determined with the voltage
clamp assay. A comparison of current versus voltage (I-V plots) for both the AgAEl
expressing oocytes and the water injected controls are compared when two different bath
solutions are applied to the oocytes. The I-V plots for the water-injected controls
displayed no transport with or without chloride (Fig. 5-18A). When the solution contains
chloride (N98), the AgAEl expressing oocytes are capable of transporting chloride, as
seen by the steep rise in current (Fig. 5-18B). When chloride is replaced by a non-ionic
equivalent (N98-C1) there is no transport (Fig. 5-18B). The transporter blockers, 4,4-
diisothiocyanodihydrostilbene-2,2 -disulfonate (DIDS) and niflumic acid (NA) both
inhibited the transport capabilities of the expressed AE1 protein (NA not shown). The
application of DIDS inhibited the transport of chloride such that the I-V plot showed an
affect similar to the removal of chloride ions from the bath solution (Fig. 5-19A). The
likeness of removing chloride and the inhibitory affect of DEDS is easily viewed by
comparing the differences in current between the DIDS inhibition and the removal of
chloride (Fig. 5-19B). When the difference between chloride and chloride removal are
compared to the blocked and chloride removal transport, a large difference can be seen
(Fig. 5-19B). Activity of the mosquito AE1 was also inhibited when the CA-specific
inhibitor acetazolamide was added to the media (data not shown). Because
acetazolamide is known to have no direct inhibitory effect on AEs, unlike other
sulfonamides, it can be inferred that the AgAEl is inhibited by acetazolamide due to its
tight coordination and regulation by endogenous CA, as was found to be the case in
mammalian systems (Sterling et al., 2001a). Endogenous CA was bound by mammalian


81
Figure 4-10. High magnification of immunoreactive muscle fibers within the Aedes
albopictus midgut. A. Labeling of muscle fibers appears to be
extracellular. B. Cross section of the same fiber demonstrates that the
localization pattern is confined to the extracellular plasma membrane of
the midgut muscle fibers. The scale bars represent 25 pm.


50
levels of activity in the anterior midgut might be too low to be detected using the 0
method, but that the enzyme is present throughout the entire length of the midgut.
In summary, our evidence demonstrates the existence of CA in Ae. aegypti larvae
and it also suggests that the gastric caeca and posterior midgut exhibit the highest levels
of CA activity. In addition, the enzyme seems to be associated with the small cuboidal
cells of the midgut epithelium. Furthermore, enzyme activity has also been detected in
membrane preparations isolated from whole midguts and could be due to the presence of
more than one isoenzyme. Carbonic anhydrase activity has previously been
demonstrated in the epithelium of the larval midgut of six species of lepidopterans, in
which it has been associated with the particulate fractions of the homogenate (Turbeck
and Foder, 1970). This is consistent with our hypothesis that there might be more than
one CA and that one of these enzymes may be associated with the plasma membrane.
What is the role of CA in the alkalization mechanism? BTB proved useful in
monitoring the impact of CA inhibition on the maintenance of gut luminal pH and the
excretion of acid/base. As mentioned earlier, Ae. aegypti larvae typically alkalize the
medium in which they are reared by secreting bicarbonate ions (Stobbart, 1971). The
ingestion of CA inhibitors altered the metabolism of the larvae to the point that the
metabolic products secreted into the medium change the pH of the environment, shifting
it towards more acidic values than those observed in the absence of inhibitors. The
lowering of the pH of the medium might be related to a decrease in the rate of secretion
of HCO3'. The effect of the ingestion of CA inhibitors on the secretion of bicarbonate
into the medium remains to be explored. However, as indicated by measurements with
ion-selective microelectrodes, inhibition of CA in the midgut has an extreme effect on the


92
In Xenopus oocyte expression tests of AgAEl, activity was decreased with a CA-
specific inhibitor, acetazolamide. Bicarbonate, rapidly formed by the hydration of carbon
dioxide by CA, is a substrate for the AE. The decreased ability to exchange ions in the
presence of acetazolamide leads to the prediction that this mosquito AE is directly
regulated by CA activity. There is evidence within the mammalian system for the
tethered coordination of anion exchangers with carbonic anhydrase. The mammalian
anion exchanger (AE1/ band3) has been shown to interact with and actually bind to CA II
at its carboxy terminus (Sterling et al., 2001b). AEs and in fact all bicarbonate
transporters (except DRA) identified to date have potential CA E-binding sites at their
carboxy termini (Sterling et al., 2002b) including this An. gambiae AE. The inhibitory
effect of acetazolamide on AgAEl expressed in Xenopus oocytes suggests that this
mosquito AE is coupled with an active CA enzyme, speculatively an endogenous
Xenopus CA protein. A protein complex consisting of a membrane-spanning AE and a
cytoplasmic CA has the ability to maintain tight pH homeostasis both inside and outside
of the cell at the same time. Furthermore, this complex brings together bicarbonate
production and transport in such a way that virtually all lag time is abolished by the
tethered coordination of the system (Sterling et al., 2001b). This type of bicarbonate
transport metabolon, if found to exist within the mosquito, may explain how the mosquito
gut is capable of driving and supporting a pH greater than 10 within the lumen while
sustaining a near neutral pH within the adjacent cells. Now that an AE has been localized
to CA active regions within the mosquito gut, namely the gastric caeca and posterior
midgut, it will be necessary to determine whether they form a bicarbonate metabolon as
proposed.


89
Antibody Localization of AE Protein
Two antigenic peptide sequences (EVRKRPPEKNPKEEIDEE and
KPKQQPVTTISVTKVAEQ) were chosen from the cytosolic framework of the amino
terminus and the anion exchange carboxy terminus respectively, of the translated AE
cDNA (accession number AAQ21364). Chickens were used to produce antibodies
against these antigenic peptides. The resultant chicken antisera were used to localize the
AgAEl protein within whole mounts of the An. gambiae fourth instar larvae. The
AgAEl protein was localized to the plasma membranes of the gastric caeca (Fig. 5-15)
and posterior midgut epithelial cells (Fig. 5-16). A three-dimensional reconstruction of
the cellular localization of AgAEl enabled us to discriminate the immunolabeled
basolateral membranes from the non-labeled apical membranes. Antisera for both
peptides displayed the same basolateral immunolocalization pattern. Neuronal cells
within the AMG displayed immunoreactivity (Fig. 5-17). Pre-immune sera displayed no
specific immunoreactivity.
AE Functional Expression in Oocytes
In order to ascertain the functional characteristics of the AgAEl, the protein was
expressed inXenopus oocytes. The AgAEl was subcloned into the pXOOM vector
(Jespersen et al., 2002; generous gift from Dr. T. Jespersen) for oocyte expression. The
Xenopus oocytes were injected with either AgAEl RNA or water to serve as the control.
Three to seven days post-injection the oocytes were tested for AgAEl expression using
two-electrode voltage clamp electrophysiology. Oocytes expressing AgAEl displayed a
decreased volume as compared to the water-injected controls (T. Sern, unpublished
observation), which correlates with the AE regulatory functions of cell pH and volume.


67
Antibody Cross-Reactivity with Other Mosquito Species
The 18 amino acid sequence from the Ae. aegypti CA, used to elicit the antibody
response, shares 14 identical residues with the homologous An. gambiae CA protein
(refer to Fig. 4-1), and therefore, the antiserum recognizes the CA IV-like isoform present
in An. gambiae as well. The immunoreactivity within the An. gambiae gut displays a
strikingly similar, yet species-distinctive pattern of CA IV-like protein expression (Fig. 4-
7B). Similar to Ae. aegypti (Fig. 4-7A), not all of the muscle fibers were localized in the
An. gambiae gut. However, in An. gambiae the immunolabeled muscle fiber network
runs down the lateral sides of the midgut, while the dorsal and ventral muscle fibers are
not immunoreactive (compare Fig. 4-7A to 4-7B).
The high sequence conservation between the chosen antigenic peptide from Ae.
aegypti and the An. gambiae CA, prompted us to check other mosquito species for
immunoreactivity. The other species tested also displayed the same strong labeling on a
subset of anterior gut muscle fibers that included both circular and longitudinal muscle
fibers. Figures 4-9 and 4-10 display the immunoreactive results obtained from Aedes
albopictus and are representative of the results from the other mosquito species including
Ae. aegypti and An. gambiae. The CA IV-like immunolabeling is clearly confined to
only a subset of actin-containing muscle fibers (Fig. 4-9D). Labeled phalloidin, which
binds to actin, labeled all of the muscle fibers within the gut, while the antibody for CA
IV-like mosquito CA only recognized a subset of the anterior muscle fibers. The CA
antibody is also specific for the extracellular plasma membrane of these muscle cells,
which is clearly shown by cross-sectional analysis (Fig. 4-10). Thus, there appear to be
two different sets of muscle fibers within the anterior mosquito midgut is an intriguing


57
Ajadae aagypti MIALFVATPSTIKAD EWHY PT PAPNGVINE PERWGGQCE TS
Oros AAF57140
Droa AXP57141 .
Droa AAF56666 MSEIATGKSCT1AVBSHVFGY8EPNQRRWARHHGHCAGKTQS PIAITTfR
Droa AAF544 94 MPLRHSVGIQ8VKLMIIANEWGYPDLDNNQDEPFPK WGGLCDM
Droa AAF4 99418 .MRRCRMTPFAIVIAPILICABLVLAQDFGYEGRHGPEHW8EDYARC8G
Droa AAF44817 MSHHWGYTEENGPAHWAKEYPQAlBG
43
O
O
49
43
48
25
Ajeadas sagypti
Daros AAF57140
Daros AAF57141
Droa AAF56666
Daros AAjr54494
Daros AAJP4 994 8
Droa AAF44817
RRQMIDLTYOAKVKGDFAPFLtSlKMNPIRHAQLTNTGH . 8IQID8T
MRQLIVPLPRIVFGHYDVKLRGPLTLLIING . . HTETAN
. .1 TTAltaffAVDMIGYHNLLPYPLKMINHGHTVSITIPKVN VTEVGE
KKOPXKI-HVKGlU^KGEFnAI*K3rE|Jnn3EHQKNI-RMVNWGH . SIQLSGF
KHQlPINIDQVSAVEKKFPKLErFlIFKWPDNLQMTNIIGHTVLVKMSYNE
HRQ#PVDITP3SAKKGSE1*NVAPUCWKYVPEHTK31jVHPG . YCHRVDVN
9G
37
43
94
90
90
73
Aedeo aegypti
Daros AAF57140
Daros AAF57141
Droa AAF56666
Daros AXP54494
Daros AAP4 9948
Daros AAF44817
130
83
89
140
130
148
120
Aadtts aegypti
Droa AAF57140
Droa AAF57141
Daros AXF56666
Daros AAF54494
Daros AXF4 9948
Daros AAF44817
ICWLFHV8NQ. DNTHMDWLET8QD
JONP .NRIFPGLSKVMSALP
rAIVRKDNAKSTPtSRLMEAW
IFFFNLDE . DEGAGtVTINRHtH
. TPNEAIQSIIKSLGA.
GD . K8TG6YEGFTNLK8
....HHAEtDKVTSLLQ
177
130
138
180
177
194
166
Aedaa a9ypti
Droa AAF57140
Daros AAP57141
Daros AAF56666
Daros AAF54494
Daros AAF49948
Daros AAF44817
IRDAACKSAWJCCKUIPHNPLP KNRTS1
RVTKYNAKTI PGGLM^QHLGNVNpRDF
BVPIEDSNAjpVFGQSltDQLIGGVSHRDF
LIADANQEAtLNVTFK, S 8L lAGVDVD*
VKSTOSMN1CPVLVADSIAVX)DLVP8|IEI
QIDRKGKSVMMTNPLP1/3EYIS 1
FVLHKGDRVfLPQGOJPGQLLP. .1
226
180
188
236
227
243
214
Asiaa a*gyp ti
Daros AXF57140
Droa AAF57141
Droa AAF56666
Dros AXF54494
Dros AAF49948
Dros AAF44817
275
229
235
285
276
291
264
Aedaa a5YPti
Dros AAFS7140
Dros AAF57141
Dros AAF56666
Dros AAF54494
Dros AAF49948
Dros AAF44817
G8GAIPKLSLTLIVAAIAALIAK
GAYLGK
TSIATLKHEGEYLKYDWFY
SG3AGLTASVSLGI>1TLIIAGQKFLL
HNMGS X PLVDAEHAAGKWRAQAAAVLLPLWLAAL3RTSIPRGF
REIGGH
298
235
235
304
302
335
270
B
Homology (%)
100 80 60 40 20
I I I I I
Figure 3-5. Comparison of the extrapolated amino acid sequences of A-CA with six
putative dipteran CA genes identified in the D. melanogaster gene
databases. (A) An alignment of A-CA with the amino acid sequences of
the six D. melanogaster genes (accession numbers listed) identified
through bioinformatics searching. Regions of exact homology across all
species are highlighted in blue (100%); regions with less homology are
highlighted in red (>75%) and green (>50%). (B) A homology tree
comparison of these seven Dipteran CAs.


134
To further regulate ion exchange, the AE has a particular region of amino acids with
similarity to mammalian AEs 2 and 3. This region was found to confer pH sensitivity
through intracellular alkalization. The ability of a mosquito AE to detect intracellular
alkalization may be the underlying mechanism through which the alkalinity remains
confined to the anterior midgut region. The AE- and CA-containing regions (gastric
caeca and posterior midgut) flank the anterior midgut and could possibly modulate their
transport rates in response to the encroaching or retreating alkaline pH. This bicarbonate
metabolon could have the ability to maintain a large pH gradient as is displayed in the
larval mosquito gut. Discovering the proteins involved in the production and
maintenance of such an alkaline pH could define a fundamental metabolon that is critical
for pH homeostasis.
New Model
The cloning and localization of CAs and an AE within the larval mosquito gut has
uncovered an unpredicted model of physiology. Our original model, based on Manduca
sexta (refer to Fig. 1-3), has evolved considerably due to our new findings within larval
mosquitoes. The most unexpected finding, was the failure to detect a CA within the
anterior midgut. Carbonic anhydrase histochemistry, l80 isotope exchange, in situ
hybridization, immunohistochemistry, and real time PCR all failed to give evidence for a
CA within the AMG region. Although these data did not support our original model of
anterior midgut alkalization, our new model describes a system in which CA is not
necessaiy within the AMG epithelial cells. Furthermore, our new data, including the
localization of AE within the mosquito gut, has provided insight as to why our new
model of AMG alkalization is more efficient in the absence of CA.


This dissertation was submitted to the Graduate Faculty of the College of
Agricultural and Life Sciences and to the Graduate School and was accepted as partial
fulfillment of the requirements for the degree of Doctor nf Philosophy
El '
May 2004
Dean, Collegof Agricul
Sciences
Dean, Graduate School


69
while the head sample contained roughly half of the gastric caeca concentration. CA
immunoreactivity was most striking on specific muscle fibers of the anterior midgut,
along with labeling of the gastric caeca and CNS ganglia. The localization of CA protein
in a distinct subset of muscle fibers was found in several different mosquito species. This
finding suggests that many mosquito species express a similar CA IV-like protein as well
as confirms the immunoreactivity by only a subset of muscles. Immunolocalization of
the CA IV-like isozyme within the mosquito gut and CNS also demonstrates that the CA
message is being translated into protein and agrees with the localization of CA mRNA in
the gastric caeca and muscle fibers. Interestingly, not all of the muscle fibers that show
CA mRNA expression also show CA protein. Therefore, only a fraction of the muscle
fibers that contain the CA mRNA are translating the message into protein. This may
represent a form of regulation, in which the muscle fibers not expressing the CA protein
could be turned on to translate the CA message if needed. This ability would be very
advantageous if indeed this particular CA isoform is involved in buffering the alkaline
gut.
The posterior midgut was also found by in situ hybridization to express the CA
mRNA. However, both the real time PCR analysis of CA mRNA expression and the
immunolocalization of CA protein failed to determine the presence of this particular CA
within the posterior midgut. One explanation for this may be the inability of our CA
antibody to permeate the posterior midgut tissue. However, the real time analysis, which
is very specific for a region of mRNA, also did not find this particular CA message in
that region. The most likely cause for the in situ hybridization showing a positive CA
result in the posterior midgut, is the existence of a very similar CA isoform within that


86
The amino terminus of the AgAEl protein contains 523 cytoplasmic residues
enriched with multiple binding sites for cytoskeletal proteins (Bairoch et al., 1997). The
carboxy terminus contains the membrane-spanning domains that are responsible for ion
transport. According to the PROSITE motif search and the hmmtop server, sites of
potential post-translational modification include 2 cAMP and cGMP-dependent protein
kinase phosphorylation sites, 16 protein kinase C phosphorylation sites, 11 casein kinase
II phosphorylation sites, 16 N-myristoylation sites, 1 prokaryotic membrane lipoprotein
lipid attachment site, and 1 leucine zipper pattern (Bairoch et al., 1997; Gupta et al.,
2002). The last 82 amino acids of the An. gambiae AE carboxy terminus are predicted to
project into the cytosol, the correct orientation that is expected for binding of a cytosolic
CA. The CAII binding site comprises a hydrophobic amino acid residue followed by at
least two acidic residues within the next four residues (Vince and Reithmeier, 2000).
Figure 5-2 displays the CA II binding sites found in several AE proteins along with the
putative CA II binding site (LDDIM) in the An. gambiae AE.
BT Sequence Comparisons
Following the sequence prediction that this An. gambiae BT is an AE, the closest
characterized protein sequence is the NDAE1 (accession number AAF98636) from D.
melanogaster. The An. gambiae AE1 shares 72% identity with NDAE1 (Fig. 5-3).
However, the greatest similarity is with an uncharacterized splice form of NDAE1
(AAF52497) that contains an inserted sequence that is also found in the An. gambiae AE
sequence (Fig. 5-4). Other BTs such as the sodium bicarbonate cotransporters (NBCs)
and AE4 show 45%-52% identity to AgAEl, AE1-3 exhibits 36% identity, and the BTs


120
present within the purified, recombinant CA fractions (data not shown). Activity was
partially inhibited by the application of methazolamide, a CA-specific inhibitor (data not
shown). These analyses confirmed that this cytosolic CA, cloned from an An. gambiae
gastric caeca cDNA library, has CA activity and therefore contributes to the CA activity
in the gastric caeca and posterior midgut regions, as determined by CA histochemistry.
Discussion
The CA isoform, cloned from the An. gambiae midgut, is a predicted cytosolic
protein that is expressed in the cardia, gastric caeca and posterior midgut regions.
Carbonic anhydrase activity was localized to these same regions through CA
histochemistry. The purified recombinant CA was shown to have CA activity through
180 isotope exchange experiments. This CA isoform is therefore responsible, at least in
part, for the CA activity displayed within the gastric caeca and posterior midgut regions
of the larval mosquito gut.
The one amino acid difference (C instead of S) within the active site of this
mosquito CA is not found in any mammalian isoform. The D. melanogaster genome
however displays an identical (C instead of S) difference in one of its putative CA
isoforms. The S residue at this position is present in every mammalian CA and therefore
may represent an evolutionary divergence within the a CA family. The phylogenetic
analysis of mammalian and dipteran CAs shows a more distant relationship between the
Dipteran and mammalian proteins than within the mammalian CA family. The dipteran
CA isoforms do not cluster with the mammalian CAs despite common functional
characteristics and sequence homology in all known a CAs. Instead, the mosquito CA
sequences cluster together, apart from the mammalian CA clusters.



PAGE 1

CARBONIC ANHYD RASES AND BICARBONATE TRANSPORT IN LARVAL MOSQUITOES By THERESA J. SERON A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by Theresa J. Seron

PAGE 3

DEDICATION I wish to dedicate this dissertation to my incredible family, Mom, Dad, Grandparents, Tracey, and George. My family has witnessed my struggles and triumphs and has helped me through it all. I am so proud to call them my family. I cannot thank them enough for all of their support. This achievement is really a reflection of all of us. I also want to dedicate this milestone to my soon to be husband, Dr. Peter Lovell. Coming home to his love, humor, and music, has given me true joy. My family would be incomplete if I did not mention our furry companions, Frodo and Princess, who remind us that a nap can solve most problems. iii

PAGE 4

ACKNOWLEDGEMENTS I would like to acknowledge my dissertation committee, Dr. Paul J. Linser, Dr. Edward J. Phlips, Dr. Leonid Moroz, Dr. Robert Greenberg, and Dr. Shirley Baker, for their suggestions and comments on the final rewriting of this document. I want to thank my project supervisor, Dr. Paul J. Linser, for allowing me to form my own project goals and the space to tackle them. There are a number of people at The Whitney Laboratory who I would like to thank for their assistance with this dissertation project as well as their friendship. Dr. Judith Ochrietor devoted her time and energy to improving every aspect of this dissertation. Judy assisted me with experimental designs, introduced me to real time PCR, provided a wealth of knowledge about molecular biology, and was a great person with which to share a laboratory and office. Dr. Andrea Kohn provided molecular biology teaching and advice along with being a fantastic person to work with and be inspired by. Leslie vanEkeris taught me how to do mosquito dissections and provided many of the mosquito guts that I photographed for this document. Dr. Bill Harvey provided insight into the ionic transport mechanisms of the mosquito and the editing of this manuscript. Dr. Dmitri Boudko was instrumental in the expression of the anion exchanger and the production of amplified cDNA libraries. Jessica Roberts-Misterly and Dr. Robert Greenberg also provided teachings and suggestions in cloning cDNAs from the mosquito. iv

PAGE 5

TABLE OF CONTENTS page DEDICATION m ACKNOWLEDGEMENTS iv LIST OF TABLES wm LIST OF FIGURES k ABSTRACT xii CHAPTER 1 INTRODUCTION 1 Alkaline Gut 2 Carbonic Anhydrase 4 Mosquito Development and Control 5 Carbonic Anhydrase Inhibition 6 Bicarbonate Transport ' Gut Alkalization Model 8 Specific Aims ' 2 MATERIALS AND METHODS I 4 Experimental Insects 14 Preparation and Fixation of Tissue 15 Bromothymol Blue Qualitative Assay 16 Effect of Methazolamide on the Alkalization of the Midgut of Live Larvae 16 ,8 0 Exchange Method to Measure Carbonic Anhydrase Activity 17 Isolation of RNA and Synthesis of cDNA 18 Bioinformatics 1 8 Cloning of CA from Aedes aegypti Larval Midgut 19 Construction of Amplified cDNA Pools 20 3' and 5' Rapid Amplification of cDNA Ends and Sequencing 23 Construction of In Situ Hybridization Probes 23 In Situ Hybridization 26 CA Histochemistry 27 Real Time PCR 27 v

PAGE 6

29 Antibody Production 3Q Immunohistocheniistry ^ CA Protein Expression Anion Exchanger Oocyte Expression ^ Anion Exchanger Physiology 3 CARBONIC ANHYDRASE IN THE MIDGUT OF LARVAL AEDES AEGYPTI: CLONING, LOCALIZATION, AND INHIBITION 39 39 Introduction Results .~ Bromothymol Blue Qualitative Assay ™ Carbonic Anhydrase Activity and Alkalization 42 l8 0 Isotope-Exchange Experiments 43 Cloning of Carbonic Anhydrase from Aedes aegypti Larvae 43 Localization of the Enzyme in the Midgut Epithelium: Carbonic Anhydrase Enzyme Histochemistry 45 In Situ Hybridization • 46 Discussion 4 A GPI-LINKED CARBONIC ANHYDRASE EXPRESSED IN THE LARVAL MOSQUITO MIDGUT 61 Introduction * Results Bioinformatics of Aedes aegypti CA b JSequence Comparisons of CA TV-like Isoforms 62 Localization of CA TV-like Isoform in the Mosquito Midgut 65 Real Time PCR Analysis of Aedes aegypti CA TV-like Transcripts 65 Immunolocalization of CA TV-like Protein in the Mosquito Gut 66 Antibody Cross-Reactivity with Other Mosquito Species 67 Phospholipase C Treatment 68 Discussion ^ 8 5 ANION EXCHANGER EXPRESSED WITHIN THE LARVAL ANOPHELES GAMBIAE MOSQUITO 83 Introduction ° J Results 84 An. gambiae AE Sequence Analysis 84 BT Sequence Comparisons 86 Localization of Anion Exchanger mRNA in An. gambiae Larvae 87 Antibody Localization of AE Protein 89 AE Functional Expression in Oocytes 89 Discussion 9 ^ vi

PAGE 7

6 C YTOSOLIC CA EXPRESSION IN LARVAL ANOPHELES GAMBIAE 115 Introduction Results 116 Anopheles gambiae CA Sequence Analysis 116 Localization of CA Activity in Anopheles gambiae Larvae 118 Localization of Cytosolic CA mRNA in Anopheles gambiae Larvae 1 1 8 Antibody Localization of CA Protein 119 Bacterial Expression and Purification of Anopheles gambiae Cytosolic CA n9 • 1 90 Discussion 7 CONCLUSIONS AND FUTURE DIRECTIONS 1 3 1 Conclusions 131 New Model 134 Future Directions 13' REFERENCES 140 BIOGRAPHICAL SKETCH I 47 vii

PAGE 8

LIST OF TABLES Table 2-1. PCR primer sequences 2-2. Composition of all solutions used in Xenopus oocyte expression of An. gambiae AE viii

PAGE 9

LIST OF FIGURES Figure E&ge 1 1 . Illustration showing the regions of the larval mosquito gut 11 1 -2. Illustration of the mosquito life cycle 1 2 1 -3 . Preliminary mosquito anterior midgut model based on M. Sexta 1 3 21 . Efficiency plots for real-time PCR primers 37 22. Three-dimensional (Cn3D) depiction of human CA P/ (1ZNC) 38 31 . Effect of CA inhibition on culture medium pH with fourth-instar Ae. aegypti. larvae 53 3-2. Effect of methazolamide on the alkalization of the midgut using Bromothymol Blue (BTB) assay of pH within living, but isolated, gut tissue 54 3-3 . Relative activity of CA in different pooled segments of the midgut of larval Ae. aegypti ^5 3-4. Carbonic anhydrase from the midgut of larval Ae. aegypti 56 3-5. Comparison of the extrapolated amino acid sequences of A-CA with six putative dipteran CA genes identified in the D. melanogaster gene databases 57 3-6. Polymerase chain reaction (PCR) analysis of Ae. aegypti amplified cDNA from different gut regions 58 3-7. Hansson' s histochemistry of whole mount Ae. aegypti gut 59 38. Localization of CA mRNA expression in larval Ae. aegypti 60 41 . Alignment of several mammalian CA IV enzymes with two mosquito CA isoforms 72 4-2. Clustal alignment of CA protein sequences 73 ix

PAGE 10

4-3 Localization of CA mRNA in a whole mount preparation of early 4 th 74 instar^e. aegypti 4-4. Expression of CA mRNA in Ae. aegypti anterior midgut 75 4-5 . Localization of CA IV-like message within Ae. aegypti CNS tissue 76 4-6. Relative quantification of CA IV-like message in Ae. aegypti larvae using real time PCR ^ 4-7. Ae. aegypti and An. gambiae CA protein labeling 78 4-8. The Ae. aegypti CNS ganglia express the CA IV-like isoform 79 4-9. Immunolocalization of mosquito CA IV-like enzyme in Aedes albopictus 80 4-10. High magnification of immunoreactive muscle fibers within the Aedes albopictus midgut 81 411. Immunoreactivity of Ae. aegypti guts for the CA IV-like isozyme 82 51 . Structural prediction of the An. gambiae AE1 96 5-2. Putative amino terminus CA II binding motif 97 5-3. Homology tree depicting the amino acid identity between several BTs 98 5-4. Alignment of carboxy terminus amino acids of An. gambiae and D. melanogaster AEs 99 5-5 . Alignment of An. gambiae and human AEs 100 5-6. Localization of AgAEl mRNA within whole mount An. gambiae larvae 101 5-7. Localization of AgAEl mRNA in muscle, nerve, and trachea in An. gambiae 102 5-8. In situ hybridization of AgAEl in whole mount An. gambiae consistently shows positive labeling of tracheal fibers along the midgut 103 5-9. Anion exchanger mRNA localization reveals trachea and nerve fibers along with neuronal cell labeling 104 5-10. Localization of AgAEl mRNA to the PMG of larval An. gambiae 105 x

PAGE 11

5-11. Larval An. gambiae displays strong AgAE 1 expression in the hindgut, the pylorus 106 5-12. Localization of AE mRNA in An. gambiae shows abundant labeling of the Malpighian tubules 107 5-13. Expression of AE mRNA was found throughout the ventral midgut ganglia 108 5-14. Sense AE probes display no specific hybridization 109 5-15. Antibody localization of AgAEl protein to the gastric caeca in An. gambiae larvae 1^ 5-16. Localization of AgAEl protein within the PMG of An. gambiae larvae 1 1 1 5-17. Neuronal cells within the AMG display immunoreactivity for our An. gambiae AE specific antibody 1L2 5-18. Current-voltage (I-V) plots depicting ion transport by the AgAEl expressing oocytes in contrast to the water injected control oocytes 113 519. Inhibition of AgAEl mediated chloride transport by DIDS 1 14 61 . Clustal alignment of active sites within An. gambiae, D. melanogaster, and human CA proteins 122 6-2. Phylogenetic analysis between mammalian (human and mouse) and dipteran {An. gambiae and D. melanogaster) CAs 123 6-3. Localization of An. gambiae CA activity 1 24 6-4. Localization of CA mRNA expression within An. gambiae whole mounts 1 25 6-5. Localization of CA mRNA expression within the posterior midgut of An. gambiae 126 6-6. Localization of CA mRNA expression within the hindgut 1 27 6-7. Localization of CA protein within gastric caeca of An. gambiae larvae 128 6-8. Localization of CA protein within the PMG of An. gambiae 129 69. Protein gels and western blots of recombinantly expressed CA protein 130 71 . New larval mosquito model 139 xi

PAGE 12

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy CARBONIC ANHYDRASES AND BICARBONATE TRANSPORT IN LARVAL MOSQUITOES By Theresa J. Seron May 2004 Chair: Edward J. Phlips Cochair: Paul J. Linser Major Department: Fisheries and Aquatic Sciences Carbonic anhydrase (CA) is an important enzyme due to its involvement in many pH-dependent, physiological processes. CA reversibly converts C0 2 and H 2 0 into bicarbonate and a proton. The anterior midgut lumen of the larval mosquito has an extremely alkaline pH; therefore, we hypothesized that an active CA within the epithelial cells surrounding this region would rapidly produce bicarbonate to buffer the high pH. A cDNA cloning strategy, followed by in situ hybridization, was employed to isolate and localize CA and anion exchanger (AE) transcripts within the mosquito gut. Localization of CA enzymatic activity was assessed via histochemical analyses. Enzymatic and electrophysiological analyses of recombinant CAs and AE were also performed. In this dissertation, cDNAs encoding three CA genes and an AE were cloned and localized within the larval mosquito gut. One isoform of mosquito CA, which was cloned from two different mosquito species, was localized to a specific subset of muscle fibers on the xii

PAGE 13

basal side of the anterior midgut. This CA resembles the mammalian CA IV isozyme in that a glycosylphosphatidylinositol (GPI)-link tethers the enzyme to the extracellular membrane. The other CA isoform, an active cytosolic enzyme, was localized to the gastric caeca and posterior midgut regions. It was also determined that the AE transports chloride and is expressed in the gastric caeca, posterior midgut, and Malpighian tubules. We were unable to detect CA within the anterior midgut epithelial cells using a variety of assays. My studies have therefore led to an alternative hypothesis that one or more CAs within the mosquito gut, but located outside of the anterior midgut epithelial cells, contribute to buffering the alkaline pH of 1 1 within the anterior midgut lumen. The localization of two CA isoforms, one extracellular and the other cytosolic, and an AE possessing a putative CA binding sequence, to the regions flanking the anterior midgut, supports the prediction of a bicarbonate transport metabolon within the gastric caeca and posterior midgut regions. Such a metabolon has only been studied in mammals, however, the colocalization of CA and AE within the mosquito gut suggests a similar network of bicarbonate production and transport. xiii

PAGE 14

CHAPTER 1 INTRODUCTION Insects represent one of the most numerous and diverse groups of animals on the planet. One particularly successful group of insects is the well-studied Pterogota (winged insect) group. This grouping includes the Lepidoptera (butterflies and moths) as well as the Diptera (flies). These insects have been extensively studied due to their huge impact on the lives of humans. For example, the Lepidopteran, Manduca sexta, is a great pest to tobacco companies that rely on abundant and healthy tobacco crops, on which M. sexta feeds. Also, mosquitoes (Dipterans) are responsible for transmitting a host of diseases to humans as well as other mammals by injecting pathogens, along with the anti-coagulants from their salivary glands, to aid in bloodletting. The pathogens that cause these diseases can be viruses or various parasites (eg. protozoans). Mosquitoes belong to the order Diptera, family Culicidae. According to the American Mosquito Control Association, there are more than 2500 different species throughout the world, with 150 species in the United States (Darsie and Morris, 2000; Spielman and D Antonio, 2001). Mosquitoes act as vectors for a wide variety of diseases such as malaria, yellow fever, west nile virus, and dengue fever. Recent reports estimate that fifty to one hundred million cases of dengue fever occur annually, along with several hundred thousand cases of the life-threatening form of the disease, dengue hemorrhagic fever (DHF; Halstead, 1997). The geographic range of dengue fever has expanded over the last two decades, primarily because of the spread of its principal vector, Aedes aegypti 1

PAGE 15

2 (Gubler, 1997). Another mosquito example, Anopheles gambiae kills millions of people each year in Africa by infecting them with the deadly Plasmodium parasite that causes malaria. Many studies have therefore been undertaken to understand the life cycle and physiology of these insects that exert such a large socio-economic impact. The mosquito's ability to acquire, harbor and transmit deadly pathogens has spurred research into the workings of the mosquito gut. Specific cells of the midgut, which express a proton pumping V-ATPase, have been found to be preferentially invaded by pathogens (Shahabuddin and Pimenta, 1998). Studies have also shown that the mosquito gut is not a static organ but is comprised of several different regions. Each region displays different characteristics and is made up of different cell types. Alkaline Gut Larval mosquitoes, as well as some caterpillars, are known to possess a highly alkaline digestive system (Dadd, 1975). The tobacco hornworm, M. sexta, has a gut lumen pH that can exceed 11, while the larval mosquito, Aedes aegypti, displays a pH greater than 10 in its anterior midgut region (Zhuang et al., 1999). These insects are not only unharmed by this caustic pH, but are able to generate this property while maintaining cellular homeostasis. The larval midgut is involved in ionic and osmotic regulation as well as digestion, absorption, and excretion (Clements, 1992). It is subdivided into four structurally distinguishable regions: cardia, gastric caeca, anterior stomach, and posterior stomach (Fig. 1-1). Each of these regions consists of one cell layer of epithelial cells, composed of large columnar cells and much smaller cuboidal cells, which vary in character somewhat from region to region. Belying this simple architecture however, the epithelial

PAGE 16

3 cells are capable of maintaining physiological homeostasis while facing a pH range of 71 1 along the length of the mosquito gut lumen (Dadd, 1 975). This range in pH, along the length of the mosquito gut, is presumed to support digestive and assimilation functions (Clements, 1992). The epithelial cells of the anterior midgut (AMG) surround a highly alkaline lumen (pH 1 1) while those of the gastric caeca (GC) and posterior midgut (PMG) surround a neutral to mildly alkaline lumen (pH 7-8; Clements, 1992; Zhuang et al., 1999). The different pH values found along the midgut may support the various metabolic functions that are active in each gut region. The gastric caeca perform ion and water transport, the anterior midgut performs alkaline digestion, the posterior midgut performs nutrient absorption, and the Malpighian tubules (part of the hindgut) actively transport potassium and fluid (Clements, 1992). The role of the alkaline pH in the anterior midgut is a point of some controversy. It has been suggested that the high pH contributes to the digestion of plant detritus and, in particular, to the dissociation of tannin-protein complexes (Martin et al., 1980). The high pH restricts the conglomeration of proteins within the anterior midgut that could interfere with the insects' normal physiology. These complexes could also interfere with insect digestion by blocking the active sites of many different digestive enzymes. Therefore, the alkaline gut serves as a proposed benefit to the insects by allowing ingested food to remain soluble. The alkalinity therefore keeps the gut free from attachable tannin-protein complexes and enhances the assimilation of proteins. Berenbaum's review (1980) of Lepidopteran insects correlated gut pH (range from 7.0-10.3) with diet. Caterpillars feeding on leaves containing tannins were found to display a more alkaline pH (average pH 8.76) than those feeding on low tannin diets (average pH 8.25; Berenbaum, 1980).

PAGE 17

Although this alkaline digestive strategy is well documented in insects, the molecular processes involved have not been clearly defined. Carbonic Anhydrase Carbonic anhydrase (CA), a blood enzyme, first described by Meldrum and Roughton in 1933, catalyzes the reversible hydration of carbon dioxide to form bicarbonate and a proton (C0 2 + H 2 0 <-> HC0 3 ' + H*; Meldrum and Roughton, 1933). Carbonic anhydrase was first characterized in erythrocytes as the result of a search for a catalytic factor that would enhance the transfer of bicarbonate from the erythrocyte to the pulmonary capillaries (Meldrum and Roughton, 1933). Since it was first described, CA has been shown to play an important role in most acid/base transporting epithelia. Fourteen different CA isoforms have been characterized to date in mammals (HewettEmmett and Tashian, 1996). These enzymes have been determined to function in pH regulation and ion balance, thereby performing a crucial role in many biological processes such as respiration, bone resorption, renal acidification, gluconeogenesis, aqueous humor production, gastric acid production, cerebrospinal fluid formation, and signal processing (Dodgson, 1991; Sly and Hu, 1995; Hewett-Emmett and Tashian, 1996; Lindskog, 1997; Sun and Alkon, 2002). Various types of epithelial cells, such as those described in the mammalian kidney, contain CAs that can provide large quantities of bicarbonate for buffering cells and their microenvironment. Polarized epithelia play an important role in partitioning physiologically distinct compartments, and in maintaining cell and tissue homeostasis. The epithelial cells found in the larval mosquito midgut may serve a similar partitioning function. Like the mammalian kidney, different regions of the mosquito gut may play

PAGE 18

5 differential roles in homeostasis and function. Elucidating the distribution of CAs along the mosquito midgut epithelium may uncover the mechanisms responsible for the unique alkaline physiology of the mosquito gut. Mosquito Development and Control Part of the success of insects can be attributed to the structural adaptation of their integument, which functions as skin, skeleton, sensory and respiratory organ, and food reserve (Rockstein, 1964). The advantage of having an extremely strong integument is offset by the disadvantage of not being able to grow significantly in size. Insects have overcome this growth-limiting problem by shedding their integument and rebuilding a new larger one. This process of ecdysis (molting) is used as a tool for marking the different stages of development in many insect species. While mosquito control can target different stages of mosquito development, this project focuses on the larval enzymes, specifically early fourth instar, which begins immediately after the third molt. Careful attention was paid to the stage of insect development in all experiments due to a previous study that showed insect enzymes to decrease or completely arrest prior to molting (Jungreis et al., 1981). The mosquito life cycle begins at hatching from the egg (Fig. 1-2). At this point the fully independent mosquito is called a first instar larva. Successive molts mark the transition to the next larval instar, four larval instars in all. In each instar, the larvae possess a series of morphological characteristics, some particular to that stage. However, there are only slight changes in internal organs such as the midgut. Within a day or two the late fourth instar larva changes into a pupa (Clements, 1992). Within twenty-four hours, the flying adult emerges from the pupa case. Adult females of most mosquito

PAGE 19

6 species require a bloodmeal in order to nourish their developing eggs. However, the males do not ingest blood but instead feed on fruit or do not feed at all (Clements, 1992). Mosquito control tactics use different methods for controlling mosquito larvae as compared to the flying adults. Mosquito larvae are confined to the water in which they develop, whereas the adults are free-flying and therefore highly mobile. Pesticide sprays are employed against the flying adult mosquitoes, but dragonflies and butterflies are also ill-affected. An arguably better strategy for mosquito control is to target the larvae before they are capable of biting and transmitting disease. Mosquito larvae are voracious eaters, incessantly consuming particulates in the water around them, taking in almost anything. Because of this non-discretional eating behavior, the wriggling larvae can potentially consume a larvacidal agent if placed in the water. Determining the physiological roles of larval mosquito gut enzymes and metabolic transporters may provide a lead for constructing mosquito larvacides. Carbonic Anhydrase Inhibition The focus of this project is to examine the distribution and expression of CAs within the fourth instar of larval development of two species of mosquito, Ae. aegypti and An. gambiae. A tangential result of characterizing mosquito CAs may be in the development of mosquito-specific inhibitors. If a CA is discovered to be essential for mosquito development or homeostasis, a specific inhibitor of precisely this mosquito CA isoform could be developed. Since virtually all organisms contain CA enzymes, an inhibitor that would compromise this mosquito CA while not affecting any other isozymes would be necessary for mosquito control so that non-target species would not be affected. Differentially specific CA inhibitors are already employed in the distinctive

PAGE 20

7 characterization of mammalian CA isoforms. For example, the acidic sulfonamide benzolamide has been used for the preferential inhibition of extracellular CA while not compromising any intracellular CA activity (Tong et al., 2000). This occurs due to the inability of benzolamide to readily penetrate cell membranes (Tong et al., 2000). The wealth of information pertaining to mammalian CA isoforms and their specific inhibitors provides a basis for comparisons with CAs that are discovered in the mosquito midgut. Sulfonamide CA inhibitors are widely used to treat a number of conditions including glaucoma, gastro-duodenal ulcers, and cancer, by lowering the production of fluids and acids. Parkkila et al. (2000) showed that the invasion of renal cancer cells in vitro could be inhibited with CA inhibitors. If larval mosquito physiology is dependent upon the generation or maintenance of the alkaline gut, and CA is a necessary component, then the possibility exists for the use of CA inhibitors as mosquito larvacides. Bicarbonate Transport The site(s) of bicarbonate production by CA may not be as important as the translocation of the bicarbonate that is produced. Transporters can facilitate the passage of bicarbonate and other ions through otherwise impermeable cell membranes. Bicarbonate transporters compose a large family of membrane proteins that includes the anion exchangers (AEs), sodium bicarbonate cotransporters (NBCs), and members of the sulfate transporter group that can also transport bicarbonate (Alper et al., 2001). Most of the BT proteins consist of a cytosolic anchoring domain as well as a 10-14 membranespanning transporter domain (Alper et al., 2001). Also, evidence exists that some AEs are capable of physically binding CA enzymes. Thus, the fourth extracellular loop of AE1 contains a glycosyl -phosphatidyl-inositol (GPI)-linked CA IV binding site and the

PAGE 21

8 intracellular carboxy terminus of AE1 was found to contain a cytosolic CA II binding site (Vince and Reithmeier, 2000; Sterling et al., 2002a). A metabolon, a complex of membrane proteins involved in regulation of bicarbonate metabolism and transport, defines the relationship between the CA and AE proteins (Sterling et al., 2001a). This bicarbonate transport metabolon, is thus capable of transporting bicarbonate as soon as it is available from the CA enzyme. Transport can be in either direction, into or out of the cell, and is therefore predictively capable of maintaining a tight hold on pH. The occurrence of such a tight bicarbonate control mechanism could be very advantageous to the mosquito. With such a large pH gradient across the membrane, a bicarbonate transport metabolon could ensure that the pH on either side of the membrane is strictly monitored. This bicarbonate transport metabolon has only been identified in a mammalian system. Despite this fact, an insect gut model that employs such a bicarbonate transport metabolon is easy to envision. Because of the strong pH gradient that is maintained in the mosquito gut, it is reasonable to propose that a bicarbonate transport metabolon could exist in this system as well. Gut Alkalization Model My first physiological model of the larval mosquito midgut was derived from the tobacco hornworm, M. sexta, which also uses an alkaline digestive strategy. In this model, several proteins contribute to the high alkalinity (Fig. 1-3). These are the CA, the FT V-ATPase, and the cation and anion exchangers. The Ff V-ATPase is thought to be the energizer of the system by using ATP, and pumping protons out into the lumen of the anterior midgut. This sets up a potential difference across the membrane of about 210 mv (Harvey, 1992). In this model, the predicted cytosolic CA within the anterior midgut

PAGE 22

combines carbon dioxide and water to produce bicarbonate and a proton ion. The bicarbonate is pushed from the epithelial cell, across to the lumen side by the anion exchanger, in trade for a chloride ion. The proton then gets stripped off of the bicarbonate and, along with the proton pumped across by the V-ATPase, is brought back into the cell in exchange for a potassium ion (Wieczorek et al., 2000). This potassium ion combines with the carbonate to produce potassium carbonate, which is hypothesized to be responsible for the high alkaline pH of the anterior gut region. This hypothesis stems from the fact that potassium ions are actively produced by the Malpighian tubules and are circulated throughout the gut via the hemolymph (Clements, 1992). Potassium carbonate also has a pKA greater than 10 and can therefore contribute to the gut alkalization. The goal of this project was to expand and adapt this model to the larval mosquito by completing several clear objectives. These objectives are outlined vvdthin the following specific aims. Specific Aims 1 . Determine whether CA is involved in buffering the high alkalization of the larval mosquito gut. A. Determine if a CA enzyme is present within the mosquito gut. Determine which regions of the larval Ae. aegypti gut display CA activity using CA histochemistry and l8 0 isotope exchange. B. Determine if CA-specific inhibitors, such as acetazolamide, can influence larval midgut alkalization. 2. Determine whether CA is expressed in the larval mosquito gut. A. Clone and characterize full length CA cDNAs from the larval midgut of Ae. aegypti and An. gambiae.

PAGE 23

10 B. Use experimental and bioinforrnatical approaches to determine if CA expressed in the mosquito gut is similar to a characterized mammalian CA isoform. Furthermore, determine the subcellular location of mosquito CA isoforms as cytosolic, membrane-bound, mitochondrial, or GPI-linked. C. Determine which regions of the mosquito gut express CA mRNA and protein using in situ hybridization, real time PCR, and immuno-localization. 3. Determine whether anion exchangers (AE) are involved in the pH regulation of the larval mosquito gut. A. Clone and characterize an AE from the larval mosquito gut that uses the bicarbonate produced by CA as a substrate. B. Determine whether the mosquito AE transports chloride using a Xenopus oocyte expression assay. C. Determine if AE is expressed in the same regions of the mosquito gut as the CA. Co-localization of CA and AE would support the existence of a bicarbonate transport metabolon. D. Determine if the AE contains the amino acid sequence predicted to be necessary for binding CA. If indeed the AE protein is predicted to bind CA, a bicarbonate transport metabolon within the larval mosquito gut could maximize bicarbonate production and transport. 4. Present a new larval mosquito model that reflects the studies in this dissertation. Bring together all localized components of mosquito gut physiology into one model.

PAGE 24

11 Figure 1-1 . Illustration showing the regions of the larval mosquito gut. The midgut is composed of the cardia, gastric caeca (GC), anterior midgut (AMG), and the posterior midgut (PMG). The hindgut is composed of the Malpighian tubules (MT) and the rectum.

PAGE 25

12 1 instar 2 instar eggs | ar va larva Hill *v ^ *V 2nd I moltf pupa Figure 1-2. Illustration of the mosquito life cycle. The four life stages are egg, larva, pupa, and adult. The larval stage consists of four different instars. Early fourth instar larvae, following the third molt, were chosen for all experiments. The female mosquito continues the cycle by laying eggs, usually after a required blood meal.

PAGE 26

13 Blood (Acidic) pH ~ 6.5 Caterpillar gut cell (Neutral) pH*7.0 2H Af b « 30 mV Hfi KCOj ATP Lumen (Alkaline) pH= 11.0 A»P T = 210 mV A*a_~240mV 1 H*V-ATPaee © Anion exchanger PUmP ^# Cation exchanger 6 Carbonic anhydraee Amino acid: K+ cotraneporter Channel Figure 1-3 Preliminary mosquito anterior midgut model based on M. sexta. This theoretical model places a C A H-like isoform within the cell cytosol where it combines carbon dioxide and water to form bicarbonate and a proton. Alkalization is driven by a proton pumping V-ATPase that resides in the apical membrane and pumps protons into the lumen. A chloride/ bicarbonate exchanger, that is also located in the apical membrane, exchanges bicarbonate from the CA, for chloride from the lumen. A cation exchanger transfers potassium to the lumen while stripping protons from the bicarbonate for the exchange. The potassium ion combines with the deprotonated carbonate ion to form potassium carbonate, which brings the pH to highly alkaline levels.

PAGE 27

CHAPTER 2 MATERIALS AND METHODS Experimental Insects Ae. aegypti eggs were obtained from a colony maintained by the United States Department of Agriculture (USD A) laboratory in Gainesville, Florida. The eggs were allowed to hatch in 20 ml of 2% artificial seawater (ASW; 8.4 mM NaCl, 1 .7 mM KC1, 0.1 mM CaCl 2 , 0.46 mM MgCl 2 , 0.51 mM MgS0 4 , and 0.04 mM NaHC0 3 ). The mosquito larvae were reared in 2% ASW at room temperature. The Ae. aegypti larvae were fed a mixture of yeast and liver powder (1:1.5 g respective dry weight; ICN Biomedicals Inc., Aurora, Ohio). Eight to ten days were required for this species to reach the early fourth instar. An. gambiae eggs were obtained from the Centers for Disease Control and Prevention (CDC) in Atlanta, Georgia. Strict handling guidelines were followed with this particular species, which does not currently inhabit Florida, due to its inherent ability to acquire and transmit the Plasmodium protozoan, which causes malaria. This Anopheles species was therefore reared in deionized water inside of a locked incubator set at 30°C. A mesh screen served as a second barrier within the incubator while the sealed (but not airtight) containers harboring the An. gambiae larvae served as the third barrier against escape. The An. gambiae larvae were fed a Wardley tropical fish flake food (The Hartz Mountain Corp., Secaucus, New Jersey). Early fourth instar larvae were chosen for all experiments. Ten to twelve days from the hatch day were required for this species to 14

PAGE 28

15 reach the early fourth instar. Late fourth instar larvae that went unused were sacrificed to prevent any chance of emerging adults. Preparation and Fixation of Tissue To dissect out the midgut, the heads of the cold-immobilized larvae were pinned down using fine stainless-steel pins to a Sylgard layer at the bottom of a Petri dish containing hemolymph substitute solution consisting of 42.5 mM NaCl, 3.0 mM KC1, 0.6 mM MgS0 4 , 5.0 mM CaCl 2 , 5.0 mM NaHC0 3 , 5.0 mM L-succinic acid, 5.0 mM L-malic acid, 5.0 mM L-proline, 9.1 mM L-glutamine, 8.7 mM L-histidine, 3.3 mM L-arginine, 10.0 mM dextrose, 25 mM Hepes and adjusted to pH 7.0 with NaOH (Clark et al., 1999). The anal segment and the saddle papillae were removed using ultra-fine scissors and forceps, and an incision was made longitudinally along the thorax. The cuticle was gently pulled apart and the midgut and gastric caeca were removed. In some cases, the gut contents enclosed in the peritrophic membrane slid out, leaving behind the empty midgut. In other cases, it was necessary to remove the peritrophic membrane and its contents manually. For enzyme histochemistry, fixation was in 3% glutaraldehyde in 0.1 M phosphate buffer, pH 7.3, overnight at 4°C (Ridgway and Moffet, 1986). For in situ hybridization and immunohistochemistry, dissected tissues were fixed overnight in 4% paraformaldehyde in 0.1 M cacodylate buffer, pH 7.2 (0.1 M phosphate buffer, pH 7.2 was used for Ch 4 and 5 in situ). In some cases, the dissected larval midguts were photographed using a Nikon FX-35DX photographic camera mounted on a Nikon SMZ10 dissecting microscope. In other cases, digital images were acquired using a Leica DMR microscope equipped with a Hammamatsu CCD camera. All images were assembled using Corel Draw-1 1 software.

PAGE 29

16 Bromothymol Blue Qualitative Assay A qualitative test to detect carbonic anhydrase activity in mosquito larval midgut homogenate was adapted from the test described by Tashian (1969). The procedure included immersing a piece of Whatman no.l paper in a solution made with 0.15% Bromothymol Blue (BTB) in ice-cold 25 mM Tris HC1, 0.1 M Na 2 S0 4 , pH 8.0. The paper was allowed to soak completely in this blue solution and was placed on ice for 30 minutes. The colored filter was then transferred to a Petri dish with a hole in the lid. Samples of mosquito larval midgut homogenate were prepared by sonicating midguts of early fourth instar larvae in ice-cold 25 mM Tris HC1, 0.1 M Na 2 S0 4 pH 8.0, with protease inhibitor cocktail (SigmaAldrich, St. Louis, MO; diluted 1:1000). An autopipette was used to spot exactly 4 ul samples on the paper. Controls were also spotted. The controls included a buffer with protease inhibitor and controls for the liver/yeast food added to the medium in which the mosquito larvae were reared. These food controls included a range of concentration from 1 to 100 ug/ml liver powder and yeast. Carbonic anhydrase from bovine erythrocytes (SigmaAldrich) dissolved in the same buffer described above was used as a positive control. A steady stream of C0 2 at 34.5 KPa was blown for 3 seconds through the opening on the lid of the Petri dish, and the dish was sealed and kept on ice. The formation of yellow spots in a few seconds was indicative of carbonic anhydrase activity. Effect of Methazolamide on the Alkalization of the Midgut of Live Larvae The effect of a CA inhibitor, methazolamide, on gut alkalization and the capacity of whole larvae to alkalize their culture medium was examined. Flat-bottomed tissue culture plates (24 well, Sarstedt Inc., Newton, North Carolina) were filled with 1 ml of 25

PAGE 30

17 mM Tris HC1, 0.1 M Na 2 S0 4 buffer, pH 8.5. BTB solution was added to each well until a 0.003% solution was achieved. Five live early fourth-instar larvae that had been placed in BTB indicator solution for 2 hours were added to each of the wells, and the larvae were allowed to adjust to their new environment for 30 minutes. Methazolamide dissolved in Dimethyl Sulfoxide (DMSO; Sigma-Aldrich) at concentrations ranging from 10" 6 M to 8xl0" 3 M was added to the wells. Controls included wells containing DMSO with BTB indicator but no inhibitor and wells containing BTB indicator but no DMSO. The plates were scanned using a Hewlett Packard ScanJet 6100C scanner before addition of the inhibitor and 5 hours later. In addition, the midguts were dissected and photographed to record the pH within the gut lumen as revealed by the color of ingested BTB. 18 0 Exchange Method to Measure Carbonic Anhydrase Activity 18 Tissue homogenate carbonic anhydrase activity was measured using the O exchange method (Silverman and Tu, 1986). Midguts were dissected, and the peritrophic membrane was removed together with its contents. Individual measurements of CA activity were performed with pooled samples of gastric caeca, anterior midgut, posterior midgut and Malpighian tubules. The method involved adding 18 0-labeled NaHC0 3 to 0.1 M Hepes buffer, pH 7.6, at 9.5°C. The disappearance of 18 0 isotopes from C0 2 and/or HC0 3 " upon addition of the enzyme preparations was monitored. Measurements of 18 0 in C0 2 were accomplished with a mass spectrometer, using a C0 2 -permeable inlet that allowed very rapid, continuous measurement of the isotopic content of C0 2 in solution. All samples were centrifuged at 14,000 rpm at room temperature prior to the assay to remove food and insoluble material. Inhibition was accomplished by adding

PAGE 31

18 methazolamide to a final concentration of 10" 6 M. Recombinant^ expressed and purified mosquito CAs were also tested for activity using this assay. Isolation of RNA and Synthesis of cDNA Total RNA was isolated from freshly dissected fourth instar mosquito larval midguts using TRI Reagent (Molecular Research Center Inc., Cincinnati, Ohio) according to the manufacturer's instructions. Briefly, 100 Ae. aegypti gut epithelial organs, including fore-, mid-, and hindgut (approximately 20 mg) were dissected in HSS and transferred to a sterile microcentrifuge tube containing TRI Reagent (600 ul). The tissue was homogenized and incubated for 5 min at room temperature. The homogenate was then extracted with chloroform (40 uL) and precipitated with isopropanol (100 uL). The RNA pellet was washed with 75% ethanol (200 uL), air-dried and resuspended in 50 uL diethylpyrocarbonate (DEPC; Sigma-Aldrich)-treated H 2 0. RNA concentrations were calculated from the absorbance at 260 nm. Total RNA (10 ug) was reversetranscribed for 2 hours at 42°C in a 20 ul reaction mixture using 5 pmol of oligo(dT)1218, RNasin (1:40 dilution), IX first strand buffer, 1 mM dNTPs, and 200 units (U) of Superscript II reverse transcriptase (Invitrogen Inc., Carlsbad, California). This cDNA was used to clone the first fragment of Ae. aegypti CA. Bioinformatics The National Center for Biotechnology Information (NCBI) website (www.ncbi.nlm.nih.gov) was used for the majority of the bioinformatical data presented in this study. The first mosquito genome, An. gambiae, was released in 2002 (Holt et al., 2002), and made accessible to the public on the NCBI website. The basic local alignment search tool (BLAST; Altschul et al, 1990) was employed for primer construction as well

PAGE 32

19 as analyzing PCR products. The NCBI Blast Flies database (www.ncbi.nrm.nih.gov/BLAST/Genome/FlyBlast.html), together with the Ensembl database (www.ensembl.org/Anopheles gambiae/) were used to predict the number of CA genes in the Drosophila melanogaster and An. gambiae genomes by inputting the Ae. aegypti CA as the search sequence. These partial sequence results were then annotated to reflect the 2 full-length CA sequences that we have cloned from An. gambiae and presented within this manuscript. Ensembl is a joint project between the European Bioinformatics Institute and the Sanger Institute to bring together genome sequences with annotated structural and functional information. The NCBI protein database (pdb) and the BLAST were used in conjunction with the 3-dimensional structure viewer (Cn3D; Hogue, 1997) for the prediction of antibody accessible peptide regions in mosquito proteins. BLAST analyses also confirmed that the chosen antigenic peptides were unique. The conserved domain database (CDD; Marchler-Bauer et al., 2002) and the conserved domain architecture retrieval tool (CDART; Geer et al., 2002) were used to predict the function of our newly cloned mosquito proteins. Alignments were produced using Clustal W (Thompson et al., 1994), as implemented in DNAman software (Lynnon Biosoft, Vaudreuil, Quebec, Canada). Cloning of CA from Aedes aegypti Larval Midgut Degenerate oligonucleotides were designed against the regions of conserved amino acids among CA proteins as determined by the BLAST analysis of several vertebrate and two putative, but annotated, CA proteins from the D. melanogaster sequence database.

PAGE 33

20 The primer sequences used initially for Ae. aegypti CA were CA5F and CA3R (see Table 2-1). PCRs were performed in a total volume of 20 ul, and the reaction mixture contained 0.1 \ig of cDNA as template, 0.2 uM of each primer, 200 |iM each of dNTPs, IX PCR buffer and 1 U of Taq polymerase (Promega; Madison, Wisconsin). The PCR cycling profile was: 94°C for 5 min, 55°C for 2 min and 72°C for 3 min, followed by six cycles of 94°C for 0.5 min, 53°C (in increments of 2°C/cycle) for 1 min and 72°C for 1 min and 35 cycles of 94°C for 0.5 min, 45°C for 1 min and 72°C for 2 min followed by a final extension at 72°C for 15 min. The PCR products were visualized on 1% agarose gels and specific products were isolated using a QIAquick gel extraction kit (Qiagen, Inc, Valencia, California), diluted 1 : 100 in water, and used as template for a second, identical PCR. The resulting 297 base-pair (bp) product was gel-purified, ligated into pGem-T (Promega) and transformed into JM109 Escherichia coli (Promega) for subcloning. This partial Ae. aegypti CA cDNA was completed using amplified cDNA pools from gastric caeca and posterior midgut. Construction of Amplified cDNA Pools Adapter-ligated, amplified cDNA pools ("libraries") were constructed from different regions of the fourth instar larval gut of both Ae. aegypti and An. gambiae using a technique optimized for invertebrate tissues (Matz et al., 1999). The gastric caeca, anterior midgut, posterior midgut, rectal salt gland, Malpighian tubules, and anal papillae of ten larvae were dissected in HSS and collected separately, resulting in six discreet tissue pools. The tissue was dissolved in Buffer D (500 u.L; 4 M guanidine thiocyanate, 30 mM sodium citrate, and 30 mM beta-mercaptoethanol). The mixture was placed on ice and combined with phenol (500 uL, pH 7.0) and chloroform (100 uL). The mixture

PAGE 34

21 was vortexed and centrifuged at 14,000 g for 30 seconds at 4°C. The upper, aqueous phase was transferred to a clean tube and 5 uL glycogen solution (Pharmacia Quick Prep Micro RNA purification kit, Piscataway, New Jersey). The RNA was precipitated by the addition of 100% ice-cold ethanol (550 uL) followed by centrifugation at 14,000 g for 6 minutes at room temperature. The supernatant was removed and 1 mL of ice-cold ethanol (80%) was added. The mixture was centrifuged at 14,000 g for 10 minutes at room temperature, the supernatant was removed, and the pellet was air-dried. For first strand synthesis, the pellet was resuspended in DEPC-treated water (5 uL) and combined with the TRsa primer (1 uM; Table 2-1). This mixture was incubated at 50°C for 3 minutes and immediately placed on ice. Then IX ligation buffer (Marathon cDNA Amplification kit, BD Biosciences, Palo Alto, California), 0.01 M DDT, 1 U Superscript II (Life Technologies; Rockville, Maryland), and 0.5 uL dNTP mix (10 mM each dNTP, Marathon cDNA Amplification kit) were added to a total volume of 10.5 uL. This reaction mixture was incubated at 42°C for 1 hour and immediately put on ice. For second strand synthesis, DEPC-treated water (49 u.L) was added to the first strand reaction mix. The mixture was then combined with 1.6 uL dNTP mix (10 mM each, Marathon cDNA Amplification kit), IX reaction buffer (Marathon cDNA Amplification kit), and 4 u.L second strand synthesis enzyme mix (Marathon cDNA Amplification kit) in 80 uL total volume. The reaction mix was then incubated at 1 6°C for 1 .5 hours. T4 DNA polymerase (1 U; Marathon cDNA Amplification kit) was added to the reaction mixture and the entire mixture was incubated at 16°C for an additional 0.5 hour. The reaction was stopped by incubation at 65 °C for 5 minutes.

PAGE 35

22 The reaction mix (80 uL) was combined with 40 uL phenol and 40 uL chloroform and centrifuged at 14,000 g for 10 minutes. The upper, aqueous phase was removed and transferred to a clean tube. The cDNA was precipitated by the addition of 3 M sodium acetate (8 uL, pH 5.0) and 100% ethanol (160 uL). The mixture was centrifuged at 14,000 g for 15 minutes at room temperature. The supernatant was removed and the pellet was air-dried. For adaptor ligation, the cDNA pellet was resuspended in DEPC-treated water (6 uL) and combined with 1 uM adaptor, IX ligase buffer, and 1 U T4 ligase (Marathon cDNA Amplification kit) in 10 uL total volume. This mixture was stored overnight at 16°C. For cDNA amplification, the ligation mixture (10 uL) was combined with 40 uL DEPC-treated water. PCR amplification was then performed using the Advantage kit (BD Biosciences). The diluted cDNA (1 uL) was combined with IX advantage buffer, 0.4 uL dNTP mixture (10 mM each), 0.1 uM DAP and TRsa primers (Table 2-1), and 0.4 UL advantage enzyme mix in 20 uL total volume. The cycling profile consisted of 94°C for 30 seconds, 66°C for 1 minute, and 72°C for 2.5 minutes. The reaction was analyzed on a 1% agarose gel after 12, 16, and 20 cycles. A final chase step was then performed to ensure that all cDNAs were completely double-stranded. Both 5' and 3' adaptor primers were added to the PCR reactions and two cycles of 77°C for 1 min, 65°C for 1 min, and 72°C for 2.5 min were performed. The resulting collections of amplified cDNA were then diluted 1 :50 and used as template for subsequent PCR experiments. Amplified cDNA pools from An. gambiae were used to clone two CA cDNAs and the AE cDNA. Exact primers were designed from conserved regions of the proteins as

PAGE 36

23 determined by BLAST analysis using characterized proteins against the An. gambiae genome. See table 2-1 for all initial primer sequences. 3' and 5' Rapid Amplification of cDNA Ends and Sequencing Full-length cDNAs were obtained by rapid amplification of cDNA ends (RACE), (Zhang and Frohman 1997, modified by Mate et al., 1999). Exact primers were defined according to the 5' adaptor (DAP primer) along with a reverse primer specific to the cloned fragment, and 3' TRsa adaptor (TRsa primer) along with a forward primer specific to the cloned fragment (see Table 2-1 for adaptor primer sequences). These ends, which included the 5' and 3' UTR sequences, were then used to design PCR primers to produce a single product with consensus start and stop codons. Plasmid DNA from individual colonies was purified using a Qiaprep Plasmid Mini kit (Qiagen). The plasmid DNA (50 ng) was then sequenced using the ABI Prism Big Dye Terminator Cycle Sequencing Kit (PE Biosystems, Foster City, California) and the reaction products were analyzed on an ABI Prism 310 Genetic Analyzer (PE Biosystems). Construction of In Situ Hybridization Probes Sense and antisense digoxygenin (DIG)-labeled cRNA probes were generated by in vitro transcription using a DIG RNA labeling kit (Roche Molecular Biochemicals, Indianapolis, Indiana). The initial in situ hybridization experiment, presented in chapter 3, used a cRNA probe derived from the original 297 bp Ae. aegypti CA sequence. The in situ experiments presented in chapters 4 and 5 utilized the full-length CA and AE sequences. For the first CA antisense probe, the pGEM-T vector containing the 297 bp CA sequence was linearized by incubating 2 |ig of plasmid with Pst I restriction enzyme

PAGE 37

24 (New England Biolabs (NEB); Beverly, Massachusetts) and IX buffer 3 (NEB) at a total volume of 20 uL for 1 hour at 37°C. For the sense probe, the pGEM-T vector containing the 297 bp CA sequence was linearized by incubating 2 ug of plasmid with Not I restriction enzyme and IX buffer 3 (NEB) in a total volume of 20 uL for 1 hour at 37°C. After digestion, the volume was brought to 1 00 \xL with the addition of 80 uL water. A phenol/ chloroform extraction was performed such that 100 uL of phenoV chloroformisoamyl alcohol was added to the linearized plasmid and the solution was centrifuged at 14,000 g for 1 minute. The upper aqueous phase was transferred to a new tube and 100 uL chloroform was added. After centrifugation at 14,000 g for 1 minute, the upper aqueous phase was transferred to a new tube and the chloroform step was repeated. The linearized plasmid DNA was precipitated by the addition of 10 uL sodium acetate (3 M, pH 2.5) and 200 uL cold ethanol (100%). The DNA was incubated at -80°C for 15 minutes and then centrifuged at 14,000 g for 10 minutes at 4°C. The supernatant was removed and the DNA pellet was washed by the addition of 500 uL ethanol (70%) followed by centrifugation at 14,000 g for 5 minutes at 4°C. The supernatant was removed and the pellet was air-dried and then resuspended in 13 uL DEPC-treated water. The full-length Ae. aegypti CA was subcloned into pCR 4-TOPO plasmid using a PCR manufactured 5' Sal I restriction site and a 3' Xho I site. Therefore, the pCR 4TOPO plasmid was linearized by incubating 2 ug of plasmid with either Sal I, IX Sal I buffer, and BSA, or Xho I, IX buffer 2, and BSA (NEB). The pCR 4-TOPO plasmid was also used for the generation of the An. gambiae CA and AE probes. For these probes, the unique restriction sites, Pme I and Not I, located within the pCR 4-TOPO plasmid were used for linearization with IX buffer 4 and BSA, or IX buffer 3 and BSA, respectively.

PAGE 38

25 These mixtures were all incubated at 37°C for 2 hours to ensure complete linearization of the plasmids. After digestion the uncut pCR 4-TOPO plasmids were compared to the cut plasmids on a 1% agarose gel to confirm linearization. The cut plasmids (10 uL) were cleansed using a Qiaquick PCR Purification kit (Qiagen Inc, Valencia, California). For in vitro translation, the resuspended pellet or purified plasmids were combined with IX transcription buffer, IX NTP labeling mixture, RNase inhibitor (20 U), and 40 U T3 RNA polymerase (or SP6 for pGEM-T plasmids) or 40 U T7 RNA polymerase. For the Ae. aegypti CA probes, T7 polymerase was used with the Sal I cut plasmid to produce the antisense probe, while T3 polymerase was used with the Xho I cut plasmid to produce the sense (control) probe. The pCR 4-TOPO plasmid used for the generation of the An. gambiae CA and AE probes contained the CA and AE sequences in the reverse configuration. Therefore, for these An. gambiae probes, T3 was used with the Not I linearized plasmids to produce the antisense probes, while T7 was used with the Pme I cut plasmids to produce the sense (control) probes. The mixtures were incubated at 37°C for 2 hours followed by the addition of 20 U DNase I and incubation at 37°C for 15 minutes. The DNase I reaction was stopped by the addition of 0.5 uL of EDTA (500 mM). The DIG-labeled cRNA was then precipitated by the addition of 2.5 uL of LiCl (4 M) and 75 uL cold ethanol (100%). The mixture was incubated overnight at -20°C and centrifuged at 14,000 g for 10 minutes at 4°C. The supernatant was removed and the pellet was washed with 50 uL cold ethanol (75%). The centrifugation step was repeated and the pellet was air-dried and resuspended in 100 uL DEPC-treated water. The probes were stored at -80°C.

PAGE 39

26 In Situ Hybridization The in situ hybridization experiments presented in chapters 4 and 5 added an additional fixation step due to a recommendation by Dr. Dmitri Boudko to increase the clarity of the in situ labeling. A glass electrode fitted to a micromanipulator was used to inject 4% paraformaldehyde into the thoracic cavity, just behind the head. Successful perfusion was easily identified by the cessation of the otherwise constant muscle twitching along the length of the body. This injection of fixative served to preserve the cellular integrity and protect against the many proteases that exist within the mosquito gut. For in situ hybridization, methods were adapted from Westerfield (1994). The midguts were washed with PBS at room temperature and then incubated in 100% methanol at -20°C for 30 minutes to ensure permeabilization of the gut tissue. The tissue was washed (5 min each wash) in 50% methanol in PBST (Dulbecco's phosphate buffered saline [Sigma-Aldrich] plus 0.1% Tween-20), followed by 30% methanol in PBST and then PBST alone. The tissue was fixed in 4% paraformaldehyde in 0.1 M sodium cacodylate buffer (or 0.1 M phosphate buffer) for 20 min. at room temperature and washed with PBST. The larval midguts were digested with proteinase K (10 ng/ml in PBST) at room temperature for 10 min, washed briefly with PBST and fixed again, as described previously. Prehybridization of the tissue was accomplished by incubation in HYB solution (50% formamide, 5X SSC [IX SSC equals 0.15 M NaCl, 0.015 M Na-citrate buffer pH 7.0], 0.1% Tween-20) for 24 hours at 55°C. The larval midguts were transferred to HYB+ solution (HYB plus 5 mg/ml tRNA, 50 ug/ml heparin) containing 5 ng/ml DIGlabeled probe and incubated overnight at 55°C. Excess probe was removed by washing at

PAGE 40

27 55°C with 50% formamide in 2X SSCT for 30 min (twice), 2X SSCT for 15 min and 0.2X SSCT for 30 min (twice). For detection, the tissue was incubated in PBST containing 1% blocking solution (Roche Molecular Biochemicals) for 1 h at room temperature. The tissue was incubated with anti-DIG-alkaline phosphatase (Roche Molecular Biochemicals) diluted 1 :5000 in blocking solution for 4 hours at room temperature. The tissue was washed with PBST and incubated in alkaline phosphatase substrate solution (Bio Rad Laboratories, Hercules, CA, USA) until the desired intensity of staining was achieved (2-3 hours). CA Histochemistry Carbonic anhydrase activity was detected in isolated Ae. aegypti midguts using Hansson's method (Hansson, 1967), as modified by Ridgway and Moffet (1986). The procedure involved the incubation of isolated, 3% glutaraldehyde-fixed midguts in 1.75 mM C0SO4, 53 mM H 2 S0 4 , 1 1 .7 mM KH 2 P0 4 , and 15.7 mM NaHC0 3 (pH 6.8). The incubation medium contains a high concentration of bicarbonate, which stimulates the production of C0 2 and hence a decrease in pH in the presence of CA. The acidic pH then stimulates the formation of insoluble black cobalt salts which were visualized using 0.5% (NHO2S in distilled water. Therefore, micro-sites of active CA liberation of C0 2 from bicarbonate dehydration become apparent with this assay. Removal of the bicarbonate substrate (NaHCOa) eliminated staining. Real Time PCR Region-specific cDNA was produced from dissected mosquito tissue using the Cells-to-cDNA standard protocol (Ambion INC, Austin, Texas). The gut regions used to make the amplified cDNA pools were incubated in 50 \iL of hot cell lysis buffer for 10

PAGE 41

28 minutes at 75°C. The lysed tissues were treated with 2 U of DNase I for 30 minutes at 37°C. The DNase I was then inactivated by heating to 75°C for 5 minutes. For the reverse transcription reaction, 10 uL of cell lysate was combined with 4 uL dNTP mix (contains 2.5 raM each dNTP) and 5 uM random decamer first strand primer in 16 uL total volume. The mixture was incubated at 70°C for 3 minutes and then chilled on ice for 1 minute. This mixture was then combined with IX RT buffer, 1 U M-MLV reverse transcriptase, and 10 U RNase inhibitor, and incubated at 42°C for 1 hour. The reverse transcriptase was then inactivated by incubation at 95°C for 10 minutes. Primers (Table 2.1) were designed using Primer Express software (Applied Biosystems; FosteT City, California). The SYBR Green PCR Master mix, which includes SYBR Green I dye, Amplitaq Gold DNA Polymerase, dNTPs, and buffer, was used for all real time PCR investigations. Each cycle of PCR was detected by measuring the increase in fluorescence caused by the binding of the SYBR Green dye to double-stranded DNA using an ABI Prism 7000 Sequence Detection System (Applied Biosystems). Initially, each primer set, including the control 18s ribosomal RNA (Genbank accession M95126), was assessed to determine the optimal concentration of primer to be used. All real time experiments used the same 2-step cycling profile: 50°C for 2 minutes followed by 95°C for 10 minutes and 40 cycles of 95°C for 15 seconds and 60°C for 1 minute. Whole gut cDNA (100 nM) was used as template with 500 nM, 300 nM, 100 nM, or 50 nM of each primer set and IX SYBR green I master mix in 25 uL total volume. Each reaction was done in triplicate. The optimal concentration was then chosen based on the amplification plots and the dissociation curves generated. Once a concentration was chosen for each primer set, the efficiency of amplification of that set was determined. Serial dilutions of

PAGE 42

29 whole gut cDNA were used as template with the appropriate concentration of primers and IX SYBR green I master mix in 25 uL total volume. The threshold cycle number (Ct) was plotted versus the log of the template concentration and the slope (m) and intercept (b) were determined (Figure 2-1). These pre-determinations were then used in the standardized comparison of the amount of 18s transcript and CA transcript in each of the cDNA samples tested. For each analysis, a control containing all of the necessary PCR components except the cDNA template was run. To determine the relative expression level for each transcript analyzed, the following equation was used: (Ct-b)/m. The average log ng for each transcript was then compared to the average log ng of 18s RNA transcript to normalize the values. Then the expression levels were determined relative to the transcript with the greatest normalized log ng value and expressed in a bar graph using Microsoft Excel software. Antibody Production An antigenic peptide consisting of eighteen amino acids was chosen from the Ae. aegypti CA sequence for antibody production. In order to increase the probability that this antibody would be specific for this particular CA sequence (in the event that other CA isoforms were isolated from the mosquito gut), attempts were made to synthesize an antigenic peptide that would be specific to this isoform. The well-characterized mammalian CA isoforms served as a model in trying to choose a unique CA peptide sequence. The comparison of the mosquito CA with the mammalian isoforms yielded a peptide sequence from the amino (N) terminus, where CA isoforms showed the most diversity, and least conservation. The N terminus of our mosquito CA was predicted to have an extended loop secondary structure. Unlike an alpha helix, an extended loop is

PAGE 43

30 more accessible to antibody probing. Furthermore, three-dimensional analyses (Cn3D v4.1 NCBI) of predicted CA IV structures (human 1ZNC and mouse 2ZNC) predicted that the N terminus is exposed and accessible (Figure 2-2). An antigenic peptide was therefore chosen from the N terminus of the Ae. aegypti CA sequence. This peptide sequence (GVINEPERWGGQCETGRR) was sent to Sigma-Genosys (Woodlands, Texas), where it was synthesized and conjugated to bovine serum albumin (BSA). The synthetic peptide-BSA construct and Freund's incomplete adjuvant were injected into two rabbits to elicit an immune response. Prior to injection, a blood sample from each rabbit was collected to serve as the control pre-immune serum. Every two weeks a blood sample was collected from the rabbits, the fraction of immunoglobulin G (IgG) pooled, and another dose of the peptide-BSA construct administered. Three months after the initial injections, the final bleeds were collected and used for all immunohistochemical analyses. We also raised antibodies against an An. gambiae cytosolic CA peptide and an anion exchanger (AE) peptide. These antibodies were produced by the Aves Labs, Inc., (Tigard, Oregon), using a similar strategy to that described above. However, these antibodies were produced in hens. The synthesized peptides were conjugated to BSA and injected into two hens each. The immunoglobulin Y (IgY) antibodies were collected from the hens' eggs, pooled, and purified. Immunohistochemistry The specificity of the antibodies in the resultant antisera was determined. The antisera were then used to localize the larval mosquito proteins. Dissected and fixed whole mount mosquito guts were washed 6 times in tris-buffered saline (TBS), placed in

PAGE 44

31 pre-incubation medium (pre-inc) for a rninimum of 1 hour, and then incubated in primary antibody (1:1 000) overnight at 4°C. The guts were then washed in pre-inc and incubated in FITC-conjugated goat anti-rabbit (GAR) or Alexa-GAR secondary antibody (Jackson ImmunoResearch, West Grove, Pennsylvania, 1 :250 dilution) overnight at 4°C. The whole mount preparations were rinsed in pre-inc and mounted onto slides using pphenylenediamine (PPD, Sigma-Aldrich) in 60% glycerol. In some cases Draq 5 (Jackson ImmunoResearch, 1:1000 dilution) was applied before mounting to visualize nuclear DNA. The samples were examined and images captured using the Leica scanning confocal microscope. Live preparations were examined, following a similar procedure, to ensure that antibodies were capable of localizing extracellular proteins only. In this case, the primary antibody was applied to the live guts for four hours. The samples were washed in TBS, and fixed as described previously, before the secondary antibody was applied. The samples were mounted on slides as described above. The live gut assays were also performed to determine whether this specific CA is tethered to the cell membrane via a GPI linkage. Ten live gut preparations were incubated with phosphoinositol-specific phospholipase C (PI-PLC, 1:100 in HSS; SigmaAldrich) for 90 minutes at 37°C. PI-PLC was used as a tool in determining the presence of a GPI link. Controls in which the guts were incubated in HSS alone were also performed. The guts were then washed in HSS, fixed, and treated with primary and secondary antibodies as described above.

PAGE 45

CA Protein Expression Recombinant Ae. aegypti and An. gambiae CAs were produced using the pETl 00 vector (Invitrogen). Specific primers were designed to amplify each cDNA. The 3' primers included the sequence 5' to and including the native stop codon. The 5' primers contain the sequence CACC preceeding the native start codon for correct frame insertion (See Table 2-1 for primer sequences). PCRs were performed using 1 U of Platinum ?Jx polymerase (Invitrogen), the gastric caeca cDNA collections as template (200 ng), IX Pfx amplification buffer, 1.2 mM dNTP mixture, 1 mM MgS04, and 0.3 uM of each primer in a total volume of 50 uL. A three-step PCR protocol was used consisting of 94°C for 2 minutes followed by 30 cycles of 94°C for 30 seconds, 55°C for 30 seconds, and 68°C for 1 minute. The resultant blunt-ended cDNAs (4 uL from PCR mix) were ligated with the pETlOO directional Topo vector (1 uL and 1 uL salt solution; Invitrogen) for 10 minutes at room temperature. Top 10 chemically competent E. coli (50 uL; Invitrogen) were transformed by incubating 3 uL of ligation mix with the cells for 30 minutes on ice, followed by a heat shock of 42°C for 30 seconds. SOC (250 uL) was added to the cells and they were then incubated at 37°C for 30 minutes with shaking. The transformation mix (100 uL) was then plated on a LB-carbenicillin plate (50 ug/mL) and incubated overnight at 37°C. Colonies were sequenced using Big Dye version 1.1 as described previously. The purified plasmids (10 ng each) were transformed into BL21 Star (DE3) cells (Invitrogen) for CA expression as described above. However, after SOC addition and incubation, the culture was transferred to fresh LB-carb (10 mL) and grown overnight at 37°C with shaking. The next day, 1 mL of culture was transferred to 100

PAGE 46

33 mL of fresh LB-carb and was grown at 37°C with shaking. Optimization experiments were performed in order to facilitate the production of the greatest quantity of CA protein. For production of CA protein, isopropylthio-p-galactoside (IPTG, 1 mM final concentration; Stratagene, La Jolla, California) was added when the culture had attained an optical density of 0.5 at a 600 nm wavelength. Achieving this density took about 1 .5 hours of growth at 37°C and 200 rpm. Zinc, in the form of zinc sulfate (0.5 mM final concentration), was added along with the IPTG to facilitate the proper conformation of an active CA protein. In order to optimize the duration of the induced growth phase, samples were collected every hour for six hours. These samples were analyzed on an SDS-Page 4-12% Bis-Tris gel to compare CA protein content. Four hours of growth was determined to be ideal for the production of the truncated Ae. aegypti CA IV-like and full-lengthen, gambiae CA II-like proteins. Total protein was collected using the Probond Purification System according to the manufacturers instructions for soluble proteins (Invitrogen). The cells were harvested by centrifugation, sonicated in native buffer (250 mM NaP0 4 , 2.5 M NaCl; Invitrogen) with lysozyme (1 mg/mL; SigmaAldrich), and centrifuged again to collect a crude protein extract. The supernatant was applied to a Probond nickel column (Invitrogen) and washed free of non-specific binding contaminants. The nickel column binds the CA protein due to the added histidine tag, a repeat of six histidine residues within the pETlOO expression vector that is inserted after the carboxy-terminus of the CA protein. CA was eluted by adding imidazole (250 mM; Invitrogen) to the column, which competes with and displaces the histidine tag. Eluted fractions were separated on an SDS-Page 4-12% Bis-Tris gel (Invitrogen).

PAGE 47

34 Anion Exchanger Oocyte Expression The full-length anion exchanger (AE) sequence was subcloned into the pXOOM vector, which is optimized for both oocyte and mammalian expression (Jespersen et al., 2002; a generous gift from Dr. T. Jespersen). In addition to a T7 RNA polymerase promoter, this vector contains XenopusspeaSc 5' and 3' UTR sequences flanking the insert in both directions. cRNA synthesis was performed using the T7 mMessage mMachine kit (Ambion, Austin, Texas), after the cDNA was linearized using PMEI. One day after surgical removal of the eggs from the frog, the eggs were injected with either AE cRNA or water (control). After injection the eggs were incubated at 16°C for 4 days, long enough for measurable protein production and expression. The oocytes were maintained in ND96 (96 mM NaCl, 2 raM KC1, 1 mM MgCl 2 , 10 mM HEPES, pH 7.4 with NaOH). The medium was changed daily and dead oocytes were removed. Anion Exchanger Physiology Expression of the An. gambiae AE was examined using 2-electrode voltage clamp electrodes. The voltage electrodes were pulled using 1.2 mm glass (M1B120F-3, World Precision Instruments), and showed resistances between 1-2 MIX Oocytes were clamped to -50 mV and stepped from -90 mV to +70 mV in 10 mV increments. The water injected eggs served as the control in evaluating any activity exerted by endogenous proteins found in the Xenopus oocytes. Several different solutions were used to determine the exchanger's functional activities (refer to table 2-2). The transporter blockers, 4,4'-diisothiocyanodihydrostilbene-2,2'-disulfonate (DIDS, Calbiochem, La Jolla, California) and niflumic acid ( SigmaAldrich) were used to inhibit the transporter capabilities of the expressed AE1 protein.

PAGE 48

35 Aedes degenerate CA primers CA5F: 5' GAR CAR TTY CAY TKY CAY TGG GG CA3R: 5' GTI ARI SWN CCY TCR TA N=G,A,T.C; K=G.T: S=G,C; W=A.T; Y=C,T, R=A,G Amplified cDNA adaptor primers: DAP: 5' CGA CGT GGA CTA TCC ATG AAC GCA TRsa: 5' CGC AGT CGG TAC TTT TTT TTT TTT Anopheles exact CA primers Ag1CA2F: 5' CAG TCA CCT ATC GAC CTA AC Ag1CA4R: 5' CTC GCG TGT TCA ATG GTT Ag4CA11F: 5' GGA GGC GTC CTT GGC AAC Ag4CA12R: 5' CTG CAC TGA CCG GAA GTT es exact AE primers: Aedes CA Real time PCR primers: 5SPCAF1 5SPCAR1 18s RIBF 18s RIBR GCA CCG CGC GTC ACA GTT TAC AAC Aedes CA expression primers CTG CTT CCG TCT ACA CGT TAA TAA CTC CAT TG TAC CGA TGG ATT ATT TAG TG TTC AGC GAT TCA AAT GTA ExCAshortF: 5 1 CACC ATG GAC GAA TGG CAC ExCAshortR: 5' TTA GTA ATC CAT ATC GGT GTG GT Anopheles CA expression primers ExCA4F: CA4end: CACC TTA ATG CAG GCA CTT TCA CGA AAA AAG ACA CAC ACA AAC AAG] GG Table 2-1 . PCR primer sequences.

PAGE 49

36 Salt for 98mM value nW #1 #2 #3 #4 98N 98K 98N-C 1 98K-CI L mM mM f mM f g/1 mM l" $ Solution: lx 4x lx 4x lx 4x lx 4x NaCl 58.44 n 5.73 22.91 2 0.12 0.47 0 0 0 0 KCI 74.55 2 0.15 0.60 98 7.31 29.22 0 0 0 0 Na Gluconate 218.1 0 0 0 0 98 21.37 85.50 2 0.44 1.74 234.2 0 0 0 0 2 0.47 1.87 M 22.95 91.81 Choline CI 139.6 0 0 0 0 0 0 0 0 MgS0 4 7H 2 0 (120.36) 246.5 0 0 0 0 0.5 0.12 0.49 0.5 0.12 0.49 MgCl 2 6H : 0 (59.7) 203 0.5 0.10 0.41 0.5 0.10 0.41 0 0 0 0 CaCl 2 2H 2 0 (110.98) 147.02 0.5 0.07 0.29 0.5 0.07 0.29 0 0 0 0 430.38 0 0 0 0 0.5 0.22 0.86 0.5 0.22 0.86 HEPES (free base) 238.3 10 2.38 9.53 10 2.38 9.53 10 2.38 9.53 10 2.38 9.53 EGTA for InsideOout 380.4 0 0 0 0 0 0 0 0 pH 7.2 (4M NaOH) 7.: (4M KOH) 7.: ! (4M NaOH) 7. 2 (4M KOH) Table 2-2. Composition of all solutions used in Xenopus oocyte expression of An. gambiae AE. Total molarity and pH were kept constant in all solutions. Expression profiles were recorded in high sodium (#1), high sodium minus chloride (#3), high potassium (#2), and high potassium minus chloride (#4).

PAGE 50

37 Aedes CA Primer Linearization BrCA equations: y = -3.2655x + 39.958 R 2 = 0.9776 18s equations: y = -3.1312x + 27.781 R 2 = 0.9931 BrCA primers GC 18s primers WG IU I 1 1 • 1 o 1 2 Logng 3 4 5 Figure 2-1. Efficiency plots for real-time PCR primers. Serially diluted cDN A samples were tested with each primer set to determine the efficiency of amplification. A linear regression was performed to determine the slope and intercept for each primer set. These values were then used in an algorithm to compare cDNA concentrations within the samples.

PAGE 51

38 Figure 2-2. Three-dimensional (Cn3D) depiction of human C A IV (1ZNC). The green barrel represents an alpha helix structure, the tan arrows represent beta sheets, and the colored strings represent extended loop structures. The yellow coloring represents the accessible, extended loop peptide region against which the homologous Ae. aegypti CA antibody was raised.

PAGE 52

CHAPTER 3 CARBONIC ANHYDRASE IN THE MIDGUT OF LARVAL AEDES AEGYPTI: CLONING, LOCALIZATION, AND INHIBITION 1 Introduction Bicarbonate (and ultimately carbonate) ions are produced in vivo primarily by the enzymatic action of carbonic anhydrase (CA). Its activity contributes to the transfer and accumulation of Yt or HC0 3 " in bacteria, plants, vertebrates and invertebrates. Although there are innumerable reports related to the isolation of CA from vertebrates, studies involving CA from invertebrates are very rare and there are no reports of the isolation of CA from adult or larval mosquitoes. There is strong immunohistochemical (Zhuang et al., 1999) and physiological (Clark et al., 1999; Boudko et al., 2001b) evidence that an electrogenic, basal H 1 " VATPase energizes luminal alkalinization in the anterior midgut of the larval mosquito by producing a net extrusion of protons out of the lumen and a hyperpolarization of the basal membrane. In contrast, H 1 " V-ATPase appears to be localized in the apical membrane of the posterior midgut and gastric caeca providing a reversed H + pumping capacity relative to the anterior midgut (Zhuang et al., 1999). A system capable of generating a high luminal pH is likely to be buffered by carbonate (C0 3 " 2 ), which has a pKa of approximately 10.5. 'This chapter was slightly modified and reprinted with permission from The Company of Biologists LTD. Corena, M. P., Seron, T. J., Lehman, H. K., Ochrietor, J. D., Kohn, A., Tu, C. and Linser, P. J. (2002). Carbonic anhydrase in the midgut of larval Aedes aegypti: cloning, localization and inhibition. J. Exp. Biol. 205, 591-602. 39

PAGE 53

40 The purpose of this study was to determine the presence and location of CA in the midgut of larval Ae. aegypti and to clone and characterize the enzyme. To investigate the role of CA in the alkalization of the larval midgut, the effects of CA inhibitors were tested. Here, we report the cloning and localization of the first CA from mosquito larvae and, in particular, from the midgut epithelium of larval Ae. aegypti. A cDNA clone isolated from fourth-instar Ae. aegypti midgut (termed A-CA) revealed sequence homology to the a-carbonic anhydrases (Hewett-Emmett, 2000). Histochemistry and in situ hybridization showed that the enzyme appears to be localized throughout the midgut, although preferentially in the gastric caeca and posterior regions. In addition, classic carbonic anhydrase inhibitors such as acetazolamide and methazolamide inhibit the mosquito enzyme in the midgut. Results Bromothymol Blue Qualitative Assay This assay allowed the identification of samples of solubilized midgut tissue containing CA activity by spotting them onto a filter paper soaked in a basic buffered solution containing a pH indicator, bromothymol blue (BTB). As stated previously, BTB changes color from yellow (at pH<7.6) to blue when the pH increases above this value. The principle behind the assay is based on the fact that CA catalyzes the conversion of CO2 into bicarbonate with the concomitant release of protons (Donaldson and Quinn, 1974). The presence of protons lowers the pH in those regions of the paper where the spotted samples contain the enzyme. As the pH falls below 7.6, these spots rapidly change color from blue to yellow. This assay is not effective for samples in acidic solution, and the tissue homogenization must be accomplished in alkaline buffer. The

PAGE 54

41 enzymatic reaction takes only a few seconds, and it can be delayed if the solutions, the paper and the samples are kept cold on ice. However, a few seconds is usually sufficient to discriminate the samples that contain CA from those lacking enzymatic activity. The assay must be performed quickly since, after approximately one minute the entire filter paper turns yellow, probably as a result of the uncatalyzed hydration of carbon dioxide absorbed by the solution at this basic pH. The test has proved useful in determining the presence of small amounts of CA m homogenates of mosquito larvae. The assay was also used to detect CA activity qualitatively, in fractions obtained from affinity chromatography (Osborne and Tashian, 1975) of larval homogenates. The affinity chromatographic procedure, which employs a bound CA inhibitor (p-aminomethyl benzyl sulfonamide (p-AMBS); Sigma), produced two peaks of CA activity upon exposure to the standard elution buffers. The amount of protein that we were able to produce by this technique was, however, very small and resisted several efforts at direct microsequencing. This change in color was inhibited by acetazolamide and methazolamide when these inhibitors (10" 5 M) were added to the samples prior to spotting on the dye-impregnated filter papers. Inhibition of the reaction resulted in blue spots that did not change color upon addition of C0 2 . The positive control containing commercial CA turned yellow when carbon dioxide was added, and this color change was also inhibited by acetazolamide and methazolamide. This finding confirmed that the yellow color of the spots was due to the action of CA and that the mosquito larva contains active CA.

PAGE 55

42 Carbonic Anhydrase Activity and Alkalization A classic CA inhibitor methazolamide, was tested in live fourth instar larvae to examine the influence of CA on the maintenance of the pH extremes inside the midgut, and the effect of the enzyme on the net alkalinization of the growth medium by the intact animals. Previous investigations have shown that living mosquito larvae excrete bicarbonate, which results in the net alkalization of their surrounding aqueous medium (Stobbart, 1971). Equal numbers of living larvae of equivalent age and size were placed in culture plate wells containing lightly buffered medium and the pH indicator BTB. The tissue culture plates used in this assay were scanned before and after addition of various concentrations of methazolamide. In the absence of methazolamide, the blue color of the medium, indicating a pH of at least 7.6, was maintained (Stobbart, 1971). Actual measurement of the pH in each well showed a slow increase over time (data not shown). Upon addition of methazolamide, the culture medium slowly became acidic, with a resulting change in color to yellow as the pH dropped below 7.6 (Figure 3-1). All of the controls that did not contain methazolamide remained blue. Addition of methazolamide, at various concentrations, to wells containing only medium with BTB (no mosquito larva control) remained blue. These data show that CA activity is present in the living larvae and that it plays some role in acid/base excretion. Moreover, fourth instar larvae cultured in BTB-containing medium ingest the dye, which can then be used as a visible indicator of the pH in the gut lumen. Treatment of the cultured larvae with methazolamide showed a direct impact of inhibited CA activity on gut luminal pH. Figure 3-2 compares the luminal pH of dissected larval midguts with and without a 5 hour exposure to methazolamide. The micrographs reveal that

PAGE 56

43 alkalinization of the midgut was inhibited by methazolamide as shown by the color change of the BTB indicator. Interestingly, the effect was most pronounced in the anterior midgut, where the pH indicator changed from blue in the midgut of larvae reared in the absence of inhibitor to yellow in as little as 30 minutes when methazolamide (10"* M) was added to the culture. The indicator also changed color progressively from blue through green to yellow in the gastric caeca (Figure 3-2). No apparent change was observed in the posterior midgut. The color of the midgut in this region was yellow both in the untreated larvae and in the larvae treated with methazolamide. Since the pH of the posterior midgut has been associated with values close to 7.6, no change in color was evident using this qualitative method. 18 0 Isotope-Exchange Experiments The relative activity of CA, normalized to total protein content, was calculated as described by Silverman and Tu (1 986). The relative activity of CA was highest in the gastric caeca, followed by the posterior midgut and Malpighian tubules (Figure 3-3). The relative activity of CA in the anterior midgut was either extremely low or non-existent, falling at or below that of the buffer blank. The specificity of the reaction was confirmed by complete inhibition with the addition of 10" 6 M methazolamide (results not shown). Cloning of Carbonic Anhydrase from Aedes Aegypti Larvae We utilized a cDNA cloning strategy to obtain a specific carbonic anhydrase cDNA from the midgut epithelial cells of the larval Ae. aegypti. A comparison of twelve CA sequences, including two putative CA sequences that had been annotated but not characterized in the Drosophila melanogaster databases, was made. We then produced degenerate PCR primers from consensus regions of the CA gene family. The initial 297

PAGE 57

44 bp partial sequence was used to derive exact PCR primers for a modified 3*and 5RACE (Frohman and Zhang, 1997, modified by Matz, 1999). Amplified cDNA pools from each region of the isolated gut, facilitated the eventual cloning of a single contiguous cDNA (Matz, 1999). The final contiguous region spanned both start and stop codons, and encoded a polypeptide of 298 residues (GenBank accession number AF395662). Figure 3-4A shows an alignment of the Ae. aegypti carbonic anhydrase (ACA) amino acid sequence with several other, previously characterized members of this extensive a gene family. Figure 3-4B shows a homology tree depicting the percentage of identical amino acids between sequences, generated using DNAman software. Figure 35 A shows the alignment between A-CA and six putative CA gene sequences from the D. melanogaster genome that our homology search (BLAST) revealed. Four of the D. melanogaster genes (AAF54494, AAF56666, AAF57140, AAF57141) had not previously been annotated. Figure 3-5B shows the homology tree generated with these sequences. A-CA has a putative molecular mass of 32.7 kDa. The translated A-CA protein sequence possesses a characteristic eukaryotic-type CA signature sequence within the polypeptide (amino acid residues 99-1 15; Fernley, 1988). To examine the possibility of regionalized expression of the A-CA, PCR using exact primers was performed on amplified cDNA pools from the various sections of the gut. Figure 3-6 shows an ethidium-bromide-stained agarose gel. PCR products of the expected, 894 nucleotide length, are readily seen in the gastric caeca and the posterior midgut regions. Anal papillae (not shown), anterior midgut, Malpighian tubules and rectal salt gland showed little or no PCR product. When the PCR products were subjected to a second round of PCR using the same primers, an appropriately sized

PAGE 58

45 product was also discernible in the anterior midgut. This PCR analysis also revealed higher molecular mass products in the anterior midgut and Malpighian tubules that may represent additional carbonic anhydrases specific to larval Ae. aegypti (Figure 3-6). This result is shown only to display the gut regions in which the A-CA clone was derived. The lack of an 894 bp product in the other gut regions may simply be due to poor quality cDNA pools from those regions. However, the cloning of A-CA from both the gastric caeca and posterior midgut regions is consistent with the location of enzyme activities described above. Localization of the Enzyme in the Midgut Epithelium: Carbonic Anhydrase Enzyme Histochemistry To further analyze the regional and cellular expression of CA in the midgut epithelium of larval mosquitoes, a modified Hansson's histochemical reaction was performed on whole mount preparations of the gut (Hansson, 1967). Figure 3-7 summarizes the results of this analysis. Carbonic anhydrase activity was detected in a non-uniform pattern along the length of the gut. The most intense staining was evident in the gastric caeca and the posterior midgut. Staining was less intense in the anterior midgut. At higher magnification, it was obvious that cellular heterogeneity with regard to CA activity also exists. This is particularly evident in the posterior midgut, where very large and regularly spaced cells appear nearly white on a background of dark CA reaction product. The larger cells have been characterized as "columnar" or ion-transporting cells (Volkman and Peters, 1989b). Surrounding these large cells are more numerous smaller cells termed "cuboidal" or resorbing/secreting cells (Zhuang et al., 1999). The CA histochemical stain clearly distinguishes these cells from one another and indicates that the large columnar cells contain relatively very little CA in comparison with the smaller

PAGE 59

46 cuboidal cells. In addition, the distal cells of each lobe of the gastric caeca, termed Cap cells, show little or no histochemical staining, suggesting further cellular heterogeneity with respect to CA distribution in the gut (Figure 3-7). In Situ Hybridization To further characterize the localization of A-CA expression, in situ hybridization was performed using a portion (approximately 300 bp) of the central coding region of the cDNA. Figure 3-8 shows typical results of this type of analysis. A strong hybridization signal was evident in the gastric caeca and the posterior midgut. Lower levels of hybridization were evident in other gut regions. As with the CA histochemical stain, higher magnification revealed that the relatively small cuboidal cells exhibit more intense labeling than do the large columnar cells (Figure 3-8B). Discussion The search for the enzyme in the midgut of the larval mosquito was triggered by the observations of a pH value around 1 1 in the anterior midgut lumen and a high bicarbonate concentration (Zhuang et al., 1999; Boudko et al., 2001b). The presence of CA in the midgut of the larval mosquito has been suggested before by investigations of the epithelium of larval lepidopteran midgut. Carbonic anhydrase has been studied in Manduca sexta, where the enzyme has been associated with the fat body, midgut and integumentary epithelium (Jungreis et al., 1981). The enzyme has also been localized in the goblet cells of the epithelium of Hyalophora cecropia using Hansson's histochemical stain. The same procedure showed that the columnar cells were devoid of activity (Turbeck and Foder, 1970).

PAGE 60

47 Even though a number of genes and their products have been isolated from the midgut ofAe. aegypti, and the role of CA in the alkalization of the midgut has been suggested (Turbeck and Foder, 1970; Haskell et al., 1965; Ridgway and Moffett, 1986; Boudko et al., 2001b), there have been no reports of the isolation or cloning of CA or of the localization of the enzyme within the midgut of larval mosquitoes. This is the first recorded cloning of a CA from a mosquito, and is also the first to be cloned from any arthropod. Our results show that at least one (and perhaps more) CA is present in the midgut of larval Ae. aegypti. The CA of larval Ae. aegypti (A-CA) is inhibited by classical carbonic anhydrase inhibitors such as methazolamide and acetazolamide. Methazolamide has the most potent effect on A-CA. Direct physiological measurements of ion fluxes from living larval mosquito midgut epithelial cells also show methazolamide to be a very potent inhibitor of ion movements and balance (Boudko et al., 2001a). To investigate the distribution of CA in the midgut of the larval mosquito, we employed both in situ hybridization and enzyme histochemistry. Our results indicate that enzymatic activity is greatest in the gastric caeca and the posterior midgut, as demonstrated by the intense staining obtained using Hansson's method and by in situ hybridization using cRNA probes. Measurements of activity using the 18 0 exchange method in pools of dissected regions of the gut corroborate these findings. In addition, the enzyme seems to be preferentially associated with the small cuboidal cells in the midgut epithelium, as determined both by enzyme histochemistry and by in situ hybridization.

PAGE 61

48 As reviewed in Clements (1992), two major cell types have been defined in the gastric caeca by inferring functional states from cytological findings. These two major cell types have been called ion-transporting cells and resorbing/secreting cells (Volkman and Peters, 1989a,b) and they correspond to the columnar and cuboidal cells mentioned above with the ion-transporting cells being equivalent to the columnar cells and the resorbing/secreting cells being the cuboidal cells (Zhuang et al., 1999). Neither of these cell types, as characterized in the larval mosquito gut, parallels the structurally unique qualities of the lepidopteran goblet cell. Nonetheless, our results indicate that, as in lepidopterans, CA activity is preferentially associated with one of two distinct cell types whose functional complementation must produce the alkalization and ionic balances regulated by the gut. These results are consistent with the observations of lepidopteran midgut by Turbeck and Foder (1970). In the larval lepidopteran midgut, two morphologically distinct cell types have been long recognized: goblet cells and columnar cells. Goblet cells posses both the proton-pumping V-ATPase and CA activity (Harvey, 1992; Ridgway and Moffet, 1986; Wieczorek et al., 1999). One of the enigmas of using the pioneering analyses of insect model systems such as M. sexta to produce testable hypotheses for gut alkalinization in mosquito larvae has been the apparent absence of goblet cells from mosquitoes. Previous investigations have inferred different functional cell types in the larval mosquito gut epithelium. We are currently developing antibody probes for A-CA. Immunocytochemical analyses of A-CA distribution in comparison with other key components of gut function, such as V-ATPase (Zhuang et al., 1999), should provide new insights into the cell biology of this intriguing epithelial system.

PAGE 62

49 It is interesting to note that the lowest concentration of CA in the midgut epithelium occurs in the region that surrounds and probably regulates the region of highest luminal pH, the anterior midgut. The pKa of C0 3 " 2 is approximately 10.5 and, hence, this anion is likely to be the primary buffer of the pH 10.5-1 1 gut contents within the anterior midgut. Our results therefore suggest that the major buffering anion in this area of the midgut is probably not produced by local CA but instead either upstream, in the gastric caeca, or downstream, in the posterior midgut, where CA levels are very high. This result, and results presented elsewhere (Boudko et al, 2001a), are consistent with a model in which a major function of the anterior midgut is to pump protons out of this region of the gut lumen, promoting the conversion of HC0 3 ' to C0 3 " 2 . A comprehensive model of the regulation of ion homeostasis and gut alkalization in the larval mosquito awaits the characterization and localization of other major components of the system in addition to CA. It will also be very important to resolve the question of whether multiple CAs are expressed in the midgut and how each is distributed in this dynamic tissue. Quantitative evidence corroborating the distribution of CA within the midgut and • 18 supporting the histochemical and in situ observations was obtained using the 0exchange mass spectrometric method. The results obtained with this method indicate that the gastric caeca exhibit the highest level of carbonic anhydrase, relative to total protein content, followed by the posterior midgut and the Malpighian tubules. The anterior midgut showed levels of activity so low that two possibilities could be considered: either the method could not detect the enzyme or it is absent from the anterior midgut. The presence of faint staining using the histochemical and in situ methods suggests that the

PAGE 63

50 levels of activity in the anterior midgut might be too low to be detected using the 0 method, but that the enzyme is present throughout the entire length of the midgut. In summary, our evidence demonstrates the existence of CA in Ae. aegypti larvae and it also suggests that the gastric caeca and posterior midgut exhibit the highest levels of CA activity. In addition, the enzyme seems to be associated with the small cuboidal cells of the midgut epithelium. Furthermore, enzyme activity has also been detected in membrane preparations isolated from whole midguts and could be due to the presence of more than one isoenzyme. Carbonic anhydrase activity has previously been demonstrated in the epithelium of the larval midgut of six species of lepidopterans, in which it has been associated with the particulate fractions of the homogenate (Turbeck and Foder, 1970). This is consistent with our hypothesis that there might be more than one CA and that one of these enzymes may be associated with the plasma membrane. What is the role of CA in the alkalization mechanism? BTB proved useful in monitoring the impact of CA inhibition on the maintenance of gut luminal pH and the excretion of acid/base. As mentioned earlier, Ae. aegypti larvae typically alkalize the medium in which they are reared by secreting bicarbonate ions (Stobbart, 1971). The ingestion of CA inhibitors altered the metabolism of the larvae to the point that the metabolic products secreted into the medium change the pH of the environment, shifting it towards more acidic values than those observed in the absence of inhibitors. The lowering of the pH of the medium might be related to a decrease in the rate of secretion of HC0 3 ". The effect of the ingestion of CA inhibitors on the secretion of bicarbonate into the medium remains to be explored. However, as indicated by measurements with ion-selective microelectrodes, inhibition of CA in the midgut has an extreme effect on the

PAGE 64

51 maintenance of an alkaline pH within the midgut lumen (Boudko et al., 2001a). It is plausible that a decrease in the rate of secretion of bicarbonate is elicited by inhibiting the CA enzyme. A simple model of bicarbonate transport fails to explain how the high pH is achieved within the anterior midgut of the larval mosquito. At a pH of approximately 11, similar to that observed within the anterior midgut, the majority of bicarbonate is present as carbonate. In fact, measurements of lepidopteran midgut fluid have shown that it contains 37 mM carbonate and 17 mM bicarbonate (Turbeck and Foder, 1970). Since the pH of a 0.1 M solution of sodium bicarbonate is only approximately 8.3, secretion of bicarbonate alone cannot be responsible for the high pH observed in the anterior midgut (Dow, 1984). It could, however, explain the pH values at the gastric caeca and posterior midgut. The mechanism for maintenance of an alkaline pH within the anterior midgut must be more complex than just a simple buffering of a physiological solution with bicarbonate. Although this mechanism has been investigated (Wieczorek et al., 1999; Zhuang et al., 1999; Boudko et al., 2001a), its details remain unclear. However, the evidence suggests that a basal, electrogenic FT" V-ATPase energizes luminal alkalization in the midgut of larval mosquitoes (Zhuang et al, 1999; Boudko et al., 2001b). Although the electrogenic transport of K + drives the pH gradient, there must also be flux of one or more weak anions in the opposite direction to maintain homeostasis. Several transporters are thought to participate in this mechanism. Another line of evidence suggests that the levels of carbon dioxide in the hemolymph of lepidopterans are lower than those within the midgut lumen. The concentration of C0 2 has been determined to be near 5 mM in the hemolymph and 50

PAGE 65

52 mM in the midgut lumen in larval Hyalophora cecropia (Turbeck and Foder, 1970). Recent measurements using capillary zone electrophoresis of larval Ae. aegypti fluids have revealed a bicarbonate/carbonate level as high as 50.8±4.21 mM in the midgut lumen compared with 3.96±2.89 mM in the hemolymph (Boudko et al, 2001a). These values correlate with those observed by Turbeck and Foder (1970). This combined evidence suggests that the C0 2 that reaches the midgut lumen in the larvae of lepidopterans is rapidly converted to a mixture of bicarbonate and carbonate. The role of CA in the alkalization process would be of great significance. The generation of antibodies against A-CA will facilitate a detailed analysis of the cellular and subcellular distribution of this key enzyme in this system.

PAGE 66

53 Figure 3-1 . Effect of CA inhibition on culture medium pH with fourth-instar Ae. aegypti larvae. Mosquito larvae typically alkalize the medium in which they are reared (Stobbart, 1971). (A) Six culture wells each containing five fourth-instar larvae incubated for 5 hours in medium containing 0.003% Bromothymol blue (BTB). The blue color is retained, indicating a pH greater than 7.6. (B) The same as A, except that each well also contains a different concentration of the CA inhibitor methazolamide ranging from 10" 6 to 10' 3 M from left to right. A yellow color indicates a pH below 7.6.

PAGE 67

54 Figure 3-2. Effect of methazolamide on the alkalization of the midgut using Bromothymol Blue (BTB) assay of pH within living, but isolated, gut tissue. Gut tubes were dissected after pre-loading with BTB and then incubated for 5 hours in hemolymph substitute (Clark et al., 1999) in the absence (A) or presence (B) of 10' 6 M methazolamide. The loss of blue coloration in B shows that the internal pH of the gut lumen has dropped below 7.6. Scale bar represents 300 um.

PAGE 68

55 12i GC AM PM MT Figure 3-3. Relative activity of CA in different pooled segments of the midgut of larval Ae. aegypti. Midguts were dissected from early fourth-instar larvae and separated into gastric caeca (GC), anterior midgut (AM), posterior midgut (PM) and Malpighian tubules (MT). The relative activity of CA was measured using the 18 0 mass spectrometry method (Silverman and Tu, 1986), normalized to total protein content. The activity of the anterior midgut was lower than that of the water blank and, thus, is set as "zero" activity.

PAGE 69

56 aagypti HorsaCU -P0091 7 2«brn£i»h-Q92053 MCM -P07 4 SI Mou».CAH-NP0359J7 C . alagana-Tl 657 S RatGAIV-HPOG2047 AAd«B aagypti HorsaCAl PO 0 91 7 Z«foir»fTi»h-Q92 0Sl Hua«nCA3-FC74Sl NMMOM 4 -MPO 35927 C . «l^ar.»-T16S75 R»fcCJCTV-KP062047 «»gypti XoriaCU-ro0917 Zabrafiah-Q920Sl HunanCA3-P07451 Mou»«CA14-NP03S927 C.«l.gann-T16575 R»tCAIV-HP062047 A«dse aogypti HorMCM -POO 9 1 7 Z«bra£i»li-092051 MmCM -P07 4 51 MouaaCAl 4 -NP03S 927 C.«l«Qran»-T1657 5 RatC»JV-HP062047 aajypti HoraaCAl-P00917 Zabc»fiuh-a92051 HuaumCJa3-P074Sl NounaCAl 4 -MPD3592 7 C . «*l«cjnrm-Tl £575 RatOOV-HPO 62047 AaKiaia ma>qyr>tzl Hor«»C*l-P00917 Zebraf ieh-Q92051 HuaanCA3-V07451 HouiaCM 4 -MP03592 7 C.olagam-TJ 6575 RatOUV->IP(l62C47 aagypt! HuraaCM -POU917 Z«brafiah-B92051 HumanCA3-P07d51 2U.4-NP03S927 . alaqana-Tlfi575 »tCUV-«PO 62047 ,kialfvatij.pstirajdewhyptpapnwineperhggq<^tbrhH MXJTF.a^l.IJ75%) and green (>50%). (B) A homology tree comparing A-CA and several other a-CAs (DNAman software).

PAGE 70

57 Dro« XKF57140 Droll MFS11J1 Dru AAF56666 Dron AAF544 94 Drug AAF19948 Droa AAFA4B17 Droa AAFS7140 DXOS AAF57141 Dro> AAF56666 Deo* AM'54494 Droi AAF4 9948 DroB AAF44817 . MIALFVATLLP3TIHADEWHYPTPAPNGVIKEPERWGGQCET6 Droa AAF57KO Droa AAF67141 Droa AAF56666 Droa AAP54494 Droa AAF4 994B Droi AkP44817 Aflrlaa oaiffyptl Droi AAFS7140 Droa AlFbTUl Droa AAV56666 Droa AAF54494 Droa AAP49946 Droa AAF44817 •oc/ypti. Droa AM'57I40 Droa AAF57141 Droa AAF56666 Droa AAPS4494 Droa AAF49946 Droa M744817 Droa AAF5714D Droa AAFS714 1 Droa AAT56666 Droa AAFS4494 Droa AAF49948 Droa AAF44817 Aaaoaas B«gypti Droa AWS7140 Droa AJIF57141 Droa AAF56666 Droa AJ\F544 94 Droa AAF4 9948 Droa MF44817 MSr. TATCXrCSaF-AVCSHVPTiy flF.n^QRKMARHHGHCAOrCTOS . P1AITTSR MPlJPMSVGIOSVKIJ4«4»NEMGYl?DI^14t4QDEI>FPK .HCGlCIliO . M^RCRFrTPF-ATVIAPTI-ICAJFiT-VTAQnFTilECmMGPKHWaKr^YTvRCaaO MBHHW»TEEHCPA)nU>XYPQAK KirTOH . . . siqidst MRQalLXVPLF^TVaTG^DVKI^GPI.'riJjBTMG . . . .HTETMi . . .»4B*4rreij™G»CX)WEDlX:SSGkHQ8PirJJJSRTVSISIJ3 . . . .IPVTSH TTAIa»»i*AraMIGYHr4XX.PYPlJQLIGGV3rmD: I.IAIxa^QEA*LNVTF*X.S 8L IJkGVDWDrCFY ' VKSxDSMMKPVI.VAD»r£AVDr>IAMPSVE QIDRKGKSV»rr»JPL.Pi«3E MS . KB^BESfcTa PVXjTKGDRVTX-POGCDPGQLI-P -DflgH' Tf: 3 1 .pv s :j>o»f l^k 1 1 IDPSOHT: BQVX^VPi TET'l laja IMS PDPI PI SPKQ X SPtfcrJJ^SDTQOCiA. »IrTF1WrTIX>4a»4>refc4 90 s>n 73 X30 83 89 140 130 148 120 177 130 138 iee 177 194 166 226 ISO 188 236 227 243 214 275 229 235 285 276 291 264 298 235 235 304 302 335 270 B 100 i_ Aedes aegypti i>asAAF54494 Drew AAT57 140 DrasAAF57141 Dros AAF56666 Dros AAF49948 £>rasAAF44817 80 _l_ Homology (%) GO I 40 20 38% 49% 28° 34° 27% Figure 3-5. Comparison of the extrapolated amino acid sequences of A-CA with six putative dipteran CA genes identified in the D. melanogaster gene databases. (A) An alignment of A-CA with the amino acid sequences of the six D. melanogaster genes (accession numbers listed) identified through bioinformatics searching. Regions of exact homology across all species are highlighted in blue (100%); regions with less homology are highlighted in red (>75%) and green (>50%). (B) A homology tree comparison of these seven Dipteran CAs.

PAGE 71

58 Figure 3-6. Polymerase chain reaction (PCR) analysis of Ae. aegypti amplified cDNA from different gut regions. PCR was performed using exact primers for the cloned A-CA. Anterior midgut (lane 2), gastric caeca (lane 3), posterior midgut (lane 4), whole gut RNA control (lane 5), Malpighian tubules (lane 6) and a water template control (lane 7) are shown. Note the primary product in gastric caeca and posterior midgut samples at the expected size of approximately 894 nucleotides. Also note the absence of this band from other gut regions but the appearance of bands of higher molecular masses. Lane 1 is a 100 bp molecular mass ladder (Promega).

PAGE 72

59 Figure 3-7. Hansson's histochemistry of whole mount Ae. aegypti gut. A. Intense dark staining is observed in the cardia, gastric caeca (GC) and posterior midgut (PMG), indicating CA activity. B. Higher magnification of the gastric caeca. The distal lobes of the gastric caeca (Cap cells) exhibit relatively low levels of reaction product, indicating lower levels of enzyme activity in these cells relative to other cells of the gastric caeca. C. Higher magnification of the PMG shows large, relatively unstained columnar cells (*) contrasted with the smaller stained cuboidal cells (arrow). Scale bars represent 1 50 nm in A and B, and 75 um in C.

PAGE 73

60 Figure 3-8. Localization of CA mRNA expression in larval Ae. aegypti. A. An isolated whole mount gut probed with DIG-labeled cRNA for A-CA. Abundant hybridization is observed in the cardia, gastric caeca (GC), and posterior midgut (PMG). B. The smaller cuboidal cells (arrow) display stronger hybridization than the larger columnar cells (*). C. Isolated midgut reacted with the sense (control) cRNA for A-CA. Scale bars represent 300 um in A, 75 um in B and C.

PAGE 74

CHAPTER 4 A GPI-LINKED CARBONIC ANHYDRASE EXPRESSED IN THE LARVAL MOSQUITO MIDGUT Introduction The CA enzyme expressed in the midgut of larval mosquitoes shares some characteristics with the mammalian CA IV isozyme, including a glycosyl-phosphatidylinositol (GPI) link to the plasma membrane. Mammalian CA IV enzymes have been found in dynamic organs such as kidney, lung, gut, brain, eye, and capillary endothelium (Chegwidden and Carter, 2000). The human CA TV isoform was found to be as active as the CA II isoform in carbon dioxide hydration and even more active in bicarbonate dehydration (Baird et al., 1997). Studies of larval mosquito CAs are being pursued to better understand the alkaline gut system. As the anterior midgut of the larval mosquito lacks a highly active cytosolic CA II-like isozyme (previous chapter; Corena et al., 2002), the presence of a highly active CA rV-like isozyme within the mosquito gut may be able to provide the buffering capacity that is needed within the highly alkaline anterior midgut. A more detailed characterization of larval Aedes aegypti CA is presented in this study as well as the sequence of a homologous CA isoform from Anopheles gambiae. New tools and techniques, such as the generation of a mosquito-specific CA antibody and real time PCR, as well as improved methodology for in situ hybridization, have enabled this further analysis. 61

PAGE 75

62 Results Bioinformatics of Aedes Aegypti CA We have previously cloned a CA cDNA from the Ae. aegypti midgut (accession number AF395662; Corena et al., 2002). Our initial structure prediction indicated that the protein was cytosolic. However, further characterization has indicated that this CA is actually membrane associated via a GPI-link. We have determined that the CA propeptide sequence encodes an extracellular protein with a hydrophobic tail region. The first 17 amino acids of the propeptide are predicted by the Simple Modular Architecture Research Tool (SMART) program to be the signal sequence (Letunic et al., 2002). This sequence "flags" the message for transport to the endoplasmic reticulum (ER). Using the PSORT H server, the prediction of membrane topology (MTOP) indicates that the Ae. aegypti CA sequence is GPI anchored. Amino acid G-276, is predicted by the GPI prediction server (Eisenhaber et al., 1999) to be the site for GPI attachment. The hydrophobic tail (L278-A289) allows translocation of the transcript through the ER plasma membrane and is also predicted to stabilize the protein with the membrane until the pre-formed GPI anchor is transferred to the protein. The hydrophobic tail is then cleaved to produce a completely extracellular protein that is tethered to the cell by the GPI link (for a review of GPI-linked proteins see Brown and Waneck, 1992). Sequence Comparisons of CA TV-like Isoforms I also cloned a CA TV-like cDNA from the gut of An. gambiae, an important vector in the spread of malaria. This CA isoform (Ensembl gene ID: ENS ANGG000000 18824, chromosome 2L) is partially predicted by the Ensembl CA protein family (ENSF00000000228) as 1 of the 14 gene members found in the An.

PAGE 76

63 gambiae genome. These cloned mosquito cDNAs from^e. aegypti and An. gambiae are 61% identical to each other in amino acid residues and show similarities to the mammalian CA IV isozyme. However, in contrast to the mammalian CA IV, which is encoded by 7 exons (Sly and Hu, 1995), only 3 exons make up the An. gambiae CA isoform. Alignment of the mosquito CA IV-like isoforms from Ae. aegypti and An. gambiae with various mammalian CA IV isozymes reveals conserved features within this CA isoform (Fig. 4-1). For example, the multiple leucine (L) residues within the amino terminus of the mammalian CA IV propeptides that comprise the signal sequence are found in the Ae. aegypti and An. gambiae CA IV-like isoforms. One important feature of the mosquito CA IV-like sequences is the conserved alignment of G-69 (human CA IV numbering) with the human, bovine, and rabbit CA IV sequences. This particular amino acid residue has been changed to glutamine (Q) in rat and mouse CA IV, which results in reduced enzyme activity (Tamai et al., 1996b). Additionally, all of the CA IV sequences, including the mosquito isoforms, display a hydrophobic tail region. In addition to the conserved CA IV-like features of GPI-linked proteins, there are also conserved cysteine residues (C28 and C21 1, human CA IV numbering) between all of these CAs (Fig. 4-1). It has been determined via cysteine labeling, proteolytic cleavage and sequencing that these two cysteine residues, in the human CA TV, form a disulfide bond (Waheed et al., 1996). A second disulfide bond is present in the mammalian CA IVs between residues C6 and CI 8 (human CA IV numbering; Waheed et al, 1996). This second pair of cysteine residues, and hence the resultant disulfide bond, is not present in either of the mosquito isoforms.

PAGE 77

64 Although the mosquito CA isoforms display features similar to mammalian CA IVs, such as a 5' signal sequence, a hydrophobic 3' tail, and extracellular GPI expression, there is one striking difference in the amino acid composition of the mosquito CAs' active sites. The active site within all of the 14 characterized mammalian CA isoforms is tightly conserved. Three histidine residues (His-94, His-96, and His-1 19) are essential for CA activity through their coordinated binding of a required zinc molecule. The absence of one or more of these histidine residues results in inactive proteins called CArelated proteins (CARPs), as found in mammalian CA isoforms Vffl, X, and XI (Tashian et al., 2000). The mosquito CA IV-like isoforms contain all three of the required histidine residues along with all of the other 13 highly conserved residues found in most other CAs (refer to Fig 4-1; Tashian, 1992; Sly and Hu, 1995; Tamai et al., 1996a). However, as the alignment shows in Figure 4-1, there is a conserved gap within the active site of the mosquito CAs that is not present in any of the mammalian active sites. Because this shortened active site was found in mosquitoes but was not found in any mammalian CA isoform, I searched the Drosophila melanogaster genome for potential CA homologs. The D. melanogaster genome was found to contain 14 putative CA genes (ENSF00000000228), the same number found mAn. gambiae. One out of the fourteen D. melanogaster CA isoforms was found to contain the identical number of deleted amino acids as the mosquito forms within the active site region. This D. melanogaster CA sequence (accession number CG3940-PA) may also be a GPI-linked isoform due to the presence of a lysine-rich 5' signal sequence and hydrophobic tail region. Figure 4-2 shows an alignment of the three Dipteran CAs with shortened active site regions and all

PAGE 78

65 of the human CAs, which show several additional amino acids within the conserved active site region. Localization of CA IV-like Isoform in the Mosquito Midgut In situ hybridization analyses indicate that the Ae. aegypti CA message is expressed most heavily within the epithelial cells of the gastric caeca and posterior midgut (Fig. 4-3). An antisense cRNA probe corresponding to the entire cDNA sequence generated strong cytoplasmic staining of the proximal gastric caeca, while the distal Cap cells showed no detectable hybridization (Fig. 4-3B). Rostral to the gastric caeca, a strong localization was evident in a small subset of cardia cells that encircle the tissue, forming a collar (Fig 4-3B). These "collar cells" are clearly different from the surrounding cells in this same area. This technique also highlighted a set of specific epithelial cells that are found only in a subset of the posterior midgut. These CA-positive cells form a ring of about 5 cells in width that circumscribe the lower-posterior gut region (Fig. 4-3C). The CA message was also localized to longitudinal and circular muscle fibers of the anterior and posterior midgut (Fig. 4-4). Following the longitudinal muscle fibers, in close association, are distinct nerve fibers that also show strong CA mRNA expression (Fig. 4-4). Epithelial cells of the anterior midgut were clearly void of signal beneath the labeled muscle and nerve cells. Specific staining was also evident however within the abdominal ganglia central nervous system (CNS) and peripheral nerve tissue (Fig. 4-5). No labeling was seen in the Malpighian tubules. Real Time PCR Analysis of Aedes aegypti CA IV-like Transcripts Real time PCR was used to compare the levels of Ae. aegypti CA mRNA within specific tissue regions of the larvae. The guts of 20 fourth instar Ae. aegypti larvae were

PAGE 79

66 dissected and the head, gastric caeca (GC), anterior midgut (AMG), posterior midgut (PMG), and Malpighian tubules (MT) were pooled. RNA was isolated from each tissue sample for subsequent real time PCR analysis. Ae. aegypti ribosomal RNA (Genbank accession number M95126) was used to normalize the quantity of transcript from each sample. The results are presented in graph format in figure 4-6. Gastric caeca contain the greatest quantity of CA message within the gut sections (Fig. 4-6). The head tissue contained roughly half as much message as the gastric caeca (Fig. 4-6). The localization of CA rV-like message within the larval head is consistent with the localization of CA message to CNS tissue by in situ hybridization. The anterior midgut, posterior midgut, and Malpighian tubule collections showed CA message only marginally greater than zero (Fig. 4-6). Immunolocalization of CA IV-like Protein in the Mosquito Gut The amino terminal peptide sequence (GVTNEPERWGGQCETGRR) was chosen from the Ae. aegypti CA sequence as an antigen for antibody production. The resultant antiserum was used to immunolocalize the CA IV-like isoform within the mosquito gut. The pre-immune serum was used as a control for all experiments. Immunoreactivity was found within the gastric caeca region of the gut as well as on muscle fibers along the anterior midgut (Fig. 4-7 A). A subset of anterior muscle fibers displays the strongest and most striking labeling on their extracellular surface, while other muscle fibers show little or none. Immunoreactivity was also found within the CNS ganglia and immunoreactive nerve fibers that traverse the gut (Fig. 4-8). There was no immunoreactivity detected in the Malpighian tubules.

PAGE 80

67 Antibody Cross-Reactivity with Other Mosquito Species The 18 amino acid sequence from the Ae. aegypti CA, used to elicit the antibody response, shares 14 identical residues with the homologous An. gambiae CA protein (refer to Fig. 4-1), and therefore, the antiserum recognizes the CA IV-like isoform present in An. gambiae as well. The immunoreactivity within the An. gambiae gut displays a strikingly similar, yet species-distinctive pattern of CA TV-like protein expression (Fig. 47B). Similar to Ae. aegypti (Fig. 4-7A), not all of the muscle fibers were localized in the An. gambiae gut. However, in An. gambiae the immunolabeled muscle fiber network runs down the lateral sides of the midgut, while the dorsal and ventral muscle fibers are not immunoreactive (compare Fig. 4-7A to 4-7B). The high sequence conservation between the chosen antigenic peptide from Ae. aegypti and the An. gambiae CA, prompted us to check other mosquito species for immunoreactivity. The other species tested also displayed the same strong labeling on a subset of anterior gut muscle fibers that included both circular and longitudinal muscle fibers. Figures 4-9 and 4-1 0 display the immunoreactive results obtained from Aedes albopictus and are representative of the results from the other mosquito species including Ae. aegypti and An. gambiae. The CA TV-like immunolabeling is clearly confined to only a subset of actin-containing muscle fibers (Fig. 4-9D). Labeled phalloidin, which binds to actin, labeled all of the muscle fibers within the gut, while the antibody for CA TV-like mosquito CA only recognized a subset of the anterior muscle fibers. The CA antibody is also specific for the extracellular plasma membrane of these muscle cells, which is clearly shown by cross-sectional analysis (Fig. 4-10). Thus, there appear to be two different sets of muscle fibers within the anterior mosquito midgut is an intriguing

PAGE 81

68 discovery. This immune-reactive subset of CA-containing muscle fibers traverses the cells that surround the highly alkaline anterior gut lumen. Determining the role of these CA-specific muscle fibers holds promise for deciphering the necessary CA component of mosquito gut alkalization. Phospholipase C Treatment In order to validate the CA IV-like isoform cloned from Ae. aegypti is indeed GPI linked to the membrane, live fourth instar Ae. aegypti larvae were subjected to phosphoinositol-specific phospholipase C (PI-PLC) treatment and subsequent immunohistochemistry. This compound is capable of breaking the GPI-anchor and therefore severs GPI-linked proteins from the plasma membrane. Larvae subjected to PIPLC treatment showed a decrease in CA immunoreactivity along the midgut muscle fibers, as compared to the non PI-PLC treated controls (Fig. 4-1 1). This evidence supports the prediction that the mosquito CA rV-like isoform is in fact GPI-linked to the outer plasma membrane. Discussion In this study, we show that two GPI-linked CAs are expressed in the midguts of two different mosquito species that rely on an alkaline digestive strategy. These mosquito CAs share characteristics with the mammalian CA PV isozyme, including the GPI link to the membrane. In situ hybridization localized CA message predominantly to the gastric caeca and posterior midgut epithelial cells, as well as muscle and nerve fibers along the anterior midgut, and CNS ventral ganglia. Real time PCR analyses confirmed the presence of CA message within the Ae. aegypti gut and CNS. The gastric caeca were found to contain the greatest amount of CA message in relation to the other gut samples

PAGE 82

69 while the head sample contained roughly half of the gastric caeca concentration. CA immunoreactivity was most striking on specific muscle fibers of the anterior midgut, along with labeling of the gastric caeca and CNS ganglia. The localization of CA protein in a distinct subset of muscle fibers was found in several different mosquito species. This finding suggests that many mosquito species express a similar CA TV-like protein as well as confirms the immunoreactivity by only a subset of muscles. Immunolocalization of the CA TV-like isozyme within the mosquito gut and CNS also demonstrates that the CA message is being translated into protein and agrees with the localization of CA mRNA in the gastric caeca and muscle fibers. Interestingly, not all of the muscle fibers that show CA mRNA expression also show CA protein. Therefore, only a fraction of the muscle fibers that contain the CA mRNA are translating the message into protein. This may represent a form of regulation, in which the muscle fibers not expressing the CA protein could be "turned on" to translate the CA message if needed. This ability would be very advantageous if indeed this particular CA isoform is involved in buffering the alkaline gut. The posterior midgut was also found by in situ hybridization to express the CA mRNA. However, both the real time PCR analysis of CA mRNA expression and the immunolocalization of CA protein failed to determine the presence of this particular CA within the posterior midgut. One explanation for this may be the inability of our CA antibody to permeate the posterior midgut tissue. However, the real time analysis, which is very specific for a region of mRNA, also did not find this particular CA message in that region. The most likely cause for the in situ hybridization showing a positive CA result in the posterior midgut, is the existence of a very similar CA isoform within that

PAGE 83

70 region. Evidence for this is supported by genome data showing 14 different CA isoforms, all with regions of high nucleotide identity. The posterior midgut region does display CA activity, but apparently not as a result of the GPI-linked CA isoform presented here. The specific isoform or number of CAs contributing to the activity of the posterior midgut is still unknown. We have previously shown that the application of CA-specific inhibitors dramatically decreases the alkaline gut pH, and in fact is lethal to the larval mosquitoes (Corena et al., 2002). We now present evidence that a CA found in the mosquito gut is most similar to the mammalian CA IV isozyme but contains a novel active site motif unlike any of the mammalian CA IV isoforms (Fig. 4-1). The finding of a novel CA active site within the mosquito may facilitate the construction of a mosquito-specific CA inhibitor for use in larval mosquito control. We are hopeful that the ongoing mosquito CA crystallization project will yield further significant structural differences from the mammalian CA IV structure. These differences may be useful in the design of a mosquito-specific CA inhibitor. Out of the 14 mammalian CAs identified thus far as cytosolic, membrane-bound, secreted, and mitochondrial, only CA IV has a GPI link to the cell membrane. The localization of this highly active mammalian isozyme to dynamic tissues such as the gut, brain, kidney, and lung supports the important catalyst role of CA. It should not be surprising that the gut of a mosquito, a highly alkaline and fluctuating system, has been found to contain a presumably active CA IV-like isoform as well. The single amino acid substitution of glycine-63 to glutamine is unique to rodents (rat and mouse) CA IV, and was found to be responsible for their reduced activity rate of only 1 0-20% of the human

PAGE 84

71 CA IV enzyme (Tamai et al., 1996b). Mutating glutamine-63 to glycine within the rodent sequence resulted in almost three times greater CA activity (Tamai et al., 1996b). Unlike the rodent sequences, both of the mosquito CA IV-like sequences display the high activity glycine residue adjacent to histidine-69 (Human CA IV numbering, refer to Fig. 4-1). The task ahead is to decipher if a GPI-linked CA is better equipped to function in a highly dynamic system than other CA isoforms. Perhaps the GPI link affords the mosquito CA enzyme a characteristic advantage in buffering such an alkaline pH through its exclusively extracellular expression. Residing at the plasma membrane mtrinsically affords this isozyme the best location for monitoring C0 2 and HC0 3 " flux. Indeed, mammalian CA TVs are expressed on membrane surfaces where large fluxes of C0 2 and /or HC0 3 ' are expected (Sly, 2000). The most compelling ability of GPI-linked proteins is that they are known to elicit second messengers for signal transduction (Brown and Waneck, 1992). The alkaline pH of the larval mosquito gut was found to drop within two to three minutes after being narcotized or just simply handled (Dadd, 1975). This "handling effect" lends itself to our prediction that larval mosquitoes may exert neuronal control over the generation of the gut lumen's pH. Since a GPI-linked CA was localized within the mosquito gut and CNS tissue we propose that a GPI-linked CA may regulate the pH of the mosquito gut by severing the GPI-link and starting a signal cascade. Further studies are being pursued within the mosquito gut to encompass the localization and characteristics of other CA isoforms as well as bicarbonate exchangers.

PAGE 85

72 Aedes CA Anoph CA Human CA XV Bovine CA TV Rabbit CA IV Murine CA IV Rat CA IV Aedes CA Anoph CA Human CA IV Bovine CA IV Rabbit CA IV Murine CA IV Rat CA IV Antigenic peptide MIALFVATLL PSTIRADEWHfflPTPA — PN )GVINEPERWGGQCETGl5fa 5|PIDLTY MKSFTLLLCYALFVLHAARGDEWN XPTPG — TNGVMSEPERWGGQCDNGRRQS PIDLTI MRMLLALLALSAARPSASfAESHWe XEVQAESSNYPCLVPVKWGGNCQKD-RC S PINIVT MRLLLALLVLAAAPPQAfIaASHWC X QIQVKPSNYTCLEPDEWEGSCQNN-RC S PVNIVT MQLLFALLALGALRPLAGEELHWC YEIQA--SKYSCLGPDKWQEDCQKS-RC S PINIVT MQLLLALLALAYVAPST-EDSGWC X EIQTKDPRSSCLGPEKWPGACKEN-QC S PINIVT MQLLLALLALAYVAPST-pDSHWC X EIQAKEPNSHCSGPEQWTGDCKKN-QQSjPINIVT 1 * * * * *** m QAAVKGDFAPFLF-SNYMNPIRNAQLlSlTGHSIQIDSTDPSVTLYGGGLPGKFVIiDpMHF AAAVRGQFAPLFF-SNYMLPLKQPRVT N rGHSIQINNRDSAITMQGGGLGGRFVLD|QMHF TKAKVDKKLGRFFFSGyDK-KQTKWQNNGHSVMMLLEN-KASISGGGLPAPYQAI^QLHL GHTVMVLLEN-KPSIAGGGLSTRYQATfflLHL GHSVMVSLGD-EISISGGGLPARYRA' QHTVEMTLGG-GACIIGGDLPARYEA' ; HSVEMSLGE-DIYIFGGDLPTQYKAIjQ|LHL AKTQLDPNLGRFSFSGYNM-KHQWWI TKAEVDHSIiGRFHFSGYDQ-REARL' ARTKVNPRLTPFILVGYDQ-KQQWPI SKTKLNPSLTPFTFVGYDQ Aedes CA Anoph CA Human CA IV Bovine CA IV Rabbit CA IV Murine CA IV Rat CAIV Aedes CA Anoph CA Human CA IV Bovine CA IV Rabbit CA IV Murine CA IV Rat CA IV Aedes CA Anoph CA Human CA IV Bovine CA IV Rabbit CA IV Murine CA IV Rat CA IV Aedes CA Anoph CA Human CA IV Bovine CA TV Rabbit CA IV Murine CA IV Rat CA rv HWG SlEHTlAGVRYGQfiLHMVHHDSRYNS — LTEAGAVKNAVAVIGVLFHVSNQD HWG S EH TLDDTRYGI E LHLVHHDTRYAS — LEDAVQARNGVAVLGVLFHVGSQP HWSDLPYKGS EH SLDGEHFAM E MH IVHEKEKGTSRNVKEAQDPEDE IAVLAFLVEAGTQV KWSRAMDRGSEH SFDGERFAM E MHIVHEKEKGLSGNASQNQFAEOEIAVLAFMVEOG-SK KWSQELDRGSEH SLDGERSAM E MHIVHQKETGTSGNEVQD — SDDSIAVLAFLVEAGPTM KWSNGNDNGSEH S IDGRHFAN E MHIVHKKLTS S KED SKDKFAVLAFMIEVGDKV KWS EE SNKGSEKS IDGKHFAMSMHVVHKKMTTG-DKVQDSDSKDKIAVLAFMVEVGNEV ** *** . .*.*.** **. NTHMDVVLETSQDIRDAAGKSAPLK-GKLSPHNPLPKNRTSYFRYEGSLTT^TCAESVIH NMHIDTILDTATE IQNEVGKEALLR-GKLSPYNLLPSNRTSFYR V EGSLT I E ACAESV1 W NEGFQPLVEALSNIPKPEMSTTMAE-SSLLDLLPKEEKLRHYFB Y LGSLff I E TCDEKWW NVNFQPLVEALSD I PRPNMNTTMKEGVSLFDLLPEEESLRHYFRY LGSLT I E TCDE NEGFQPLVTALSAI S I PGTNTTMAPS SLHDLLPAEEELRHYFRY MGSLT I E ACSE NKGFQPLVEALPS I SKPHSTSTVRE SSLQDMLPPSTKMYTYFRY NGSLT I E NCDETVIW NEGFQPLVEALSRLSKPFTNSTVSE-SCLQDMLPEKKKLSAYFF^QGSLTWHGCDETVIW * * . ** ****** * * * * TVFTESLPVSLDQVELFKT IHDPSGHELVI TVFTESISVSLEQVERFKA IHDQTGRELV TVFREPIQLHREQILAFSQKLYYDKEQTVSK TVFQKP I QLHRD2I LAFSQKLFYDDQ2KVNMTI3IHVIR1PVQS LGQRQVFRdGAPGLLLAQPL TVFQEPIRLHRDQILEFSSKLYYDQERKMNMKD NVR PLQRLGDRSVFKS QAAGQLLPLPL TVYKQPIKIHKNQFLEFSKNLYYDEDQKLNMKD NV R PLQPLGKRCVFKS HAPGQLLSLPL TVFEEPIKIHKDQFIjEFSIvKLYYDQEQKLNMKIWJVRJPLQPLGNRQVFR^HASGRLLSLPL ** * * * * * * PKLSLTLIVAAIAALLAK TKMTSNWFLGAIVLLVITSRLSYH PALLGPMLACLLAGFLR PTLLAPVLACLTVGFLR PTLLVPTLACVMAGLLR PTLLVPTLTCLVANFLQ PTLLVPTLTCLVASFLH !LQPLNARALVYHTDMDYSG£|gAI IVQPUITRALVYATEWDQHGNNFA >LQQLGQRTVIK£|GAPGRPLPWAL Figure 4-1 . Alignment of several mammalian CA IV enzymes with two mosquito CA isoforms. The leucine-rich signal sequences are found in all aligned isoforms, along with the 3 essential zinc-binding histidines (red), and cysteine residues (green) that form disulfide bonds. The reduced activity in rodent CA IVs is caused by the glycine-69 mutation to glutamine (orange; Tamai et al., 1996b), which the mosquito CAs do not display. Important conserved residues are boxed. The position of mammalian signal sequence cleavage is shown and the following amino acid is residue #1 in the functional protein. Antigenic peptide sequence is also displayed.

PAGE 86

73 AAL72625 Aedes CA AAQ21365 Anoph CA CG3940-PA Dros CA P00915 Human CA I P00918 Human CA II P07451 Human CA III P22748 Human CA IV AAB47048 Human CA V CAC42429 Human CA VI P43166 Human CA VII JN057 6 Human CA VIII AAH14950 Human CA IX Q9NS85 Human CA X AAH02662 Human CA XI AAH23981 Human CA XII BAA85002 Human CA XIV FVLDQMHFHWG SEHTIAGVRYGQELHMVHHDS FVLDQMHFHWG SEHTLDDTRYGLELHLVHHDT FWEQIHMHWW SEHTINDIRYPLEVHIVHRNT YRLFQFHFHWG-STNEHGSEHTVDGVKYSAELHVAHWNS YRLIQFHFHWGSLDGQGSEHTVDKKKYAAELHLVHWNT YRLRQFHLHWG — S SDDHG SEHTVD GVKYAAELHLVHWNP YQAKQLHLHWS — DLPYKGSEHSLDGEHFAMEMHIVHEKE YRLKQFHFHWG-AVNEGGSEHTVDGHAYPAELHLVHWNS YIAQQMHFHWGGASSEISGSEHTVDGIRHVIEIHIVHYNS YRLKQFHFHWG-KKHDVGSEHTVDGKSFPSELHLVHWMA FELYEVRFHWG-RENQRGSEHTVNFKAFPMELHLIHWNS YRALQLHLHWG-AAGRPGSEHTVEGHRFPAEIHWHLST HRLEE IRLHFG — SEDSQGSEHLLNGQAFSGEVQLIHYNH HRLSELRLLFG — ARDGAGSEHQINHQGFSAEVQLIHFNQ YSATQLHLHWG-NPNDPHGSEHTVSGQHFAAELHIVHYNS YVAAQLHLHHG-QKGSPGGSEHQINSEATFAELHIVHYDS *** * * Figure 4-2. Clustal alignment of CA protein sequences. All characterized human CA isoforms are presented along with putative GPI-linked isoforms from Ae. aegypti, An. gambiae, and D. melanogaster. Three histidine residues that are required for the essential binding of zinc are shaded in blue. Note that 1 or more of these histidine residues are missing from the inactive human CArelated proteins VTA, X, and XI while all three histidines are present within the Dipteran sequences. The three CAs from Dipterans contain a shortened active site region (marked by red dashes) when compared to any of the human or other mammalian CA sequences. This difference may provide a potential target for mosquito-specific CA inhibitors, for use as larvacides.

PAGE 87

74 Figure 4-3 . Localization of C A mRNA in a wholemount preparation of early 4 instar Ae. aegypti. A. The wholemount gut preparation localizes CA message to specific cells of the gastric caeca (GC) and posterior midgut (PMG). B. A subset of cardia (arrows) and gastric caeca cells display the CA message. The distal lobes of the caeca, called Cap cells, display no staining (*). C. There is a distinctive labeling pattern of CA message within a specific band of posterior epithelial cells. In addition, numerous trachea (arrows) are heavily labeled along the length of the midgut. Scale bar represents 300 urn in A, 150 um in B, 75 urn in C.

PAGE 88

75 Figure 4-4. Expression of CA mRNA in Ae. aegypti anterior midgut. While the cardia. gastric caeca and posterior midgut display heavy epithelial hybridization, there is also specific CA mRNA expression seen in muscle and nerve cells. A. A representative whole mount Ae. aegypti larvae displaying the strong CA expression in epithelia, along with muscle fiber staining that can be overlooked at low magnification. B. The beginning of the anterior midgut shows hybridization to both muscle (arrowheads) and nerve fibers (arrows). The labeled fibers reveal striated muscle running longitudinally down the length of the anterior gut and circularly around the girth of the gut. C. The anterior midgut (AMG) displayed strong hybridization in muscle (arrowheads) and nerve fibers (arrows) while displaying no epithelial cell labeling. The posterior midgut (PMG) shows intense fiber labeling as well as epithelial cell labeling (*). The scale bar represents 300 um in A, 25 um in B, 50 um in C.

PAGE 89

76 A. B. Figure 4-5. Localization of CA IV-like message within Ae. aegypti CNS tissue. A. In situ hybridization localized the CA IV-like mRNA within all ventral ganglia CNS clusters (arrows) as well as hair sensory cells (*) and longitudinal nerve fibers (arrowheads). B. The sense control probe displayed no specific hybridization. Scale bar represents 300 um.

PAGE 90

77 Aedes aegypti Carbonic Anhydrase Ae. aegypti Tissue Sections Figure 4-6. Relative quantification of CA IV-like message in Ae. aegypti larvae using real time PCR. The gastric caeca tissue displays the greatest amount of CA IV-like message. Data was normalized to the gastric caeca (GC) sample. The anterior and posterior midgut along with the Malpighian tubules display very little CA message. The head section displays roughly half the amount of message found in the gastric caeca.

PAGE 91

78 Figure 4-7. Ae. aegypti and An. gambiae CA protein labeling. The antibody generated against the Ae. aegypti CA can also be used to localize the homologous CA isoform within An. gambiae. The larvae were incubated with phalloidin (red) and the CA-specific antiserum (green). Colocalization of the red and green signals appears yellow. A. The antibody localization shows the strongest labeling in Ae. aegypti for a subset of muscle fibers in the anterior midgut and the proximal portions of the gastric caeca. B. Antibody localization of An. gambiae CA is depicted by the yellow muscle fibers while the red muscle fibers are not recognized by the antibody. The scale bar represents 300 urn in A, 150 urn in B.

PAGE 92

79 Fig 4-8. The Ae. aegypti CNS ganglia express the CA lY-like isoform. A. Pre-immune serum does not show any detectable labeling of the CNS tissue. B. Strong immunolabeling for the mosquito CA IV-like isoform is displayed in the ventral ganglion clusters, as displayed by the fluorescent green coloring as compared to the yellow control (pre-immune) ganglia. The scale bars represent 100 um.

PAGE 93

80 Figure 4-9. Iinmuiiolocalization of mosquito CA IV-like enzyme ^ A ^ db ^?^\ A Muscle and nerve fibers within the anterior midgut region are heavily labeled. A Selective labeling of particular muscle fibers (green). B. Labf ed phalloidin (red) was used to localize actin and labels all muscle fibers, including those that were not recognized by the antibody against mosquito CA C A nuclear label (blue) was used to distinguish cell numbers present. D Overlay of all three signals. Colocalization of the green and red signals appear yellow. The CA antibody recognizes only a subset of anterior muscle fibers, and is seen in several different mosquito species. The scale bar represents 50 um.

PAGE 94

SI B Figure 4-10. High magnification of immunoreactive muscle fibers within the Aedes albopictus midgut. A. Labeling of muscle fibers appears to be extracellular. B. Cross section of the same fiber demonstrates that the localization pattern is confined to the extracellular plasma membrane of the midgut muscle fibers. The scale bars represent 25 um.

PAGE 95

82 Figure 4-11. Immunoreactivity of Ae. aegypti guts for the CA IV-like isozyme. The red labeling is specific for muscle fibers. The green labeling shows localization of the mosquito CA IV-like protein. The yellow labeling shows colocalization of the mosquito CA IV-like isoform and actin. Prior to immunolabeling, the guts were treated with PI-PLC to determine if the CA IV-like isoforms are GPI-linked to the cell membrane. A. Immunolabeling of the gastric caeca, without the PI-PLC treatment, displays heavy yellow labeling of the GPI-linked CA isoform. B. After PI-PLC treatment there is no CA IV-like immunolabeling of the gastric caeca. C. The anterior midgut displays the immunolocalization of the CA IV-like isoform along a subset of muscle fibers. D. After PI-PLC treatment the yellow immunolabeling for the GPI-linked mosquito CA is greatly reduced. The decreased immunolocalization within the gastric caeca and anterior midgut signifies that the PI-PLC was successful in severing the GPI-link and releasing the CA from the membrane association. The scale bars represent 1 00 um.

PAGE 96

CHAPTER 5 ANION EXCHANGER EXPRESSED WITHIN THE LARVAL ANOPHELES GAMBIAE MOSQUITO Introduction The larval mosquito gut provides an ideal model for studying epithelial transport due to its cellular simplicity, being only one cell layer thick. The transport of bicarbonate within the larval mosquito gut was prompted by several studies (Boudko et al., 2001 a,b; Corena et al., 2002). Bicarbonate is the main pH buffer in most complex organisms (Sterling and Casey, 2002) so it was predicted that de-protonated bicarbonate (ie. carbonate) is necessary for attaining the highly alkaline lumen of the larval mosquito midgut. However, the anterior midgut was found to apparently lack an active CA enzyme (Corena et al., 2002). An alternative to bicarbonate being rapidly produced within the anterior midgut, is the transport of bicarbonate into the anterior midgut. This alternative was supported by a previous study which implicated a chloride/ bicarbonate anion exchanger within the larval mosquito gut through the use of self-referencing ionselective (SERIS) microelectrodes (Boudko et al., 2001b). However, the molecular identity of the chloride/ bicarbonate anion exchanger (AE) was not determined. We now present the first anion exchanger (AE) to be cloned and characterized from the An. gambiae mosquito, in an attempt to unravel the physiology of an extremely alkaline digestive system. Localizing an AE within the larval mosquito gut is important because intracellular pH is known to be regulated by exchangers of bicarbonate and chloride (Phillips and 83

PAGE 97

84 Baltz, 1999). AE localization will also distinguish whether the AE is co-expressed within the same epithelial cells as the V-ATPase and if the polarity of basal or apical expression is also the same. The localization of a it V-ATPase within the larval mosquito gut was found to be apical in the gastric caeca, basal in the AMG, and then apical again in the PMG (Zhuang et al., 1999). The highly dynamic system of alkaline digestion in the larval mosquito gut does not exist in any known mammalian system. However, mammalian organs such as the kidney are able to perform many parallel functions of the mosquito gut, such as water regulation, filtration, and ionic homeostasis. The CA, AE, and it V-ATPase proteins, in particular, have been extensively studied and localized within the mammalian kidney due to their dynamic roles in acid-base balance (Huber et al., 1999; Schwartz, 2002). The colocalization and polar expression of an AE with a it V-ATPase will define the epithelial cells of the mosquito gut as resembling the mammalian kidney A-intercalated cell type, the B-intercalated cell type, or the non-A non-B intercalated cell type (types as defined by Brown and Breton, 1996 and Kim et al., 1999). Results An. gambiae AE Sequence Analysis The full length An. gambiae AE1 (AgAEl) cDNA was cloned from midgut tissue and contains 3309 bases (accession number AY28061 1) with a molecular weight of 123 kDa for the predicted protein. The NCBI conserved domain search tool (CDD) determined that the protein sequence was part of a family of bicarbonate transporters (BT) and cotransporters (PF00955) as well as sodium-independent chloride/bicarbonate exchangers and related sodium/bicarbonate cotransporters (KOG1 172; Geer et al., 2002).

PAGE 98

85 This complex of BTs contains both the solute carrier 4A (SLC4) and solute carrier 26A (SLC26A) proteins. More specifically, the EnsEMBL database places this particular AE cDNA sequence within the anion exchange/ band3 protein family (ENSF00000000189) as 1 of the 8 putative anion exchange band3 transcripts (ENSANGP00000010112) encoded in the An. gambiae genome. These 8 transcripts arise from 3 different genes (ENSANGG00000007623, ENSANGG00000004501, and ENSANG000000 12483). The gene that gives rise to the cloned AE that we are presenting (ENSANGG00000007623) is located on chromosome 3R. The other 2 genes (ENSANGG00000004501 and ENSANG000000 12483) are located on chromosome 2L (Hubbard et al., 2002; Clamp et al., 2003). The 1 102 amino acids comprising the An. gambiae AE form a cytosolic framework at the amino terminus while the carboxy terminus is composed of 12 transmembrane spanning domains also with an intracellular cytosolic terminus (hmmtop v.2; Tusnady and Simon 1998; Tusnady and Simon 2001). This hmmtop prediction was generated based on two assumptions: 1), that the CA binding site is within the carboxy terminus; and 2), that the C-terminus is intracellular, as is found for all known AEs. This structure is consistent with the predicted structure of the Drosophila melanogaster sodium dependent anion exchanger (NDAE1), which consists of a 12 membranespanning pattern with intracellular carboxy and amino termini (Romero et al., 2000). The highly conserved sequence identity of the An. gambiae AE1 with respect to the D. melanogaster NDAE1 allows the predictive 12 transmembrane-spanning domains of the D. melanogaster protein to be superimposed upon the An. gambiae protein (Fig. 5-1).

PAGE 99

86 The amino terminus of the AgAEl protein contains 523 cytoplasmic residues enriched with multiple binding sites for cytoskeletal proteins (Bairoch et al., 1997). The carboxy terminus contains the membrane-spanning domains that are responsible for ion transport. According to the PROSITE motif search and the hmmtop server, sites of potential post-translational modification include 2 cAMP and cGMP-dependent protein kinase phosphorylation sites, 16 protein kinase C phosphorylation sites, 1 1 casein kinase II phosphorylation sites, 16 N-myristoylation sites, 1 prokaryotic membrane lipoprotein lipid attachment site, and 1 leucine zipper pattern (Bairoch et al., 1 997; Gupta et al., 2002). The last 82 amino acids of the An. gambiae AE carboxy terminus are predicted to project into the cytosol, the correct orientation that is expected for binding of a cytosolic CA. The CA II binding site comprises a hydrophobic amino acid residue followed by at least two acidic residues within the next four residues (Vince and Reithmeier, 2000). Figure 5-2 displays the CA II binding sites found in several AE proteins along with the putative CA II binding site (LDDIM) in the An. gambiae AE. BT Sequence Comparisons Following the sequence prediction that this An. gambiae BT is an AE, the closest characterized protein sequence is the NDAE1 (accession number AAF98636) from D. melanogaster. The An. gambiae AE1 shares 72% identity with NDAE1 (Fig. 5-3). However, the greatest similarity is with an uncharacterized splice form of NDAE1 (AAF52497) that contains an inserted sequence that is also found in the An. gambiae AE sequence (Fig. 5-4). Other BTs such as the sodium bicarbonate cotransporters (NBCs) and AE4 show 45%-52% identity to AgAEl, AE1-3 exhibits 36% identity, and the BTs

PAGE 100

87 that are also capable of transporting sulfate (SLC26A group) show only 1 1% amino acid identity (Fig. 5-3). Anion exchangers, specifically AE2 and AE3, were determined to be pH sensitive. AE3 is stimulated by intracellular alkalization whereas AE1 is not. More specifically, a region of amino acids (WRETARWIKFEE) within the carboxy terminus is responsible for the pH sensitivity seen in AE2 (Vince et al., 2000). The An. gambiae AE sequence in this same region contains 14 of the 16 residues found within AE2 while the other two amino acids are conserved (Fig. 5-5). AE1 shares only 8 of the 16 amino acids in this region. Localization of Anion Exchanger mRNA in An. gambiae Larvae A DIG-labeled antisense cRNA probe comprising the full length AE cDNA was employed to localize the AgAEl mRNA. A DIG-labeled sense probe was used as a control. The AgAEl mRNA was found in every region of the larval gut including gastric caeca, anterior midgut, posterior midgut, Malpighian tubules, and rectum (Fig. 5-6). In the gastric caeca and posterior midgut regions, the probe was localized to epithelial cells. Within the gastric caeca the labeling is most intense in the area where the lobes face the lumen. The gastric caeca labeling was confined to the proximal cells, whereas the Cap cells displayed no label (Fig. 5-6A,B). The rectum displayed staining in a small subset of epithelial cells along with tracheoles (Fig. 5-6D). Muscle, nerve, and trachea cells that traverse the outer plasma membrane of the anterior gut epithelial cells were labeled with the AE antisense probe (Fig. 5-7). Labeled tracheal fibers are displayed in close association with the gastric caeca (Fig. 5-8). These

PAGE 101

88 trachea extend from the gastric caecal region and become incorporated with the anterior midgut where they are intimately associated with nerve fibers (Fig. 5-9). The third abdominal segment marks the beginning of the posterior midgut and is located by the fourth pair of trachea that connect to this part of the midgut. This junction displays a marked contrast in AgAEl mRNA transcript expression. Strong staining of the epithelial cells begins here, coinciding with the beginning of the posterior midgut and change in epithelial cell morphology (Fig. 5-10). Along with small epithelial cell labeling, tracheal fibers also display strong label for AgAEl mRNA within the posterior midgut. The larger type of epithelial cell within the posterior midgut, the columnar cell, shows extensive labeling for AgAEl mRNA expression, with signal localized near the plasma membranes (Fig. 5-1 OB). The extreme end of the posterior midgut displays strong labeling within a cluster of small epithelial cells known as cuboidal cells (Fig. 511). Cellular processes that extend rostral and lateral from these cells are also labeled. All cells of the Malpighian tubules label positively for AgAEl transcript expression (Fig. 5-12). AgAEl mRNA expression is also localized to the ventral midgut ganglion. Each ventral ganglion displays specific labeling within one or two longitudinally directed neurons that traverse the same plane (Fig. 5-13). No other neuronal cells display signal for AE mRNA expression. The labeling is very specific for precisely one neuronal pathway within each ganglion (Fig. 5-13). The sense (control) DIG probes display no hybridization (Fig. 5-14).

PAGE 102

Antibody Localization of AE Protein Two antigenic peptide sequences (EVRKRPPEKNPKEEIDEE and KPKQQPVTTISVTKVAEQ) were chosen from the cytosolic framework of the amino terminus and the anion exchange carboxy terminus respectively, of the translated AE cDNA (accession number AAQ21364). Chickens were used to produce antibodies against these antigenic peptides. The resultant chicken antisera were used to localize the AgAEl protein within whole mounts of the An. gambiae fourth instar larvae. The AgAEl protein was localized to the plasma membranes of the gastric caeca (Fig. 5-15) and posterior midgut epithelial cells (Fig. 5-16). A three-dimensional reconstruction of the cellular localization of AgAEl enabled us to discriminate the immunolabeled basolateral membranes from the non-labeled apical membranes. Antisera for both peptides displayed the same basolateral immunolocalization pattern. Neuronal cells within the AMG displayed immunoreactivity (Fig. 5-17). Pre-immune sera displayed no specific immunoreactivity. AE Functional Expression in Oocytes In order to ascertain the functional characteristics of the AgAEl, the protein was expressed mXenopus oocytes. The AgAEl was subcloned into the pXOOM vector (Jespersen et al., 2002; generous gift from Dr. T. Jespersen) for oocyte expression. The Xenopus oocytes were injected with either AgAEl RNA or water to serve as the control. Three to seven days post-injection the oocytes were tested for AgAEl expression using two-electrode voltage clamp electrophysiology. Oocytes expressing AgAEl displayed a decreased volume as compared to the water-injected controls (T. Seron, unpublished observation), which correlates with the AE regulatory functions of cell pH and volume.

PAGE 103

The AgAEl was determined to be a functional protein with the capacity to transport chloride. No sodium ion or potassium ion dependence was determined with the voltage clamp assay. A comparison of current versus voltage (I-V plots) for both the AgAEl expressing oocytes and the water injected controls are compared when two different bath solutions are applied to the oocytes. The I-V plots for the water-injected controls displayed no transport with or without chloride (Fig. 5-1 8A). When the solution contains chloride (N98), the AgAEl expressing oocytes are capable of transporting chloride, as seen by the steep rise in current (Fig. 5-1 8B). When chloride is replaced by a non-ionic equivalent (N98-C1) there is no transport (Fig. 5-1 8B). The transporter blockers, 4,4'diisothiocyanodihydrostilbene-2,2'-disulfonate (DIDS) and niflumic acid (NA) both inhibited the transport capabilities of the expressed AE1 protein (NA not shown). The application of DIDS inhibited the transport of chloride such that the I-V plot showed an affect similar to the removal of chloride ions from the bath solution (Fig. 5-1 9A). The likeness of removing chloride and the inhibitory affect of DIDS is easily viewed by comparing the differences in current between the DIDS inhibition and the removal of chloride (Fig. 5-19B). When the difference between chloride and chloride removal are compared to the blocked and chloride removal transport, a large difference can be seen (Fig. 5-19B). Activity of the mosquito AE1 was also inhibited when the CA-specific inhibitor acetazolamide was added to the media (data not shown). Because acetazolamide is known to have no direct inhibitory effect on AEs, unlike other sulfonamides, it can be inferred that the AgAEl is inhibited by acetazolamide due to its tight coordination and regulation by endogenous CA, as was found to be the case in mammalian systems (Sterling et al., 2001a). Endogenous CA was bound by mammalian

PAGE 104

91 AEs when they were expressed in HEK293 cells, and was shown to increase the rate of bicarbonate transport. Furthermore, co-expression with mutant CA II (non-active) was shown to result in decreased bicarbonate transport due to the displacement of the active endogenous CA. The CA II binding motif found in this mosquito AE could similarly bind the oocyte's endogenous CA, also resulting in an increased transport rate. This may explain the inhibition seen in AE transport when acetazolamide was applied. AgAEl expression studies in oocytes are presently ongoing to further assess the function and inhibition of this protein. Discussion An AE cDNA was cloned from fourth instar, larval An. gambiae gut tissue. The translated 123 kDa protein is predicted to consist of intracellular amino and carboxy termini and 12 transmembrane segments. In situ hybridization and antibody immunolocalization identified AE mRNA message expression and protein localization within epithelial cells of cardia, gastric caeca, posterior midgut, rectum, and Malpighian tubules as well as tracheal, nerve, and muscle cells. Expression of AgAEl in Xenopus oocytes displayed a reversible transport of chloride. The most similar characterized protein sequence is the NDAE1 from D. melanogaster. Xenopus oocyte expression with pH analysis determined that this protein was sodium ion dependent. The An. gambiae AE1 protein has 72% sequence identity to the NDAE1 but does not display sodium ion dependence. The carboxy termini of these proteins show little similarity and therefore the carboxy terminus of NDAE1, unlike AgAEl, may contain the necessary domain for sodium ion dependence (refer to Fig. 5-4).

PAGE 105

92 InXenopus oocyte expression tests of AgAEl, activity was decreased with a CAspecific inhibitor, acetazolamide. Bicarbonate, rapidly formed by the hydration of carbon dioxide by CA, is a substrate for the AE. The decreased ability to exchange ions in the presence of acetazolamide leads to the prediction that this mosquito AE is directly regulated by CA activity. There is evidence within the mammalian system for the tethered coordination of anion exchangers with carbonic anhydrase. The mammalian anion exchanger (AE1/ band3) has been shown to interact with and actually bind to CA II at its carboxy terminus (Sterling et al., 2001b). AEs and in fact all bicarbonate transporters (except DRA) identified to date have potential CA E-binding sites at their carboxy termini (Sterling et al., 2002b) including this An. gambiae AE. The inhibitory effect of acetazolamide on AgAEl expressed mXenopus oocytes suggests that this mosquito AE is coupled with an active CA enzyme, speculatively an endogenous Xenopus CA protein. A protein complex consisting of a membrane-spanning AE and a cytoplasmic CA has the ability to maintain tight pH homeostasis both inside and outside of the cell at the same time. Furthermore, this complex brings together bicarbonate production and transport in such a way that virtually all lag time is abolished by the tethered coordination of the system (Sterling et al., 2001b). This type of bicarbonate transport metabolon, if found to exist within the mosquito, may explain how the mosquito gut is capable of driving and supporting a pH greater than 1 0 within the lumen while sustaining a near neutral pH within the adjacent cells. Now that an AE has been localized to CA active regions within the mosquito gut, namely the gastric caeca and posterior midgut, it will be necessary to determine whether they form a bicarbonate metabolon as proposed.

PAGE 106

93 Several acid-base controlling proteins have now been identified within the mosquito gut (Zhuang et al., 1999; Corena et al, 2002). The distribution of these proteins along the length of the mosquito gut is both heterogeneous and discontinuous. Along the length of the gut, non-adjacent regions such as the gastric caeca and posterior midgut display similar protein expression and CA activity, while the region that separates them, the anterior midgut, displays a different pattern of protein expression. This is not surprising as the highly alkaline pH of the anterior midgut also contrasts with the nearly neutral pH of the flanking gut regions. The most intriguing part of the novel expression profile of these mosquito proteins is the parallel expression profile for the same proteins within the mosquito midgut and the well-characterized mammalian kidney. As characteristic of most epithelia, the epithelial cells found in both the mammalian kidney and the mosquito midgut share several specific morphological features. Both populations are mitochondria-rich, display apical microvilli, contain active cytosolic CA activity, and express a proton translocating PI* V-ATPase on specific domains of their plasma membranes (Clements, 1992; Sterling et al., 2001a). Similarly, cell polarity proteins can also be compared between the A (alpha) intercalated cells of the collecting tubules of the mammalian kidney and the mosquito epithelial cells of the gastric caeca and posterior midgut. The A cell subpopulation of kidney intercalated cells expresses, and is defined by, an apical FT V-ATPase, and a basolateral AE (Matsumoto et al., 1994). The epithelial cells of the gastric caeca and posterior midgut also express an apical Ff VATPase (Zhuang et al., 1999), and a basolateral AE. Furthermore, B (beta) intercalated cells of the mammalian kidney are defined by a PT f V-ATPase expressed within the basolateral plasma membrane and an apical AE. This apical AE is different from AE1,

PAGE 107

94 but has yet to be cloned and characterized (Kim et al., 1999). Like B cells, anterior mosquito midgut cells have been shown to express a basolateral V-ATPase whereas expression of AgAEl was absent from this region. The closely associated A and B intercalated cells of the mammalian kidney are known to function in acid secretion and bicarbonate secretion, respectively. Interestingly, the bicarbonate-secreting moiety, for which the function of the B cell is defined, is unknown. The co-occurrence of these A and B cell types provides for the tight regulatory control of the maintenance of near-neutral pH in the kidney. Perhaps the pH differential of 4 units, as seen in the mosquito anterior midgut, is achieved by the decoupling of these cell types. In the mosquito gut, the A-like cells of the gastric caeca and posterior midgut translocate protons toward the lumen and reduce the alkaline pH to near neutral levels. The B-like cells of the alkaline anterior midgut secrete bicarbonate and protect the cells from the high lumenal pH. Although the presented AgAEl is not that moiety, the parallels between cells of the mammalian kidney and the mosquito midgut are evident. The similarities between epithelial cells of the mammalian kidney and the mosquito midgut support the use of the mosquito midgut as a simple model in which to study cell polarization, pH balance, protein targeting and trafficking, as well as disease states. The decoupling of A and B intercalated cells, as they may exist in the mosquito midgut, provides an excellent model for studying human diseases such as distal renal tubular acidosis, which is caused by mutations in either the basolateral AE1 or different subunits of the apical If" V-ATPase (Alper, 2002). The simple epithelium of the mosquito midgut may continue to reveal mechanisms and pathways that also function within the complex metabolic network

PAGE 108

95 comprising the mammalian kidney. Many studies have sought to determine specific amino acids responsible for pH sensitivity within the anion exchangers. The sixteen amino acid pH sensitive region of AE2 is almost identical to the comparable region within the AgAEl sequence (refer to Fig. 5-5). The mosquito midgut provides an excellent model for studying precisely the pH dependent moieties and proteins due to the large pH gradient that it supports. The one cell layer epithelium that divides the alkalinity of the lumen (pH 1 1) from the neutral pH of 7-8 within the cell cytosol has yet to reveal the cell polarity that is capable of maintaining this system. The elucidation of such a metabolon within the mosquito gut, in which an AE is directly tethered to one or more CAs, such as in the mammalian system (Sterling et al., 2001a), would provide a mosquito model which could be used as a simple framework for uncovering metabolic networks within complicated mammalian systems such as the kidney.

PAGE 109

96 Multiple Alignment 1 281 561 841 1121 B Figure 5-1 . Structural prediction of the An. gambiae AE1 . A. Hydrophobicity plot of the DNAman-aligned D. melanogasterNDAEl and^«. gambiae AE1 sequences suggests a nearly identical protein topology of 12 membrane-spanning domains in both proteins. B. Illustration depicting the intracellular location of both protein termini as well as the predicted 12 transmembrane domains.

PAGE 110

97 AgAEl [AAQ21364] LDYIFTKRELKILDDIMPEMTKRARADDLHQLEDGEVG Human AE1 [P02730] LPLIFRNVELQCLDADDAKATFDEEEGRDEYDEVAMPV Human AE2 [P04920] LTRI FTDREMKCLD ANEAE PVFDEREGVDE YNEMPMPV Human AE3 [NP005061] LPRLFQDRELQALDSEDAEPNFDE-DGQDEYNELHMPV Human AE4[Q96Q91] LERVFS PQELLWLDELMPEEERS I PEKGLEPEHSFSGS * * *_ ** Figure 5-2. Putative amino terminus CA II binding motif. The highlighted conserved leucine (L) was shown to be necessary for the specificity of CA binding in AE1 (Vince et al., 2000) and is also present in AgAEl . The motif consists of at least two acidic amino acids within the four residues following the conserved non-polar L. The ability of AEs to bind CA enzymes greatly raises their ability to regulate ionic homeostasis. Identical residues (*) are noted in the alignment as well as conserved residues (.).

PAGE 111

98 100% I 80% I 60% I 40% 20% 0% Human AE2 Human AEl Human AE3 Human NBC4a Human AE4 Dros NDAE1 Anoph AEl Human NBC8 Human SLC26A7Human DRA Human SLC26A660% 58% 72% 53% 45% 36% 31°/1 22% 11% Figure 5-3. Homology tree depicting the amino acid identity between several BTs. The An. gambiae AEl amino acid sequence displays the closest identity to the D. melanogasterNDAEl (AAF98636) and human NBC8 (NP004849) sequences with 72% and 53% identity, respectively. The human AEs display 36% to 45% identity while the sulfate transporters (SLC26 group) display only 1 1% identity to AgAEl. Accession numbers: human AE2 (P04920), human AEl (P02730), AE3 (NP005061), NBC4a (NP067019), AE4 (Q96Q91), SLC26A7 (NP439897), DRA (P40879), and SLC26A6 (Q9BXS9).

PAGE 112

99 AgAEl AAQ21364 MMDHGWDEEAPIDPRLKMRTFTADQD FEGHRAHTVFVGVHIPGSSR 47 Dros AAF52497 MAEKNEYIELPMTMNSSSGDDEAPKDPRTGGEDFTQQFTEKDFEGHRAHTVYVGVHVPGG-R 61 NDAE1 AAF98636 MAEKNEYIELPWTMNSSSGDDEAPKDPRTGGEDFTQQFTENDFE 24 AgAEl AAQ21364 RHSQRRRHKHHQASRENGDKGSTG SEAERPVTPPAQRVQFILG 90 Dros AAF52497 rhSQRRRKHHHSGPGGGGGGGGGGGSIGGSGSVGGGAGKDNVSEKQQEVERPVTPPAQRVQFILG 126 NDAE1 AAF98636 — VTPPAQRVQFILG 57 Figure 5-4. Alignment of carboxy terminus amino acids of An. gambiae and D. melanogaster AEs. The characterized D. melanogasterNDAEl (AAF98636) is 72% identical to our An. gambiae AE1 sequence. The greatest number of amino acid differences occurs at the carboxy terminus, the regulation domain. However, an uncharacterized splice variant of NDAE1 (AAF52497) displays an inserted sequence at the carboxy terminus that the characterized protein does not. This inserted sequence shows similarity to the AgAEl sequence and therefore is the closest predicted protein to AgAEl.

PAGE 113

100 Anoph AE1 Human AE1 Human AE2 Human AE3 Human AE4 GDEMAWKETARWVKFEEDVEEGG NOEL RWMEAARWVQLEENLGEN G NQEPdWRETARWIKFEEDVEdET S OF. P tijwRE TARW I KFEE DVEEj E T SITLSTHLHHRWVLFEEKLEVAA * Figure 5-5. Alignment of An. gambiae and human AEs. pH sensitivity of AE2 (P04920) and AE3 (NP005061) was mapped to the boxed 16 amino acids shown (Vince et al., 2000). Unlike AE3, AE1 (P02730) was not stimulated by intracellular alkalization (Vince et al., 2000). The An. gambiae AE1 (AAQ21364) sequence shows a strong similarity to the identified amino acid sequence and has 14 identical residues. The human AE1 sequence has only 8 identical residues. pH sensitivity of an AE within the mosquito gut would be an important attribute due to the regional compartmentalization of the pH flux. The recently identified AE4 (Q96Q91) was included in this alignment for completeness. It displays the least conservation with only 6 identical residues. Stars indicate identical residues within all aligned sequences.

PAGE 114

101 Figure 5-6. Localization of AgAEl mRNA within whole mount An. gambiae larvae. A. Gastric caeca (GC), posterior midgut (PMG), and Malpighian tubules (MT) show extensive expression while other gut regions display more restricted hybridization. B. The gastric caeca as well as the cardia region (arrows) display label. C. Extensive expression of AgAEl mRNA in the Malpighian tubules. D. Specific cells (arrows) and trachea (arrowheads) of the rectum show expression of AgAEl . Scale bars represent 300 um in A, 100 um in B, and 25 um in C and D.

PAGE 115

102 Figure 5-7. Localization of AgAEl mRNA in muscle, nerve, and trachea in An. gambiae. A. The whole mount gut preparation localizes AE message to specific muscle, nerve (long arrow), and trachea (short arrow) fibers of the anterior midgut (AMG). B. A high magnification of the AMG region detailing the tracheal fibers (short arrows) and neuronal cells (long arrow) that express the AgAEl message. Scale bar represents 75 (am in A and B.

PAGE 116

103 Figure 5-8. In situ hybridization of AgAEl in whole mount An. gambiae consistently shows positive labeling of tracheal fibers along the midgut. A. A thick tracheal stalk penetrates the gastric caeca while thinner branches join the AMG. This main tracheal stalk that is closely associated with the gastric caeca consistently displays labeled particles that may be secretory vesicles (arrow). B. Labeled trachea (arrows) traverse the AMG and coincide with labeled nerve fibers (arrowheads) that extend down the midgut. C and D. Labeled trachea (arrows) are randomly associated with the area surrounding the gastric caeca. Scale bars represent 25 um.

PAGE 117

104 Figure 5-9. Anion exchanger mRNA localization reveals trachea and nerve fibers along with neuronal cell labeling. A. Trachea (large arrow) display a random pattern of distribution on the An. gambiae midgut along with pairs of neuronal cells (small arrows). B. A labeled tracheal stalk (large arrow) shows abundant labeling where it joins with the midgut (*). The finer branches of the trachea can be seen joining (small arrow) the parallel nerve fibers (arrowhead) that also display label. C. Neuronal cells (small arrows) scattered over the AMG display AE label along with the nerve (arrowheads) and tracheal fibers (large arrows). D. Strong labeling is consistently seen where the thick trachea connects to the midgut (*) and sends out smaller trachiole fibers (arrows). Scale bars represent 25 urn in A, 50 urn in B, 50 urn in C, and 50 urn in D.

PAGE 118

105 Figure 5-10. Localization of AgAEl mRNA to the PMG of larval An. gambiae. A. The whole mount gut preparation displays the unlabeled AMG on the left side of the photo as compared to the labeled PMG on the right side of the photo. The arrows point to the tracheal stalks that join the gut at the third body segment, corresponding to the beginning of the PMG region. B. Outer margins of large columnar PMG cells display AE mRNA labeling (*) along with labeling of tracheal fibers (arrows) and small cuboidal cells (arrowheads). Scale bar represents 50 um in A, 25 um in B.

PAGE 119

106 Figure 5-1 1 . Larval An. gambiae displays strong AgAEl expression in the hindgut, the pylorus. A. Distal to the joining of the Malpighian tubules with the gut, the pylorus displays pronounced labeling of small epithelial cells (arrowhead) and closely-associated muscle fibers. B. Higher magnification of the boxed region displays labeled epithelial cells (arrowheads) of the pylorus with closely associated circular muscle fibers (arrows) that form a pyloric sphincter. The pylorus, a part of the hindgut, functions in ionic and osmotic regulation. Scale bars represent 25 um in A and B.

PAGE 120

107 Figure 5-12. Localization of AE mRNA in An. gambiae shows abundant labeling of the Malpighian tubules. A. The entire length of the Malpighian tubules displayed labeling of AE mRNA. B. Labeled tracheal fibers are also associated with the Malpighian tubules. C. Labeled tracheal fibers also extend from the tips of Malpighian tubules and may contain secretory vesicles (arrow). Scale bar represents 25 um in A, B, and C.

PAGE 121

108 Figure 5-13. Expression of AE mRNA was found throughout the ventral midgut ganglia. Between the ventral midgut and the ventral integument a labeled neuronal pathway connected each ganglia to the next. A. Two neurons (arrows) label positively for the AE mRNA, rendering them highly visible above the other neuronal cells within the unstained ganglia. B. Between each ganglia cluster, a tissue crossbridge (*) passes in a 90° angle between the ganglia and the gut. The neuronal pathway showing AE mRNA expression makes contact with this junction and continues on to the next ganglia cluster on the other side. C. Two neuronal cells within the following ganglia show clear expression of AE mRNA. D. This panel shows an unobstructed view of a single triangular-shaped ganglia cluster. It is clear that the AE mRNA is located within a distinct population of neuronal cells in each ganglion. Scale bars represent 50 um.

PAGE 122

109 Figure 5-14. Sense AE probes display no specific hybridization. A. These whole mount preparations show no AE sense (control) label in the integument, midgut or hindgut (B). A higher magnification of the gastric caeca (C) and posterior midgut (D) with Malpighian tubules (MT) shows no hybridization with the sense AE probe. These experiments were performed side by side with the antisense probes and therefore length of exposure was identical. Scale bars represent 600 um in A, 300 um in B, 100 um in C, and 100 um in D.

PAGE 123

110 A c E 1 B| D m m f \ WW' ' W r f|| |H -d Figure 5-15. Antibody localization of AgAEl protein to the gastric caeca in^n. gambiae larvae. A. Our AE specific antibody displays immunoreactivity within the cardia (*) and gastric caeca. B. Phalloidin was used to label the actincontaining muscle fibers throughout the mosquito gut. C. Draq-5 was used to label nuclear DNA. D. Three signal overlay depicting the AE protein in relation to muscles and cell nuclei. E-H shows higher magnification views of the gastric caeca with the same labeling profile as in A-D. AE protein expression can be seen on plasma membranes (arrows) of the gastric caeca (E). Scale bars represent 25 um in A-D and 100 um in E-H.

PAGE 124

Ill Figure 5-16. Localization of AgAEl protein within the PMG of An. gambiae larvae. A. Our AE specific antibody displayed immunoreactivity within the PMG; most prominantly within a specific band of cells that encircle the PMG region. B. Phalloidin was used to label muscle fibers. C. Draq-5 was used to localize nuclear DNA. D. Overlay of AE labeled cell membranes in relation to muscle fibers and nuclei. E. High magnification of AE protein localization within the PMG. Cell membranes of both large and small cells are clearly labeled with our antibody. Scale bars represent 150 um in A-D and 75 um in E.

PAGE 125

112 Figure 5-17. Neuronal cells within the AMG display immunoreactivity for our An. gambiae AE specific antibody. These neuronal cells also displayed AE mRNA expression (refer to Fig. 513 A) and are most often seen in pairs (*). Scale bar represents 50 um.

PAGE 126

113 i 1 1 1 1 — 29, 4 I 1 1 1 1 -100 -80 -60 -40 -20 0 20 40 60 80 Voltage (mV) B. AgAE1 Expressing Oocyte 60 N98 N98-CI I — 20 — i — 40 -20 0 Voltage (mV) 80 Figure 5-18. Current-voltage (I-V) plots depicting ion transport by the AgAEl expressing oocytes in contrast to the water injected control oocytes. A. When chloride is removed from the solution bathing the control oocyte and replaced with a nonionic equivalent there is no change in the slope of the curve, signifying no ionic transport. B. When chloride is removed from the media surrounding the AgAEl expressing oocyte, the ionic transport is eliminated, as seen by the decrease in slope.

PAGE 127

114 Figure 5-19. Inhibition of AgAEl mediated chloride transport by DIDS. A. DIDS blocks the transport of chloride by AgAEl . The result is similar to taking chloride out of the bath solution. B. There is almost no difference between blocking chloride transport with DIDS and removing chloride. In contrast, removing chloride from the uninhibited exchanger shows a large difference in chloride transport.

PAGE 128

CHAPTER 6 CYTOSOLIC CA EXPRESSION IN LARVAL ANOPHELES GAMBIAE Introduction Carbonic anhydrase (CA) represents a superfamily of enzymes that reversibly hydrates carbon dioxide to form bicarbonate and a proton. There are three families of CAs (a, p\ and y). Mammals have fourteen different CA isoforms, all belonging to the a CA family. The vast characterizations of the mammalian CAs have revealed cytosolic, membrane-bound, membrane-spanning, and mitochondrial isoforms. Interestingly, it is becoming increasingly apparent with the recent availability of genome sequences, that less complex organisms also contain a large array of CAs. The Anopheles gambiae genome contains at least 14 putative CA genes (EnsEMBL protein family ENSF00000000228). Whether the 14 well-characterized mammalian CAs serve the same functions as the 14 mosquito genes is intriguing. This question will remain unanswered until all of the CAs from An. gambiae or a similarly distant species are characterized. We can, however, speculate as to their relatedness through sequence identity and conservation. Carbonic anhydrase is an interesting enzyme to study in the context of the mosquito for several reasons. Unlike most animals, mosquitoes use a highly alkaline digestive strategy instead of an acid environment. Additionally, the functions of the mosquito gut are similar to the mammalian kidney in filtering wastes and maintaining ionic homeostasis. Uncovering the function of CAs in an insect, such as the mosquito, 115

PAGE 129

116 will lead to evolutionary clues within the family of a CAs. The relationship between the three distinct CA families is believed to represent convergent evolution. The relationship between a CAs from distantly related species is unknown, mostly due to the lack of characterized CAs from non-mammalian species. We report here the first cytosolic CA that has been cloned and characterized from the .4«. gambiae mosquito, in an attempt to unravel the physiology of an extremely alkaline digestive system. There are at least two different CA isoforms expressed within the larval mosquito gut. One is a GPI-linked CA isoform expressed in a specific subset of muscle and nerve fibers that traverse the anterior midgut and gastric caeca regions (refer to Chapter 4). The other is a cytosolic CA isoform expressed primarily within the gastric caeca and posterior midgut regions. Results Anopheles gambiae CA Sequence Analysis A CA cDNA was cloned from An. gambiae gut tissue. The full-length CA cDNA sequence (accession number AY280613) represents 1 of 14 putative CA genes in the An. gambiae genome predicted by the Ensembl CA protein family (ENSF00000000228). This CA is comprised of 257 amino acids and is predicted to be a cytosolic isoform (no signal sequence or hydrophobic transmembrane domains; Letunic et al., 2002). The molecular weight is predicted to be 29 kDa (DNAman software). Multiple sites of potential post-translational modification include 5 protein kinase C phosphorylation sites, 4 casein kinase II phosphorylation sites, and 5 N-myristoylation sites (Bairoch et al., 1997). The numerous potential sites for protein modification may contribute to regulatory control.

PAGE 130

117 This CA displays all of the 13 highly conserved residues found in most other active CA proteins, including the three necessary histidine residues required for the binding of a zinc atom (Tashian, 1992; Sly and Hu, 1995; Tamai et al., 1996a). Figure 61 shows an alignment of the active sites of An. gambiae, D. melanogaster, and human CA proteins. The active site region of the cytosolic CA from An. gambiae contains one amino acid difference from all of the mammalian isoforms. This particular residue (C89; An. gambiae CA numbering), is replaced by a serine (S) residue in all of the known human CA isoforms (including mouse CA XIII in lieu of the uncharacterized human CA XIH). This C to S amino acid change is also found in 1 of the 14 CA isoforms predicted from the D. melanogaster genome (accession number CGI 1284-PA; Fig. 6-1). The fourteen CA genes of humans are apparently required to perform the many metabolic functions that complex organisms require. The recent release of the sequenced genomes of two insect species, An. gambiae and D. melanogaster has revealed that these organisms also have fourteen CAs. Comparisons based on amino acid composition revealed that these 14 dipteran CAs are probably not homologs to each of the fourteen human CAs. A phylogenetic analysis of the CA protein sequences was performed using the NeighborJoining method (Saitou and Nei, 1987) as implemented in DNAman software. A rooted tree shows the relationship between the human, mouse, and dipteran CAs (Fig. 6-2). The human and mouse CAs cluster together, and the dipteran CAs cluster together, however the mammalian and the dipteran proteins cluster separately. A bootstrapping test was performed to determine the confidence value of the phylogenetic tree. The An. gambiae CA that has the S to C difference within the active site pairs with the D. melanogaster CA with the same difference (Fig. 6-2). These two dipteran CAs

PAGE 131

118 appear to be CA homologs. Further studies must be performed to determine whether any differences in the activity or inhibitory profile of these dipteran CA proteins exists due to this active site difference. This conserved amino acid change found in An. gambiae and D. melanogaster, but not humans, may suggest an evolutionary importance that could be exploited in future mosquito larvacide production. Ongoing efforts aimed at crystallization and x-ray analyses with collaborators at the University of Florida will also reveal the structural identity of this An. gambiae cytosolic CA. Localization of CA Activity in Anopheles gambiae Larvae A modified version of Hansson's CA histochemistry method (Hansson, 1967) was used to localize CA activity within the An. gambiae larvae. Precipitated cobalt salts marked the regions of CA activity within the larvae. The regions of CA activity include the cardia, gastric caeca and posterior midgut (Fig. 6-3). A small subset of specific cells within the rectum also stained positively for CA activity (Fig. 6-3E). Localization of Cytosolic CA mRNA in Anopheles gambiae Larvae A DIG-labeled antisense RNA probe was utilized to localize the CA message. The most intense labeling was consistently viewed within epithelial cells of the gastric caeca and posterior midgut (Fig. 6-4). The localization of cytosolic CA mRNA to the gastric caeca and posterior midgut correlates with the location of C A activity within the An. gambiae gut, as determined by CA histochemistry (refer to Fig. 6-3). The cardia was also labeled, as well as a subset of nerve cells and fibers that traverse the AMG longitudinally (Fig. 6-4B,C). Within the posterior midgut, the labeling was confined to the outer edges of a subset of large columnar cells and the small cuboidal cells (Fig.6-5).

PAGE 132

119 Labeling was also seen within the rectum and the last distal cell of the Malpighian tubules (MT; Fig. 6-6). Antibody Localization of CA Protein The antigenic peptide sequence QYIRSPDAQTEIDAD was chosen from the^w. gambiae translated CA cDNA sequence (accession number AAQ21366) to elicit the production of antibody. This peptide was chosen for its antigenic capacity as well as its uniqueness among the 14 putative An. gambiae CA genes. The resultant chicken antiserum was used to localize the CA protein within whole mount preparations of fourth instar An. gambiae larvae. Immunoreactivity for the cytosolic CA was predominantly displayed within the gastric caeca (Fig.6-7). The PMG also displayed immunolabeling along the periphery of both the large and small epithelial cells. Immunoreactivity was also evident within a small population of neuronal cells scattered along the midgut, most often seen in pairs (Fig.6-8). The pre-immune antisera displayed no specific immunoreactivity. Bacterial Expression and Purification of Anopheles gambiae Cytosolic CA The full-length CA cDNA was subcloned into a pETlOO directional expression vector (pETlOO/D-TOPO; Invitrogen) for expression of the recombinant protein with an N-terminal tag containing an Xpress epitope and a polyhistidine (6X His) tag. The 6X His tag was utilized when purifying the CA protein. Antibodies against both the Xpress epitope and the CA peptide recognized a band of the predicted molecular weight (33 kDa) for the recombinant CA protein (Fig. 6-9). Purified CA fractions were tested for CA activity using 18 0 isotope exchange experiments (Silverman and Tu, 1 986). This technique showed that CA activity was

PAGE 133

120 present within the purified, recombinant CA fractions (data not shown). Activity was partially inhibited by the application of methazolamide, a CA-specific inhibitor (data not shown). These analyses confirmed that this cytosolic CA, cloned from an An. gambiae gastric caeca cDNA library, has CA activity and therefore contributes to the CA activity in the gastric caeca and posterior midgut regions, as determined by CA histochemistry. Discussion The CA isoform, cloned from the An. gambiae midgut, is a predicted cytosolic protein that is expressed in the cardia, gastric caeca and posterior midgut regions. Carbonic anhydrase activity was localized to these same regions through CA histochemistry. The purified recombinant CA was shown to have CA activity through 18 0 isotope exchange experiments. This CA isoform is therefore responsible, at least in part, for the CA activity displayed within the gastric caeca and posterior midgut regions of the larval mosquito gut. The one amino acid difference (C instead of S) within the active site of this mosquito CA is not found in any mammalian isoform. The D. melanogaster genome however displays an identical (C instead of S) difference in one of its putative CA isoforms. The S residue at this position is present in every mammalian CA and therefore may represent an evolutionary divergence within the a CA family. The phylogenetic analysis of mammalian and dipteran CAs shows a more distant relationship between the Dipteran and mammalian proteins than within the mammalian CA family. The dipteran CA isoforms do not cluster with the mammalian CAs despite common functional characteristics and sequence homology in all known a CAs. Instead, the mosquito CA sequences cluster together, apart from the mammalian CA clusters.

PAGE 134

121 The mammalian anion exchanger (AE1) has been shown to physically bind a cytosolic CA isoform (Sterling et al., 2001a). An AE has been cloned from the An. gambiae midgut (refer to chapter 5) and protein expression has been localized to the gastric caeca and posterior midgut regions. This AE also has a putative CA binding site within its intracellular carboxy terminus. The membrane localization of this predicted cytosolic CA may be caused by its interaction with a membrane protein such as an AE. Future experiments will reveal if this cytosolic CA is indeed coupled to an AE within the gastric caeca and posterior midgut regions of the mosquito gut, forming a bicarbonate transport metabolon. The existence of such a metabolon within gut regions flanking the highly alkaline anterior midgut may be the mechanism through which cellular homeostasis is maintained.

PAGE 135

122 AAQ21365 CAIV-like AAQ21366 CAII-like ENSANGP00000001812 ENSANGPOOO 00018 999 ENSANGP00000029518 ENSANGP0OO00 011908 ENSANGP00000001574 ENSANGP00000011013 ENSANGP00000012957 ENSANGP00000016412 ENSANGP00000010017 ENSANGP00000014 948 ENSANGP00000014919 ENSANGP00000001278 CG9235-PA Dros CG18672-PA Dros CG10899-PA Dros CG11284-PA Dros CG12309-PA Dros CG3940-PA Dros CG32698-PA Dros CG6074-PA Dros CG18673-PA Dros CG1402-PA Dros CG3669-PA Dros CG6906-PA Dros CG5379-PA Dros CG7820-PA Dros P00915 CA-I P00918 CA-I I P07451 CA-III P22748 CA-IV AAB47048 CA-V CAC42429 CA-VI P43166 CA-VII JN0576 CA-VIII AAH14950 CA-IX Q9NS85 CA-X AAH02662 CA-XI AAH23981 CA-XII AAK16672 mCA-XIII BAA85002 CA-XIV QMHFHWG SEHTLDDTRYGLELHLVH QLHFHWGIGDGS— GCEHTLEGSTYSMEAHAVH QFHFHAP SENLIKGHSYPLEGHLVH QLHFHWGLSALD-GSEHTIDGYRLPLELHVIH QFHCHWGCSDSR-GSEHTVDGESFAGELHLVH EIHVHYGLHDQF-GSEHSVEGYTFPAEARHIQ EI YFHYGTDNNQ-GSEHHIHGYSFPGEIQLYG QLHFHWGDNDTF — GSEDMIDNHRFPMELHWF GLHFHWGDKNNR — GAEHVLNDIRYPLEMHIIH QFHCHWGCSDSR — GSEHTVDGESFAGELHLVH QLHFHWGPDDAV — GSEHLLDGRAHSMEAHLVH QLHFHHGADMGR-GSEHTFDGVAWAAEAHFVF QFHFHWGVNSTV — GSEHVYDYQRYPMEIHLVF QMHFHHGPNNSE— GSEHRINGERFPLEVHLVF EISFRWSWASSL— GSEHTLDHHHSPLEMQCLH GLHFHWGSYKSR-GSEHLINKRRFDAEIHIVH QLHFHWGSALSK-GSEHCLDGNYYDGEVHIVH QLHFHHSDCDES — GCEHTLEGMKYSMEAHAVH ELRFHWGWCNSE — GSEHTINHRKFPLEMQVMH QIHHHHW SEHTINDIRYPLEVHIVH EIHMHYGLKDQF-GSEHSVEGYTFPAEIQIFG GLHFHWGDKNNR-GSEHVINDIRYTMEHHIVB SVHFHWGSREAK-GSEHAINFQRYDVEHHIVH EIYIHYGTENVR-GSEHFIQGYSFPGEIQIYG AFHFHWGSPSSR— GSEHSINQQRFDVEMHIVH QFHFHWGENDTI— GSEDLINNRAYPAELHWL AVHFHWGSPESK-GSEHLLNGRRFDLEMHIVB QFHCHWGCTDSKGSEHTVDGVS YS GELKLVH QFHFHWGSTNEH — GSEHTVDGVKYSAELHVAH QFHFHWGSLDGQ — GSEHTVDKKKYAAELHLVH QFHLHWGSSDDH — GSEHTVDGVKYAAELHLVH QLHLHWSDLPYK — GSEHSLDGEHFAMEMHIVH QFHFHWGAVNEG — GS E HTVDGHAY PAELHLVH QMHFHWGGAS SEI SGSEHTVDGI RHVIEI EI VH QFH FHWGKKHD V — GSEHTVDGKSFPSELHLVH EVRFHWGRENQR-GSEHTVNFKAFPMELHLIH QLHLHWGAAGRP-GSEHTVEGHRFPAEIHWH EIRLHFGSEDSQ — GSEHLLNGQAFSGEVQLIH ELRLLFGARDGAGSEHQ I NHQGFS AEVQLIH QLHLHWGNPNDPH-GSEHTVSGQHFAAELHIVH QFHLHWGSADDH — GSEHWDGVRYAAELHWH QLHLHWGQKGSPG-GSEHQINSEATFAELKIVH Figure 6-1. Clustal alignment of active sites within An. gambiae, D. melanogaster, and human CA proteins. All active a CAs are tethered to a zinc molecule by the coordination with three histidine (H) residues. There are three human CAs which do not possess H residues in this orientation and have been shown to lack CA activity. Within the active site region, some of the An. gambiae CAs display novel differences, as compared to the tightly-conserved human CAs. Without exception in the human isoforms, a conserved glutamine (E; marked with *) always follows a serine (S). In contrast, the An. gambiae alignment shows a change from the S to a cysteine (C) in one of their 14 putative isoforms. The An. gambiae alignment also displays a gap within the active site regions of two of the CAs. One of these sequences (AAQ21365) was cloned and determined to be a GPI-linked CA TV-like isoform that was discussed in a previous chapter. These novel active site differences may be exploited in the formulation of a specific mosquito larvacide. The same active site differences are also displayed within the D. melanogaster genome and therefore may also provide clues to the evolutionary mechanism of these proteins.

PAGE 136

123 0.05 92 95 Tof IOC IOC 94 96 IOC IOC 10Q10C IOC IOC IOC IOC BAA85002 CA-XIV i NP035927 MCA-XIV • AAH23981 CA-XII NP848483 MCA-XII AAH14950 CA-IX — NP647466 MCA-IX CAC42429 CA-VI — NP033932 MCA-VI — P22748 CA-IV NP031633 MCA-IV Anoph CA II -LIKE — CG11284-PA DROS CA 10C| XP134293 MCA-VII 10C 10C P43166 CA-VII — NP078771 MCA-XIII — XP372051 CA-XIII P00915 CA-I ^— PI 3 634 MCA1 P07451 CA-III P16015 MCA-III P00918 CA-II — NP033931 MCA-II AAB47048 CA-V NP031634 MCA-VA r— NP031618 MCA-VIII JK0576 CA-VIII 10C Anoph CA IV-LIKE Aedes CA IV-LIKE CG3940-PA Dros CA Figure 6-2. Phylogenetic analysis between mammalian (human and mouse) and dipteran (An. gambiae and D. melanogaster) CAs. This phylogenetic tree was assembled using the NeighborJoining method. The tree has been rooted and the bootstrapping confidence values are shown. The mammalian sequences are shown in blue, MCA designates the mouse, while the dipteran sequences are shown in green. The mammalian CAs cluster separately from the dipteran CA clusters, signifying a high divergence between mammalian and dipteran CAs. There is also a high divergence between the dipteran CA II-like sequences and the dipteran CA IV-like sequences.

PAGE 137

124 Figure 6-3. Localization of An. gambiae CA activity. Using CA histochemistry, CA activity was localized to the cardia, gastric caeca, and posterior midgut regions. A. A whole mount view of the entire gut region. B. A higher magnification view of the gastric caeca and cardia regions that displayed CA activity. Trachea are indicated with arrowheads. C. A higher magnification view of the AMG to PMG boundary. The PMG displayed epithelial CA activity while the AMG does not. The trachea are indicated by arrowheads. D. Specific cells of the cardia along with cells of the AMG (arrows) displayed CA staining. E. Small cells within the rectum also displayed CA activity (arrows). Scale bars represent 300 um in A, 100 um in B and C, and 150 urn in D and E.

PAGE 138

125 A m W MT Figure 6-4. Localization of CA mRNA expression within An. gambiae whole mounts. A. A view of the entire gut region hybridized with the CA cRNA probe. The most prominent hybridization was found in the gastric caeca and posterior midgut. B. A higher magnification view of the gastric caeca. Both the proximal cells and distal cap cells of the gastric caeca show CA mRNA expression. Rostral to the gastric caeca the cardia region also displayed intense label (arrow). C. Top view of the cardia displays consistent labeling around the circumference of the cardia region (arrows). Scale bars represent 300 um in A, 75 um in B and C.

PAGE 139

126 Figure 6-5. Localization of CA mRNA expression within the posterior midgut of An. gambiae. A. The PMG displays staining for CA mRNA within the peripheral borders of both the large and small epithelial cells. B. Higher magnification of the PMG shows the stained large columnar cells (*) and the labeled small cuboidal cells (arrows). C. Side view of the PMG reveals CA mRNA expression near the plasma membranes (arrows) and not throughout the cytoplasm. Scale bars represent 1 50 um in A, 25 um in B, and 50 um in C.

PAGE 140

127 A B Figure 6-6. Localization of CA mRNA expression within the hindgut. A. The hindgut displays CA label within the Malpighian tubules (MT, arrow) and the rectum. B. Higher magnification of the Malpighian tubules shows staining for CA expression confined to the most distal one or two cells. All other cells of the MT show no hybridization. Scale bars represent 300 um in A and 25 um in B.

PAGE 141

128 Figure 6-7. Localization of CA protein within gastric caeca of An. gambiae larvae. A. The CA specific antibody prominently labels cells of the gastric caeca. B. Phalloidin was used to localize muscle fibers. C. Draq-5 was used to localize nuclear DNA. D. Overlay of the three signals shows the location of CA in relation to muscle fibers and nuclei. Scale bars represent 100 um.

PAGE 142

129 Figure 6-8. Localization of CA protein within the PMG of An. gambiae. The CAspecific antibody labels cell membranes of both large (double arrows) and small cells (small arrows), mimicking the protein localization pattern for the AgAEl protein. Since this particular CA isoform is predicted to be a cytosolic isoform, its localization to cell membranes suggests an interaction with a membrane protein. The AgAEl protein, known to be a membrane protein, capable of binding CA II, and the localization of the AE1 protein to the same cells supports the hypothesis of a bicarbonate metabolon within the mosquito gut. Scale bar represents 25 urn.

PAGE 143

Figure 6-9. Protein gels and western blots of recombinantly expressed CA protein. A. Brilliant blue staining of CA protein induction (33 kDa) from 0 to 24 hours. B. Fast green staining of recombinant CA protein expression (33 kDa). C. The XPRESS (XP) antibody detected a 33 kDa band. Pre-immune sera from two chickens inoculated with a conjugated CA peptide (PI) did not detect a band at 33 kDa. The antisera collected from one of the inoculated chickens, detects a visible protein band at the expected 33 kDa (94; red arrow). The antisera collected from the other chicken (93), does not recognize the 33 kDa protein band on the western blot. The lane containing the molecular weight standard is marked with an M.

PAGE 144

CHAPTER 7 CONCLUSIONS AND FUTURE DIRECTIONS Conclusions Molecular cloning techniques using isolated mosquito guts from Aedes aegypti and Anopheles gambiae resulted in the full-length cDNA cloning of three carbonic anhydrase (CA) genes. Our study of CA(s) within the mosquito gut began with a simple understanding that the anterior midgut lumen has a pH of 1 1 . The ability of C A to greatly enhance the production of bicarbonate (and carbonate), a strong buffer, made it a likely candidate for buffering the mosquito gut. Mammalian CA isoforms have been studied extensively throughout the past several decades. However, the relationship between a CAs from mammals, and those of less complex species such as mosquitoes, is unknown. This is partly due to the lack in characterization of multiple CA isoforms from a single non-mammalian species. Fourteen different a CA isoforms from mammals have been characterized. The An. gambiae and D. melanogaster genomes also display fourteen CAs. However, a phylogenetic analysis of amino acid sequences has shown that the fourteen mammalian CA isoforms and the fourteen dipteran CA isoforms are not direct homologs. When more CA isoforms are characterized from insects it will be extremely interesting to determine which isoforms are represented or omitted from the insect divergence of CAs. Before the fourteen CA isoforms in mammals and mosquitoes can be truly compared, every mosquito C A isoform must be characterized to determine their 131

PAGE 145

132 functional identities. Two different CA isoforms from larval mosquitoes were partially characterized within this study. One GPI-linked CA isoform was found in both An. gambiae andAe. aegypti, and the other cytosolic CA isoform was found in An. gambiae. The additional characterization of more mosquito CAs in the future, will present a great opportunity for studying the evolution of CAs within the same a C A family. A GPI-linked CA isoform, similar to the mammalian CA TV isoform was localized to the gastric caeca and a specific subset of muscle fibers in the anterior midgut region of both Ae. aegypti and An. gambiae. This isoform is different from the other a CA isoforms (characterized in mammals) in that the entire CA protein is located extracellularly, with only the GPI-link maintaining an association to the plasma membrane. These GPI-linked CA isoforms from two different mosquito species also have a direct homolog in the D. melanogaster genome database. These three dipteran CA IV-like isoforms all display a shortened active site region. How this novel active site affects the activity of the CA protein is unknown. However, it is known that mammalian CA IV proteins are oriented by the GPI attachment so that the active site is directed away from the membrane, thereby affording the greatest accessibility of substrate to the active site. The GPI-linked CA of the anterior midgut muscle fibers would be in the prime location and conformation for taking up substrate from the hemolymph. The other full-length CA cDNA isolated from An. gambiae gut tissue is a cytosolic CA isoform. The expression of mRNA was localized to the cardia, gastric caeca and posterior midgut epithelial cells. These regions were also shown to contain an active CA enzyme through CA histochemistry. Recombinant expression of this An. gambiae CA protein in bacteria, produced a purified CA-active eluate, as measured by

PAGE 146

133 18 0 exchange. The activity was shown to be sensitive to the CA-specific inhibitor, methazolamide. This cytosolic CA isoform therefore contributes to the CA activity present within the cardia, gastric caeca and posterior midgut epithelial cells. A full-length anion exchanger (AE) cDNA was also cloned from the gut tissue of An. gambiae. The AEs are a small group within a large bicarbonate transporter (BT) superfamily. Studies of mammalian BTs are ongoing and many new forms are still being discovered and characterized to date. AEs reversibly transport chloride for bicarbonate in a 1:1 electroneutral exchange. Cloning and localizing an AE within the mosquito gut was therefore a direct progression from localizing the CAs, relating the production and transport of bicarbonate within the alkaline gut. The larval mosquito AE was expressed in Xenopus oocytes and was shown to have electrophysiological characteristics of known AEs (i.e. chloride transport and DIDS inhibition). The AE contains an intracellular carboxy terminus that is predicted to moderate ion exchange and an amino terminus that performs ion exchange via the twelve membrane-spanning domains. The antibodies we produced, specific to carboxy and amino terminal peptides, label the membranes of both gastric caeca and posterior midgut epithelial cells. To maximize ion exchange, the An. gambiae AE contains an amino terminus CA binding motif. If indeed the AE binds a cytosolic CA, the gastric caeca and posterior midgut regions would have control over cellular pH, via both intracellular and extracellular means. Since both of these regions contain active cytosolic CA enzyme(s), we propose that the AE spans the membrane in the GC and PMG and binds a cytosolic CA, forming a bicarbonate transport metabolon to transport the bicarbonate that is made by the CA(s).

PAGE 147

134 To further regulate ion exchange, the AE has a particular region of amino acids with similarity to mammalian AEs 2 and 3. This region was found to confer pH sensitivity through intracellular alkalization. The ability of a mosquito AE to detect intracellular alkalization may be the underlying mechanism through which the alkalinity remains confined to the anterior midgut region. The AEand CA-containing regions (gastric caeca and posterior midgut) flank the anterior midgut and could possibly modulate their transport rates in response to the encroaching or retreating alkaline pH. This bicarbonate metabolon could have the ability to maintain a large pH gradient as is displayed in the larval mosquito gut. Discovering the proteins involved in the production and maintenance of such an alkaline pH could define a fundamental metabolon that is critical for pH homeostasis. New Model The cloning and localization of C As and an AE within the larval mosquito gut has uncovered an unpredicted model of physiology. Our original model, based on Manduca sexta (refer to Fig. 1-3), has evolved considerably due to our new findings within larval mosquitoes. The most unexpected finding, was the failure to detect a CA within the anterior midgut. Carbonic anhydrase histochemistry, I8 0 isotope exchange, in situ hybridization, immunohistochemistry, and real time PCR all failed to give evidence for a CA within the AMG region. Although these data did not support our original model of anterior midgut alkalization, our new model describes a system in which CA is not necessary within the AMG epithelial cells. Furthermore, our new data, including the localization of AE within the mosquito gut, has provided insight as to why our new model of AMG alkalization is more efficient in the absence of CA.

PAGE 148

Although CA is a reversible enzyme, the mammalian CA IV is even faster than CA II at bicarbonate dehydration. This ability, along with prior studies that have found a very low concentration of bicarbonate in the hemo lymph surrounding the mosquito anterior midgut, have led to a prediction of the role of the CA IV-like enzyme in the mosquito midgut. The high alkalinity of the AMG lumen is energized by a H* VATPase, which pumps protons from the epithelial cells into the hemolymph surrounding the anterior midgut. Only the AMG contains a basally-oriented H* V-ATPase, whereas the other regions of the midgut (gastric caeca and PMG) express an apically-oriented V-ATPase (Zhuang et al., 1999). The CA IV-like enzymes that are suspended in the hemolymph are in the perfect position to provide a sink for these protons. Through the action of the CA IV-like enzyme in the hemolymph, protons are combined with bicarbonate to form carbon dioxide and water. Due to this coordinated effort, the H 4 " VATPase can keep pumping protons out into the hemolymph without creating a concentration gradient. The CA activity on the hemolymph side of the AMG therefore allows the fT V-ATPase to function more efficiently. In contrast to the benefit that the CA IV-like enzyme provides to the If 4 " VATPase, the existence of CA within the epithelial cells of the AMG would actually hinder the PT V-ATPase. The benefits of not having a CA in the AMG are twofold, due to the reversibility of CA. First, because there is no CA to dehydrate bicarbonate, there is no competition with the V-ATPase for protons. Secondly but most importantly, because there is no CA to hydrate carbon dioxide to provide protons to the H + V-ATPase, the protons must come into the cell from the lumen to replace the protons that are being pumped out to the hemolymph. This need for proton replacement may be the driving

PAGE 149

136 force that pulls the proton from bicarbonate in the lumen to leave a carbonate ion. The proton may enter the cell through a channel or be exchanged for a cation, such as potassium. This carbonate or potassium carbonate could then drive the pH up to the high alkalinity. The absence of an anion exchanger in the AMG would also contribute to the alkalinity in the lumen by not providing a route for bicarbonate to leave the AMG lumen. The localization of the AE to the basolateral membranes in both gastric caeca and posterior midgut correlates with the localization of a cytosolic CA isoform to the same regions. This AE brings bicarbonate into the cell in exchange for dumping chloride into the hemolymph. This AE could account for the high levels of chloride and low levels of bicarbonate that have been measured in the hemolymph (Boudko et al., 2001a). The bicarbonate that enters the cell provides a substrate for the cytosolic CA. Due to the CA binding motif at the amino terminal of the AE, the binding of these two proteins would form a bicarbonate transport metabolon. This complex would maximize bicarbonate transport due to the elimination of diffusion, normally relied upon to get substrate from one protein to the next. Our new model reflects the absence of CA within the AMG epithelial cells and instead shows a GPI-linked CA bound to specific muscle fibers traversing the AMG and GC, as well as a cytosolic CA within epithelial cells of the cardia, GC and PMG (Fig. 71). The AE is found in regions that express CA and flank the alkaline anterior midgut. Therefore, the cloned AE is shown in the GC and PMG. This AE has the ability to bind a cytosolic CA, whereby the coordinated efforts of producing and transporting bicarbonate are enhanced. The new model reflects a bicarbonate transport metabolon within the gastric caeca and posterior midgut regions.

PAGE 150

137 Future Directions There are fourteen predicted CA isoforms in An. gambiae; however, this dissertation research only dealt with the cloning and characterization of two. Although both were found to have expression in the larval midgut, there may be more that can influence our model. However, if any additional active CAs are localized within the larval mosquito midgut, they most likely will also be localized to the gastric caeca and/or PMG. All of our CA studies failed to show any evidence for a CA in the AMG epithelial cells. An apically expressed CA in the AMG may have avoided the CA histochemical staining but the 0 isotope exchange assay should have uncovered any CA activity present within this gut region. The hypothesis that a bicarbonate transport metabolon exists within the regions flanking the alkaline anterior midgut needs to be investigated. If indeed the AE is found to bind CA, then the putative CA binding sequence located within the intracellular carboxy terminus needs to be examined. The amino acids necessary for binding should be studied to determine whether the binding motif of mammalian AEs is valid for the binding of mosquito CA(s) to the mosquito AE, as proposed. The localization of two different CA isoforms and an AE within the larval mosquito gut has established a new model upon which to build. Preceeding this investigation, only a FT V-ATPase was localized within the larval mosquito gut. Other channels and transporters need to be identified and localized, such as chloride channels and sodium/ hydrogen exchangers. The results of these studies may be used to formulate specific mosquito larvacides. Both of the cloned mosquito CA isoforms display novel active site regions.

PAGE 151

138 CA inhibitors can be manufactured to potentially block only CAs containing this novel active site. Although these novel active sites are not found in any mammalian CA, they are also found in other insects such as D. melanogaster. A larvacide specific for mosquitoes can still be made due to the incorporation of an alkaline pH trigger. Species that do not have a highly alkaline digestive strategy, such as D. melanogaster, will therefore not be exposed to the specific CA inhibitor(s). The larval mosquito gut provides a simple model for kidney epithelial transport. An important finding was that the mosquito epithelial cells resemble the a and P intercalated cells of the mammalian kidney. The addition of more ion transporters and exchangers to our mosquito model will enable further comparisons between the mosquito gut and the mammalian kidney. Diseases related to anion exchangers and V-ATPases can also be studied using the mosquito gut due to the one cell layer epithelium that allows for easier tracking of ions, as compared to the complex kidney.

PAGE 152

139 A. Midgut Hindgut 1 II 1 GC 1 AMG PMG MT cardia H+ ciHCO, HCO, B. lumen H HCO, + CO, 2 H < AMG 2H>K + PMG PH 10-11 J pH 7-8 HCO, +H + :o 2 + H 2 0 HCO, .co 2 + H 2 6 Hemolymph H + +" HCO, P" 7 ' 8 Figure 7-1 . New larval mosquito model. A. The larval mosquito gut is divided into the foregut, midgut, and hindgut. The gastric caeca (GC) and anterior midgut (AMG) express a GPI-linked CA isoform on muscle fibers (shown in yellow). The cardia, GC, posterior midgut (PMG), rectum, and last distal cell of Malpighian tubules (MT) express a cytosolic CA isoform. The GC, PMG, MT and rectum express a chloride/ bicarbonate anion exchanger (AE). In the GC, and PMG, the AE may bind a cytosolic CA isoform forming a metabolon. A V-ATPase is expressed in GC, AMG, and PMG. B. Diagram of a representative cell from GC, AMG, and PMG displaying the cell polarity. The V-ATPase is expressed apically in GC and PMG, and basally in AMG. The AE is expressed basally in the GC and PMG. The GPI-linked CA isoform is expressed extracellularly on muscle fibers in the GC and AMG.

PAGE 153

REFERENCES Alper, S. L. (2002). Genetic diseases of acid-base transporters. Annu. Rev. Physiol. 64, 899-923. Alper, S. L., Chernova, M. N. and Stewart, A. K. (2001). Regulation of Na+independent C1-/HC03exchangers by pH. Jop. 2, 171-175. Altschul, S. F., Gish, W., Miller, W., Myers, E. W. and Lipman, D. J. (1990). Basic local alignment search tool. J. Mol. Biol. 215, 403-410. Baird, T. T., Jr., Waheed, A., Okuyama, T., Sly, W. S. and Fierke, C. A. (1997). Catalysis and inhibition of human carbonic anhydrase IV. Biochemistry 36, 2669-2678. Bairoch, A., Bucher, P. and Hofmann, K. (1997). The PROSITE database, its status in 1997. Nucleic Acids Res. 25, 217-221. Berenbaum, M. (1980). Adaptive significance of midgut pH in larval Lepidoptera. Am. Nat. 115, 138-146. Boudko, D. Y., Moroz, L. L., Harvey, W. R. and Linser, P. J. (2001a). Alkalinization by chloride/ bicarbonate pathway. Proc. Natl. Acad. Sci. USA 98, 15354-15359. Boudko, D. Y., Moroz, L. L., Linser, P. J., Trimarchi, J. R., Smith, P. J. S., Harvey, W. R. (2001b). In situ analysis of pH gradients in mosquito larvae using non-invasive, self-referencing, pH-sensitive microelectrodes. J. Exp. Biol. 204, 691-699. Brown, D. and Breton, S. (1996). Mitochondria-rich, proton-secreting epithelial cells. J Exp. Biol. 199 (Pt 11), 2345-2358. Brown, D. and Waneck, G. L. (1992). Glycosyl-phosphatidylinositol-anchored membrane proteins. J. Am. Soc. Nephrol. 3, 895-906. Chegwidden, W. R. and Carter, N. D. (2000). Introduction to the carbonic anhydrases Exs. 14-28. Clamp, M., Andrews D., Barker, D., Beven, P., Cameron, G., Chen, Y., Clark, L., Cox, T., Cuff, J., Curwen, V., et al. (2003). Ensembl 2002: accommodating comparative genomics. Nucleic acids res. 31 (1), 38-42. 140

PAGE 154

141 Clark, T. M., Koch, A. and Moffett, D. F. (1999). The anterior and posterior "stomach" regions of larval Aedes aegypti midgut: regional specialization of ion transport and stimulation by 5-hydroxytryptamine. J. Exp. Biol. 202 (Pt 3), 247-252. Clements, A. N. (1992). The biology of mosquitoes. Chapman & Hall, London, UK. Corena, M. P., Seron, T. J., Lehman, H. K, Ochrietor, J. D., Kohn, A., Tu, C. and Linser, P. J. (2002). Carbonic anhydrase in the midgut of larval Aedes aegypti: cloning, localization and inhibition. J. Exp. Biol. 205, 591-602. Dadd, R. H. (1975). Alkalinity within the midgut of mosquito larvae with alkaline-active digestive enzymes. J. Insect Physiol. 21, 1847-1853. Darsie, R. F., Jr. and Morris, C. D. (2000). Keys to the adult females and fourth instar larvae of the mosquitoes of Florida (Diptera, Culicidae). Bull. Fla. Mosq. Control Assoc. Dodgson, S. J. (1991). Why are there carbonic anhydrases in the liver? Biochem. Cell Biol. 69, 761-763. Donaldson, T. L., and Quinn, J. A. (1974). Kinetic constants determined from membrane transport measurements: carbonic anhydrase activity at high concentrations. Proc. Natl. Acad. Sci. USA 71, 4995-4999. Dow, J. A. (1984). Extremely high pH in biological systems: a model for carbonate transport. Am. J. Physiol. 24, (4 Pt 2), R633-636. Eisenhaber, B., Bork, P. and Eisenhaber, F. (1999). Prediction of potential GPImodification sites in proprotein sequences. J. Mol. Biol. 292, 741-758. Fernley, R. T. (1988). Non-cytoplasmic carbonic anhydrases. Trends Biochem. Sci. 13, 356-359. Frohman, M. A. and Zhang, Y. (1997). Using rapid amplification of cDNA ends (RACE) to obtain full-length cDNAs. Methods Mol. Biol. 69, 61-87. Geer, L. Y., Domrachev, M., Lipman, D. J. and Bryant, S. H. (2002). CD ART: protein homology by domain architecture. Genome Res. 12, 1619-1623. Gubler, D. J. (1997). in: Dengue and dengue hemorrhagic fever, (ed. Gubler, D. J. & Kuno, G.), CAB International, Wallingford, U.K. pp. 1-22. Gupta, R, Jung, E., and Bmnak S. (In preparation 2002) Prediction of N-glycosylation sites in human proteins.

PAGE 155

142 Halstead, S. B. (1997). in: Dengue and dengue hemorrhagic fever, (ed. Gubler, D. J. & Kuno, G.), CAB International, Wallingford, U.K. pp. 23-44. Hansson, H. P. J. (1967). Histochemical demonstration of carbonic anhydrase activity. Histochemie 11, 112-128. Harvey, W. R. (1992). Physiology of V-ATPases. J. Exp. Biol. 172, 1-17. Haskell, J. A., demons, R. D. and Harvey, W. R. (1965). Active transport by the Cecropia midgut. L Inhibitors, stimulants and potassium-transport. J. Cell. Comp. Physiol. 65, 45-56. Hewett-Emmett, D. (2000). Evolution and distribution of the carbonic anhydrase gene families. Exs. 90, 29-76. Hewett-Emmett, D. and Tashian, R. E. (1996). Functional diversity, conservation, and convergence in the evolution of the alpha-, beta-, and gamma-carbonic anhydrase gene families. Mol. Phylogenet. Evol. 5, 50-77. Holt, R A. Subramanian, G. M. Halpern, A. Sutton, G. G. Charlab, R. Nusskern, D. R. Wincker, P. Clark, A. G. Ribeiro, J. M. Wides, R. et al. (2002). The genome sequence of the malaria mosquito Anopheles gambiae. Science 298, 129-149. Hogue, C. W. (1997). Cn3D: a new generation of three-dimensional molecular structure viewer. Trends Biochem. Sci. 22, 314-316. Hubbard, T., Barker, D., Birney, E., Cameron, G., Chen, Y., Clark, L., Cox, T., Cuff, J., Curwen, V., Down, T. et al. (2002). The ensembl genome database project. Nucleic Acids Res. 30, 38-41. Huber, S., Asan, E., Jons, T., Kerscher, C, Puscbel, B. and Drenckhahn, D. (1999). Expression of rat kidney anion exchanger 1 in type A intercalated cells in metabolic acidosis and alkalosis. Am. J. Physiol. 277, F841-849. Jespersen, T., Grunnet, M., Angelo, K., Klaerke, D. A. & Olesen, S. P. (2002). Dualfunction vector for protein expression in both mammaliancells and Xenopus laevis oocytes. Biotechniques 32, 536-538, 540. Jungreis, A. M., Barron, D. N. and Johnston, J. W. (1981). Comparative properties of tobacco hornworm Manduca sexta, carbonic anhydrases. Am. J. Physiol. 241, R92-99. Kim, J., Kim, Y. H., Cha, J. H., Tisher, C. C, and Madsen, K. M. (1999). Intercalated cell subtypes in collecting tubule and cortical collecting duct of rat and mouse. J. Amer. Soc. Nephrol. 10, 1-12.

PAGE 156

143 Letunic, I., Goodstadt, L., Dickens, N. J., Doerks, T., Schultz, J., Mott, R., Ciccarelli, F., Copley, R. R., Ponting, C. P. and Bork, P. (2002). Recent improvements to the SMART domain-based sequence annotation resource. Nucleic Acids Res. 30, 242-4. Lindskog, S. (1997). Structure and mechanism of carbonic anhydrase. Pharmacol. Ther. 74, 1-20. Marchler-Bauer, A., Panchenko, A. R., Shoemaker, B. A., Thiessen, P. A., Geer, L. Y. and Bryant, S. H. (2002). CDD: a database of conserved domain alignments with links to domain three-dimensional structure. Nucleic Acids Res. 30, 281-283. Martin, M. M., Martin, J. S., Kukor, J. J. and Merrit, R. W. (1980). The digestion of protein and carbohydrate by the stream detritivore, Tipula abdominalis (Diptera, Tipulidae). Oecologia 46, 360-364. Matsumoto, T., Winkler, C. A., Brion, L. P. and Schwartz, G. J. (1994). Expression of acid-base-related proteins in mesonephric kidney of the rabbit. Am. J. Physiol. 267, F987-997. Matz, M., Shagin, D., Bogdanova, E., Britanova, O., Lukyanov, S., Diatchenko, L. and Chenchik, A. (1999). Amplification of cDNA ends based on template-switching effect and step-out PCR. Nucleic Acids Res. 27, 1558-1560. Meldrum, N. U. and Roughton, F. J. W. (1933). Carbonic anhydrase: Its preparation and properties. J. Physiol. London 80, 113-142. Osborne, W. R. A. and Tashian, R. E. (1975). An improved method for the purification of carbonic anhydrase isozymes by affinity chromatography. Anal. Biochem. 64, 297303. Parkkila, S., Rajaniemi, H., Parkkila, A. K., Kivela, J., Waheed, A., Pastorekova, S., Pastorek, J. and Sly, W. S. (2000). Carbonic anhydrase inhibitor suppresses invasion of renal cancer cells in vitro. Proc. Natl. Acad. Sci. USA 97, 2220-2224. Phillips, K. P. and Baltz, J. M. (1999). Intracellular pH regulation by HC03-/C1exchange is activated during early mouse zygote development. Dev. Biol. 208, 392-405. Ridgway, R L. and Moffett, D. F. (1986). Regional differences in the histochemical localization of carbonic anhydrase in the midgut of tobacco hornworm (Manduca sexta). J. Exp. Zool. 237, 407-412. Rockstein, M. (1964). The physiology of insecta. (ed. Rockstein, M.), Academic Press, New York, NY. pp. 380-387.

PAGE 157

144 Romero, M. F., Henry, D., Nelson, S., Harte, P. J., Dillon, A. K. and Sciortino, C. M. (2000). Cloning and characterization of aNa+-driven anion exchanger (NDAE1). A new bicarbonate transporter. J. Biol. Chem. 275, 24552-24559. Saitou, N. and Nei, M. (1987). The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4, 406-25. Schwartz, G. J. (2002). Physiology and molecular biology of renal carbonic anhydrase. J. Nephrol. 15 Suppl 5, S61-74. Shahabuddin, M. and Pimenta, P. F. (1998). Plasmodium gallinaceum preferentially invades vesicular ATPase-expressing cells in Aedes aegypti midgut. Proc. Natl. Acad. Sci. USA 95, 3385-3389. Silverman, D. N. and Tu, C. K. (1986). Molecular basis of the oxygen exchange from C02 catalyzed by carbonic anhydrase HI from bovine skeletal muscle. Biochemistry 25, 8402-8408. Sly, W. S. (2000). The membrane carbonic anhydrases: from C02 transport to tumor markers. Exs. 95-104. Sly, W. S. and Hu, P. Y. (1995). Human carbonic anhydrases and carbonic anhydrase deficiencies. Annu. Rev. Biochem. 64, 375-401. Spielman, A. and D'Antonio, M. (2001). Mosquito: A natural history of our most persistent and deadly foe. Hyperion Press, New York, NY. Sterling, D., Alvarez, B. V. and Casey, J. R. (2002a). The extracellular component of a transport metabolon. Extracellular loop 4 of the human AE1 C1-/HC03exchanger binds carbonic anhydrase TV. J. Biol. Chem. 277, 25239-25246. Sterling, D., Brown, N. J., Supuran, C. T. and Casey, J. R. (2002b). The functional and physical relationship between the DRA bicarbonate transporter and carbonic anhydrase H. Am. J. Physiol. Cell Physiol. 283, CI 522-1 529. Sterling, D. and Casey, J. R. (2002). Bicarbonate transport proteins. Biochem. Cell Biol. 80, 483-97. Sterling, D., Reithmeier, R. A. and Casey, J. R. (2001a). A transport metabolon. Functional interaction of carbonic anhydrase II and chloride/bicarbonate exchangers. J. Biol. Chem. 276, 47886-47894. Sterling, D., Reithmeier, R. A. and Casey, J. R. (2001b). Carbonic anhydrase: in the driver's seat for bicarbonate transport. Jop. 2, 165-170.

PAGE 158

Stobbart, R. H. (1971). Evidence of NaVrf" and C17HC0 3 " exchanges during independent sodium and chloride uptake by the larvae of the mosquito Aedes aegypti. J. Exp. Biol. 54,19-27. Sun, M. K. and Alkon, D. L. (2002). Carbonic anhydrase gating of attention: memory therapy and enhancement. Trends Pharmacol. Sci. 23, 83-89. Tamai, S., Cody, L. B. and Sly, W. S. (1996a). Molecular cloning of the mouse gene coding for carbonic anhydrase IV. Biochem. Genet. 34, 31-43. Tamai, S., Waheed, A., Cody, L. B. and Sly, W. S. (1996b). Gly-63->Gln substitution adjacent to His-64 in rodent carbonic anhydrase IVs largely explains their reduced activity. Proc. Natl. Acad. Sci. USA 93, 13647-13652. Tashian, R. E. (1969). in: Biochemical methods in red cell genetics, (ed. Yunis, J. J.), Academic Press, New York, NY. pp. 307-336. Tashian, R. E. (1992). Genetics of the mammalian carbonic anhydrases. Adv. Genet. 30, 321-356. Tashian, R E., Hewett-Emmett, D., Carter, N. and Bergenhem, N. C. (2000). Carbonic anhydrase (CA)-related proteins (CA-RPs), and transmembrane proteins with CA or CA-RP domains. Exs. 105-120. Thompson, J. D., Higgins, D. G. and Gibson, T. J. (1994). CLUSTAL W: Improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22, 46734680. Tong, C. K., Br ion, L. P., Suarez, C. and Chesler, M. (2000). Interstitial carbonic anhydrase (CA) activity in brain is attributable to membrane-bound CA type IV. J. Neurosci. 20, 8247-8253. Turbeck, O. B. and Foder, B. (1970). Studies on a carbonic anhydrase from the midgut epithelium of larvae of lepidoptera. Biochim. Biophys. Acta. 212, 139-149. Tusnady, G. E. and Simon, I. (1998). Principles governing amino acid composition of integral membrane proteins: application to topology prediction. J. Mol. Biol. 283, 489506. Tusnady, G. E. and Simon, I. (2001). The HMMTOP transmembrane topology prediction server. Bioinformatics 17, 849-50. Vince, J. W. and Reithmeier, R. A. (2000). Identification of the carbonic anhydrase II binding site in the Cl(-)/HCO(3)(-) anion exchanger AE1. Biochemistry 39, 5527-5533.

PAGE 159

146 Vince, J. W., Carlsson, U. and Reithmeier, R. A. (2000). Localization of the Cl/HC03anion exchanger binding site to the amino-terminal region of carbonic anhydrase II. Biochemistry 39, 13344-13349. Volkmann, A. and Peters, W. (1989a). Investigations on the midgut caeca of mosquito larvae I. Fine structure. Tissue & Cell 21, 243-251. Volkmann, A. and Peters, W. (1989b). Investigations on the midgut caeca of mosquito larvae II. Functional aspects. Tissue & Cell 21, 253-261. Waheed, A., Okuyama, T., Heyduk, T. and Sly, W. S. (1996). Carbonic anhydrase rV: purification of a secretory form of the recombinant human enzyme and identification of the positions and importance of its disulfide bonds. Arch. Biochem. Biophys. 333, 432438. Westerfield, M. (1994). The zebrafish book: A guide for the laboratory use of zebrafish (Brachydanio rerio). (ed. Westerfield, M.), University of Oregon Press, Eugene, OR. pp. 9.16-9.21. Wieczorek, H., Gruber, G., Harvey, W.R., Huss, M. and Merzendorfer, H. (1999). The plasma membrane Ff-V-ATPase from tobacco hornworm midgut. J. Bioenerg. Biomembr. 31, 67-74. Wieczorek, H., Gruber, G., Harvey, W. R., Huss, M., Merzendorfer, H. and Zeiske, W. (2000). Structure and regulation of insect plasma membrane H(+)V-ATPase. J. Exp. Biol. 203 (Pt 1), 127-135. Zhang, Y. and Frohman, M. A. (1997). Using rapid amplification of cDNA ends (RACE) to obtain full-length cDNAs. Methods Mol. Biol. 69, 61-87. Zhuang, Z., Linser, P. J. and Harvey, W. R. (1999). Antibody to H(+) V-ATPase subunit E colocalizes with portasomes in alkaline larval midgut of a freshwater mosquito (Aedes aegypti). J. Exp. Biol. 202 (Pt 18), 2449-2460.

PAGE 160

BIOGRAPHICAL SKETCH Theresa J. Seron was born and raised in Connecticut with her older sister and younger brother. She enjoyed the outdoors and wildlife from a very young age and continues to pursue activities such as scuba diving and kayaking. She received her Bachelor of Science degree from the University of Connecticut in 1995. She then ventured off to Saint Thomas in the U.S. Virgin Islands, where she lived for twelve months to satisfy her enthusiasm for traveling and diving. Upon returning to the United States, she was employed by Boehringer-Ingelheim Pharmaceuticals in Danbury, Connecticut, where she was introduced to the world of scientific research and discovery. This path was strengthened by a relocation to Miami, Florida, and a second research position at Noven Pharmaceuticals. In 1997, she moved to Gainesville, Florida, to pursue an advanced degree at the University of Florida. She became acquainted with The Whitney Laboratory in Saint Augustine, Florida, when she was chosen to participate in a summer research program. The following fall she was admitted to the University of Florida graduate school in the Department of Fisheries and Aquatic Sciences. Here she was able to combine her research background with her enthusiasm for aquatic animals. Her research project was carried out at The Whitney Laboratory under the direction of Dr. Paul Linser. The project focused on the enzyme carbonic anhydrase and its unknown role in larval mosquito physiology. Upon completion of her doctorate degree she plans to continue scientific research within the aquatic realm. 147

PAGE 161

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Edward J. Philips, Chair Professor of Fisheries and Aquatic Sciences I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, injicope and quality, as a dissertation for the degree of Doctor of Philosophy. Professor of Anatomy and Cell Biology I certify that I have read this study and that in my opinion it confoi standards of scholarly presentation and is fully adequate, in scot, dissertation for the degree of Doctor of Philosophy. cceptable as a L. Moroz Assistant Professor of Neuroscience I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Shirley M. Baket Assistant Professor of Fisheries and Aquatic Sciences I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy^/ Robert M. Greeniberg Associate Professor of Neuroscience

PAGE 162

This dissertation was submitted to the Graduate Faculty of the College of Agricultural and Life Sciences and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doct^rjif-Ehilosophy^ May 2004 Dean, Collegeof Agriculture Sciences Dean, Graduate School


ACKNOWLEDGEMENTS
I would like to acknowledge my dissertation committee, Dr. Paul J. Linser, Dr.
Edward J. Phlips, Dr. Leonid Moroz, Dr. Robert Greenberg, and Dr. Shirley Baker, for
their suggestions and comments on the final rewriting of this document. I want to thank
my project supervisor, Dr. Paul J. Linser, for allowing me to form my own project goals
and the space to tackle them.
There are a number of people at The Whitney Laboratory who I would like to
thank for their assistance with this dissertation project as well as their friendship. Dr.
Judith Ochrietor devoted her time and energy to improving every aspect of this
dissertation. Judy assisted me with experimental designs, introduced me to real time
PCR, provided a wealth of knowledge about molecular biology, and was a great person
with which to share a laboratory and office. Dr. Andrea Kohn provided molecular
biology teaching and advice along with being a fantastic person to work with and be
inspired by. Leslie vanEkeris taught me how to do mosquito dissections and provided
many of the mosquito guts that I photographed for this document. Dr. Bill Harvey
provided insight into the ionic transport mechanisms of the mosquito and the editing of
this manuscript. Dr. Dmitri Boudko was instrumental in the expression of the anion
exchanger and the production of amplified cDNA libraries. Jessica Roberts-Misterly and
Dr. Robert Greenberg also provided teachings and suggestions in cloning cDNAs from
the mosquito.
iv


5
differential roles in homeostasis and function. Elucidating the distribution of CAs along
the mosquito midgut epithelium may uncover the mechanisms responsible for the unique
alkaline physiology of the mosquito gut.
Mosquito Development and Control
Part of the success of insects can be attributed to the structural adaptation of their
integument, which functions as skin, skeleton, sensory and respiratory organ, and food
reserve (Rockstein, 1964). The advantage of having an extremely strong integument is
offset by the disadvantage of not being able to grow significantly in size. Insects have
overcome this growth-limiting problem by shedding their integument and rebuilding a
new larger one. This process of ecdysis (molting) is used as a tool for marking the
different stages of development in many insect species. While mosquito control can
target different stages of mosquito development, this project focuses on the larval
enzymes, specifically early fourth instar, which begins immediately after the third molt.
Careful attention was paid to the stage of insect development in all experiments due to a
previous study that showed insect enzymes to decrease or completely arrest prior to
molting (Jungreis et ah, 1981).
The mosquito life cycle begins at hatching from the egg (Fig. 1-2). At this point
the fully independent mosquito is called a first instar larva. Successive molts mark the
transition to the next larval instar, four larval instars in all. In each instar, the larvae
possess a series of morphological characteristics, some particular to that stage. However,
there are only slight changes in internal organs such as the midgut. Within a day or two
the late fourth instar larva changes into a pupa (Clements, 1992). Within twenty-four
hours, the flying adult emerges from the pupa case. Adult females of most mosquito


129
Figure 6-8. Localization of CA protein within the PMG of An. gambiae. The CA-
specific antibody labels cell membranes of both large (double arrows) and
small cells (small arrows), mimicking the protein localization pattern for the
AgAEl protein. Since this particular CA isoform is predicted to be a
cytosolic isoform, its localization to cell membranes suggests an interaction
with a membrane protein. The AgAEl protein, known to be a membrane
protein, capable of binding CA II, and the localization of the AE1 protein to
the same cells supports the hypothesis of a bicarbonate metabolon within the
mosquito gut. Scale bar represents 25 pm.


49
It is interesting to note that the lowest concentration of CA in the midgut
epithelium occurs in the region that surrounds and probably regulates the region of
highest luminal pH, the anterior midgut. The pKa of CO3'2 is approximately 10.5 and,
hence, this anion is likely to be the primary buffer of the pH 10.5-11 gut contents within
the anterior midgut. Our results therefore suggest that the major buffering anion in this
area of the midgut is probably not produced by local CA but instead either upstream, in
the gastric caeca, or downstream, in the posterior midgut, where CA levels are very high.
This result, and results presented elsewhere (Boudko et al., 2001a), are consistent with a
model in which a major function of the anterior midgut is to pump protons out of this
region of the gut lumen, promoting the conversion of HCO3 to CO32. A comprehensive
model of the regulation of ion homeostasis and gut alkalization in the larval mosquito
awaits the characterization and localization of other major components of the system in
addition to CA. It will also be very important to resolve the question of whether multiple
CAs are expressed in the midgut and how each is distributed in this dynamic tissue.
Quantitative evidence corroborating the distribution of CA within the midgut and
10
supporting the histochemical and in situ observations was obtained using the O-
exchange mass spectrometric method. The results obtained with this method indicate that
the gastric caeca exhibit the highest level of carbonic anhydrase, relative to total protein
content, followed by the posterior midgut and the Malpighian tubules. The anterior
midgut showed levels of activity so low that two possibilities could be considered: either
the method could not detect the enzyme or it is absent from the anterior midgut. The
presence of faint staining using the histochemical and in situ methods suggests that the


73
AAL72625 Aedes CA
AAQ21365 Anoph CA
CG3940-PA Dros CA
P00915 Human CA I
P00918 Human CA II
P07451 Human CA III
P22748 Human CA IV
AAB47048 Human CA V
CAC42429 Human CA VI
P43166 Human CA VII
JN0576 Human CA VIII
AAH14950 Human CA IX
Q9NS85 Human CA X
AAH02662 Human CA XI
AAH23981 Human CA XII
BAA85002 Human CA XIV
FVLDQMHFHWG SEHTIAGVRYGQELHMVHHDS
FVLDQMHFHWG SEHTLDDTRYGLELHLVHHDT
FWEQIHMHWW SEHTINDIRYPLEVHIVHRNT
YRLFQFHFHWGSTNEHGSEHTVDGVKYSAELHVAHWNS
YRLIQFHFHWG--SLDGQGSEHTVDKKKYAAELHLVHWNT
YRLRQFHLHWGS SDDHG SEHTVDGVKYAAELHLVHWNP
YQAKQLHLHWSDLPYKGSEHSLDGEHFAMEMHIVHEKE
YRLKQFHFHWG--AVNEGGSEHTVDGHAYPAELHLVHWNS
YIAQQMHFHWGGASSEISGSEHTVDGIRHVIEIHIVHYNS
YRLKQFHFHWG--KKHDVGSEHTVDGKSFPSELHLVHWNA
FELYEVRFHWGRENQRGSEHTVNFKAFPMELHLIHWNS
YRALQLHLHWG--AAGRPGSEHTVEGHRFPAEIHWHLST
HRLEEIRLHFGSEDSQGSEHLLNGQAFSGEVQLIHYNH
HRLSELRLLFG--ARDGAGSEHQINHQGFSAEVQLIHFNQ
YSATQLHLHWG-NPNDPHGSEHTVSGQHFAAELHIVHYNS
YVAAQLHLHWG-QKGSPGGSEHQINSEATFAELHIVHYDS
*** *... *
Figure 4-2. Clustal alignment of CA protein sequences. All characterized human CA
isoforms are presented along with putative GPI-linked isoforms from Ae.
aegypti, An. gambiae, and D. melanogaster. Three histidine residues that are
required for the essential binding of zinc are shaded in blue. Note that 1 or
more of these histidine residues are missing from the inactive human CA-
related proteins VIH, X, and XI while all three histidines are present within
the Dipteran sequences. The three CAs from Dipterans contain a shortened
active site region (marked by red dashes) when compared to any of the human
or other mammalian CA sequences. This difference may provide a potential
target for mosquito-specific CA inhibitors, for use as larvacides.


94
but has yet to be cloned and characterized (Kim et al., 1999). Like B cells, anterior
mosquito midgut cells have been shown to express a basolateral bT V-ATPase whereas
expression of AgAEl was absent from this region. The closely associated A and B
intercalated cells of the mammalian kidney are known to function in acid secretion and
bicarbonate secretion, respectively. Interestingly, the bicarbonate-secreting moiety, for
which the function of the B cell is defined, is unknown.
The co-occurrence of these A and B cell types provides for the tight regulatory
control of the maintenance of near-neutral pH in the kidney. Perhaps the pH differential
of 4 units, as seen in the mosquito anterior midgut, is achieved by the decoupling of these
cell types. In the mosquito gut, the A-like cells of the gastric caeca and posterior midgut
translocate protons toward the lumen and reduce the alkaline pH to near neutral levels.
The B-like cells of the alkaline anterior midgut secrete bicarbonate and protect the cells
from the high lumenal pH. Although the presented AgAEl is not that moiety, the
parallels between cells of the mammalian kidney and the mosquito midgut are evident.
The similarities between epithelial cells of the mammalian kidney and the mosquito
midgut support the use of the mosquito midgut as a simple model in which to study cell
polarization, pH balance, protein targeting and trafficking, as well as disease states. The
decoupling of A and B intercalated cells, as they may exist in the mosquito midgut,
provides an excellent model for studying human diseases such as distal renal tubular
acidosis, which is caused by mutations in either the basolateral AE1 or different subunits
of the apical fT V-ATPase (Alper, 2002).
The simple epithelium of the mosquito midgut may continue to reveal
mechanisms and pathways that also function within the complex metabolic network


126
Figure 6-5. Localization of CA mRNA expression within the posterior midgut o An.
gambiae. A. The PMG displays staining for CA mRNA within the
peripheral borders of both the large and small epithelial cells. B. Higher
magnification of the PMG shows the stained large columnar cells (*) and
the labeled small cuboidal cells (arrows). C. Side view of the PMG reveals
CA mRNA expression near the plasma membranes (arrows) and not
throughout the cytoplasm. Scale bars represent 150 pm in A, 25 pm in B,
and 50 pm in C.


45
product was also discernible in the anterior midgut. This PCR analysis also revealed
higher molecular mass products in the anterior midgut and Malpighian tubules that may
represent additional carbonic anhydrases specific to larval Ae. aegypti (Figure 3-6). This
result is shown only to display the gut regions in which the A-CA clone was derived.
The lack of an 894 bp product in the other gut regions may simply be due to poor quality
cDNA pools from those regions. However, the cloning of A-CA from both the gastric
caeca and posterior midgut regions is consistent with the location of enzyme activities
described above.
Localization of the Enzyme in the Midgut Epithelium: Carbonic Anhydrase Enzyme
Histochemistry
To further analyze the regional and cellular expression of CA in the midgut
epithelium of larval mosquitoes, a modified Hanssons histochemical reaction was
performed on whole mount preparations of the gut (Hansson, 1967). Figure 3-7
summarizes the results of this analysis. Carbonic anhydrase activity was detected in a
non-uniform pattern along the length of the gut. The most intense staining was evident in
the gastric caeca and the posterior midgut. Staining was less intense in the anterior
midgut. At higher magnification, it was obvious that cellular heterogeneity with regard
to CA activity also exists. This is particularly evident in the posterior midgut, where very
large and regularly spaced cells appear nearly white on a background of dark CA reaction
product. The larger cells have been characterized as columnar or ion-transporting cells
(Volkman and Peters, 1989b). Surrounding these large cells are more numerous smaller
cells termed cuboidal or resorbing/secreting cells (Zhuang et ah, 1999). The CA
histochemical stain clearly distinguishes these cells from one another and indicates that
the large columnar cells contain relatively veiy little CA in comparison with the smaller


56
A
Aselas aagypti
HoraaCAl-P00917
Zbra£ih-Q92051
HuunCX3-P07451
MouasCAl 4 -NP035927
C.alaganfl-Tl6575
RatGAIV-NPO62047
MAH
46
2 8
2 8
28
45
24
50
Aftdos aagypti
Hora*CAl-P00917
2abrafih-Q92051
HimanCA3 P0 7 4 51
MOU8OCA14-NP035927
C.ttl*gana-T16575
RatCAIV-NP062047
94
77
77
77
92
74
<>8
Ascias Agypti
HoraaCAl-P00917
Zibrafxah-Q92051
HuoanCA3-P07451
MouaCA14-HP035927
C.lganB-T16B75
RatCAIV-HPO62047
137
125
125
125
142
121
146
Aadss aagypti
HorsaCAl -POOS17
Zbra£iah-Q92051
HuBanCA3-P0745l
MouseCAl4 -HP035927
C.alagana-Tl6575
RatCAIV-NP062047
186
175
174
174
192
162
195
Aadas aagypti.
HoraCAI-POO 917
ZabrafiahQ92051
HuaanCA3-P07451
MouaaCAl4-NP035927
C.alagana-Tl6575
RatCAJV-NP062047
236
222
221
221
240
210
245
Aadas aagypti
HoreaCAl-P00917
Zebrafieh-Q92051
HuunCA3-P07451
MouaaCA14-NP035927
C.alagana-Tl6575
RatCAIV-MP062047
282
260
260
260
290
246
294
Aadaa aagypti
HoraaCAl-P00917
Zbrmfiah>Q920Sl
HuaanCA3-P07451
MouaoCA14-NP035927
C.alogans-T1657B
RatGATV-KP062047
L3LTLIVAAIAXLLAK.
LGLGVGILAGCLCLLIAVY FI AQKI R1 LLVPTLTCLVASFLH
298
260
260
260
337
246
30 9
B
Homology (%)
Figure 3-4. Carbonic anhydrase from the midgut of larval Ae. aegypti. (A) Alignment
(BLAST) of the predicted amino acid sequence of Ae. aegypti cDNA with
several known a-carbonic anhydrases. Regions of exact homology across
all species are highlighted in blue (100%); regions with less homology are
highlighted in red (>75%) and green (>50%). (B) A homology tree
comparing A-CA and several other a-CAs (DNAman software).


32
CA Protein Expression
Recombinant Ae. aegypti and An. gambiae CAs were produced using the pETlOO
vector (Invitrogen). Specific primers were designed to amplify each cDNA. The 3
primers included the sequence 5 to and including the native stop codon. The 5 primers
contain the sequence CACC preceeding the native start codon for correct frame insertion
(See Table 2-1 for primer sequences). PCRs were performed using 1 U of Platinum P/x
polymerase (Invitrogen), the gastric caeca cDNA collections as template (200 ng), IX
Pfx amplification buffer, 1.2 mM dNTP mixture, 1 mM MgS04, and 0.3 pM of each
primer in a total volume of 50 pL. A three-step PCR protocol was used consisting of
94C for 2 minutes followed by 30 cycles of 94C for 30 seconds, 55C for 30 seconds,
and 68C for 1 minute.
The resultant blunt-ended cDNAs (4 pL from PCR mix) were ligated with the
pETlOO directional Topo vector (1 pL and 1 pL salt solution; Invitrogen) for 10 minutes
at room temperature. Top 10 chemically competent E. coli (50 pL; Invitrogen) were
transformed by incubating 3 pL of ligation mix with the cells for 30 minutes on ice,
followed by a heat shock of 42C for 30 seconds. SOC (250 pL) was added to the cells
and they were then incubated at 37C for 30 minutes with shaking. The transformation
mix (100 pL) was then plated on a LB-carbenicillin plate (50 pg/mL) and incubated
overnight at 37C. Colonies were sequenced using Big Dye version 1.1 as described
previously. The purified plasmids (10 ng each) were transformed into BL21 Star (DE3)
cells (Invitrogen) for CA expression as described above. However, after SOC addition
and incubation, the culture was transferred to fresh LB-carb (10 mL) and grown
overnight at 37C with shaking. The next day, 1 mL of culture was transferred to 100


41
enzymatic reaction takes only a few seconds, and it can be delayed if the solutions, the
paper and the samples are kept cold on ice. However, a few seconds is usually sufficient
to discriminate the samples that contain CA from those lacking enzymatic activity. The
assay must be performed quickly since, after approximately one minute the entire filter
paper turns yellow, probably as a result of the uncatalyzed hydration of carbon dioxide
absorbed by the solution at this basic pH.
The test has proved useful in determining the presence of small amounts of CA in
homogenates of mosquito larvae. The assay was also used to detect CA activity
qualitatively, in fractions obtained from affinity chromatography (Osborne and Tashian,
1975) of larval homogenates. The affinity chromatographic procedure, which employs a
bound CA inhibitor (p-aminomethyl benzyl sulfonamide (p-AMBS); Sigma), produced
two peaks of CA activity upon exposure to the standard elution buffers. The amount of
protein that we were able to produce by this technique was, however, very small and
resisted several efforts at direct microsequencing. This change in color was inhibited by
acetazolamide and methazolamide when these inhibitors (105 M) were added to the
samples prior to spotting on the dye-impregnated filter papers. Inhibition of the reaction
resulted in blue spots that did not change color upon addition of CO2. The positive
control containing commercial CA turned yellow when carbon dioxide was added, and
this color change was also inhibited by acetazolamide and methazolamide. This finding
confirmed that the yellow color of the spots was due to the action of CA and that the
mosquito larva contains active CA.


27
55C with 50% formamide in 2X SSCT for 30 min (twice), 2X SSCT for 15 min and
0.2X SSCT for 30 min (twice). For detection, the tissue was incubated in PBST
containing 1% blocking solution (Roche Molecular Biochemicals) for 1 h at room
temperature. The tissue was incubated with anti-DIG-alkaline phosphatase (Roche
Molecular Biochemicals) diluted 1:5000 in blocking solution for 4 hours at room
temperature. The tissue was washed with PBST and incubated in alkaline phosphatase
substrate solution (Bio Rad Laboratories, Hercules, CA, USA) until the desired intensity
of staining was achieved (2-3 hours).
CA Histochemistry
Carbonic anhydrase activity was detected in isolated Ae. aegypti midguts using
Hansson's method (Hansson, 1967), as modified by Ridgway and Moffet (1986). The
procedure involved the incubation of isolated, 3% glutaraldehyde-fixed midguts in 1.75
mM C0SO4, 53 mM H2S04, 11.7 mM KH2P04, and 15.7 mM NaHC03 (pH 6.8). The
incubation medium contains a high concentration of bicarbonate, which stimulates the
production of C02 and hence a decrease in pH in the presence of CA. The acidic pH then
stimulates the formation of insoluble black cobalt salts which were visualized using 0.5%
(NHi)2S in distilled water. Therefore, micro-sites of active CA liberation of C02 from
bicarbonate dehydration become apparent with this assay. Removal of the bicarbonate
substrate (NaHC03) eliminated staining.
Real Time PCR
Region-specific cDNA was produced from dissected mosquito tissue using the
Cells-to-cDNA standard protocol (Ambion INC, Austin, Texas). The gut regions used to
make the amplified cDNA pools were incubated in 50 pL of hot cell lysis buffer for 10


101
Figure 5-6. Localization of AgAEl mRNA within whole mount An. gambiae larvae. A.
Gastric caeca (GC), posterior midgut (PMG), and Malpighian tubules (MT)
show extensive expression while other gut regions display more restricted
hybridization. B. The gastric caeca as well as the cardia region (arrows)
display label. C. Extensive expression of AgAEl mRNA in the Malpighian
tubules. D. Specific cells (arrows) and trachea (arrowheads) of the rectum
show expression of AgAEl. Scale bars represent 300 pm in A, 100 pm in B,
and 25 pm in C and D.


47
Even though a number of genes and their products have been isolated from the
midgut of Ae. aegypti, and the role of CA in the alkalization of the midgut has been
suggested (Turbeck and Foder, 1970; Haskell et al., 1965; Ridgway and Moffett, 1986;
Boudko et al., 2001b), there have been no reports of the isolation or cloning of CA or of
the localization of the enzyme within the midgut of larval mosquitoes. This is the first
recorded cloning of a CA from a mosquito, and is also the first to be cloned from any
arthropod. Our results show that at least one (and perhaps more) CA is present in the
midgut of larval Ae. aegypti. The CA of larval Ae. aegypti (A-CA) is inhibited by
classical carbonic anhydrase inhibitors such as methazolamide and acetazolamide.
Methazolamide has the most potent effect on A-CA. Direct physiological measurements
of ion fluxes from living larval mosquito midgut epithelial cells also show
methazolamide to be a very potent inhibitor of ion movements and balance (Boudko et
al., 2001a).
To investigate the distribution of CA in the midgut of the larval mosquito, we
employed both in situ hybridization and enzyme histochemistry. Our results indicate that
enzymatic activity is greatest in the gastric caeca and the posterior midgut, as
demonstrated by the intense staining obtained using Hansson's method and by in situ
hybridization using cRNA probes. Measurements of activity using the l80 exchange
method in pools of dissected regions of the gut corroborate these findings. In addition,
the enzyme seems to be preferentially associated with the small cuboidal cells in the
midgut epithelium, as determined both by enzyme histochemistry and by in situ
hybridization.


CHAPTER 1
INTRODUCTION
Insects represent one of the most numerous and diverse groups of animals on the
planet. One particularly successful group of insects is the well-studied Pterogota (winged
insect) group. This grouping includes the Lepidoptera (butterflies and moths) as well as
the Dptera (flies). These insects have been extensively studied due to their huge impact
on the lives of humans. For example, the Lepidopteran, Manduca sexta, is a great pest to
tobacco companies that rely on abundant and healthy tobacco crops, on which M. sexta
feeds. Also, mosquitoes (Dipterans) are responsible for transmitting a host of diseases to
humans as well as other mammals by injecting pathogens, along with the anti-coagulants
from their salivary glands, to aid in bloodletting. The pathogens that cause these diseases
can be viruses or various parasites (eg. protozoans).
Mosquitoes belong to the order Dptera, family Culicidae. According to the
American Mosquito Control Association, there are more than 2500 different species
throughout the world, with 150 species in the United States (Darsie and Morris, 2000;
Spielman and D'Antonio, 2001). Mosquitoes act as vectors for a wide variety of diseases
such as malaria, yellow fever, west nile virus, and dengue fever. Recent reports estimate
that fifty to one hundred million cases of dengue fever occur annually, along with several
hundred thousand cases of the life-threatening form of the disease, dengue hemorrhagic
fever (DHF; Halstead, 1997). The geographic range of dengue fever has expanded over
the last two decades, primarily because of the spread of its principal vector, Aedes aegypti
1


122
AAQ21365 CAIV-like
AAQ21366 CAII-like
ENSANGPO0000001812
ENSANGP00000018999
ENSANGPO0000029518
ENSANGP00000011908
ENSANGP00000001574
ENSANGP00000011013
ENSANGPO0000012957
ENSANGP00000016412
ENSANGPO0000010017
ENSANGP00000014948
ENSANGPO0000014919
ENSANGP00000001278
CG9235-PA Dros
CG18672-PA Dros
CG10899-PA Dros
CG11284-PA Dros
CG12309-PA Dros
CG3940-PA Dros
CG32698-PA Dros
CG6074-PA Dros
CG18673-PA Dros
CG1402-PA Dros
CG3669-PA Dros
CG6906-PA Dros
CG5379-PA Dros
CG7820-PA Dros
P00915 CA-I
P00918 CA-II
P07451 CA-III
P22748 CA-IV
AAB47048 CA-V
CAC42429 CA-VI
P43166 CA-VII
JN0576 CA-VIII
AAH14950 CA-IX
Q9NS85 CA-X
AAH02662 CA-XI
AAH23981 CA-XII
AAK16672 mCA-XIII
BAA85002 CA-XIV
QMHFHWG SEHTLDDTRYGLELHLVH
QLHFHWGIGDGSG JEHTLEGSTYSMEAHAVH
QFHFHAP SENLIKGHSYPLEGHLVH
QLHFHWGLSALDGSEHTIDGYRLPLELHVIH
QFHCHWGCSDSR- -GSEHTVDGESFAGELHLVH
EIHVHYGLHDQF--GSEHSVEGYTFPAEARHIQ
EIYFHYGTDNNQGSEHHIHGYSFPGEIQLYG
QLHFHWGDNDTF- -GSEDMIDNHRFPMELHWF
GLHFHWGDKNNR--GAEHVLNDIRYPLEMHIIH
QFHCHWGCSDSR- -GSEHTVDGESFAGELHLVH
QLHFHWGPDDAV- -GSEHLLDGRAHSMEAHLVH
QLHFHWGADNGRGSEHTFDGVAWAAEAHFVF
QFHFHWGVNSTVGSEHVYDYQRYPMEIHLVF
QMHFHWGPNNSEGSEHRINGERFPLEVHLVF
EISFRWSWASSLGSEHTLDHHHSPLEMQCLH
GLHFHWGSYKSR--GSEHLINKRRFDAEIHIVH
QLHFHWGSALSK--GSEHCLDGNYYDGEVHIVH
QLHFHWSDCDESG JEHTLEGMKYSMEAHAVH
ELRFHWGWCNSE- -GSEHTINHRKFPLEMQVMH
QIHMHWW SEHTINDIRYPLEVHIVH
EIHMHYGLNDQF--GSEHSVEGYTFPAEIQIFG
GLHFHWGDKNNR--GSEHVINDIRYTMEMHIVH
SVHFHWGSREAK- -GSEHAINFQRYDVEMHIVH
EIYIHYGTENVR--GSEHFIQGYSFPGEIQIYG
AFHFHWGSPSSRGSEHSINQQRFDVEMHIVH
QFHFHWGENDTI -GSEDLINNRAYPAELHWL
AVHFHWGSPESK--GSEHLLNGRRFDLEMHIVH
QFHCHWGCTDSK--GSEHTVDGVS YSGELHLVH
QFHFHWGSTNEH--GSEHTVDGVKYSAELHVAH
QFHFHWGSLDGQ--GSEHTVDKKKYAAELHLVH
QFHLHWGSSDDH--GSEHTVDGVKYAAELHLVH
QLHLHWSDLPYK--GSEHSLDGEHFAMEMHIVH
QFHFHWGAVNEG--GSEHTVDGHAYPAELHLVH
QMHFHWGGAS SEISGSEHTVDGIRHVIEIHIVH
QFHFHWGKKHDV--GSEHTVDGKSFPSELHLVH
EVRFHWGRENQR- -GSEHTVNFKAFPMELHLIH
QLHLHWGAAGRP--GSEHTVEGHRFPAEIHWH
EIRLHFGSEDSQGSEHLLNGQAFSGEVQLIH
ELRLLFGARDGA--GSEHQINHQGFSAEVQLIH
QLHLHWGNPNDPH-GSEHTVSGQHFAAELHIVH
QFHLHWGSADDHGSEHWDGVRYAAELHWH
QLHLHWGQKGSPG-GSEHQINSEATFAELHIVH
*
Figure 6-1. Clustal alignment of active sites within An. gambiae, D. melanogaster, and
human CA proteins. All active a CAs are tethered to a zinc molecule by the
coordination with three histidine (H) residues. There are three human CAs
which do not possess H residues in this orientation and have been shown to
lack CA activity. Within the active site region, some of the An. gambiae CAs
display novel differences, as compared to the tightly-conserved human CAs.
Without exception in the human isoforms, a conserved glutamine (E; marked
with *) always follows a serine (S). In contrast, the An. gambiae alignment
shows a change from the S to a cysteine (C) in one of their 14 putative
isoforms. The An. gambiae alignment also displays a gap within the active
site regions of two of the CAs. One of these sequences (AAQ21365) was
cloned and determined to be a GPI-linked CA IV-like isoform that was
discussed in a previous chapter. These novel active site differences may be
exploited in the formulation of a specific mosquito larvacide. The same active
site differences are also displayed within the D. melanogaster genome and
therefore may also provide clues to the evolutionary mechanism of these
proteins.


143
Letunic, I., Goodstadt, L., Dickens, N. J., Doerks, T., Schultz, J., Mott, R., Ciccarelli,
F., Copley, R. R., Ponting, C. P. and Bork, P. (2002). Recent improvements to the
SMART domain-based sequence annotation resource. Nucleic Acids Res. 30, 242-4.
Lindskog, S. (1997). Structure and mechanism of carbonic anhydrase. Pharmacol. Ther.
74,1-20.
Marchler-Bauer, A., Panchenko, A. R., Shoemaker, B. A., Thiessen, P. A., Geer, L.
Y. and Bryant, S. H. (2002). CDD: a database of conserved domain alignments with
links to domain three-dimensional structure. Nucleic Acids Res. 30,281-283.
Martin, M. M., Martin, J. S., Kukor, J. J. and Merrit, R. W. (1980). The digestion of
protein and carbohydrate by the stream detritivore, Tpula abdominalis (Dptera,
Tipulidae). Oecologia 46,360-364.
Matsumoto, T., Winkler, C. A., Brion, L. P. and Schwartz, G. J. (1994). Expression
of acid-base-related proteins in mesonephric kidney of the rabbit. Am. J. Physiol. 267,
F987-997.
Matz, M., Shagin, D., Bogdanova, E., Britanova, O., Lukyanov, S., Diatchenko, L.
and Chenchik, A. (1999). Amplification of cDNA ends based on template-switching
effect and step-out PCR. Nucleic Acids Res. 27, 1558-1560.
Meldrum, N. U. and Roughton, F. J. W. (1933). Carbonic anhydrase: Its preparation
and properties. J. Physiol. London 80, 113-142.
Osborne, W. R. A. and Tashian, R. E. (1975). An improved method for the purification
of carbonic anhydrase isozymes by affinity chromatography. Anal. Biochem. 64, 297-
303.
Parkkila, S., Rajaniemi, H., Parkkila, A. K., Kivela, J., Waheed, A., Pastorekova, S.,
Pastorek, J. and Sly, W. S. (2000). Carbonic anhydrase inhibitor suppresses invasion
of renal cancer cells in vitro. Proc. Natl. Acad. Sci. USA 97, 2220-2224.
Phillips, K. P. and Baltz, J. M. (1999). Intracellular pH regulation by HC03-/C1-
exchange is activated during early mouse zygote development. Dev. Biol. 208, 392-405.
Ridgway, R. L, and Moffett, D. F. (1986). Regional differences in the histochemical
localization of carbonic anhydrase in the midgut of tobacco homworm (Manduca
sexta). J. Exp. Zool. 237, 407-412.
Rockstein, M, (1964). The physiology of insecta, (ed. Rockstein, M.), Academic Press,
New York, NY. pp. 380-387.


36
Salt for 98mM value
mW
#1
#2
#3
#4
98N
98k
98N-CI
98K-CL
mM
p/1
p/I
mM
mM
mM
g/l
S/I
Solution:
lx
4x
lx
4x
lx
4x
lx
4x
NaCl
58.44
n
5.73
22.91
2
0.12
0.47
0
0
0
0
KC1
74.55
2
0.15
0.60
98
7.31
29.22
0
0
0
0
Na Gluconate
218.1
0
0
0
0
98
21.37
85.50
2
0.44
1.74
K Gluconate
234.2
0
0
0
0
2
0.47
1.87
98
22.95
91.81
Choline Cl
139.6
0
0
0
0
0
0
0
0
MgS047H20 (120.36)
246.5
0
0
0
0
0.5
0.12
0.49
0.5
0.12
0.49
MgCl26H20(59.7)
203
0.5
0.10
0.41
0.5
0.10
0.41
0
0
0
0
CaCl22H20 (110.98)
147.02
0.5
0.07
0.29
0.5
0.07
0.29
0
0
0
0
Ca Gluconate
430.38
0
0
0
0
0.5
0.22
0.86
0.5
0.22
0.86
HEPES (free base)
238.3
10
2.38
9.53
10
2.38
9.53
10
2.38
9.53
10
2.38
9.53
EGTA for InsideOout
380.4
0
0
0
0
0
0
0
0
pH
| 7.2 (4M NaOH)
7.2 (4M KOH)
7.2 (4M NaOH)
7.2 (4M KOH)
Table 2-2. Composition of all solutions used in Xenopus oocyte expression of An.
gambiae AE. Total molarity and pH were kept constant in all solutions.
Expression profiles were recorded in high sodium (#1), high sodium minus
chloride (#3), high potassium (#2), and high potassium minus chloride (#4).


6
species require a bloodmeal in order to nourish their developing eggs. However, the
males do not ingest blood but instead feed on fruit or do not feed at all (Clements, 1992).
Mosquito control tactics use different methods for controlling mosquito larvae as
compared to the flying adults. Mosquito larvae are confined to the water in which they
develop, whereas the adults are free-flying and therefore highly mobile. Pesticide sprays
are employed against the flying adult mosquitoes, but dragonflies and butterflies are also
ill-affected. An arguably better strategy for mosquito control is to target the larvae before
they are capable of biting and transmitting disease. Mosquito larvae are voracious eaters,
incessantly consuming particulates in the water around them, taking in almost anything.
Because of this non-discretional eating behavior, the wriggling larvae can potentially
consume a larvacidal agent if placed in the water. Determining the physiological roles of
larval mosquito gut enzymes and metabolic transporters may provide a lead for
constructing mosquito larvacides.
Carbonic Anhydrase Inhibition
The focus of this project is to examine the distribution and expression of CAs
within the fourth instar of larval development of two species of mosquito, Ae. aegypti and
An. gambiae. A tangential result of characterizing mosquito CAs may be in the
development of mosquito-specific inhibitors. If a CA is discovered to be essential for
mosquito development or homeostasis, a specific inhibitor of precisely this mosquito CA
isoform could be developed. Since virtually all organisms contain CA enzymes, an
inhibitor that would compromise this mosquito CA while not affecting any other
isozymes would be necessary for mosquito control so that non-target species would not
be affected. Differentially specific CA inhibitors are already employed in the distinctive


77
Aedes aegypti Carbonic Anhydrase
Ae. aegypti Tissue Sections
Figure 4-6. Relative quantification of CA IV-like message in Ae. aegypti larvae using
real time PCR. The gastric caeca tissue displays the greatest amount of CA
IV-like message. Data was normalized to the gastric caeca (GC) sample. The
anterior and posterior midgut along with the Malpighian tubules display very
little CA message. The head section displays roughly half the amount of
message found in the gastric caeca.


64
Although the mosquito CA isoforms display features similar to mammalian CA
TVs, such as a 5 signal sequence, a hydrophobic 3 tail, and extracellular GPI expression,
there is one striking difference in the amino acid composition of the mosquito CAs'
active sites. The active site within all of the 14 characterized mammalian CA isoforms is
tightly conserved. Three histidine residues (His-94, His-96, and His-119) are essential
for CA activity through their coordinated binding of a required zinc molecule. The
absence of one or more of these histidine residues results in inactive proteins called CA-
related proteins (CARPs), as found in mammalian CA isoforms VIII, X, and XI (Tashian
et al., 2000). The mosquito CA IV-like isoforms contain all three of the required
histidine residues along with all of the other 13 highly conserved residues found in most
other CAs (refer to Fig 4-1; Tashian, 1992; Sly and Hu, 1995; Tamai et al., 1996a).
However, as the alignment shows in Figure 4-1, there is a conserved gap within the active
site of the mosquito CAs that is not present in any of the mammalian active sites.
Because this shortened active site was found in mosquitoes but was not found in any
mammalian CA isoform, I searched the Drosophila melanogaster genome for potential
CA homologs. The D. melanogaster genome was found to contain 14 putative CA genes
(ENSF00000000228), the same number found in An. gambiae. One out of the fourteen
D. melanogaster CA isoforms was found to contain the identical number of deleted
amino acids as the mosquito forms within the active site region. This D. melanogaster
CA sequence (accession number CG3940-PA) may also be a GPI-linked iso form due to
the presence of a lysine-rich 5 signal sequence and hydrophobic tail region. Figure 4-2
shows an alignment of the three Dipteran CAs with shortened active site regions and all


137
Future Directions
There are fourteen predicted CA isoforms in An. gambiae; however, this
dissertation research only dealt with the cloning and characterization of two. Although
both were found to have expression in the larval midgut, there may be more that can
influence our model. However, if any additional active CAs are localized within the
larval mosquito midgut, they most likely will also be localized to the gastric caeca and/or
PMG. All of our CA studies failed to show any evidence for a CA in the AMG epithelial
cells. An apically expressed CA in the AMG may have avoided the CA histochemical
staining but the 0 isotope exchange assay should have uncovered any CA activity
present within this gut region.
The hypothesis that a bicarbonate transport metabolon exists within the regions
flanking the alkaline anterior midgut needs to be investigated. If indeed the AE is found
to bind CA, then the putative CA binding sequence located within the intracellular
carboxy terminus needs to be examined. The amino acids necessary for binding should
be studied to determine whether the binding motif of mammalian AEs is valid for the
binding of mosquito CA(s) to the mosquito AE, as proposed.
The localization of two different CA isoforms and an AE within the larval
mosquito gut has established a new model upon which to build. Preceeding this
investigation, only a fT V-ATPase was localized within the larval mosquito gut. Other
channels and transporters need to be identified and localized, such as chloride channels
and sodium/ hydrogen exchangers.
The results of these studies may be used to formulate specific mosquito
larvacides. Both of the cloned mosquito CA isoforms display novel active site regions.


15
reach the early fourth instar. Late fourth instar larvae that went unused were sacrificed to
prevent any chance of emerging adults.
Preparation and Fixation of Tissue
To dissect out the midgut, the heads of the cold-immobilized larvae were pinned
down using fine stainless-steel pins to a Sylgard layer at the bottom of a Petri dish
containing hemolymph substitute solution consisting of 42.5 mM NaCl, 3.0 mM KC1, 0.6
mM MgS04, 5.0 mM CaCl2, 5.0 mM NaHC03, 5.0 mM L-succinic acid, 5.0 mM L-malic
acid, 5.0 mM L-proline, 9.1 mM L-glutamine, 8.7 mM L-histidine, 3.3 mM L-arginine,
10.0 mM dextrose, 25 mM Hepes and adjusted to pH 7.0 with NaOH (Clark et al., 1999).
The anal segment and the saddle papillae were removed using ultra-fine scissors and
forceps, and an incision was made longitudinally along the thorax. The cuticle was
gently pulled apart and the midgut and gastric caeca were removed. In some cases, the
gut contents enclosed in the peritrophic membrane slid out, leaving behind the empty
midgut. In other cases, it was necessary to remove the peritrophic membrane and its
contents manually. For enzyme histochemistry, fixation was in 3% glutaraldehyde in 0.1
M phosphate buffer, pH 7.3, overnight at 4C (Ridgway and Moffet, 1986). For in situ
hybridization and immunohistochemistry, dissected tissues were fixed overnight in 4%
paraformaldehyde in 0.1 M cacodylate buffer, pH 7.2 (0.1 M phosphate buffer, pH 7.2
was used for Ch 4 and 5 in situ). In some cases, the dissected larval midguts were
photographed using a Nikon FX-35DX photographic camera mounted on a Nikon SMZ-
10 dissecting microscope. In other cases, digital images were acquired using a Leica
DMR microscope equipped with a Hammamatsu CCD camera. All images were
assembled using Corel Draw-11 software.


145
Stobbart, R. H. (1971). Evidence ofNa+/H4 and C17HC03' exchanges during
independent sodium and chloride uptake by the larvae of the mosquito Aedes aegypti. J.
Exp. Biol. 54,19-27.
Sun, M. K. and Alkon, D. L. (2002). Carbonic anhydrase gating of attention: memory
therapy and enhancement. Trends Pharmacol. Sci. 23, 83-89.
Tamai, S., Cody, L. B. and Sly, W. S. (1996a). Molecular cloning of the mouse gene
coding for carbonic anhydrase IV. Biochem. Genet. 34, 31-43.
Tamai, S., Waheed, A., Cody, L. B. and Sly, W. S. (1996b). Gly-63-->Gln substitution
adjacent to His-64 in rodent carbonic anhydrase IVs largely explains their reduced
activity. Proc. Natl. Acad. Sci. USA 93, 13647-13652.
Tashian, R. E. (1969). in: Biochemical methods in red cell genetics, (ed. Yunis, J. J.),
Academic Press, New York, NY. pp. 307-336.
Tashian, R. E. (1992). Genetics of the mammalian carbonic anhydrases. Adv. Genet. 30,
321-356.
Tashian, R. E., Hewett-Emmett, D., Carter, N. and Bergenhem, N. C. (2000).
Carbonic anhydrase (CA)-related proteins (CA-RPs), and transmembrane proteins with
CA or CA-RP domains. Exs. 105-120.
Thompson, J. D., Higgins, D. G. and Gibson, T. J. (1994). CLUSTAL W: Improving
the sensitivity of progressive multiple sequence alignment through sequence weighting,
position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22, 4673-
4680.
Tong, C. K., Brion, L. P., Suarez, C. and Chesler, M. (2000). Interstitial carbonic
anhydrase (CA) activity in brain is attributable to membrane-bound CA type IV. J.
Neurosci. 20, 8247-8253.
Turbeck, O. B. and Foder, B. (1970). Studies on a carbonic anhydrase from the midgut
epithelium of larvae of lepidoptera. Biochim. Biophys. Acta. 212, 139-149.
Tusnady, G. E. and Simon, I. (1998). Principles governing amino acid composition of
integral membrane proteins: application to topology prediction. J. Mol. Biol. 283. 489-
506.
Tusnady, G. E. and Simon, I. (2001). The HMMTOP transmembrane topology
prediction server. Bioinformatics 17, 849-50.
Vince, J. W. and Reithmeier, R. A. (2000). Identification of the carbonic anhydrase II
binding site in the Cl(-)/HCO(3)(-) anion exchanger AE1. Biochemistry 39, 5527-5533.


138
CA inhibitors can be manufactured to potentially block only CAs containing this novel
active site. Although these novel active sites are not found in any mammalian CA, they
are also found in other insects such as D. melanogaster. A larvacide specific for
mosquitoes can still be made due to the incorporation of an alkaline pH trigger. Species
that do not have a highly alkaline digestive strategy, such as D. melanogaster, will
therefore not be exposed to the specific CA inhibitor(s).
The larval mosquito gut provides a simple model for kidney epithelial transport.
An important finding was that the mosquito epithelial cells resemble the a and p
intercalated cells of the mammalian kidney. The addition of more ion transporters and
exchangers to our mosquito model will enable further comparisons between the mosquito
gut and the mammalian kidney. Diseases related to anion exchangers and V-ATPases
can also be studied using the mosquito gut due to the one cell layer epithelium that allows
for easier tracking of ions, as compared to the complex kidney.


78
B
Figure 4-7. Ae. aegypti and An. gambiae CA protein labeling. The antibody generated
against the Ae. aegypti CA can also be used to localize the homologous CA
isoform within An. gambiae. The larvae were incubated with phalloidin
(red) and the CA-specific antiserum (green). Colocalization of the red and
green signals appears yellow. A. The antibody localization shows the
strongest labeling in Ae. aegypti for a subset of muscle fibers in the anterior
midgut and the proximal portions of the gastric caeca. B. Antibody
localization of An. gambiae CA is depicted by the yellow muscle fibers,
while the red muscle fibers are not recognized by the antibody. The scale
bar represents 300 pm in A, 150 pm in B.


38
Figure 2-2. Three-dimensional (Cn3D) depiction of human CA IV (1ZNC). The green
barrel represents an alpha helix structure, the tan arrows represent beta
sheets, and the colored strings represent extended loop structures. The
yellow coloring represents the accessible, extended loop peptide region
against which the homologous Ae. aegypti CA antibody was raised.


82
Figure 4-11. Immunoreactivity of Ae. aegypti guts for the CA IV-like isozyme. The red
labeling is specific for muscle fibers. The green labeling shows localization
of the mosquito CA IV-like protein. The yellow labeling shows co
localization of the mosquito CA IV-like isoform and actin. Prior to
immunolabeling, the guts were treated with PI-PLC to determine if the CA
IV-like isoforms are GPI-linked to the cell membrane. A. Immunolabeling
of the gastric caeca, without the PI-PLC treatment, displays heavy yellow
labeling of the GPI-linked CA isoform. B. After PI-PLC treatment there is
no CA IV-like immunolabeling of the gastric caeca. C. The anterior midgut
displays the immunolocalization of the CA IV-like isoform along a subset of
muscle fibers. D. After PI-PLC treatment the yellow immunolabeling for
the GPI-linked mosquito CA is greatly reduced. The decreased
immunolocalization within the gastric caeca and anterior midgut signifies
that the PI-PLC was successful in severing the GPI-link and releasing the
CA from the membrane association. The scale bars represent 100 pm.


75
Figure 4-4. Expression of CA mRNA in Ae. aegypti anterior midgut. While the cardia,
gastric caeca and posterior midgut display heavy epithelial hybridization,
there is also specific CA mRNA expression seen in muscle and nerve cells.
A. A representative whole mount Ae. aegypti larvae displaying the strong
CA expression in epithelia, along with muscle fiber staining that can be
overlooked at low magnification. B. The beginning of the anterior midgut
shows hybridization to both muscle (arrowheads) and nerve fibers (arrows).
The labeled fibers reveal striated muscle running longitudinally down the
length of the anterior gut and circularly around the girth of the gut. C. The
anterior midgut (AMG) displayed strong hybridization in muscle
(arrowheads) and nerve fibers (arrows) while displaying no epithelial cell
labeling. The posterior midgut (PMG) shows intense fiber labeling as well
as epithelial cell labeling (*). The scale bar represents 300 pm in A, 25 pm
in B, 50 pm in C.


65
of the human CAs, which show several additional amino acids within the conserved
active site region.
Localization of CA IV-like Isoform in the Mosquito Midgut
In situ hybridization analyses indicate that the Ae. aegypti CA message is
expressed most heavily within the epithelial cells of the gastric caeca and posterior
midgut (Fig. 4-3). An antisense cRNA probe corresponding to the entire cDNA sequence
generated strong cytoplasmic staining of the proximal gastric caeca, while the distal Cap
cells showed no detectable hybridization (Fig. 4-3B). Rostral to the gastric caeca, a
strong localization was evident in a small subset of cardia cells that encircle the tissue,
forming a collar (Fig 4-3B). These collar cells are clearly different from the
surrounding cells in this same area. This technique also highlighted a set of specific
epithelial cells that are found only in a subset of the posterior midgut. These CA-positive
cells form a ring of about 5 cells in width that circumscribe the lower-posterior gut region
(Fig. 4-3C). The CA message was also localized to longitudinal and circular muscle
fibers of the anterior and posterior midgut (Fig. 4-4). Following the longitudinal muscle
fibers, in close association, are distinct nerve fibers that also show strong CA mRNA
expression (Fig. 4-4). Epithelial cells of the anterior midgut were clearly void of signal
beneath the labeled muscle and nerve cells. Specific staining was also evident however
within the abdominal ganglia central nervous system (CNS) and peripheral nerve tissue
(Fig. 4-5). No labeling was seen in the Malpighian tubules.
Real Time PCR Analysis of Aedes aegypti CA IV-like Transcripts
Real time PCR was used to compare the levels of Ae. aegypti CA mRNA within
specific tissue regions of the larvae. The guts of 20 fourth instar Ae. aegypti larvae were


30
more accessible to antibody probing. Furthermore, three-dimensional analyses (Cn3D
v4.1 NCBI) of predicted CA IV structures (human 1ZNC and mouse 2ZNC) predicted
that the N terminus is exposed and accessible (Figure 2-2). An antigenic peptide was
therefore chosen from the N terminus of the Ae. aegypti CA sequence. This peptide
sequence (GVINEPERWGGQCETGRR) was sent to Sigma-Genosys (Woodlands,
Texas), where it was synthesized and conjugated to bovine serum albumin (BSA). The
synthetic peptide-BSA construct and Freunds incomplete adjuvant were injected into
two rabbits to elicit an immune response. Prior to injection, a blood sample from each
rabbit was collected to serve as the control pre-immune serum. Every two weeks a blood
sample was collected from the rabbits, the fraction of immunoglobulin G (IgG) pooled,
and another dose of the peptide-BSA construct administered. Three months after the
initial injections, the final bleeds were collected and used for all immunohistochemical
analyses.
We also raised antibodies against an An. gambiae cytosolic CA peptide and an
anion exchanger (AE) peptide. These antibodies were produced by the Aves Labs, Inc.,
(Tigard, Oregon), using a similar strategy to that described above. However, these
antibodies were produced in hens. The synthesized peptides were conjugated to BSA and
injected into two hens each. The immunoglobulin Y (IgY) antibodies were collected
from the hens eggs, pooled, and purified.
Immunohistochemistry
The specificity of the antibodies in the resultant antisera was determined. The
antisera were then used to localize the larval mosquito proteins. Dissected and fixed
whole mount mosquito guts were washed 6 times in tris-buffered saline (TBS), placed in


63
gambiae genome. These cloned mosquito cDNAs from Ae. aegypti and An. gambiae are
61% identical to each other in amino acid residues and show similarities to the
mammalian CAIV isozyme. However, in contrast to the mammalian CA IV, which is
encoded by 7 exons (Sly and Hu, 1995), only 3 exons make up the An. gambiae CA
isoform. Alignment of the mosquito CA IV-like isoforms from Ae. aegypti and An.
gambiae with various mammalian CA IV isozymes reveals conserved features within this
CA isoform (Fig. 4-1). For example, the multiple leucine (L) residues within the amino
terminus of the mammalian CA IV propeptides that comprise the signal sequence are
found in the Ae. aegypti and An. gambiae CA IV-like isoforms. One important feature of
the mosquito CA IV-like sequences is the conserved alignment of G-69 (human CA IV
numbering) with the human, bovine, and rabbit CA IV sequences. This particular amino
acid residue has been changed to glutamine (Q) in rat and mouse CA IV, which results in
reduced enzyme activity (Tamai et al., 1996b). Additionally, all of the CA IV sequences,
including the mosquito isoforms, display a hydrophobic tail region. In addition to the
conserved CA IV-like features of GPI-linked proteins, there are also conserved cysteine
residues (C28 and C211, human CA IV numbering) between all of these CAs (Fig. 4-1).
It has been determined via cysteine labeling, proteolytic cleavage and sequencing that
these two cysteine residues, in the human CA IV, form a disulfide bond (Waheed et ah,
1996). A second disulfide bond is present in the mammalian CA IVs between residues
C6 and Cl 8 (human CA IV numbering; Waheed et ah, 1996). This second pair of
cysteine residues, and hence the resultant disulfide bond, is not present in either of the
mosquito isoforms.


23
determined by BLAST analysis using characterized proteins against the An. gambiae
genome. See table 2-1 for all initial primer sequences.
3' and 5' Rapid Amplification of cDNA Ends and Sequencing
Full-length cDNAs were obtained by rapid amplification of cDNA ends (RACE),
(Zhang and Frohman 1997, modified by Matz et al., 1999). Exact primers were defined
according to the 5 adaptor (DAP primer) along with a reverse primer specific to the
cloned fragment, and 3 TRsa adaptor (TRsa primer) along with a forward primer specific
to the cloned fragment (see Table 2-1 for adaptor primer sequences). These ends, which
included the 5 and 3 UTR sequences, were then used to design PCR primers to produce
a single product with consensus start and stop codons.
Plasmid DNA from individual colonies was purified using a Qiaprep
Plasmid Mini kit (Qiagen). The plasmid DNA (50 ng) was then sequenced using the ABI
Prism Big Dye Terminator Cycle Sequencing Kit (PE Biosystems, Foster City,
California) and the reaction products were analyzed on an ABI Prism 310 Genetic
Analyzer (PE Biosystems).
Construction of In Situ Hybridization Probes
Sense and antisense digoxygenin (DIG)-labeled cRNA probes were generated by
in vitro transcription using a DIG RNA labeling kit (Roche Molecular Biochemicals,
Indianapolis, Indiana). The initial in situ hybridization experiment, presented in chapter
3, used a cRNA probe derived from the original 297 bp Ae. aegypti CA sequence. The in
situ experiments presented in chapters 4 and 5 utilized the full-length CA and AE
sequences. For the first CA antisense probe, the pGEM-T vector containing the 297 bp
CA sequence was linearized by incubating 2 pg of plasmid with Pst I restriction enzyme


29
whole gut cDNA were used as template with the appropriate concentration of primers and
IX SYBR green I master mix in 25 jxL total volume. The threshold cycle number (Ct)
was plotted versus the log of the template concentration and the slope (m) and intercept
(b) were determined (Figure 2-1). These pre-determinations were then used in the
standardized comparison of the amount of 18s transcript and CA transcript in each of the
cDNA samples tested. For each analysis, a control containing all of the necessary PCR
components except the cDNA template was run. To determine the relative expression
level for each transcript analyzed, the following equation was used: (Ct-b)/m. The
average log ng for each transcript was then compared to the average log ng of 18s RNA
transcript to normalize the values. Then the expression levels were determined relative to
the transcript with the greatest normalized log ng value and expressed in a bar graph
using Microsoft Excel software.
Antibody Production
An antigenic peptide consisting of eighteen amino acids was chosen from the Ae.
aegypti CA sequence for antibody production. In order to increase the probability that
this antibody would be specific for this particular CA sequence (in the event that other
CA isoforms were isolated from the mosquito gut), attempts were made to synthesize an
antigenic peptide that would be specific to this isoform. The well-characterized
mammalian CA isoforms served as a model in trying to choose a unique CA peptide
sequence. The comparison of the mosquito CA with the mammalian isoforms yielded a
peptide sequence from the amino (N) terminus, where CA isoforms showed the most
diversity, and least conservation. The N terminus of our mosquito CA was predicted to
have an extended loop secondary structure. Unlike an alpha helix, an extended loop is


58
Figure 3-6. Polymerase chain reaction (PCR) analysis of Ae. aegypti amplified cDNA
from different gut regions. PCR was performed using exact primers for the
cloned A-CA. Anterior midgut (lane 2), gastric caeca (lane 3), posterior
midgut (lane 4). whole gut RNA control (lane 5), Malpighian tubules (lane
6) and a water template control (lane 7) are shown. Note the primary
product in gastric caeca and posterior midgut samples at the expected size
of approximately 894 nucleotides. Also note the absence of this band from
other gut regions but the appearance of bands of higher molecular masses.
Lane 1 is a 100 bp molecular mass ladder (Promega).


40
The purpose of this study was to determine the presence and location of CA in the
midgut of larval Ae. aegypti and to clone and characterize the enzyme. To investigate the
role of CA in the alkalization of the larval midgut, the effects of CA inhibitors were
tested. Here, we report the cloning and localization of the first CA from mosquito larvae
and, in particular, from the midgut epithelium of larval Ae. aegypti. A cDNA clone
isolated from fourth-instar Ae. aegypti midgut (termed A-CA) revealed sequence
homology to the a-carbonic anhydrases (Hewett-Emmett, 2000). Histochemistry and in
situ hybridization showed that the enzyme appears to be localized throughout the midgut,
although preferentially in the gastric caeca and posterior regions. In addition, classic
carbonic anhydrase inhibitors such as acetazolamide and methazolamide inhibit the
mosquito enzyme in the midgut.
Results
Bromothymol Blue Qualitative Assay
This assay allowed the identification of samples of solubilized midgut tissue
containing CA activity by spotting them onto a filter paper soaked in a basic buffered
solution containing a pH indicator, bromothymol blue (BTB). As stated previously, BTB
changes color from yellow (at pH<7.6) to blue when the pH increases above this value.
The principle behind the assay is based on the fact that CA catalyzes the conversion of
CO2 into bicarbonate with the concomitant release of protons (Donaldson and Quinn,
1974). The presence of protons lowers the pH in those regions of the paper where the
spotted samples contain the enzyme. As the pH falls below 7.6, these spots rapidly
change color from blue to yellow. This assay is not effective for samples in acidic
solution, and the tissue homogenization must be accomplished in alkaline buffer. The


113
AgAE1 Expressing Oocyte
Figure 5-18. Current-voltage (I-V) plots depicting ion transport by the AgAEl
expressing oocytes in contrast to the water injected control oocytes. A.
When chloride is removed from the solution bathing the control oocyte and
replaced with a nonionic equivalent there is no change in the slope of the
curve, signifying no ionic transport. B. When chloride is removed from the
media surrounding the AgAEl expressing oocyte, the ionic transport is
eliminated, as seen by the decrease in slope.


99
AgAEl AAQ21364
Dros AAF52497
NDAE1 AAF98636
MMDHGWDEEAP ID PRLKNRTFTAD QD FEGHRAHTVFVGVHIPGSSR 47
MAEKNEyiELPWTMNSSSGDDEAPKDPRTGGEDFTQQFTENDFEGHRAHTVYVGVHVPGG-R 61
MAEKNEYIELPWTMNSSSGDDEAPKDPRTGGEDFTQQFTENDFE 24
AgAEl AAQ21364
Dros AAF52497
NDAE1 AAF98636
RHSQRRRHKHHQASKENGDKGSTG SEAERPVTPPAQRVQFILG
RHSQRRRKHHHSGPGGGGGGGGGGGSIGGSGSVGGGAGKDNVSEKQQEVERPVTPPAQRVQFILG
VTPPAQRVQFILG
90
126
57
Figure 5-4. Alignment of carboxy terminus amino acids of An. gambiae and D.
melanogaster AEs. The characterized D. melanogaster NDAE1 (AAF98636)
is 72% identical to our An. gambiae AE1 sequence. The greatest number of
amino acid differences occurs at the carboxy terminus, the regulation domain.
However, an uncharacterized splice variant of NDAE1 (AAF52497) displays
an inserted sequence at the carboxy terminus that the characterized protein
does not. This inserted sequence shows similarity to the AgAEl sequence and
therefore is the closest predicted protein to AgAEl.


71
CAIV enzyme (Tamai et al., 1996b). Mutating glutamine-63 to glycine within the
rodent sequence resulted in almost three times greater CA activity (Tamai et al., 1996b).
Unlike the rodent sequences, both of the mosquito CA IV-like sequences display the high
activity glycine residue adjacent to histidine-69 (Human CA IV numbering, refer to Fig.
4-1).
The task ahead is to decipher if a GPI-linked CA is better equipped to function in
a highly dynamic system than other CA isoforms. Perhaps the GPI link affords the
mosquito CA enzyme a characteristic advantage in buffering such an alkaline pH through
its exclusively extracellular expression. Residing at the plasma membrane intrinsically
affords this isozyme the best location for monitoring CO2 and HCO3' flux. Indeed,
mammalian CA IVs are expressed on membrane surfaces where large fluxes of CO2 and
/or HCO3' are expected (Sly, 2000). The most compelling ability of GPI-linked proteins
is that they are known to elicit second messengers for signal transduction (Brown and
Waneck, 1992). The alkaline pH of the larval mosquito gut was found to drop within two
to three minutes after being narcotized or just simply handled (Dadd, 1975). This
handling effect lends itself to our prediction that larval mosquitoes may exert neuronal
control over the generation of the gut lumens pH. Since a GPI-linked CA was localized
within the mosquito gut and CNS tissue we propose that a GPI-linked CA may regulate
the pH of the mosquito gut by severing the GPI-link and starting a signal cascade.
Further studies are being pursued within the mosquito gut to encompass the
localization and characteristics of other CA isoforms as well as bicarbonate exchangers.


CHAPTER 7
CONCLUSIONS AND FUTURE DIRECTIONS
Conclusions
Molecular cloning techniques using isolated mosquito guts from Aedes aegypti
and Anopheles gambiae resulted in the full-length cDNA cloning of three carbonic
anhydrase (CA) genes. Our study of CA(s) within the mosquito gut began with a simple
understanding that the anterior midgut lumen has a pH of 11. The ability of C A to
greatly enhance the production of bicarbonate (and carbonate), a strong buffer, made it a
likely candidate for buffering the mosquito gut. Mammalian CA isoforms have been
studied extensively throughout the past several decades. However, the relationship
between a CAs from mammals, and those of less complex species such as mosquitoes, is
unknown. This is partly due to the lack in characterization of multiple CA isoforms from
a single non-mammalian species.
Fourteen different a CA isoforms from mammals have been characterized. The
An. gambiae and D. melanogaster genomes also display fourteen CAs. However, a
phylogenetic analysis of amino acid sequences has shown that the fourteen mammalian
CA isoforms and the fourteen dipteran CA isoforms are not direct homologs. When more
CA isoforms are characterized from insects it will be extremely interesting to determine
which isoforms are represented or omitted from the insect divergence of CAs.
Before the fourteen CA isoforms in mammals and mosquitoes can be truly
compared, every mosquito CA isoform must be characterized to determine their
131


Copyright 2004
by
Theresa J. Sern


76
Figure 4-5. Localization of CA IV-like message within Ae. aegypti CNS tissue. A. In
situ hybridization localized the CA IV-like mRNA within all ventral ganglia
CNS clusters (arrows) as well as hair sensory cells (*) and longitudinal
nerve fibers (arrowheads). B. The sense control probe displayed no specific
hybridization. Scale bar represents 300 pm.


112
Figure 5-17. Neuronal cells within the AMG display immunoreactivity for our An.
gambiae AE specific antibody. These neuronal cells also displayed AE
mRNA expression (refer to Fig. 5-13A) and are most often seen in pairs (*).
Scale bar represents 50 pm.


25
These mixtures were all incubated at 37C for 2 hours to ensure complete linearization of
the plasmids. After digestion the uncut pCR 4-TOPO plasmids were compared to the cut
plasmids on a 1% agarose gel to confirm linearization. The cut plasmids (10 fiL) were
cleansed using a Qiaquick PCR Purification kit (Qiagen Inc, Valencia, California).
For in vitro translation, the resuspended pellet or purified plasmids were
combined with IX transcription buffer, IX NTP labeling mixture, RNase inhibitor (20
U), and 40 U T3 RNA polymerase (or SP6 for pGEM-T plasmids) or 40 U T7 RNA
polymerase. For the Ae. aegypti CA probes, T7 polymerase was used with the Sal I cut
plasmid to produce the antisense probe, while T3 polymerase was used with the Xho I cut
plasmid to produce the sense (control) probe. The pCR 4-TOPO plasmid used for the
generation of the An. gambiae CA and AE probes contained the CA and AE sequences in
the reverse configuration. Therefore, for these An. gambiae probes, T3 was used with the
Not I linearized plasmids to produce the antisense probes, while T7 was used with the
Pme I cut plasmids to produce the sense (control) probes. The mixtures were incubated
at 37C for 2 hours followed by the addition of 20 U DNase I and incubation at 37C for
15 minutes. The DNase I reaction was stopped by the addition of 0.5 pL of EDTA (500
mM). The DIG-labeled cRNA was then precipitated by the addition of 2.5 pL of LiCl (4
M) and 75 pL cold ethanol (100%). The mixture was incubated overnight at -20C and
centrifuged at 14,000 g for 10 minutes at 4C. The supernatant was removed and the
pellet was washed with 50 pL cold ethanol (75%). The centrifugation step was repeated
and the pellet was air-dried and resuspended in 100 pL DEPC-treated water. The probes
were stored at -80C.


22
The reaction mix (80 pL) was combined with 40 pL phenol and 40 pL
chloroform and centrifuged at 14,000 g for 10 minutes. The upper, aqueous phase was
removed and transferred to a clean tube. The cDNA was precipitated by the addition of 3
M sodium acetate (8 pL, pH 5.0) and 100% ethanol (160 pL). The mixture was
centrifuged at 14,000 g for 15 minutes at room temperature. The supernatant was
removed and the pellet was air-dried.
For adaptor ligation, the cDNA pellet was resuspended in DEPC-treated water
(6 pL) and combined with 1 pM adaptor, IX ligase buffer, and 1 U T4 ligase (Marathon
cDNA Amplification kit) in 10 pL total volume. This mixture was stored overnight at
16C. For cDNA amplification, the ligation mixture (10 pL) was combined with 40 pL
DEPC-treated water. PCR amplification was then performed using the Advantage kit
(BD Biosciences). The diluted cDNA (1 pL) was combined with IX advantage buffer,
0.4 pL dNTP mixture (10 mM each), 0.1 pM DAP and TRsa primers (Table 2-1), and 0.4
pL advantage enzyme mix in 20 pL total volume. The cycling profile consisted of 94C
for 30 seconds, 66C for 1 minute, and 72C for 2.5 minutes. The reaction was analyzed
on a 1% agarose gel after 12,16, and 20 cycles. A final chase step was then performed to
ensure that all cDNAs were completely double-stranded. Both 5 and 3 adaptor primers
were added to the PCR reactions and two cycles of 77C for 1 min, 65C for 1 min, and
72C for 2.5 min were performed. The resulting collections of amplified cDNA were
then diluted 1:50 and used as template for subsequent PCR experiments.
Amplified cDNA pools from An. gambiae were used to clone two CA cDNAs and
the AE cDNA. Exact primers were designed from conserved regions of the proteins as


Ill
Figure 5-16. Localization of AgAEl protein within the PMG of An. gambiae larvae. A.
Our AE specific antibody displayed immunoreactivity within the PMG;
most prominantly within a specific band of cells that encircle the PMG
region. B. Phalloidin was used to label muscle fibers. C. Draq-5 was used
to localize nuclear DNA. D. Overlay of AE labeled cell membranes in
relation to muscle fibers and nuclei. E. High magnification of AE protein
localization within the PMG. Cell membranes of both large and small cells
are clearly labeled with our antibody. Scale bars represent 150 pm in A-D
and 75 pm in E.


24
(New England Biolabs (NEB); Beverly, Massachusetts) and IX buffer 3 (NEB) at a total
volume of 20 pL for 1 hour at 37C. For the sense probe, the pGEM-T vector containing
the 297 bp CA sequence was linearized by incubating 2 pg of plasmid with Not I
restriction enzyme and IX buffer 3 (NEB) in a total volume of 20 pL for 1 hour at 37C.
After digestion, the volume was brought to 100 pL with the addition of 80 pL water. A
phenol/ chloroform extraction was performed such that 100 pL of phenol/ chloroform-
isoamyl alcohol was added to the linearized plasmid and the solution was centrifuged at
14,000 g for 1 minute. The upper aqueous phase was transferred to a new tube and 100
pL chloroform was added. After centrifugation at 14,000 g for 1 minute, the upper
aqueous phase was transferred to a new tube and the chloroform step was repeated. The
linearized plasmid DNA was precipitated by the addition of 10 pL sodium acetate (3 M,
pH 2.5) and 200 pL cold ethanol (100%). The DNA was incubated at -80C for 15
minutes and then centrifuged at 14,000 g for 10 minutes at 4C. The supernatant was
removed and the DNA pellet was washed by the addition of 500 pL ethanol (70%)
followed by centrifugation at 14,000 g for 5 minutes at 4C. The supernatant was
removed and the pellet was air-dried and then resuspended in 13 pL DEPC-treated water.
The full-length Ae. aegypti CA was subcloned into pCR 4-TOPO plasmid using a
PCR manufactured 5 Sal I restriction site and a 3 Xho I site. Therefore, the pCR 4-
TOPO plasmid was linearized by incubating 2 pg of plasmid with either Sal I, IX Sal I
buffer, and BSA, or Xho I, IX buffer 2, and BSA (NEB). The pCR 4-TOPO plasmid was
also used for the generation of the An. gambiae CA and AE probes. For these probes, the
unique restriction sites, Pme I and Not I, located within the pCR 4-TOPO plasmid were
used for linearization with IX buffer 4 and BSA, or IX buffer 3 and BSA, respectively.


60
B.
Figure 3-8. Localization of CA mRNA expression in larval Ae. aegypti. A. An isolated
whole mount gut probed with DIG-labeled cRNA for A-CA. Abundant
hybridization is observed in the cardia. gastric caeca (GC), and posterior
midgut (PMG). B. The smaller cuboidal cells (arrow) display stronger
hybridization than the larger columnar cells (*). C. Isolated midgut reacted
with the sense (control) cRNA for A-CA. Scale bars represent 300 pm in
A, 75 pm in B and C.


119
Labeling was also seen within the rectum and the last distal cell of the Malpighian
tubules (MT; Fig. 6-6).
Antibody Localization of CA Protein
The antigenic peptide sequence QYIRSPDAQTEIDAD was chosen from the An.
gambiae translated CA cDNA sequence (accession number AAQ21366) to elicit the
production of antibody. This peptide was chosen for its antigenic capacity as well as its
uniqueness among the 14 putative An. gambiae CA genes. The resultant chicken
antiserum was used to localize the CA protein within whole mount preparations of fourth
instar An. gambiae larvae.
Immunoreactivity for the cytosolic CA was predominantly displayed within the
gastric caeca (Fig.6-7). The PMG also displayed immunolabeling along the periphery of
both the large and small epithelial cells. Immunoreactivity was also evident within a
small population of neuronal cells scattered along the midgut, most often seen in pairs
(Fig.6-8). The pre-immune antisera displayed no specific immunoreactivity.
Bacterial Expression and Purification oiAnopheles gambiae Cytosolic CA
The full-length CA cDNA was subcloned into a pETlOO directional expression
vector (pETlOO/D-TOPO; Invitrogen) for expression of the recombinant protein with an
N-terminal tag containing an Xpress epitope and a polyhistidine (6X His) tag. The 6X
His tag was utilized when purifying the CA protein. Antibodies against both the Xpress
epitope and the CA peptide recognized a band of the predicted molecular weight (33
kDa) for the recombinant CA protein (Fig. 6-9).
Purified CA fractions were tested for CA activity using 180 isotope exchange
experiments (Silverman and Tu, 1986). This technique showed that CA activity was


9
combines carbon dioxide and water to produce bicarbonate and a proton ion. The
bicarbonate is pushed from the epithelial cell, across to the lumen side by the anion
exchanger, in trade for a chloride ion. The proton then gets stripped off of the
bicarbonate and, along with the proton pumped across by the V-ATPase, is brought back
into the cell in exchange for a potassium ion (Wieczorek et ah, 2000). This potassium
ion combines with the carbonate to produce potassium carbonate, which is hypothesized
to be responsible for the high alkaline pH of the anterior gut region. This hypothesis
stems from the fact that potassium ions are actively produced by the Malpighian tubules
and are circulated throughout the gut via the hemolymph (Clements, 1992). Potassium
carbonate also has a pKA greater than 10 and can therefore contribute to the gut
alkalization.
The goal of this project was to expand and adapt this model to the larval mosquito
by completing several clear objectives. These objectives are outlined within the
following specific aims.
Specific Aims
1. Determine whether CA is involved in buffering the high alkalization of the larval
mosquito gut.
A. Determine if a CA enzyme is present within the mosquito gut. Determine which
regions of the larval Ae. aegypti gut display CA activity using CA histochemistry
and l80 isotope exchange.
B. Determine if CA-specific inhibitors, such as acetazolamide, can influence larval
midgut alkalization.
2. Determine whether CA is expressed in the larval mosquito gut.
A. Clone and characterize full length CA cDNAs from the larval midgut of Ae.
aegypti and An. gambiae.


26
In Situ Hybridization
The in situ hybridization experiments presented in chapters 4 and 5 added an
additional fixation step due to a recommendation by Dr. Dmitri Boudko to increase the
clarity of the in situ labeling. A glass electrode fitted to a micromanipulator was used to
inject 4% paraformaldehyde into the thoracic cavity, just behind the head. Successful
perfusion was easily identified by the cessation of the otherwise constant muscle
twitching along the length of the body. This injection of fixative served to preserve the
cellular integrity and protect against the many proteases that exist within the mosquito
gut. For in situ hybridization, methods were adapted from Westerfield (1994). The
midguts were washed with PBS at room temperature and then incubated in 100%
methanol at -20C for 30 minutes to ensure permeabilization of the gut tissue. The tissue
was washed (5 min each wash) in 50% methanol in PBST (Dulbecco's phosphate
buffered saline [Sigma-Aldrich] plus 0.1% Tween-20), followed by 30% methanol in
PBST and then PBST alone. The tissue was fixed in 4% paraformaldehyde in 0.1 M
sodium cacodylate buffer (or 0.1 M phosphate buffer) for 20 min. at room temperature
and washed with PBST. The larval midguts were digested with proteinase K (10 |ig/ml
in PBST) at room temperature for 10 min, washed briefly with PBST and fixed again, as
described previously.
Prehybridization of the tissue was accomplished by incubation in HYB solution
(50% formamide, 5X SSC [IX SSC equals 0.15 M NaCl, 0.015 M Na-citrate buffer pH
7.0], 0.1% Tween-20) for 24 hours at 55C. The larval midguts were transferred to
HYB+ solution (HYB plus 5 mg/ml tRNA, 50 pg/ml heparin) containing 5 ng/ml DIG-
labeled probe and incubated overnight at 55C. Excess probe was removed by washing at


133
180 exchange. The activity was shown to be sensitive to the CA-specific inhibitor,
methazolamide. This cytosolic CA isoform therefore contributes to the CA activity
present within the cardia, gastric caeca and posterior midgut epithelial cells.
A full-length anion exchanger (AE) cDNA was also cloned from the gut tissue of
An. gambiae. The AEs are a small group within a large bicarbonate transporter (BT)
superfamily. Studies of mammalian BTs are ongoing and many new forms are still being
discovered and characterized to date. AEs reversibly transport chloride for bicarbonate in
a 1:1 electroneutral exchange. Cloning and localizing an AE within the mosquito gut was
therefore a direct progression from localizing the CAs, relating the production and
transport of bicarbonate within the alkaline gut.
The larval mosquito AE was expressed in Xenopus oocytes and was shown to
have electrophysiological characteristics of known AEs (i.e. chloride transport and DIDS
inhibition). The AE contains an intracellular carboxy terminus that is predicted to
moderate ion exchange and an amino terminus that performs ion exchange via the twelve
membrane-spanning domains.
The antibodies we produced, specific to carboxy and amino terminal peptides,
label the membranes of both gastric caeca and posterior midgut epithelial cells. To
maximize ion exchange, the An. gambiae AE contains an amino terminus CA binding
motif. If indeed the AE binds a cytosolic CA, the gastric caeca and posterior midgut
regions would have control over cellular pH, via both intracellular and extracellular
means. Since both of these regions contain active cytosolic CA enzyme(s), we propose
that the AE spans the membrane in the GC and PMG and binds a cytosolic CA, forming a
bicarbonate transport metabolon to transport the bicarbonate that is made by the CA(s).


103
Figure 5-8. In situ hybridization of AgAEl in whole mount An. gambiae consistently
shows positive labeling of tracheal fibers along the midgut. A. A thick
tracheal stalk penetrates the gastric caeca while thinner branches join the
AMG. This main tracheal stalk that is closely associated with the gastric
caeca consistently displays labeled particles that may be secretory vesicles
(arrow). B. Labeled trachea (arrows) traverse the AMG and coincide with
labeled nerve fibers (arrowheads) that extend down the midgut. C and D.
Labeled trachea (arrows) are randomly associated with the area surrounding
the gastric caeca. Scale bars represent 25 pm.


95
comprising the mammalian kidney. Many studies have sought to determine specific
amino acids responsible for pH sensitivity within the anion exchangers. The sixteen
amino acid pH sensitive region of AE2 is almost identical to the comparable region
within the AgAEl sequence (refer to Fig. 5-5). The mosquito midgut provides an
excellent model for studying precisely the pH dependent moieties and proteins due to the
large pH gradient that it supports. The one cell layer epithelium that divides the
alkalinity of the lumen (pH 11) from the neutral pH of 7-8 within the cell cytosol has yet
to reveal the cell polarity that is capable of maintaining this system.
The elucidation of such a metabolon within the mosquito gut, in which an AE is
directly tethered to one or more CAs, such as in the mammalian system (Sterling et al.,
2001a), would provide a mosquito model which could be used as a simple framework for
uncovering metabolic networks within complicated mammalian systems such as the
kidney.


4
Although this alkaline digestive strategy is well documented in insects, the molecular
processes involved have not been clearly defined.
Carbonic Anhydrase
Carbonic anhydrase (CA), a blood enzyme, first described by Meldrum and
Roughton in 1933, catalyzes the reversible hydration of carbon dioxide to form
bicarbonate and a proton (CO2 + H2O <> HCO3' + H+; Meldrum and Roughton, 1933).
Carbonic anhydrase was first characterized in erythrocytes as the result of a search for a
catalytic factor that would enhance the transfer of bicarbonate from the erythrocyte to the
pulmonary capillaries (Meldrum and Roughton, 1933). Since it was first described, CA
has been shown to play an important role in most acid/base transporting epithelia.
Fourteen different CA isoforms have been characterized to date in mammals (Hewett-
Emmett and Tashian, 1996). These enzymes have been determined to function in pH
regulation and ion balance, thereby performing a crucial role in many biological
processes such as respiration, bone resorption, renal acidification, gluconeogenesis,
aqueous humor production, gastric acid production, cerebrospinal fluid formation, and
signal processing (Dodgson, 1991; Sly and Hu, 1995; Hewett-Emmett and Tashian, 1996;
Lindskog, 1997; Sun and Alkon, 2002).
Various types of epithelial cells, such as those described in the mammalian
kidney, contain CAs that can provide large quantities of bicarbonate for buffering cells
and their microenvironment. Polarized epithelia play an important role in partitioning
physiologically distinct compartments, and in maintaining cell and tissue homeostasis.
The epithelial cells found in the larval mosquito midgut may serve a similar partitioning
function. Like the mammalian kidney, different regions of the mosquito gut may play


105
Figure 5-10. Localization of AgAEl mRNA to the PMG of larval An. gambiae. A. The
whole mount gut preparation displays the unlabeled AMG on the left side of
the photo as compared to the labeled PMG on the right side of the photo. The
arrows point to the tracheal stalks that join the gut at the third body segment,
corresponding to the beginning of the PMG region. B. Outer margins of large
columnar PMG cells display AE mRNA labeling (*) along with labeling of
tracheal fibers (arrows) and small cuboidal cells (arrowheads). Scale bar
represents 50 pm in A, 25 pm in B.


46
cuboidal cells. In addition, the distal cells of each lobe of the gastric caeca, termed Cap
cells, show little or no histochemical staining, suggesting further cellular heterogeneity
with respect to CA distribution in the gut (Figure 3-7).
In Situ Hybridization
To further characterize the localization of A-CA expression, in situ hybridization
was performed using a portion (approximately 300 bp) of the central coding region of the
cDNA. Figure 3-8 shows typical results of this type of analysis. A strong hybridization
signal was evident in the gastric caeca and the posterior midgut. Lower levels of
hybridization were evident in other gut regions. As with the CA histochemical stain,
higher magnification revealed that the relatively small cuboidal cells exhibit more intense
labeling than do the large columnar cells (Figure 3-8B).
Discussion
The search for the enzyme in the midgut of the larval mosquito was triggered by
the observations of a pH value around 11 in the anterior midgut lumen and a high
bicarbonate concentration (Zhuang et al., 1999; Boudko et al., 2001b). The presence of
CA in the midgut of the larval mosquito has been suggested before by investigations of
the epithelium of larval lepidopteran midgut. Carbonic anhydrase has been studied in
Manduca sexta, where the enzyme has been associated with the fat body, midgut and
integumentary epithelium (Jungreis et al., 1981). The enzyme has also been localized in
the goblet cells of the epithelium of Hyalophora cecropia using Hanssons histochemical
stain. The same procedure showed that the columnar cells were devoid of activity
(Turbeck and Foder, 1970).


74
Figure 4-3. Localization of CA mRNA in a wholemount preparation of early 4th instar
Ae. aegypti. A. The wholemount gut preparation localizes CA message to
specific cells of the gastric caeca (GC) and posterior midgut (PMG). B. A
subset of cardia (arrows) and gastric caeca cells display the CA message.
The distal lobes of the caeca, called Cap cells, display no staining (*). C.
There is a distinctive labeling pattern of CA message within a specific band
of posterior epithelial cells. In addition, numerous trachea (arrows) are
heavily labeled along the length of the midgut. Scale bar represents 300 pm
in A, 150 pm in B, 75 pm in C.


48
As reviewed in Clements (1992), two major cell types have been defined in the
gastric caeca by inferring functional states from cytological findings. These two major
cell types have been called ion-transporting cells and resorbing/secreting cells (Volkman
and Peters, 1989a,b) and they correspond to the columnar and cuboidal cells mentioned
above with the ion-transporting cells being equivalent to the columnar cells and the
resorbing/secreting cells being the cuboidal cells (Zhuang et al., 1999). Neither of these
cell types, as characterized in the larval mosquito gut, parallels the structurally unique
qualities of the lepidopteran goblet cell. Nonetheless, our results indicate that, as in
lepidopterans, CA activity is preferentially associated with one of two distinct cell types
whose functional complementation must produce the alkalization and ionic balances
regulated by the gut. These results are consistent with the observations of lepidopteran
midgut by Turbeck and Foder (1970). In the larval lepidopteran midgut, two
morphologically distinct cell types have been long recognized: goblet cells and columnar
cells. Goblet cells posses both the proton-pumping V-ATPase and CA activity (Harvey,
1992; Ridgway and Moffet, 1986; Wieczorek et al., 1999). One of the enigmas of using
the pioneering analyses of insect model systems such as M. sexta to produce testable
hypotheses for gut alkalinization in mosquito larvae has been the apparent absence of
goblet cells from mosquitoes. Previous investigations have inferred different functional
cell types in the larval mosquito gut epithelium. We are currently developing antibody
probes for A-CA. Immunocytochemical analyses of A-CA distribution in comparison
with other key components of gut function, such as V-ATPase (Zhuang et al., 1999),
should provide new insights into the cell biology of this intriguing epithelial system.


31
pre-incubation medium (pre-inc) for a minimum of 1 hour, and then incubated in primary
antibody (1:1000) overnight at 4C. The guts were then washed in pre-inc and incubated
in FITC-conjugated goat anti-rabbit (GAR) or Alexa-GAR secondary antibody (Jackson
ImmunoResearch, West Grove, Pennsylvania, 1:250 dilution) overnight at 4C. The
whole mount preparations were rinsed in pre-inc and mounted onto slides using p-
phenylenediamine (PPD, Sigma-Aldrich) in 60% glycerol. In some cases Draq 5
(Jackson ImmunoResearch, 1:1000 dilution) was applied before mounting to visualize
nuclear DNA. The samples were examined and images captured using the Leica
scanning confocal microscope.
Live preparations were examined, following a similar procedure, to ensure that
antibodies were capable of localizing extracellular proteins only. In this case, the
primary antibody was applied to the live guts for four hours. The samples were washed
in TBS, and fixed as described previously, before the secondary antibody was applied.
The samples were mounted on slides as described above.
The live gut assays were also performed to determine whether this specific CA is
tethered to the cell membrane via a GPI linkage. Ten live gut preparations were
incubated with phosphoinositol-specific phospholipase C (PI-PLC, 1:100 in HSS; Sigma-
Aldrich) for 90 minutes at 37C. PI-PLC was used as a tool in determining the presence
of a GPI link. Controls in which the guts were incubated in HSS alone were also
performed. The guts were then washed in HSS, fixed, and treated with primary and
secondary antibodies as described above.


108
Figure 5-13. Expression of AE mRNA was found throughout the ventral midgut ganglia.
Between the ventral midgut and the ventral integument a labeled neuronal
pathway connected each ganglia to the next. A. Two neurons (arrows) label
positively for the AE mRNA, rendering them highly visible above the other
neuronal cells within the unstained ganglia. B. Between each ganglia
cluster, a tissue crossbridge (*) passes in a 90 angle between the ganglia
and the gut. The neuronal pathway showing AE mRNA expression makes
contact with this junction and continues on to the next ganglia cluster on the
other side. C. Two neuronal cells within the following ganglia show clear
expression of AE mRNA. D. This panel shows an unobstructed view of a
single triangular-shaped ganglia cluster. It is clear that the AE mRNA is
located within a distinct population of neuronal cells in each ganglion.
Scale bars represent 50 pm.


10
B. Use experimental and bioinformatical approaches to determine if CA expressed in
the mosquito gut is similar to a characterized mammalian CA isoform.
Furthermore, determine the subcellular location of mosquito CA isoforms as
cytosolic, membrane-bound, mitochondrial, or GPI-linked.
C. Determine which regions of the mosquito gut express CA mRNA and protein
using in situ hybridization, real time PCR, and immuno-localization.
3. Determine whether anion exchangers (AE) are involved in the pH regulation of the
larval mosquito gut.
A. Clone and characterize an AE from the larval mosquito gut that uses the
bicarbonate produced by CA as a substrate.
B. Determine whether the mosquito AE transports chloride using a Xenopus oocyte
expression assay.
C. Determine if AE is expressed in the same regions of the mosquito gut as the CA.
Co-localization of CA and AE would support the existence of a bicarbonate
transport metabolon.
D. Determine if the AE contains the amino acid sequence predicted to be necessary
for binding C A. If indeed the AE protein is predicted to bind CA, a bicarbonate
transport metabolon within the larval mosquito gut could maximize bicarbonate
production and transport.
4. Present a new larval mosquito model that reflects the studies in this dissertation.
Bring together all localized components of mosquito gut physiology into one model.


33
m, of fresh LB-carb and was grown at 37C with shaking. Optimization experiments
were performed in order to facilitate the production of the greatest quantity of CA
protein. For production of CA protein, isopropylthio-JJ-galactoside (IPTG, 1 mM final
concentration; Stratagene, La Jolla, California) was added when the culture had attained
an optical density of 0.5 at a 600 nm wavelength. Achieving this density took about 1.5
hours of growth at 37C and 200 rpm. Zinc, in the form of zinc sulfate (0.5 mM final
concentration), was added along with the IPTG to facilitate the proper conformation of an
active CA protein. In order to optimize the duration of the induced growth phase,
samples were collected every hour for six hours. These samples were analyzed on an
SDS-Page 4-12% Bis-Tris gel to compare CA protein content. Four hours of growth was
determined to be ideal for the production of the truncated Ae. aegypti CA IV-like and
lull-length An. gambiae CA II-like proteins.
Total protein was collected using the Probond Purification System according to
the manufacturers instructions for soluble proteins (Invitrogen). The cells were harvested
by centrifugation, sonicated in native buffer (250 mM NaPCL, 2.5 M NaCl; Invitrogen)
with lysozyme (1 mg/mL; Sigma-Aldrich), and centrifuged again to collect a crude
protein extract. The supernatant was applied to a Probond nickel column (Invitrogen)
and washed free of non-specific binding contaminants. The nickel column binds the CA
protein due to the added histidine tag, a repeat of six histidine residues within the pETlOO
expression vector that is inserted after the carboxy-terminus of the CA protein. CA was
eluted by adding imidazole (250 mM; Invitrogen) to the column, which competes with
and displaces the histidine tag. Eluted fractions were separated on an SDS-Page 4-12%
Bis-Tris gel (Invitrogen).


16
Bromothymol Blue Qualitative Assay
A qualitative test to detect carbonic anhydrase activity in mosquito larval midgut
homogenate was adapted from the test described by Tashian (1969). The procedure
included immersing a piece of Whatman no.l paper in a solution made with 0.15%
Bromothymol Blue (BTB) in ice-cold 25 mM Tris HC1, 0.1 M Na2SC>4, pH 8.0. The
paper was allowed to soak completely in this blue solution and was placed on ice for 30
minutes. The colored filter was then transferred to a Petri dish with a hole in the lid.
Samples of mosquito larval midgut homogenate were prepared by sonicating midguts of
early fourth instar larvae in ice-cold 25 mM Tris HC1, 0.1 M Na2SC>4 pH 8.0, with
protease inhibitor cocktail (Sigma-Aldrich, St. Louis, MO; diluted 1:1000). An
autopipette was used to spot exactly 4 pi samples on the paper. Controls were also
spotted. The controls included a buffer with protease inhibitor and controls for the
liver/yeast food added to the medium in which the mosquito larvae were reared. These
food controls included a range of concentration from 1 to 100 pg/ml liver powder and
yeast. Carbonic anhydrase from bovine erythrocytes (Sigma-Aldrich) dissolved in the
same buffer described above was used as a positive control.
A steady stream of CO2 at 34.5 KPa was blown for 3 seconds through the
opening on the lid of the Petri dish, and the dish was sealed and kept on ice. The
formation of yellow spots in a few seconds was indicative of carbonic anhydrase activity.
Effect of Methazolamide on the Alkalization of the Midgut of Live Larvae
The effect of a CA inhibitor, methazolamide, on gut alkalization and the capacity
of whole larvae to alkalize their culture medium was examined. Flat-bottomed tissue
culture plates (24 well, Sarstedt Inc., Newton, North Carolina) were filled with 1 ml of 25


141
Clark, T. M., Koch, A. and Moffett, D. F. (1999). The anterior and posterior stomach
regions of larval Aedes aegypti midgut: regional specialization of ion transport and
stimulation by 5-hydroxytryptamine. J. Exp. Biol. 202 (Pt 3), 247-252.
Clements, A. N. (1992). The biology of mosquitoes. Chapman & Hall, London, UK.
Corena, M. P., Sern, T. J., Lehman, H. K., Ochrietor, J. D., Kohn, A., Tu, C. and
Linser, P. J. (2002). Carbonic anhydrase in the midgut of larval Aedes aegypti:
cloning, localization and inhibition. J. Exp. Biol. 205, 591-602.
Dadd, R. H. (1975). Alkalinity within the midgut of mosquito larvae with alkaline-active
digestive enzymes. J. Insect Physiol. 21,1847-1853.
Darsie, R. F., Jr. and Morris, C. D. (2000). Keys to the adult females and fourth instar
larvae of the mosquitoes of Florida (Dptera, Culicidae). Bull. Fla. Mosq. Control
Assoc.
Dodgson, S. J. (1991). Why are there carbonic anhydrases in the liver? Biochem. Cell
Biol. 69, 761-763.
Donaldson, T. L., and Quinn, J. A. (1974). Kinetic constants determined from
membrane transport measurements: carbonic anhydrase activity at high concentrations.
Proc. Natl. Acad. Sci. USA 71, 4995-4999.
Dow, J. A. (1984). Extremely high pH in biological systems: a model for carbonate
transport. Am. J. Physiol. 24, (4 Pt 2), R633-636.
Eisenhaber, B., Bork, P. and Eisenhaber, F. (1999). Prediction of potential GPI-
modification sites in proprotein sequences. J. Mol. Biol. 292, 741-758.
Fernley, R. T. (1988). Non-cytoplasmic carbonic anhydrases. Trends Biochem. Sci. 13,
356-359.
Frohman, M. A. and Zhang, Y. (1997). Using rapid amplification of cDNA ends
(RACE) to obtain full-length cDNAs. Methods Mol. Biol. 69, 61-87.
Geer, L. Y., Domrachev, M., Lipman, D. J. and Bryant, S. H. (2002). CDART:
protein homology by domain architecture. Genome Res. 12, 1619-1623.
Gubler, D. J. (1997). in: Dengue and dengue hemorrhagic fever, (ed. Gubler, D. J. &
Kuno, G.), CAB International, Wallingford, U.K. pp. 1-22.
Gupta, R., Jung, E., and Brunak S. (In preparation 2002) Prediction of N-glycosylation
sites in human proteins.


CARBONIC ANHYDRASES AND BICARBONATE TRANSPORT
IN LARVAL MOSQUITOES
By
THERESA J. SERON
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
2004

Copyright 2004
by
Theresa J. Sern

DEDICATION
I wish to dedicate this dissertation to my incredible family, Mom, Dad,
Grandparents, Tracey, and George. My family has witnessed my struggles and triumphs
and has helped me through it all. I am so proud to call them my family. I cannot thank
them enough for all of their support. This achievement is really a reflection of all of us.
I also want to dedicate this milestone to my soon to be husband, Dr. Peter Lovell.
Coming home to his love, humor, and music, has given me true joy. My family would be
incomplete if I did not mention our furry companions, Frodo and Princess, who remind us
that a nap can solve most problems.
m

ACKNOWLEDGEMENTS
I would like to acknowledge my dissertation committee, Dr. Paul J. Linser, Dr.
Edward J. Phlips, Dr. Leonid Moroz, Dr. Robert Greenberg, and Dr. Shirley Baker, for
their suggestions and comments on the final rewriting of this document. I want to thank
my project supervisor, Dr. Paul J. Linser, for allowing me to form my own project goals
and the space to tackle them.
There are a number of people at The Whitney Laboratory who I would like to
thank for their assistance with this dissertation project as well as their friendship. Dr.
Judith Ochrietor devoted her time and energy to improving every aspect of this
dissertation. Judy assisted me with experimental designs, introduced me to real time
PCR, provided a wealth of knowledge about molecular biology, and was a great person
with which to share a laboratory and office. Dr. Andrea Kohn provided molecular
biology teaching and advice along with being a fantastic person to work with and be
inspired by. Leslie vanEkeris taught me how to do mosquito dissections and provided
many of the mosquito guts that I photographed for this document. Dr. Bill Harvey
provided insight into the ionic transport mechanisms of the mosquito and the editing of
this manuscript. Dr. Dmitri Boudko was instrumental in the expression of the anion
exchanger and the production of amplified cDNA libraries. Jessica Roberts-Misterly and
Dr. Robert Greenberg also provided teachings and suggestions in cloning cDNAs from
the mosquito.
iv

TABLE OF CONTENTS
Eige
DEDICATION iii
ACKNOWLEDGEMENTS iv
LIST OF TABLES viii
LIST OF FIGURES ix
ABSTRACT xii
CHAPTER
1 INTRODUCTION 1
Alkaline Gut 2
Carbonic Anhydrase 4
Mosquito Development and Control 5
Carbonic Anhydrase Inhibition 6
Bicarbonate Transport 7
Gut Alkalization Model 8
Specific Aims 9
2 MATERIALS AND METHODS 14
Experimental Insects 14
Preparation and Fixation of Tissue 15
Bromothymol Blue Qualitative Assay 16
Effect of Methazolamide on the Alkalization of the Midgut of Live Larvae 16
ID # # #
0 Exchange Method to Measure Carbonic Anhydrase Activity 17
Isolation of RNA and Synthesis of cDNA 18
Bioinformatics 18
Cloning of CA from Aedes aegypti Larval Midgut 19
Construction of Amplified cDNA Pools 20
3' and 5' Rapid Amplification of cDNA Ends and Sequencing 23
Construction of In Situ Hybridization Probes 23
In Situ Hybridization 26
CA Histochemistry 27
Real Time PCR 27
v

Antibody Production 29
Immunohistochemistry 30
CA Protein Expression 32
Anion Exchanger Oocyte Expression 34
Anion Exchanger Physiology 34
3 CARBONIC ANHYDRASE IN THE MIDGUT OF LARVAL AEDES
AEGYPTI: CLONING, LOCALIZATION, AND INHIBITION 39
Introduction 39
Results 40
Bromothymol Blue Qualitative Assay 40
Carbonic Anhydrase Activity and Alkalization 42
l80 Isotope-Exchange Experiments 43
Cloning of Carbonic Anhydrase from Aedes aegypti Larvae 43
Localization of the Enzyme in the Midgut Epithelium: Carbonic
Anhydrase Enzyme Histochemistry 45
In Situ Hybridization 46
Discussion 46
4 A GPI-LINKED CARBONIC ANHYDRASE EXPRESSED IN THE
LARVAL MOSQUITO MIDGUT 61
Introduction 61
Results 62
Bioinformatics of Aedes aegypti CA 62
Sequence Comparisons of CA IV-like Isoforms 62
Localization of CA IV-like Isoform in the Mosquito Midgut 65
Real Time PCR Analysis of Aedes aegypti CA IV-like Transcripts 65
Immunolocalization of CA IV-like Protein in the Mosquito Gut 66
Antibody Cross-Reactivity with Other Mosquito Species 67
Phospholipase C Treatment 68
Discussion 68
5 ANION EXCHANGER EXPRESSED WITHIN THE LARVAL
ANOPHELES GAMBIAE MOSQUITO 83
Introduction 83
Results 84
An. gambiae AE Sequence Analysis 84
BT Sequence Comparisons 86
Localization of Anion Exchanger mRNA in An. gambiae Larvae 87
Antibody Localization of AE Protein 89
AE Functional Expression in Oocytes 89
Discussion 91
vi

6 CYTOSOLIC CA EXPRESSION IN LARVAL ANOPHELES GAMBIAE 115
Introduction 115
Results 116
Anopheles gambiae CA Sequence Analysis 116
Localization of CA Activity in Anopheles gambiae Larvae 118
Localization of Cytosolic CA mRNA in Anopheles gambiae Larvae 118
Antibody Localization of CA Protein 119
Bacterial Expression and Purification of Anopheles gambiae
Cytosolic CA 119
Discussion 120
7 CONCLUSIONS AND FUTURE DIRECTIONS 131
Conclusions 131
New Model 134
Future Directions 137
REFERENCES 140
BIOGRAPHICAL SKETCH 147
vii

LIST OF TABLES
Table page
2-1. PCR primer sequences 35
2-2. Composition of all solutions used in Xenopus oocyte expression of
An. gambiae AE 36
vm

LIST OF FIGURES
Figure E§££
1-1. Illustration showing the regions of the larval mosquito gut 11
1 -2. Illustration of the mosquito life cycle 12
1 -3. Preliminary mosquito anterior midgut model based on M. Sexta 13
2-1. Efficiency plots for real-time PCR primers 37
2-2. Three-dimensional (Cn3D) depiction of human CAIV (1ZNC) 38
3-1. Effect of CA inhibition on culture medium pH with fourth-instar Ae.
aegypti. larvae 53
3-2. Effect of methazolamide on the alkalization of the midgut using
Bromothymol Blue (BTB) assay of pH within living, but isolated, gut tissue 54
3-3. Relative activity of CA in different pooled segments of the midgut of larval
Ae. aegypti 55
3-4. Carbonic anhydrase from the midgut of larval Ae. aegypti 56
3-5. Comparison of the extrapolated amino acid sequences of A-CA with
six putative dipteran CA genes identified in the D. melanogaster
gene databases 57
3-6. Polymerase chain reaction (PCR) analysis of Ae. aegypti amplified cDNA
from different gut regions 58
3-7. Hanssons histochemistry of whole mount Ae. aegypti gut 59
3-8. Localization of CA mRNA expression in larval Ae. aegypti 60
4-1. Alignment of several mammalian CA IV enzymes with two mosquito CA
isoforms 72
4-2. Clustal alignment of CA protein sequences 73
IX

4-3. Localization of CA mRNA in a whole mount preparation of early 4th
instarle, aegypti 74
4-4. Expression of CA mRNA in Ae. aegypti anterior midgut 75
4-5. Localization of CA IV-like message within Ae. aegypti CNS tissue 76
4-6. Relative quantification of CA IV-like message in Ae. aegypti larvae
using real time PCR 77
4-7. Ae. aegypti and An. gambiae CA protein labeling 78
4-8. The Ae. aegypti CNS ganglia express the CA IV-like isoform 79
4-9. Immunolocalization of mosquito CA IV-like enzyme in Aedes albopictus 80
4-10. High magnification of immunoreactive muscle fibers within the Aedes
albopictus midgut 81
4-11. Immunoreactivity of Ae. aegypti guts for the CA IV-like isozyme 82
5-1. Structural prediction of the An. gambiae AE1 96
5-2. Putative amino terminus CA II binding motif 97
5-3. Homology tree depicting the amino acid identity between several BTs 98
5-4. Alignment of carboxy terminus amino acids of An. gambiae and D.
melanogaster AEs 99
5-5. Alignment of An. gambiae and human AEs 100
5-6. Localization of AgAEl mRNA within whole mount An. gambiae
larvae 101
5-7. Localization of AgAEl mRNA in muscle, nerve, and trachea in
An. gambiae 102
5-8. In situ hybridization of AgAEl in whole mount An. gambiae consistently
shows positive labeling of tracheal fibers along the midgut 103
5-9. Anion exchanger mRNA localization reveals trachea and nerve fibers
along with neuronal cell labeling 104
5-10. Localization of AgAEl mRNA to the PMG of larval An. gambiae 105
x

106
5-11. Larval An. gambiae displays strong AgAEl expression in the hindgut,
the pylorus
5-12. Localization of AE mRNA in An. gambiae shows abundant labeling of
the Malpighian tubules 107
5-13. Expression of AE mRNA was found throughout the ventral midgut
ganglia 108
5-14. Sense AE probes display no specific hybridization 109
5-15. Antibody localization of AgAEl protein to the gastric caeca in An.
gambiae larvae 110
5-16. Localization of AgAEl protein within the PMG of An. gambiae larvae 111
5-17. Neuronal cells within the AMG display immunoreactivity for our
An. gambiae AE specific antibody 112
5-18. Current-voltage (I-V) plots depicting ion transport by the AgAEl expressing
oocytes in contrast to the water injected control oocytes 113
5-19. Inhibition of AgAEl mediated chloride transport by DIDS 114
6-1. Clustal alignment of active sites within An. gambiae, D. melanogaster,
and human CA proteins 122
6-2. Phylogenetic analysis between mammalian (human and mouse) and
dipteran (An. gambiae and D. melanogaster) CAs 123
6-3. Localization of An. gambiae CA activity 124
6-4. Localization of CA mRNA expression within An. gambiae whole mounts 125
6-5. Localization of CA mRNA expression within the posterior midgut of An.
gambiae 126
6-6. Localization of CA mRNA expression within the hindgut 127
6-7. Localization of CA protein within gastric caeca of An. gambiae larvae 128
6-8. Localization of CA protein within the PMG of An. gambiae 129
6-9. Protein gels and western blots of recombinantly expressed CA protein 130
7-1. New larval mosquito model 139
xi

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
CARBONIC ANHYDRASES AND BICARBONATE TRANSPORT IN
LARVAL MOSQUITOES
By
Theresa J. Sern
May 2004
Chair: Edward J. Phlips
Cochair: Paul J. Linser
Major Department: Fisheries and Aquatic Sciences
Carbonic anhydrase (CA) is an important enzyme due to its involvement in many
pH-dependent, physiological processes. CA reversibly converts CO2 and H2O into
bicarbonate and a proton. The anterior midgut lumen of the larval mosquito has an
extremely alkaline pH; therefore, we hypothesized that an active CA within the epithelial
cells surrounding this region would rapidly produce bicarbonate to buffer the high pH. A
cDNA cloning strategy, followed by in situ hybridization, was employed to isolate and
localize CA and anion exchanger (AE) transcripts within the mosquito gut. Localization
of CA enzymatic activity was assessed via histochemical analyses. Enzymatic and
electrophysiological analyses of recombinant CAs and AE were also performed. In this
dissertation, cDNAs encoding three CA genes and an AE were cloned and localized
within the larval mosquito gut. One isoform of mosquito CA, which was cloned from
two different mosquito species, was localized to a specific subset of muscle fibers on the
Xll

basal side of the anterior midgut. This CA resembles the mammalian CA IV isozyme in
that a glycosylphosphatidylinositol (GPI)-link tethers the enzyme to the extracellular
membrane. The other CA isoform, an active cytosolic enzyme, was localized to the
gastric caeca and posterior midgut regions. It was also determined that the AE transports
chloride and is expressed in the gastric caeca, posterior midgut, and Malpighian tubules.
We were unable to detect CA within the anterior midgut epithelial cells using a variety of
assays. My studies have therefore led to an alternative hypothesis that one or more CAs
within the mosquito gut, but located outside of the anterior midgut epithelial cells,
contribute to buffering the alkaline pH of 11 within the anterior midgut lumen. The
localization of two CA isoforms, one extracellular and the other cytosolic, and an AE
possessing a putative CA binding sequence, to the regions flanking the anterior midgut,
supports the prediction of a bicarbonate transport metabolon within the gastric caeca and
posterior midgut regions. Such a metabolon has only been studied in mammals, however,
the colocalization of CA and AE within the mosquito gut suggests a similar network of
bicarbonate production and transport.

CHAPTER 1
INTRODUCTION
Insects represent one of the most numerous and diverse groups of animals on the
planet. One particularly successful group of insects is the well-studied Pterogota (winged
insect) group. This grouping includes the Lepidoptera (butterflies and moths) as well as
the Dptera (flies). These insects have been extensively studied due to their huge impact
on the lives of humans. For example, the Lepidopteran, Manduca sexta, is a great pest to
tobacco companies that rely on abundant and healthy tobacco crops, on which M. sexta
feeds. Also, mosquitoes (Dipterans) are responsible for transmitting a host of diseases to
humans as well as other mammals by injecting pathogens, along with the anti-coagulants
from their salivary glands, to aid in bloodletting. The pathogens that cause these diseases
can be viruses or various parasites (eg. protozoans).
Mosquitoes belong to the order Dptera, family Culicidae. According to the
American Mosquito Control Association, there are more than 2500 different species
throughout the world, with 150 species in the United States (Darsie and Morris, 2000;
Spielman and D'Antonio, 2001). Mosquitoes act as vectors for a wide variety of diseases
such as malaria, yellow fever, west nile virus, and dengue fever. Recent reports estimate
that fifty to one hundred million cases of dengue fever occur annually, along with several
hundred thousand cases of the life-threatening form of the disease, dengue hemorrhagic
fever (DHF; Halstead, 1997). The geographic range of dengue fever has expanded over
the last two decades, primarily because of the spread of its principal vector, Aedes aegypti
1

2
(Gubler, 1997). Another mosquito example, Anopheles gambiae kills millions of people
each year in Africa by infecting them with the deadly Plasmodium parasite that causes
malaria. Many studies have therefore been undertaken to understand the life cycle and
physiology of these insects that exert such a large socio-economic impact.
The mosquitos ability to acquire, harbor and transmit deadly pathogens has
spurred research into the workings of the mosquito gut. Specific cells of the midgut,
which express a proton pumping V-ATPase, have been found to be preferentially invaded
by pathogens (Shahabuddin and Pimenta, 1998). Studies have also shown that the
mosquito gut is not a static organ but is comprised of several different regions. Each
region displays different characteristics and is made up of different cell types.
Alkaline Gut
Larval mosquitoes, as well as some caterpillars, are known to possess a highly
alkaline digestive system (Dadd, 1975). The tobacco homworm, M. sexta, has a gut
lumen pH that can exceed 11, while the larval mosquito, Aedes aegypti, displays a pH
greater than 10 in its anterior midgut region (Zhuang et al., 1999). These insects are not
only unharmed by this caustic pH, but are able to generate this property' while
maintaining cellular homeostasis.
The larval midgut is involved in ionic and osmotic regulation as well as digestion,
absorption, and excretion (Clements, 1992). It is subdivided into four structurally
distinguishable regions: cardia, gastric caeca, anterior stomach, and posterior stomach
(Fig. 1-1). Each of these regions consists of one cell layer of epithelial cells, composed
of large columnar cells and much smaller cuboidal cells, which vary in character
somewhat from region to region. Belying this simple architecture however, the epithelial

3
cells are capable of maintaining physiological homeostasis while facing a pH range of 7-
11 along the length of the mosquito gut lumen (Dadd, 1975). This range in pH, along the
length of the mosquito gut, is presumed to support digestive and assimilation functions
(Clements, 1992). The epithelial cells of the anterior midgut (AMG) surround a highly
alkaline lumen (pH 11) while those of the gastric caeca (GC) and posterior midgut
(PMG) surround a neutral to mildly alkaline lumen (pH 7-8; Clements, 1992; Zhuang et
al., 1999). The different pH values found along the midgut may support the various
metabolic functions that are active in each gut region. The gastric caeca perform ion and
water transport, the anterior midgut performs alkaline digestion, the posterior midgut
performs nutrient absorption, and the Malpighian tubules (part of the hindgut) actively
transport potassium and fluid (Clements, 1992).
The role of the alkaline pH in the anterior midgut is a point of some controversy.
It has been suggested that the high pH contributes to the digestion of plant detritus and, in
particular, to the dissociation of tannin-protein complexes (Martin et al., 1980). The high
pH restricts the conglomeration of proteins within the anterior midgut that could interfere
with the insects normal physiology. These complexes could also interfere with insect
digestion by blocking the active sites of many different digestive enzymes. Therefore,
the alkaline gut serves as a proposed benefit to the insects by allowing ingested food to
remain soluble. The alkalinity therefore keeps the gut free from attachable tannin-protein
complexes and enhances the assimilation of proteins. Berenbaums review (1980) of
Lepidopteran insects correlated gut pH (range from 7.0-10.3) with diet. Caterpillars
feeding on leaves containing tannins were found to display a more alkaline pH (average
pH 8.76) than those feeding on low tannin diets (average pH 8.25; Berenbaum, 1980).

4
Although this alkaline digestive strategy is well documented in insects, the molecular
processes involved have not been clearly defined.
Carbonic Anhydrase
Carbonic anhydrase (CA), a blood enzyme, first described by Meldrum and
Roughton in 1933, catalyzes the reversible hydration of carbon dioxide to form
bicarbonate and a proton (CO2 + H2O <> HCO3' + H+; Meldrum and Roughton, 1933).
Carbonic anhydrase was first characterized in erythrocytes as the result of a search for a
catalytic factor that would enhance the transfer of bicarbonate from the erythrocyte to the
pulmonary capillaries (Meldrum and Roughton, 1933). Since it was first described, CA
has been shown to play an important role in most acid/base transporting epithelia.
Fourteen different CA isoforms have been characterized to date in mammals (Hewett-
Emmett and Tashian, 1996). These enzymes have been determined to function in pH
regulation and ion balance, thereby performing a crucial role in many biological
processes such as respiration, bone resorption, renal acidification, gluconeogenesis,
aqueous humor production, gastric acid production, cerebrospinal fluid formation, and
signal processing (Dodgson, 1991; Sly and Hu, 1995; Hewett-Emmett and Tashian, 1996;
Lindskog, 1997; Sun and Alkon, 2002).
Various types of epithelial cells, such as those described in the mammalian
kidney, contain CAs that can provide large quantities of bicarbonate for buffering cells
and their microenvironment. Polarized epithelia play an important role in partitioning
physiologically distinct compartments, and in maintaining cell and tissue homeostasis.
The epithelial cells found in the larval mosquito midgut may serve a similar partitioning
function. Like the mammalian kidney, different regions of the mosquito gut may play

5
differential roles in homeostasis and function. Elucidating the distribution of CAs along
the mosquito midgut epithelium may uncover the mechanisms responsible for the unique
alkaline physiology of the mosquito gut.
Mosquito Development and Control
Part of the success of insects can be attributed to the structural adaptation of their
integument, which functions as skin, skeleton, sensory and respiratory organ, and food
reserve (Rockstein, 1964). The advantage of having an extremely strong integument is
offset by the disadvantage of not being able to grow significantly in size. Insects have
overcome this growth-limiting problem by shedding their integument and rebuilding a
new larger one. This process of ecdysis (molting) is used as a tool for marking the
different stages of development in many insect species. While mosquito control can
target different stages of mosquito development, this project focuses on the larval
enzymes, specifically early fourth instar, which begins immediately after the third molt.
Careful attention was paid to the stage of insect development in all experiments due to a
previous study that showed insect enzymes to decrease or completely arrest prior to
molting (Jungreis et ah, 1981).
The mosquito life cycle begins at hatching from the egg (Fig. 1-2). At this point
the fully independent mosquito is called a first instar larva. Successive molts mark the
transition to the next larval instar, four larval instars in all. In each instar, the larvae
possess a series of morphological characteristics, some particular to that stage. However,
there are only slight changes in internal organs such as the midgut. Within a day or two
the late fourth instar larva changes into a pupa (Clements, 1992). Within twenty-four
hours, the flying adult emerges from the pupa case. Adult females of most mosquito

6
species require a bloodmeal in order to nourish their developing eggs. However, the
males do not ingest blood but instead feed on fruit or do not feed at all (Clements, 1992).
Mosquito control tactics use different methods for controlling mosquito larvae as
compared to the flying adults. Mosquito larvae are confined to the water in which they
develop, whereas the adults are free-flying and therefore highly mobile. Pesticide sprays
are employed against the flying adult mosquitoes, but dragonflies and butterflies are also
ill-affected. An arguably better strategy for mosquito control is to target the larvae before
they are capable of biting and transmitting disease. Mosquito larvae are voracious eaters,
incessantly consuming particulates in the water around them, taking in almost anything.
Because of this non-discretional eating behavior, the wriggling larvae can potentially
consume a larvacidal agent if placed in the water. Determining the physiological roles of
larval mosquito gut enzymes and metabolic transporters may provide a lead for
constructing mosquito larvacides.
Carbonic Anhydrase Inhibition
The focus of this project is to examine the distribution and expression of CAs
within the fourth instar of larval development of two species of mosquito, Ae. aegypti and
An. gambiae. A tangential result of characterizing mosquito CAs may be in the
development of mosquito-specific inhibitors. If a CA is discovered to be essential for
mosquito development or homeostasis, a specific inhibitor of precisely this mosquito CA
isoform could be developed. Since virtually all organisms contain CA enzymes, an
inhibitor that would compromise this mosquito CA while not affecting any other
isozymes would be necessary for mosquito control so that non-target species would not
be affected. Differentially specific CA inhibitors are already employed in the distinctive

7
characterization of mammalian CA isoforms. For example, the acidic sulfonamide
benzolamide has been used for the preferential inhibition of extracellular CA while not
compromising any intracellular CA activity (Tong et al., 2000). This occurs due to the
inability of benzolamide to readily penetrate cell membranes (Tong et al., 2000). The
wealth of information pertaining to mammalian CA isoforms and their specific inhibitors
provides a basis for comparisons with CAs that are discovered in the mosquito midgut.
Sulfonamide CA inhibitors are widely used to treat a number of conditions including
glaucoma, gastro-duodenal ulcers, and cancer, by lowering the production of fluids and
acids. Parkkila et al. (2000) showed that the invasion of renal cancer cells in vitro could
be inhibited with CA inhibitors. If larval mosquito physiology is dependent upon the
generation or maintenance of the alkaline gut, and CA is a necessary component, then the
possibility exists for the use of CA inhibitors as mosquito larvacides.
Bicarbonate Transport
The site(s) of bicarbonate production by CA may not be as important as the
translocation of the bicarbonate that is produced. Transporters can facilitate the passage
of bicarbonate and other ions through otherwise impermeable cell membranes.
Bicarbonate transporters compose a large family of membrane proteins that includes the
anion exchangers (AEs), sodium bicarbonate cotransporters (NBCs), and members of the
sulfate transporter group that can also transport bicarbonate (Alper et al., 2001). Most of
the BT proteins consist of a cytosolic anchoring domain as well as a 10-14 membrane-
spanning transporter domain (Alper et al., 2001). Also, evidence exists that some AEs
are capable of physically binding CA enzymes. Thus, the fourth extracellular loop of
AE1 contains a glycosyl-phosphatidyl-inositol (GPI)-linked CA IV binding site and the

8
intracellular carboxy terminus of AE1 was found to contain a cytosolic CAII binding site
(Vince and Reithmeier, 2000; Sterling et al., 2002a). A metabolon, a complex of
membrane proteins involved in regulation of bicarbonate metabolism and transport,
defines the relationship between the CA and AE proteins (Sterling et al., 2001a). This
bicarbonate transport metabolon, is thus capable of transporting bicarbonate as soon as it
is available from the CA enzyme. Transport can be in either direction, into or out of the
cell, and is therefore predictively capable of maintaining a tight hold on pH. The
occurrence of such a tight bicarbonate control mechanism could be very advantageous to
the mosquito. With such a large pH gradient across the membrane, a bicarbonate
transport metabolon could ensure that the pH on either side of the membrane is strictly
monitored. This bicarbonate transport metabolon has only been identified in a
mammalian system. Despite this fact, an insect gut model that employs such a
bicarbonate transport metabolon is easy to envision. Because of the strong pH gradient
that is maintained in the mosquito gut, it is reasonable to propose that a bicarbonate
transport metabolon could exist in this system as well.
Gut Alkalization Model
My first physiological model of the larval mosquito midgut was derived from the
tobacco homworm, M. sexta, which also uses an alkaline digestive strategy. In this
model, several proteins contribute to the high alkalinity (Fig. 1-3). These are the CA, the
H+ V-ATPase, and the cation and anion exchangers. The FT V-ATPase is thought to be
the energizer of the system by using ATP, and pumping protons out into the lumen of the
anterior midgut. This sets up a potential difference across the membrane of about 210 mv
(Harvey, 1992). In this model, the predicted cytosolic CA within the anterior midgut

9
combines carbon dioxide and water to produce bicarbonate and a proton ion. The
bicarbonate is pushed from the epithelial cell, across to the lumen side by the anion
exchanger, in trade for a chloride ion. The proton then gets stripped off of the
bicarbonate and, along with the proton pumped across by the V-ATPase, is brought back
into the cell in exchange for a potassium ion (Wieczorek et ah, 2000). This potassium
ion combines with the carbonate to produce potassium carbonate, which is hypothesized
to be responsible for the high alkaline pH of the anterior gut region. This hypothesis
stems from the fact that potassium ions are actively produced by the Malpighian tubules
and are circulated throughout the gut via the hemolymph (Clements, 1992). Potassium
carbonate also has a pKA greater than 10 and can therefore contribute to the gut
alkalization.
The goal of this project was to expand and adapt this model to the larval mosquito
by completing several clear objectives. These objectives are outlined within the
following specific aims.
Specific Aims
1. Determine whether CA is involved in buffering the high alkalization of the larval
mosquito gut.
A. Determine if a CA enzyme is present within the mosquito gut. Determine which
regions of the larval Ae. aegypti gut display CA activity using CA histochemistry
and l80 isotope exchange.
B. Determine if CA-specific inhibitors, such as acetazolamide, can influence larval
midgut alkalization.
2. Determine whether CA is expressed in the larval mosquito gut.
A. Clone and characterize full length CA cDNAs from the larval midgut of Ae.
aegypti and An. gambiae.

10
B. Use experimental and bioinformatical approaches to determine if CA expressed in
the mosquito gut is similar to a characterized mammalian CA isoform.
Furthermore, determine the subcellular location of mosquito CA isoforms as
cytosolic, membrane-bound, mitochondrial, or GPI-linked.
C. Determine which regions of the mosquito gut express CA mRNA and protein
using in situ hybridization, real time PCR, and immuno-localization.
3. Determine whether anion exchangers (AE) are involved in the pH regulation of the
larval mosquito gut.
A. Clone and characterize an AE from the larval mosquito gut that uses the
bicarbonate produced by CA as a substrate.
B. Determine whether the mosquito AE transports chloride using a Xenopus oocyte
expression assay.
C. Determine if AE is expressed in the same regions of the mosquito gut as the CA.
Co-localization of CA and AE would support the existence of a bicarbonate
transport metabolon.
D. Determine if the AE contains the amino acid sequence predicted to be necessary
for binding C A. If indeed the AE protein is predicted to bind CA, a bicarbonate
transport metabolon within the larval mosquito gut could maximize bicarbonate
production and transport.
4. Present a new larval mosquito model that reflects the studies in this dissertation.
Bring together all localized components of mosquito gut physiology into one model.

11
Midgut
I
GC
AMG
PMG
Hindgut
1 1
MT
Figure 1-1. Illustration showing the regions of the larval mosquito gut. The midgut is
composed of the cardia, gastric caeca (GC), anterior midgut (AMG), and
the posterior midgut (PMG). The hindgut is composed of the Malpighian
tubules (MT) and the rectum.

12
st
1 instar
larva
2nd instar
larva
adult
emerge
3rd instar
larva
3rd
molt y
th
4 instar
larva
pupa
Figure 1-2. Illustration of the mosquito life cycle. The four life stages are egg, larva,
pupa, and adult. The larval stage consists of four different instars. Early
fourth instar larvae, following the third molt, were chosen for all
experiments. The female mosquito continues the cycle by laying eggs,
usually after a required blood meal.

Blood
(Acidic)
pH 6.5
Caterpillar gut call
(Neutral)
pH* 7.0
Lumen
(Alkaline)
pH* 11.0
13
H*V-ATPaee Q Anion exchanger
Cation exchanger Carbonic anhydraae
Amino add: K+ cotranaportor
Channel
Figure 1-3. Preliminary mosquito anterior midgut model based on M. sexta. This
theoretical model places a CA II-like isoform within the cell cytosol where
it combines carbon dioxide and water to form bicarbonate and a proton.
Alkalization is driven by a proton pumping V-ATPase that resides in the
apical membrane and pumps protons into the lumen. A chloride/
bicarbonate exchanger, that is also located in the apical membrane,
exchanges bicarbonate from the CA, for chloride from the lumen. A cation
exchanger transfers potassium to the lumen while stripping protons from the
bicarbonate for the exchange. The potassium ion combines with the de-
protonated carbonate ion to form potassium carbonate, which brings the pH
to highly alkaline levels.
K*pump

CHAPTER 2
MATERIALS AND METHODS
Experimental Insects
Ae. aegypti eggs were obtained from a colony maintained by the United States
Department of Agriculture (USDA) laboratory in Gainesville, Florida. The eggs were
allowed to hatch in 20 ml of 2% artificial seawater (ASW; 8.4 mM NaCl, 1.7 mM KC1,
0.1 mM CaCl2,0.46 mM MgCl2, 0.51 mM MgS04, and 0.04 mM NaHC03). The
mosquito larvae were reared in 2% ASW at room temperature. The Ae. aegypti larvae
were fed a mixture of yeast and liver powder (1:1.5 g respective dry weight; ICN
Biomedicals Inc., Aurora, Ohio). Eight to ten days were required for this species to reach
the early fourth instar.
An. gambiae eggs were obtained from the Centers for Disease Control and
Prevention (CDC) in Atlanta, Georgia. Strict handling guidelines were followed with this
particular species, which does not currently inhabit Florida, due to its inherent ability to
acquire and transmit the Plasmodium protozoan, which causes malaria. This Anopheles
species was therefore reared in deionized water inside of a locked incubator set at 30C.
A mesh screen served as a second barrier within the incubator while the sealed (but not
airtight) containers harboring the An. gambiae larvae served as the third barrier against
escape. The An. gambiae larvae were fed a Wardley tropical fish flake food (The Hartz
Mountain Corp., Secaucus, New Jersey). Early fourth instar larvae were chosen for all
experiments. Ten to twelve days from the hatch day were required for this species to
14

15
reach the early fourth instar. Late fourth instar larvae that went unused were sacrificed to
prevent any chance of emerging adults.
Preparation and Fixation of Tissue
To dissect out the midgut, the heads of the cold-immobilized larvae were pinned
down using fine stainless-steel pins to a Sylgard layer at the bottom of a Petri dish
containing hemolymph substitute solution consisting of 42.5 mM NaCl, 3.0 mM KC1, 0.6
mM MgS04, 5.0 mM CaCl2, 5.0 mM NaHC03, 5.0 mM L-succinic acid, 5.0 mM L-malic
acid, 5.0 mM L-proline, 9.1 mM L-glutamine, 8.7 mM L-histidine, 3.3 mM L-arginine,
10.0 mM dextrose, 25 mM Hepes and adjusted to pH 7.0 with NaOH (Clark et al., 1999).
The anal segment and the saddle papillae were removed using ultra-fine scissors and
forceps, and an incision was made longitudinally along the thorax. The cuticle was
gently pulled apart and the midgut and gastric caeca were removed. In some cases, the
gut contents enclosed in the peritrophic membrane slid out, leaving behind the empty
midgut. In other cases, it was necessary to remove the peritrophic membrane and its
contents manually. For enzyme histochemistry, fixation was in 3% glutaraldehyde in 0.1
M phosphate buffer, pH 7.3, overnight at 4C (Ridgway and Moffet, 1986). For in situ
hybridization and immunohistochemistry, dissected tissues were fixed overnight in 4%
paraformaldehyde in 0.1 M cacodylate buffer, pH 7.2 (0.1 M phosphate buffer, pH 7.2
was used for Ch 4 and 5 in situ). In some cases, the dissected larval midguts were
photographed using a Nikon FX-35DX photographic camera mounted on a Nikon SMZ-
10 dissecting microscope. In other cases, digital images were acquired using a Leica
DMR microscope equipped with a Hammamatsu CCD camera. All images were
assembled using Corel Draw-11 software.

16
Bromothymol Blue Qualitative Assay
A qualitative test to detect carbonic anhydrase activity in mosquito larval midgut
homogenate was adapted from the test described by Tashian (1969). The procedure
included immersing a piece of Whatman no.l paper in a solution made with 0.15%
Bromothymol Blue (BTB) in ice-cold 25 mM Tris HC1, 0.1 M Na2SC>4, pH 8.0. The
paper was allowed to soak completely in this blue solution and was placed on ice for 30
minutes. The colored filter was then transferred to a Petri dish with a hole in the lid.
Samples of mosquito larval midgut homogenate were prepared by sonicating midguts of
early fourth instar larvae in ice-cold 25 mM Tris HC1, 0.1 M Na2SC>4 pH 8.0, with
protease inhibitor cocktail (Sigma-Aldrich, St. Louis, MO; diluted 1:1000). An
autopipette was used to spot exactly 4 pi samples on the paper. Controls were also
spotted. The controls included a buffer with protease inhibitor and controls for the
liver/yeast food added to the medium in which the mosquito larvae were reared. These
food controls included a range of concentration from 1 to 100 pg/ml liver powder and
yeast. Carbonic anhydrase from bovine erythrocytes (Sigma-Aldrich) dissolved in the
same buffer described above was used as a positive control.
A steady stream of CO2 at 34.5 KPa was blown for 3 seconds through the
opening on the lid of the Petri dish, and the dish was sealed and kept on ice. The
formation of yellow spots in a few seconds was indicative of carbonic anhydrase activity.
Effect of Methazolamide on the Alkalization of the Midgut of Live Larvae
The effect of a CA inhibitor, methazolamide, on gut alkalization and the capacity
of whole larvae to alkalize their culture medium was examined. Flat-bottomed tissue
culture plates (24 well, Sarstedt Inc., Newton, North Carolina) were filled with 1 ml of 25

17
mM Tris HC1, 0.1 M Na2S04 buffer, pH 8.5. BTB solution was added to each well until
a 0.003% solution was achieved. Five live early fourth-instar larvae that had been placed
in BTB indicator solution for 2 hours were added to each of the wells, and the larvae
were allowed to adjust to their new environment for 30 minutes. Methazolamide
dissolved in Dimethyl Sulfoxide (DMSO; Sigma-Aldrich) at concentrations ranging from
10"6 M to 8x10'3 M was added to the wells. Controls included wells containing DMSO
with BTB indicator but no inhibitor and wells containing BTB indicator but no DMSO.
The plates were scanned using a Hewlett Packard ScanJet 6100C scanner before addition
of the inhibitor and 5 hours later. In addition, the midguts were dissected and
photographed to record the pH within the gut lumen as revealed by the color of ingested
BTB.
180 Exchange Method to Measure Carbonic Anhydrase Activity
Tissue homogenate carbonic anhydrase activity was measured using the O
exchange method (Silverman and Tu, 1986). Midguts were dissected, and the peritrophic
membrane was removed together with its contents. Individual measurements of CA
activity were performed with pooled samples of gastric caeca, anterior midgut, posterior
midgut and Malpighian tubules. The method involved adding O-labeled NaHCCb to
0.1 M Hepes buffer, pH 7.6, at 9.5C. The disappearance of 180 isotopes from C02
and/or HCO3' upon addition of the enzyme preparations was monitored. Measurements
10
of O in C02 were accomplished with a mass spectrometer, using a C02-permeable inlet
that allowed very rapid, continuous measurement of the isotopic content of C02 in
solution. All samples were centrifuged at 14,000 rpm at room temperature prior to the
assay to remove food and insoluble material. Inhibition was accomplished by adding

18
methazolamide to a final concentration of 10'6 M. Recombinantly expressed and purified
mosquito CAs were also tested for activity using this assay.
Isolation of RNA and Synthesis of cDNA
Total RNA was isolated from freshly dissected fourth instar mosquito larval
midguts using TRI Reagent (Molecular Research Center Inc., Cincinnati, Ohio)
according to the manufacturer's instructions. Briefly, 100 Ae. aegypti gut epithelial
organs, including fore-, mid-, and hindgut (approximately 20 mg) were dissected in HSS
and transferred to a sterile microcentrifuge tube containing TRI Reagent (600 pi). The
tissue was homogenized and incubated for 5 min at room temperature. The homogenate
was then extracted with chloroform (40 pL) and precipitated with isopropanol (100 pL).
The RNA pellet was washed with 75% ethanol (200 pL), air-dried and resuspended in 50
pL diethylpyrocarbonate (DEPC; Sigma-Aldrich)-treated H2O. RNA concentrations
were calculated from the absorbance at 260 nm. Total RNA (10 pg) was reverse-
transcribed for 2 hours at 42C in a 20 pi reaction mixture using 5 pmol of oligo(dT)12-
18, RNasin (1:40 dilution), IX first strand buffer, 1 mM dNTPs, and 200 units (U) of
Superscript II reverse transcriptase (Invitrogen Inc., Carlsbad, California). This cDNA
was used to clone the first fragment of Ae. aegypti CA.
Bioinformatics
The National Center for Biotechnology Information (NCBI) website
(www.ncbi.nlm.nih.gov) was used for the majority of the bioinformatical data presented
in this study. The first mosquito genome, An. gambiae, was released in 2002 (Holt et al.,
2002), and made accessible to the public on the NCBI website. The basic local alignment
search tool (BLAST; Altschul et ah, 1990) was employed for primer construction as well

19
as analyzing PCR products. The NCBI Blast Flies database
(www.ncbi.nlm.nih.gov/BLAST/Genome/FlyBlast.html), together with the Ensembl
database (www.ensembl.org/Anopheles gambiae/) were used to predict the number of CA
genes in the Drosophila melanogaster and An. gambiae genomes by inputting the Ae.
aegypti CA as the search sequence. These partial sequence results were then annotated to
reflect the 2 full-length CA sequences that we have cloned from An. gambiae and
presented within this manuscript.
Ensembl is a joint project between the European Bioinformatics Institute and the
Sanger Institute to bring together genome sequences with annotated structural and
functional information. The NCBI protein database (pdb) and the BLAST were used in
conjunction with the 3-dimensional structure viewer (Cn3D; Hogue, 1997) for the
prediction of antibody accessible peptide regions in mosquito proteins. BLAST analyses
also confirmed that the chosen antigenic peptides were unique. The conserved domain
database (CDD; Marchler-Bauer et al., 2002) and the conserved domain architecture
retrieval tool (CDART; Geer et al., 2002) were used to predict the function of our newly
cloned mosquito proteins. Alignments were produced using Clustal W (Thompson et al.,
1994), as implemented in DNAman software (Lynnon Biosoft, Vaudreuil, Quebec,
Canada).
Cloning of CA from Aedes aegypti Larval Midgut
Degenerate oligonucleotides were designed against the regions of conserved
amino acids among CA proteins as determined by the BLAST analysis of several
vertebrate and two putative, but annotated, CA proteins from the D. melanogaster
sequence database.

20
The primer sequences used initially for Ae. aegypti CA were CA5F and CA3R
(see Table 2-1). PCRs were performed in a total volume of 20 pi, and the reaction
mixture contained 0.1 pg of cDNA as template, 0.2 pM of each primer, 200 pM each of
dNTPs, IX PCR buffer and 1 U of Taq polymerase (Promega; Madison, Wisconsin).
The PCR cycling profile was: 94C for 5 min, 55C for 2 min and 72C for 3 min,
followed by six cycles of 94C for 0.5 min, 53C (in increments of 2C/cycle) for 1 min
and 72C for 1 min and 35 cycles of 94C for 0.5 min, 45C for 1 min and 72C for 2
min followed by a final extension at 72C for 15 min. The PCR products were visualized
on 1 % agarose gels and specific products were isolated using a QIAquick gel extraction
kit (Qiagen, Inc, Valencia, California), diluted 1:100 in water, and used as template for a
second, identical PCR. The resulting 297 base-pair (bp) product was gel-purified, ligated
into pGem-T (Promega) and transformed into JM109 Escherichia coli (Promega) for
subcloning. This partial Ae. aegypti CA cDNA was completed using amplified cDNA
pools from gastric caeca and posterior midgut.
Construction of Amplified cDNA Pools
Adapter-ligated, amplified cDNA pools (libraries) were constructed from
different regions of the fourth instar larval gut of both Ae. aegypti and An. gambiae using
a technique optimized for invertebrate tissues (Matz et al., 1999). The gastric caeca,
anterior midgut, posterior midgut, rectal salt gland, Malpighian tubules, and anal papillae
of ten larvae were dissected in HSS and collected separately, resulting in six discreet
tissue pools. The tissue was dissolved in Buffer D (500 pL; 4 M guanidine thiocyanate,
30 mM sodium citrate, and 30 mM beta-mercaptoethanol). The mixture was placed on
ice and combined with phenol (500 pL, pH 7.0) and chloroform (100 pL). The mixture

21
was vortexed and centrifuged at 14,000 g for 30 seconds at 4C. The upper, aqueous
phase was transferred to a clean tube and 5 pL glycogen solution (Pharmacia Quick Prep
Micro RNA purification kit, Piscataway, New Jersey). The RNA was precipitated by the
addition of 100% ice-cold ethanol (550 pL) followed by centrifugation at 14,000 g for 6
minutes at room temperature. The supernatant was removed and 1 mL of ice-cold
ethanol (80%) was added. The mixture was centrifuged at 14,000 g for 10 minutes at
room temperature, the supernatant was removed, and the pellet was air-dried.
For first strand synthesis, the pellet was resuspended in DEPC-treated water (5
pL) and combined with the TRsa primer (1 pM; Table 2-1). This mixture was incubated
at 50C for 3 minutes and immediately placed on ice. Then IX ligation buffer (Marathon
cDNA Amplification kit, BD Biosciences, Palo Alto, California), 0.01 M DDT, 1 U
Superscript II (Life Technologies; Rockville, Maryland), and 0.5 pL dNTP mix (10 mM
each dNTP, Marathon cDNA Amplification kit) were added to a total volume of 10.5 pL.
This reaction mixture was incubated at 42C for 1 hour and immediately put on ice.
For second strand synthesis, DEPC-treated water (49 pL) was added to the first
strand reaction mix. The mixture was then combined with 1.6 pL dNTP mix (10 mM
each, Marathon cDNA Amplification kit), IX reaction buffer (Marathon cDNA
Amplification kit), and 4 pL second strand synthesis enzyme mix (Marathon cDNA
Amplification kit) in 80 pL total volume. The reaction mix was then incubated at 16C
for 1.5 hours. T4 DNA polymerase (1 U; Marathon cDNA Amplification kit) was added
to the reaction mixture and the entire mixture was incubated at 16C for an additional 0.5
hour. The reaction was stopped by incubation at 65C for 5 minutes.

22
The reaction mix (80 pL) was combined with 40 pL phenol and 40 pL
chloroform and centrifuged at 14,000 g for 10 minutes. The upper, aqueous phase was
removed and transferred to a clean tube. The cDNA was precipitated by the addition of 3
M sodium acetate (8 pL, pH 5.0) and 100% ethanol (160 pL). The mixture was
centrifuged at 14,000 g for 15 minutes at room temperature. The supernatant was
removed and the pellet was air-dried.
For adaptor ligation, the cDNA pellet was resuspended in DEPC-treated water
(6 pL) and combined with 1 pM adaptor, IX ligase buffer, and 1 U T4 ligase (Marathon
cDNA Amplification kit) in 10 pL total volume. This mixture was stored overnight at
16C. For cDNA amplification, the ligation mixture (10 pL) was combined with 40 pL
DEPC-treated water. PCR amplification was then performed using the Advantage kit
(BD Biosciences). The diluted cDNA (1 pL) was combined with IX advantage buffer,
0.4 pL dNTP mixture (10 mM each), 0.1 pM DAP and TRsa primers (Table 2-1), and 0.4
pL advantage enzyme mix in 20 pL total volume. The cycling profile consisted of 94C
for 30 seconds, 66C for 1 minute, and 72C for 2.5 minutes. The reaction was analyzed
on a 1% agarose gel after 12,16, and 20 cycles. A final chase step was then performed to
ensure that all cDNAs were completely double-stranded. Both 5 and 3 adaptor primers
were added to the PCR reactions and two cycles of 77C for 1 min, 65C for 1 min, and
72C for 2.5 min were performed. The resulting collections of amplified cDNA were
then diluted 1:50 and used as template for subsequent PCR experiments.
Amplified cDNA pools from An. gambiae were used to clone two CA cDNAs and
the AE cDNA. Exact primers were designed from conserved regions of the proteins as

23
determined by BLAST analysis using characterized proteins against the An. gambiae
genome. See table 2-1 for all initial primer sequences.
3' and 5' Rapid Amplification of cDNA Ends and Sequencing
Full-length cDNAs were obtained by rapid amplification of cDNA ends (RACE),
(Zhang and Frohman 1997, modified by Matz et al., 1999). Exact primers were defined
according to the 5 adaptor (DAP primer) along with a reverse primer specific to the
cloned fragment, and 3 TRsa adaptor (TRsa primer) along with a forward primer specific
to the cloned fragment (see Table 2-1 for adaptor primer sequences). These ends, which
included the 5 and 3 UTR sequences, were then used to design PCR primers to produce
a single product with consensus start and stop codons.
Plasmid DNA from individual colonies was purified using a Qiaprep
Plasmid Mini kit (Qiagen). The plasmid DNA (50 ng) was then sequenced using the ABI
Prism Big Dye Terminator Cycle Sequencing Kit (PE Biosystems, Foster City,
California) and the reaction products were analyzed on an ABI Prism 310 Genetic
Analyzer (PE Biosystems).
Construction of In Situ Hybridization Probes
Sense and antisense digoxygenin (DIG)-labeled cRNA probes were generated by
in vitro transcription using a DIG RNA labeling kit (Roche Molecular Biochemicals,
Indianapolis, Indiana). The initial in situ hybridization experiment, presented in chapter
3, used a cRNA probe derived from the original 297 bp Ae. aegypti CA sequence. The in
situ experiments presented in chapters 4 and 5 utilized the full-length CA and AE
sequences. For the first CA antisense probe, the pGEM-T vector containing the 297 bp
CA sequence was linearized by incubating 2 pg of plasmid with Pst I restriction enzyme

24
(New England Biolabs (NEB); Beverly, Massachusetts) and IX buffer 3 (NEB) at a total
volume of 20 pL for 1 hour at 37C. For the sense probe, the pGEM-T vector containing
the 297 bp CA sequence was linearized by incubating 2 pg of plasmid with Not I
restriction enzyme and IX buffer 3 (NEB) in a total volume of 20 pL for 1 hour at 37C.
After digestion, the volume was brought to 100 pL with the addition of 80 pL water. A
phenol/ chloroform extraction was performed such that 100 pL of phenol/ chloroform-
isoamyl alcohol was added to the linearized plasmid and the solution was centrifuged at
14,000 g for 1 minute. The upper aqueous phase was transferred to a new tube and 100
pL chloroform was added. After centrifugation at 14,000 g for 1 minute, the upper
aqueous phase was transferred to a new tube and the chloroform step was repeated. The
linearized plasmid DNA was precipitated by the addition of 10 pL sodium acetate (3 M,
pH 2.5) and 200 pL cold ethanol (100%). The DNA was incubated at -80C for 15
minutes and then centrifuged at 14,000 g for 10 minutes at 4C. The supernatant was
removed and the DNA pellet was washed by the addition of 500 pL ethanol (70%)
followed by centrifugation at 14,000 g for 5 minutes at 4C. The supernatant was
removed and the pellet was air-dried and then resuspended in 13 pL DEPC-treated water.
The full-length Ae. aegypti CA was subcloned into pCR 4-TOPO plasmid using a
PCR manufactured 5 Sal I restriction site and a 3 Xho I site. Therefore, the pCR 4-
TOPO plasmid was linearized by incubating 2 pg of plasmid with either Sal I, IX Sal I
buffer, and BSA, or Xho I, IX buffer 2, and BSA (NEB). The pCR 4-TOPO plasmid was
also used for the generation of the An. gambiae CA and AE probes. For these probes, the
unique restriction sites, Pme I and Not I, located within the pCR 4-TOPO plasmid were
used for linearization with IX buffer 4 and BSA, or IX buffer 3 and BSA, respectively.

25
These mixtures were all incubated at 37C for 2 hours to ensure complete linearization of
the plasmids. After digestion the uncut pCR 4-TOPO plasmids were compared to the cut
plasmids on a 1% agarose gel to confirm linearization. The cut plasmids (10 fiL) were
cleansed using a Qiaquick PCR Purification kit (Qiagen Inc, Valencia, California).
For in vitro translation, the resuspended pellet or purified plasmids were
combined with IX transcription buffer, IX NTP labeling mixture, RNase inhibitor (20
U), and 40 U T3 RNA polymerase (or SP6 for pGEM-T plasmids) or 40 U T7 RNA
polymerase. For the Ae. aegypti CA probes, T7 polymerase was used with the Sal I cut
plasmid to produce the antisense probe, while T3 polymerase was used with the Xho I cut
plasmid to produce the sense (control) probe. The pCR 4-TOPO plasmid used for the
generation of the An. gambiae CA and AE probes contained the CA and AE sequences in
the reverse configuration. Therefore, for these An. gambiae probes, T3 was used with the
Not I linearized plasmids to produce the antisense probes, while T7 was used with the
Pme I cut plasmids to produce the sense (control) probes. The mixtures were incubated
at 37C for 2 hours followed by the addition of 20 U DNase I and incubation at 37C for
15 minutes. The DNase I reaction was stopped by the addition of 0.5 pL of EDTA (500
mM). The DIG-labeled cRNA was then precipitated by the addition of 2.5 pL of LiCl (4
M) and 75 pL cold ethanol (100%). The mixture was incubated overnight at -20C and
centrifuged at 14,000 g for 10 minutes at 4C. The supernatant was removed and the
pellet was washed with 50 pL cold ethanol (75%). The centrifugation step was repeated
and the pellet was air-dried and resuspended in 100 pL DEPC-treated water. The probes
were stored at -80C.

26
In Situ Hybridization
The in situ hybridization experiments presented in chapters 4 and 5 added an
additional fixation step due to a recommendation by Dr. Dmitri Boudko to increase the
clarity of the in situ labeling. A glass electrode fitted to a micromanipulator was used to
inject 4% paraformaldehyde into the thoracic cavity, just behind the head. Successful
perfusion was easily identified by the cessation of the otherwise constant muscle
twitching along the length of the body. This injection of fixative served to preserve the
cellular integrity and protect against the many proteases that exist within the mosquito
gut. For in situ hybridization, methods were adapted from Westerfield (1994). The
midguts were washed with PBS at room temperature and then incubated in 100%
methanol at -20C for 30 minutes to ensure permeabilization of the gut tissue. The tissue
was washed (5 min each wash) in 50% methanol in PBST (Dulbecco's phosphate
buffered saline [Sigma-Aldrich] plus 0.1% Tween-20), followed by 30% methanol in
PBST and then PBST alone. The tissue was fixed in 4% paraformaldehyde in 0.1 M
sodium cacodylate buffer (or 0.1 M phosphate buffer) for 20 min. at room temperature
and washed with PBST. The larval midguts were digested with proteinase K (10 |ig/ml
in PBST) at room temperature for 10 min, washed briefly with PBST and fixed again, as
described previously.
Prehybridization of the tissue was accomplished by incubation in HYB solution
(50% formamide, 5X SSC [IX SSC equals 0.15 M NaCl, 0.015 M Na-citrate buffer pH
7.0], 0.1% Tween-20) for 24 hours at 55C. The larval midguts were transferred to
HYB+ solution (HYB plus 5 mg/ml tRNA, 50 pg/ml heparin) containing 5 ng/ml DIG-
labeled probe and incubated overnight at 55C. Excess probe was removed by washing at

27
55C with 50% formamide in 2X SSCT for 30 min (twice), 2X SSCT for 15 min and
0.2X SSCT for 30 min (twice). For detection, the tissue was incubated in PBST
containing 1% blocking solution (Roche Molecular Biochemicals) for 1 h at room
temperature. The tissue was incubated with anti-DIG-alkaline phosphatase (Roche
Molecular Biochemicals) diluted 1:5000 in blocking solution for 4 hours at room
temperature. The tissue was washed with PBST and incubated in alkaline phosphatase
substrate solution (Bio Rad Laboratories, Hercules, CA, USA) until the desired intensity
of staining was achieved (2-3 hours).
CA Histochemistry
Carbonic anhydrase activity was detected in isolated Ae. aegypti midguts using
Hansson's method (Hansson, 1967), as modified by Ridgway and Moffet (1986). The
procedure involved the incubation of isolated, 3% glutaraldehyde-fixed midguts in 1.75
mM C0SO4, 53 mM H2S04, 11.7 mM KH2P04, and 15.7 mM NaHC03 (pH 6.8). The
incubation medium contains a high concentration of bicarbonate, which stimulates the
production of C02 and hence a decrease in pH in the presence of CA. The acidic pH then
stimulates the formation of insoluble black cobalt salts which were visualized using 0.5%
(NHi)2S in distilled water. Therefore, micro-sites of active CA liberation of C02 from
bicarbonate dehydration become apparent with this assay. Removal of the bicarbonate
substrate (NaHC03) eliminated staining.
Real Time PCR
Region-specific cDNA was produced from dissected mosquito tissue using the
Cells-to-cDNA standard protocol (Ambion INC, Austin, Texas). The gut regions used to
make the amplified cDNA pools were incubated in 50 pL of hot cell lysis buffer for 10

28
minutes at 75C. The lysed tissues were treated with 2 U of DNase I for 30 minutes at
37C. The DNase I was then inactivated by heating to 75C for 5 minutes. For the
reverse transcription reaction, 10 pL of cell lysate was combined with 4 pL dNTP mix
(contains 2.5 mM each dNTP) and 5 pM random decamer first strand primer in 16 pL
total volume. The mixture was incubated at 70C for 3 minutes and then chilled on ice
for 1 minute. This mixture was then combined with IX RT buffer, 1 U M-MLV reverse
transcriptase, and 10 U RNase inhibitor, and incubated at 42C for 1 hour. The reverse
transcriptase was then inactivated by incubation at 95C for 10 minutes. Primers (Table
2.1) were designed using Primer Express software (Applied Biosystems; Foster City,
California). The SYBR Green PCR Master mix, which includes SYBR Green I dye,
Amplitaq Gold DNA Polymerase, dNTPs, and buffer, was used for all real time PCR
investigations. Each cycle of PCR was detected by measuring the increase in
fluorescence caused by the binding of the SYBR Green dye to double-stranded DNA
using an ABI Prism 7000 Sequence Detection System (Applied Biosystems). Initially,
each primer set, including the control 18s ribosomal RNA (Genbank accession M95126),
was assessed to determine the optimal concentration of primer to be used. All real time
experiments used the same 2-step cycling profile: 50C for 2 minutes followed by 95C
for 10 minutes and 40 cycles of 95C for 15 seconds and 60C for 1 minute. Whole gut
cDNA (100 nM) was used as template with 500 nM, 300 nM, 100 nM, or 50 nM of each
primer set and IX SYBR green I master mix in 25 pL total volume. Each reaction was
done in triplicate. The optimal concentration was then chosen based on the amplification
plots and the dissociation curves generated. Once a concentration was chosen for each
primer set, the efficiency of amplification of that set was determined. Serial dilutions of

29
whole gut cDNA were used as template with the appropriate concentration of primers and
IX SYBR green I master mix in 25 jxL total volume. The threshold cycle number (Ct)
was plotted versus the log of the template concentration and the slope (m) and intercept
(b) were determined (Figure 2-1). These pre-determinations were then used in the
standardized comparison of the amount of 18s transcript and CA transcript in each of the
cDNA samples tested. For each analysis, a control containing all of the necessary PCR
components except the cDNA template was run. To determine the relative expression
level for each transcript analyzed, the following equation was used: (Ct-b)/m. The
average log ng for each transcript was then compared to the average log ng of 18s RNA
transcript to normalize the values. Then the expression levels were determined relative to
the transcript with the greatest normalized log ng value and expressed in a bar graph
using Microsoft Excel software.
Antibody Production
An antigenic peptide consisting of eighteen amino acids was chosen from the Ae.
aegypti CA sequence for antibody production. In order to increase the probability that
this antibody would be specific for this particular CA sequence (in the event that other
CA isoforms were isolated from the mosquito gut), attempts were made to synthesize an
antigenic peptide that would be specific to this isoform. The well-characterized
mammalian CA isoforms served as a model in trying to choose a unique CA peptide
sequence. The comparison of the mosquito CA with the mammalian isoforms yielded a
peptide sequence from the amino (N) terminus, where CA isoforms showed the most
diversity, and least conservation. The N terminus of our mosquito CA was predicted to
have an extended loop secondary structure. Unlike an alpha helix, an extended loop is

30
more accessible to antibody probing. Furthermore, three-dimensional analyses (Cn3D
v4.1 NCBI) of predicted CA IV structures (human 1ZNC and mouse 2ZNC) predicted
that the N terminus is exposed and accessible (Figure 2-2). An antigenic peptide was
therefore chosen from the N terminus of the Ae. aegypti CA sequence. This peptide
sequence (GVINEPERWGGQCETGRR) was sent to Sigma-Genosys (Woodlands,
Texas), where it was synthesized and conjugated to bovine serum albumin (BSA). The
synthetic peptide-BSA construct and Freunds incomplete adjuvant were injected into
two rabbits to elicit an immune response. Prior to injection, a blood sample from each
rabbit was collected to serve as the control pre-immune serum. Every two weeks a blood
sample was collected from the rabbits, the fraction of immunoglobulin G (IgG) pooled,
and another dose of the peptide-BSA construct administered. Three months after the
initial injections, the final bleeds were collected and used for all immunohistochemical
analyses.
We also raised antibodies against an An. gambiae cytosolic CA peptide and an
anion exchanger (AE) peptide. These antibodies were produced by the Aves Labs, Inc.,
(Tigard, Oregon), using a similar strategy to that described above. However, these
antibodies were produced in hens. The synthesized peptides were conjugated to BSA and
injected into two hens each. The immunoglobulin Y (IgY) antibodies were collected
from the hens eggs, pooled, and purified.
Immunohistochemistry
The specificity of the antibodies in the resultant antisera was determined. The
antisera were then used to localize the larval mosquito proteins. Dissected and fixed
whole mount mosquito guts were washed 6 times in tris-buffered saline (TBS), placed in

31
pre-incubation medium (pre-inc) for a minimum of 1 hour, and then incubated in primary
antibody (1:1000) overnight at 4C. The guts were then washed in pre-inc and incubated
in FITC-conjugated goat anti-rabbit (GAR) or Alexa-GAR secondary antibody (Jackson
ImmunoResearch, West Grove, Pennsylvania, 1:250 dilution) overnight at 4C. The
whole mount preparations were rinsed in pre-inc and mounted onto slides using p-
phenylenediamine (PPD, Sigma-Aldrich) in 60% glycerol. In some cases Draq 5
(Jackson ImmunoResearch, 1:1000 dilution) was applied before mounting to visualize
nuclear DNA. The samples were examined and images captured using the Leica
scanning confocal microscope.
Live preparations were examined, following a similar procedure, to ensure that
antibodies were capable of localizing extracellular proteins only. In this case, the
primary antibody was applied to the live guts for four hours. The samples were washed
in TBS, and fixed as described previously, before the secondary antibody was applied.
The samples were mounted on slides as described above.
The live gut assays were also performed to determine whether this specific CA is
tethered to the cell membrane via a GPI linkage. Ten live gut preparations were
incubated with phosphoinositol-specific phospholipase C (PI-PLC, 1:100 in HSS; Sigma-
Aldrich) for 90 minutes at 37C. PI-PLC was used as a tool in determining the presence
of a GPI link. Controls in which the guts were incubated in HSS alone were also
performed. The guts were then washed in HSS, fixed, and treated with primary and
secondary antibodies as described above.

32
CA Protein Expression
Recombinant Ae. aegypti and An. gambiae CAs were produced using the pETlOO
vector (Invitrogen). Specific primers were designed to amplify each cDNA. The 3
primers included the sequence 5 to and including the native stop codon. The 5 primers
contain the sequence CACC preceeding the native start codon for correct frame insertion
(See Table 2-1 for primer sequences). PCRs were performed using 1 U of Platinum P/x
polymerase (Invitrogen), the gastric caeca cDNA collections as template (200 ng), IX
Pfx amplification buffer, 1.2 mM dNTP mixture, 1 mM MgS04, and 0.3 pM of each
primer in a total volume of 50 pL. A three-step PCR protocol was used consisting of
94C for 2 minutes followed by 30 cycles of 94C for 30 seconds, 55C for 30 seconds,
and 68C for 1 minute.
The resultant blunt-ended cDNAs (4 pL from PCR mix) were ligated with the
pETlOO directional Topo vector (1 pL and 1 pL salt solution; Invitrogen) for 10 minutes
at room temperature. Top 10 chemically competent E. coli (50 pL; Invitrogen) were
transformed by incubating 3 pL of ligation mix with the cells for 30 minutes on ice,
followed by a heat shock of 42C for 30 seconds. SOC (250 pL) was added to the cells
and they were then incubated at 37C for 30 minutes with shaking. The transformation
mix (100 pL) was then plated on a LB-carbenicillin plate (50 pg/mL) and incubated
overnight at 37C. Colonies were sequenced using Big Dye version 1.1 as described
previously. The purified plasmids (10 ng each) were transformed into BL21 Star (DE3)
cells (Invitrogen) for CA expression as described above. However, after SOC addition
and incubation, the culture was transferred to fresh LB-carb (10 mL) and grown
overnight at 37C with shaking. The next day, 1 mL of culture was transferred to 100

33
m, of fresh LB-carb and was grown at 37C with shaking. Optimization experiments
were performed in order to facilitate the production of the greatest quantity of CA
protein. For production of CA protein, isopropylthio-JJ-galactoside (IPTG, 1 mM final
concentration; Stratagene, La Jolla, California) was added when the culture had attained
an optical density of 0.5 at a 600 nm wavelength. Achieving this density took about 1.5
hours of growth at 37C and 200 rpm. Zinc, in the form of zinc sulfate (0.5 mM final
concentration), was added along with the IPTG to facilitate the proper conformation of an
active CA protein. In order to optimize the duration of the induced growth phase,
samples were collected every hour for six hours. These samples were analyzed on an
SDS-Page 4-12% Bis-Tris gel to compare CA protein content. Four hours of growth was
determined to be ideal for the production of the truncated Ae. aegypti CA IV-like and
lull-length An. gambiae CA II-like proteins.
Total protein was collected using the Probond Purification System according to
the manufacturers instructions for soluble proteins (Invitrogen). The cells were harvested
by centrifugation, sonicated in native buffer (250 mM NaPCL, 2.5 M NaCl; Invitrogen)
with lysozyme (1 mg/mL; Sigma-Aldrich), and centrifuged again to collect a crude
protein extract. The supernatant was applied to a Probond nickel column (Invitrogen)
and washed free of non-specific binding contaminants. The nickel column binds the CA
protein due to the added histidine tag, a repeat of six histidine residues within the pETlOO
expression vector that is inserted after the carboxy-terminus of the CA protein. CA was
eluted by adding imidazole (250 mM; Invitrogen) to the column, which competes with
and displaces the histidine tag. Eluted fractions were separated on an SDS-Page 4-12%
Bis-Tris gel (Invitrogen).

34
Anion Exchanger Oocyte Expression
The full-length anion exchanger (AE) sequence was subcloned into the pXOOM
vector, which is optimized for both oocyte and mammalian expression (Jespersen et al.,
2002; a generous gift from Dr. T. Jespersen). In addition to a T7 RNA polymerase
promoter, this vector contains Xenopus-sptciTic 5 and 3 UTR sequences flanking the
insert in both directions. cRNA synthesis was performed using the T7 mMessage
mMachine kit (Ambion, Austin, Texas), after the cDNA was linearized using PMEI.
One day after surgical removal of the eggs from the frog, the eggs were injected
with either AE cRNA or water (control). After injection the eggs were incubated at 16C
for 4 days, long enough for measurable protein production and expression. The oocytes
were maintained in ND96 (96 mM NaCl, 2 mM KC1, 1 mM MgC^, 10 mM HEPES, pH
7.4 with NaOH). The medium was changed daily and dead oocytes were removed.
Anion Exchanger Physiology
Expression of the An. gambiae AE was examined using 2-electrode voltage clamp
electrodes. The voltage electrodes were pulled using 1.2 mm glass (M1B120F-3, World
Precision Instruments), and showed resistances between 1 -2 Mil Oocytes were clamped
to -50 mV and stepped from -90 mV to +70 mV in 10 mV increments. The wTater
injected eggs served as the control in evaluating any activity exerted by endogenous
proteins found in the Xenopus oocytes. Several different solutions were used to
determine the exchangers functional activities (refer to table 2-2). The transporter
blockers, 4,4-diisothiocyanodihydrostilbene-2,2-disulfonate (DIDS, Calbiochem, La
Jolla, California) and niflumic acid ( Sigma-Aldrich) were used to inhibit the transporter
capabilities of the expressed AE1 protein.

35
Aedes degenerate CA primers:
CA5F:
5'
GAR
CAR
TTY
CAY
TKY
CAY
TGG
GG
CA3R:
5'
GTI
ARI
SWN
CCY
TCR
TA
N=G,A,T,C; K=G,T; S=G,C; VI
/=A,T;
Y=C,
T; R=,
A,G
Amplified cDNA adaptor primers:
DAP:
5'
CGA
CGT
GGA
CTA
TCC
ATG
AAC
GCA
TRsa:
5'
CGC
AGT
CGG
TAC
TTT
TTT
TTT
TTT
T
Anopheles exai
ct
ZA prim
lers:
Ag1CA2F:
5'
CAG
TCA
CCT
ATC
GAC
CTA
AC
Ag1CA4R:
5'
CTC
GCG
TGT
TCA
ATG
GTT
G
Ag4CA11F:
5'
GGA
GGC
GTC
CTT
GGC
AAC
Ag4CA12R:
5'
CTG
CAC
TGA
CCG
GAA
GTT
G
Anopheles exa(
:ti
\E prim
ers:
AgAEIF:
5
CCT
GGA
AGG
AAA
CGG
CAC
G
AgAE4R:
5'
CCT
CGA
GCT
GGT
GCA
GAT
C
Aedes CA Real
tin
fie PCR
prime
rs:
5SPCAF1:
5
GCA
ACA
CTG
CTT
CCG
TCT
ACA
A
5SPCAR1:
5'
CCG
GTT
CGT
TAA
TAA
CTC
CAT
TG
18s RIBF:
5'
CGC
TAC
TAC
CGA
TGG
ATT
ATT
TAG
TG
18s RIBR:
5'
GTC
AAC
TTC
AGC
GAT
TCA
AAT
GTA
A
Aedes CA expn
ISl
lion pri
mers:
ExCAshortF:
5'
CACC
ATG
GAC
GAA
TGG
CAC
T
ExCAshortR:
5'
TTA
GTA
ATC
CAT
ATC
GGT
GTG
GT
Anopheles CA
ex
>ressioi
n prim
ers:
ExCA4F:
5'
CACC
ATG
GCA
TCA
AAA
ACA
ACA
AAG
CA4end:
5'
TTA
CAG
CTT
CGA
AAG
CAC
AAC
GG
Table 2-1. PCR primer sequences.

36
Salt for 98mM value
mW
#1
#2
#3
#4
98N
98k
98N-CI
98K-CL
mM
p/1
p/I
mM
mM
mM
g/l
S/I
Solution:
lx
4x
lx
4x
lx
4x
lx
4x
NaCl
58.44
n
5.73
22.91
2
0.12
0.47
0
0
0
0
KC1
74.55
2
0.15
0.60
98
7.31
29.22
0
0
0
0
Na Gluconate
218.1
0
0
0
0
98
21.37
85.50
2
0.44
1.74
K Gluconate
234.2
0
0
0
0
2
0.47
1.87
98
22.95
91.81
Choline Cl
139.6
0
0
0
0
0
0
0
0
MgS047H20 (120.36)
246.5
0
0
0
0
0.5
0.12
0.49
0.5
0.12
0.49
MgCl26H20(59.7)
203
0.5
0.10
0.41
0.5
0.10
0.41
0
0
0
0
CaCl22H20 (110.98)
147.02
0.5
0.07
0.29
0.5
0.07
0.29
0
0
0
0
Ca Gluconate
430.38
0
0
0
0
0.5
0.22
0.86
0.5
0.22
0.86
HEPES (free base)
238.3
10
2.38
9.53
10
2.38
9.53
10
2.38
9.53
10
2.38
9.53
EGTA for InsideOout
380.4
0
0
0
0
0
0
0
0
pH
| 7.2 (4M NaOH)
7.2 (4M KOH)
7.2 (4M NaOH)
7.2 (4M KOH)
Table 2-2. Composition of all solutions used in Xenopus oocyte expression of An.
gambiae AE. Total molarity and pH were kept constant in all solutions.
Expression profiles were recorded in high sodium (#1), high sodium minus
chloride (#3), high potassium (#2), and high potassium minus chloride (#4).

37
Aedes CA Primer Linearization
BrCA equations:
y = -3.2655k 39.958
R2 = 0.9776
18s equations:
y -3.1312x + 27.781
R2 = 0.9931
BrCA primers GC
18s primers WG
Figure 2-1. Efficiency plots for real-time PCR primers. Serially diluted cDNA samples
were tested with each primer set to determine the efficiency of
amplification. A linear regression was performed to determine the slope and
intercept for each primer set. These values were then used in an algorithm
to compare cDNA concentrations within the samples.

38
Figure 2-2. Three-dimensional (Cn3D) depiction of human CA IV (1ZNC). The green
barrel represents an alpha helix structure, the tan arrows represent beta
sheets, and the colored strings represent extended loop structures. The
yellow coloring represents the accessible, extended loop peptide region
against which the homologous Ae. aegypti CA antibody was raised.

CHAPTER 3
CARBONIC ANHYDRASE IN THE MIDGUT OF LARVAL AEDES AEGYPTI:
CLONING, LOCALIZATION, AND INHIBITION1
Introduction
Bicarbonate (and ultimately carbonate) ions are produced in vivo primarily by the
enzymatic action of carbonic anhydrase (CA). Its activity contributes to the transfer and
accumulation of FT1' or HCO3' in bacteria, plants, vertebrates and invertebrates. Although
there are innumerable reports related to the isolation of CA from vertebrates, studies
involving CA from invertebrates are very rare and there are no reports of the isolation of
CA from adult or larval mosquitoes.
There is strong immunohistochemical (Zhuang et al., 1999) and physiological
(Clark et al., 1999; Boudko et al., 2001b) evidence that an electrogenic, basal H+ V-
ATPase energizes luminal alkalinization in the anterior midgut of the larval mosquito by
producing a net extrusion of protons out of the lumen and a hyperpolarization of the basal
membrane. In contrast, ET V-ATPase appears to be localized in the apical membrane of
the posterior midgut and gastric caeca providing a reversed FT*"- pumping capacity relative
to the anterior midgut (Zhuang et al., 1999). A system capable of generating a high
luminal pH is likely to be buffered by carbonate (CO32), which has a pKa of
approximately 10.5.
This chapter was slightly modified and reprinted with permission from The Company of
Biologists LTD. Corena, M. P., Sern, T. J., Lehman, H. K., Ochrietor, J. D., Kohn,
A., Tu, C. and Linser, P. J. (2002). Carbonic anhydrase in the midgut of larval Aedes
aegypti: cloning, localization and inhibition. J. Exp. Biol. 205, 591-602.
39

40
The purpose of this study was to determine the presence and location of CA in the
midgut of larval Ae. aegypti and to clone and characterize the enzyme. To investigate the
role of CA in the alkalization of the larval midgut, the effects of CA inhibitors were
tested. Here, we report the cloning and localization of the first CA from mosquito larvae
and, in particular, from the midgut epithelium of larval Ae. aegypti. A cDNA clone
isolated from fourth-instar Ae. aegypti midgut (termed A-CA) revealed sequence
homology to the a-carbonic anhydrases (Hewett-Emmett, 2000). Histochemistry and in
situ hybridization showed that the enzyme appears to be localized throughout the midgut,
although preferentially in the gastric caeca and posterior regions. In addition, classic
carbonic anhydrase inhibitors such as acetazolamide and methazolamide inhibit the
mosquito enzyme in the midgut.
Results
Bromothymol Blue Qualitative Assay
This assay allowed the identification of samples of solubilized midgut tissue
containing CA activity by spotting them onto a filter paper soaked in a basic buffered
solution containing a pH indicator, bromothymol blue (BTB). As stated previously, BTB
changes color from yellow (at pH<7.6) to blue when the pH increases above this value.
The principle behind the assay is based on the fact that CA catalyzes the conversion of
CO2 into bicarbonate with the concomitant release of protons (Donaldson and Quinn,
1974). The presence of protons lowers the pH in those regions of the paper where the
spotted samples contain the enzyme. As the pH falls below 7.6, these spots rapidly
change color from blue to yellow. This assay is not effective for samples in acidic
solution, and the tissue homogenization must be accomplished in alkaline buffer. The

41
enzymatic reaction takes only a few seconds, and it can be delayed if the solutions, the
paper and the samples are kept cold on ice. However, a few seconds is usually sufficient
to discriminate the samples that contain CA from those lacking enzymatic activity. The
assay must be performed quickly since, after approximately one minute the entire filter
paper turns yellow, probably as a result of the uncatalyzed hydration of carbon dioxide
absorbed by the solution at this basic pH.
The test has proved useful in determining the presence of small amounts of CA in
homogenates of mosquito larvae. The assay was also used to detect CA activity
qualitatively, in fractions obtained from affinity chromatography (Osborne and Tashian,
1975) of larval homogenates. The affinity chromatographic procedure, which employs a
bound CA inhibitor (p-aminomethyl benzyl sulfonamide (p-AMBS); Sigma), produced
two peaks of CA activity upon exposure to the standard elution buffers. The amount of
protein that we were able to produce by this technique was, however, very small and
resisted several efforts at direct microsequencing. This change in color was inhibited by
acetazolamide and methazolamide when these inhibitors (105 M) were added to the
samples prior to spotting on the dye-impregnated filter papers. Inhibition of the reaction
resulted in blue spots that did not change color upon addition of CO2. The positive
control containing commercial CA turned yellow when carbon dioxide was added, and
this color change was also inhibited by acetazolamide and methazolamide. This finding
confirmed that the yellow color of the spots was due to the action of CA and that the
mosquito larva contains active CA.

42
Carbonic Anhydrase Activity and Alkalization
A classic CA inhibitor methazolamide, was tested in live fourth instar larvae to
examine the influence of CA on the maintenance of the pH extremes inside the midgut,
and the effect of the enzyme on the net alkalinization of the growth medium by the intact
animals. Previous investigations have shown that living mosquito larvae excrete
bicarbonate, which results in the net alkalization of their surrounding aqueous medium
(Stobbart, 1971). Equal numbers of living larvae of equivalent age and size were placed
in culture plate wells containing lightly buffered medium and the pH indicator BTB. The
tissue culture plates used in this assay were scanned before and after addition of various
concentrations of methazolamide. In the absence of methazolamide, the blue color of the
medium, indicating a pH of at least 7.6, was maintained (Stobbart, 1971). Actual
measurement of the pH in each well showed a slow increase over time (data not shown).
Upon addition of methazolamide, the culture medium slowly became acidic, with a
resulting change in color to yellow as the pH dropped below 7.6 (Figure 3-1). All of the
controls that did not contain methazolamide remained blue. Addition of methazolamide,
at various concentrations, to wells containing only medium with BTB (no mosquito larva
control) remained blue. These data show that CA activity is present in the living larvae
and that it plays some role in acid/base excretion.
Moreover, fourth instar larvae cultured in BTB-containing medium ingest the dye,
which can then be used as a visible indicator of the pH in the gut lumen. Treatment of
the cultured larvae with methazolamide showed a direct impact of inhibited CA activity
on gut luminal pH. Figure 3-2 compares the luminal pH of dissected larval midguts with
and without a 5 hour exposure to methazolamide. The micrographs reveal that

43
alkalinization of the midgut was inhibited by methazolamide as shown by the color
change of the BTB indicator. Interestingly, the effect was most pronounced in the
anterior midgut, where the pH indicator changed from blue in the midgut of larvae reared
in the absence of inhibitor to yellow in as little as 30 minutes when methazolamide (10"6
M) was added to the culture. The indicator also changed color progressively from blue
through green to yellow in the gastric caeca (Figure 3-2). No apparent change was
observed in the posterior midgut. The color of the midgut in this region was yellow both
in the untreated larvae and in the larvae treated with methazolamide. Since the pH of the
posterior midgut has been associated with values close to 7.6, no change in color was
evident using this qualitative method.
lsO Isotope-Exchange Experiments
The relative activity of CA, normalized to total protein content, was calculated as
described by Silverman and Tu (1986). The relative activity of CA was highest in the
gastric caeca, followed by the posterior midgut and Malpighian tubules (Figure 3-3). The
relative activity of CA in the anterior midgut was either extremely low or non-existent,
falling at or below that of the buffer blank. The specificity of the reaction was confirmed
by complete inhibition with the addition of 1 O'6 M methazolamide (results not shown).
Cloning of Carbonic Anhydrase from Aedes Aegypti Larvae
We utilized a cDNA cloning strategy to obtain a specific carbonic anhydrase
cDNA from the midgut epithelial cells of the larval Ae. aegypti. A comparison of twelve
CA sequences, including two putative CA sequences that had been annotated but not
characterized in the Drosophila melanogaster databases, was made. We then produced
degenerate PCR primers from consensus regions of the CA gene family. The initial 297

44
bp partial sequence was used to derive exact PCR primers for a modified 3'- and 5'-
RACE (Frohman and Zhang, 1997, modified by Matz, 1999). Amplified cDNA pools
from each region of the isolated gut, facilitated the eventual cloning of a single
contiguous cDNA (Matz, 1999). The final contiguous region spanned both start and stop
codons, and encoded a polypeptide of 298 residues (GenBank accession number
AF395662). Figure 3-4A shows an alignment of the Ae. aegypti carbonic anhydrase (A-
CA) amino acid sequence with several other, previously characterized members of this
extensive a gene family. Figure 3-4B shows a homology tree depicting the percentage of
identical amino acids between sequences, generated using DNAman software. Figure 3-
5A shows the alignment between A-CA and six putative CA gene sequences from the D.
melanogaster genome that our homology search (BLAST) revealed. Four of the D.
melanogaster genes (AAF54494, AAF56666, AAF57140, AAF57141) had not
previously been annotated. Figure 3-5B shows the homology tree generated with these
sequences. A-CA has a putative molecular mass of 32.7 kDa. The translated A-CA
protein sequence possesses a characteristic eukaryotic-type CA signature sequence within
the polypeptide (amino acid residues 99-115; Femley, 1988).
To examine the possibility of regionalized expression of the A-CA, PCR using
exact primers was performed on amplified cDNA pools from the various sections of the
gut. Figure 3-6 shows an ethidium-bromide-stained agarose gel. PCR products of the
expected, 894 nucleotide length, are readily seen in the gastric caeca and the posterior
midgut regions. Anal papillae (not shown), anterior midgut, Malpighian tubules and
rectal salt gland showed little or no PCR product. When the PCR products were
subjected to a second round of PCR using the same primers, an appropriately sized

45
product was also discernible in the anterior midgut. This PCR analysis also revealed
higher molecular mass products in the anterior midgut and Malpighian tubules that may
represent additional carbonic anhydrases specific to larval Ae. aegypti (Figure 3-6). This
result is shown only to display the gut regions in which the A-CA clone was derived.
The lack of an 894 bp product in the other gut regions may simply be due to poor quality
cDNA pools from those regions. However, the cloning of A-CA from both the gastric
caeca and posterior midgut regions is consistent with the location of enzyme activities
described above.
Localization of the Enzyme in the Midgut Epithelium: Carbonic Anhydrase Enzyme
Histochemistry
To further analyze the regional and cellular expression of CA in the midgut
epithelium of larval mosquitoes, a modified Hanssons histochemical reaction was
performed on whole mount preparations of the gut (Hansson, 1967). Figure 3-7
summarizes the results of this analysis. Carbonic anhydrase activity was detected in a
non-uniform pattern along the length of the gut. The most intense staining was evident in
the gastric caeca and the posterior midgut. Staining was less intense in the anterior
midgut. At higher magnification, it was obvious that cellular heterogeneity with regard
to CA activity also exists. This is particularly evident in the posterior midgut, where very
large and regularly spaced cells appear nearly white on a background of dark CA reaction
product. The larger cells have been characterized as columnar or ion-transporting cells
(Volkman and Peters, 1989b). Surrounding these large cells are more numerous smaller
cells termed cuboidal or resorbing/secreting cells (Zhuang et ah, 1999). The CA
histochemical stain clearly distinguishes these cells from one another and indicates that
the large columnar cells contain relatively veiy little CA in comparison with the smaller

46
cuboidal cells. In addition, the distal cells of each lobe of the gastric caeca, termed Cap
cells, show little or no histochemical staining, suggesting further cellular heterogeneity
with respect to CA distribution in the gut (Figure 3-7).
In Situ Hybridization
To further characterize the localization of A-CA expression, in situ hybridization
was performed using a portion (approximately 300 bp) of the central coding region of the
cDNA. Figure 3-8 shows typical results of this type of analysis. A strong hybridization
signal was evident in the gastric caeca and the posterior midgut. Lower levels of
hybridization were evident in other gut regions. As with the CA histochemical stain,
higher magnification revealed that the relatively small cuboidal cells exhibit more intense
labeling than do the large columnar cells (Figure 3-8B).
Discussion
The search for the enzyme in the midgut of the larval mosquito was triggered by
the observations of a pH value around 11 in the anterior midgut lumen and a high
bicarbonate concentration (Zhuang et al., 1999; Boudko et al., 2001b). The presence of
CA in the midgut of the larval mosquito has been suggested before by investigations of
the epithelium of larval lepidopteran midgut. Carbonic anhydrase has been studied in
Manduca sexta, where the enzyme has been associated with the fat body, midgut and
integumentary epithelium (Jungreis et al., 1981). The enzyme has also been localized in
the goblet cells of the epithelium of Hyalophora cecropia using Hanssons histochemical
stain. The same procedure showed that the columnar cells were devoid of activity
(Turbeck and Foder, 1970).

47
Even though a number of genes and their products have been isolated from the
midgut of Ae. aegypti, and the role of CA in the alkalization of the midgut has been
suggested (Turbeck and Foder, 1970; Haskell et al., 1965; Ridgway and Moffett, 1986;
Boudko et al., 2001b), there have been no reports of the isolation or cloning of CA or of
the localization of the enzyme within the midgut of larval mosquitoes. This is the first
recorded cloning of a CA from a mosquito, and is also the first to be cloned from any
arthropod. Our results show that at least one (and perhaps more) CA is present in the
midgut of larval Ae. aegypti. The CA of larval Ae. aegypti (A-CA) is inhibited by
classical carbonic anhydrase inhibitors such as methazolamide and acetazolamide.
Methazolamide has the most potent effect on A-CA. Direct physiological measurements
of ion fluxes from living larval mosquito midgut epithelial cells also show
methazolamide to be a very potent inhibitor of ion movements and balance (Boudko et
al., 2001a).
To investigate the distribution of CA in the midgut of the larval mosquito, we
employed both in situ hybridization and enzyme histochemistry. Our results indicate that
enzymatic activity is greatest in the gastric caeca and the posterior midgut, as
demonstrated by the intense staining obtained using Hansson's method and by in situ
hybridization using cRNA probes. Measurements of activity using the l80 exchange
method in pools of dissected regions of the gut corroborate these findings. In addition,
the enzyme seems to be preferentially associated with the small cuboidal cells in the
midgut epithelium, as determined both by enzyme histochemistry and by in situ
hybridization.

48
As reviewed in Clements (1992), two major cell types have been defined in the
gastric caeca by inferring functional states from cytological findings. These two major
cell types have been called ion-transporting cells and resorbing/secreting cells (Volkman
and Peters, 1989a,b) and they correspond to the columnar and cuboidal cells mentioned
above with the ion-transporting cells being equivalent to the columnar cells and the
resorbing/secreting cells being the cuboidal cells (Zhuang et al., 1999). Neither of these
cell types, as characterized in the larval mosquito gut, parallels the structurally unique
qualities of the lepidopteran goblet cell. Nonetheless, our results indicate that, as in
lepidopterans, CA activity is preferentially associated with one of two distinct cell types
whose functional complementation must produce the alkalization and ionic balances
regulated by the gut. These results are consistent with the observations of lepidopteran
midgut by Turbeck and Foder (1970). In the larval lepidopteran midgut, two
morphologically distinct cell types have been long recognized: goblet cells and columnar
cells. Goblet cells posses both the proton-pumping V-ATPase and CA activity (Harvey,
1992; Ridgway and Moffet, 1986; Wieczorek et al., 1999). One of the enigmas of using
the pioneering analyses of insect model systems such as M. sexta to produce testable
hypotheses for gut alkalinization in mosquito larvae has been the apparent absence of
goblet cells from mosquitoes. Previous investigations have inferred different functional
cell types in the larval mosquito gut epithelium. We are currently developing antibody
probes for A-CA. Immunocytochemical analyses of A-CA distribution in comparison
with other key components of gut function, such as V-ATPase (Zhuang et al., 1999),
should provide new insights into the cell biology of this intriguing epithelial system.

49
It is interesting to note that the lowest concentration of CA in the midgut
epithelium occurs in the region that surrounds and probably regulates the region of
highest luminal pH, the anterior midgut. The pKa of CO3'2 is approximately 10.5 and,
hence, this anion is likely to be the primary buffer of the pH 10.5-11 gut contents within
the anterior midgut. Our results therefore suggest that the major buffering anion in this
area of the midgut is probably not produced by local CA but instead either upstream, in
the gastric caeca, or downstream, in the posterior midgut, where CA levels are very high.
This result, and results presented elsewhere (Boudko et al., 2001a), are consistent with a
model in which a major function of the anterior midgut is to pump protons out of this
region of the gut lumen, promoting the conversion of HCO3 to CO32. A comprehensive
model of the regulation of ion homeostasis and gut alkalization in the larval mosquito
awaits the characterization and localization of other major components of the system in
addition to CA. It will also be very important to resolve the question of whether multiple
CAs are expressed in the midgut and how each is distributed in this dynamic tissue.
Quantitative evidence corroborating the distribution of CA within the midgut and
10
supporting the histochemical and in situ observations was obtained using the O-
exchange mass spectrometric method. The results obtained with this method indicate that
the gastric caeca exhibit the highest level of carbonic anhydrase, relative to total protein
content, followed by the posterior midgut and the Malpighian tubules. The anterior
midgut showed levels of activity so low that two possibilities could be considered: either
the method could not detect the enzyme or it is absent from the anterior midgut. The
presence of faint staining using the histochemical and in situ methods suggests that the

50
levels of activity in the anterior midgut might be too low to be detected using the 0
method, but that the enzyme is present throughout the entire length of the midgut.
In summary, our evidence demonstrates the existence of CA in Ae. aegypti larvae
and it also suggests that the gastric caeca and posterior midgut exhibit the highest levels
of CA activity. In addition, the enzyme seems to be associated with the small cuboidal
cells of the midgut epithelium. Furthermore, enzyme activity has also been detected in
membrane preparations isolated from whole midguts and could be due to the presence of
more than one isoenzyme. Carbonic anhydrase activity has previously been
demonstrated in the epithelium of the larval midgut of six species of lepidopterans, in
which it has been associated with the particulate fractions of the homogenate (Turbeck
and Foder, 1970). This is consistent with our hypothesis that there might be more than
one CA and that one of these enzymes may be associated with the plasma membrane.
What is the role of CA in the alkalization mechanism? BTB proved useful in
monitoring the impact of CA inhibition on the maintenance of gut luminal pH and the
excretion of acid/base. As mentioned earlier, Ae. aegypti larvae typically alkalize the
medium in which they are reared by secreting bicarbonate ions (Stobbart, 1971). The
ingestion of CA inhibitors altered the metabolism of the larvae to the point that the
metabolic products secreted into the medium change the pH of the environment, shifting
it towards more acidic values than those observed in the absence of inhibitors. The
lowering of the pH of the medium might be related to a decrease in the rate of secretion
of HCO3'. The effect of the ingestion of CA inhibitors on the secretion of bicarbonate
into the medium remains to be explored. However, as indicated by measurements with
ion-selective microelectrodes, inhibition of CA in the midgut has an extreme effect on the

51
maintenance of an alkaline pH within the midgut lumen (Boudko et al., 2001a). It is
plausible that a decrease in the rate of secretion of bicarbonate is elicited by inhibiting the
CA enzyme.
A simple model of bicarbonate transport fails to explain how the high pH is
achieved within the anterior midgut of the larval mosquito. At a pH of approximately 11,
similar to that observed within the anterior midgut, the majority of bicarbonate is present
as carbonate. In fact, measurements of lepidopteran midgut fluid have shown that it
contains 37 mM carbonate and 17 mM bicarbonate (Turbeck and Foder, 1970). Since the
pH of a 0.1 M solution of sodium bicarbonate is only approximately 8.3, secretion of
bicarbonate alone cannot be responsible for the high pH observed in the anterior midgut
(Dow, 1984). It could, however, explain the pH values at the gastric caeca and posterior
midgut. The mechanism for maintenance of an alkaline pH within the anterior midgut
must be more complex than just a simple buffering of a physiological solution with
bicarbonate. Although this mechanism has been investigated (Wieczorek et al., 1999;
Zhuang et al., 1999; Boudko et al., 2001a), its details remain unclear. However, the
evidence suggests that a basal, electrogenic FT V-ATPase energizes luminal alkalization
in the midgut of larval mosquitoes (Zhuang et al., 1999; Boudko et al., 2001b). Although
the electrogenic transport of K+ drives the pH gradient, there must also be flux of one or
more weak anions in the opposite direction to maintain homeostasis. Several transporters
are thought to participate in this mechanism.
Another line of evidence suggests that the levels of carbon dioxide in the
hemolymph of lepidopterans are lower than those within the midgut lumen. The
concentration of CCb has been determined to be near 5 mM in the hemolymph and 50

52
mM in the midgut lumen in larval Hyalophora cecropia (Turbeck and Foder, 1970).
Recent measurements using capillary zone electrophoresis of larval Ae. aegypti fluids
have revealed a bicarbonate/carbonate level as high as 50.84.21 mM in the midgut
lumen compared with 3.962.89 mM in the hemolymph (Boudko et al., 2001a). These
values correlate with those observed by Turbeck and Foder (1970). This combined
evidence suggests that the CO2 that reaches the midgut lumen in the larvae of
lepidopterans is rapidly converted to a mixture of bicarbonate and carbonate. The role of
CA in the alkalization process would be of great significance. The generation of
antibodies against A-CA will facilitate a detailed analysis of the cellular and subcellular
distribution of this key enzyme in this system.

53
Figure 3-1. Effect of CA inhibition on culture medium pH with fourth-instar Ae.
aegypti larvae. Mosquito larvae typically alkalize the medium in which
they are reared (Stobbart, 1971). (A) Six culture wells each containing five
fourth-instar larvae incubated for 5 hours in medium containing 0.003%
Bromothymol blue (BTB). The blue color is retained, indicating a pH
greater than 7.6. (B) The same as A, except that each well also contains a
different concentration of the CA inhibitor methazolamide ranging from
10"6 to 10'3 M from left to right. A yellow color indicates a pH below 7.6.

54
Figure 3-2. Effect of methazolamide on the alkalization of the midgut using
Bromothymol Blue (BTB) assay of pH within living, but isolated, gut
tissue. Gut tubes were dissected after pre-loading with BTB and then
incubated for 5 hours in hemolymph substitute (Clark et al., 1999)
in the absence (A) or presence (B) of 1 O'6 M methazolamide. The loss of
blue coloration in B shows that the internal pH of the gut lumen has
dropped below 7.6. Scale bar represents 300 pm.

55
12
GC AM PM MT
Figure 3-3. Relative activity of CA in different pooled segments of the midgut of larval
Ae. aegypti. Midguts were dissected from early fourth-instar larvae and
separated into gastric caeca (GC), anterior midgut (AM), posterior midgut
(PM) and Malpighian tubules (MT). The relative activity of CA was
measured using the l80 mass spectrometry method (Silverman and Tu,
1986), normalized to total protein content. The activity of the anterior
midgut was lower than that of the water blank and, thus, is set as zero
activity.

56
A
Aselas aagypti
HoraaCAl-P00917
Zbra£ih-Q92051
HuunCX3-P07451
MouasCAl 4 -NP035927
C.alaganfl-Tl6575
RatGAIV-NPO62047
MAH
46
2 8
2 8
28
45
24
50
Aftdos aagypti
Hora*CAl-P00917
2abrafih-Q92051
HimanCA3 P0 7 4 51
MOU8OCA14-NP035927
C.ttl*gana-T16575
RatCAIV-NP062047
94
77
77
77
92
74
<>8
Ascias Agypti
HoraaCAl-P00917
Zibrafxah-Q92051
HuoanCA3-P07451
MouaCA14-HP035927
C.lganB-T16B75
RatCAIV-HPO62047
137
125
125
125
142
121
146
Aadss aagypti
HorsaCAl -POOS17
Zbra£iah-Q92051
HuBanCA3-P0745l
MouseCAl4 -HP035927
C.alagana-Tl6575
RatCAIV-NP062047
186
175
174
174
192
162
195
Aadas aagypti.
HoraCAI-POO 917
ZabrafiahQ92051
HuaanCA3-P07451
MouaaCAl4-NP035927
C.alagana-Tl6575
RatCAJV-NP062047
236
222
221
221
240
210
245
Aadas aagypti
HoreaCAl-P00917
Zebrafieh-Q92051
HuunCA3-P07451
MouaaCA14-NP035927
C.alagana-Tl6575
RatCAIV-MP062047
282
260
260
260
290
246
294
Aadaa aagypti
HoraaCAl-P00917
Zbrmfiah>Q920Sl
HuaanCA3-P07451
MouaoCA14-NP035927
C.alogans-T1657B
RatGATV-KP062047
L3LTLIVAAIAXLLAK.
LGLGVGILAGCLCLLIAVY FI AQKI R1 LLVPTLTCLVASFLH
298
260
260
260
337
246
30 9
B
Homology (%)
Figure 3-4. Carbonic anhydrase from the midgut of larval Ae. aegypti. (A) Alignment
(BLAST) of the predicted amino acid sequence of Ae. aegypti cDNA with
several known a-carbonic anhydrases. Regions of exact homology across
all species are highlighted in blue (100%); regions with less homology are
highlighted in red (>75%) and green (>50%). (B) A homology tree
comparing A-CA and several other a-CAs (DNAman software).

57
Ajadae aagypti MIALFVATPSTIKAD EWHY PT PAPNGVINE PERWGGQCE TS
Oros AAF57140
Droa AXP57141 .
Droa AAF56666 MSEIATGKSCT1AVBSHVFGY8EPNQRRWARHHGHCAGKTQS PIAITTfR
Droa AAF544 94 MPLRHSVGIQ8VKLMIIANEWGYPDLDNNQDEPFPK WGGLCDM
Droa AAF4 99418 .MRRCRMTPFAIVIAPILICABLVLAQDFGYEGRHGPEHW8EDYARC8G
Droa AAF44817 MSHHWGYTEENGPAHWAKEYPQAlBG
43
O
O
49
43
48
25
Ajeadas sagypti
Daros AAF57140
Daros AAF57141
Droa AAF56666
Daros AAjr54494
Daros AAJP4 994 8
Droa AAF44817
RRQMIDLTYOAKVKGDFAPFLtSlKMNPIRHAQLTNTGH . 8IQID8T
MRQLIVPLPRIVFGHYDVKLRGPLTLLIING . . HTETAN
. .1 TTAltaffAVDMIGYHNLLPYPLKMINHGHTVSITIPKVN VTEVGE
KKOPXKI-HVKGlU^KGEFnAI*K3rE|Jnn3EHQKNI-RMVNWGH . SIQLSGF
KHQlPINIDQVSAVEKKFPKLErFlIFKWPDNLQMTNIIGHTVLVKMSYNE
HRQ#PVDITP3SAKKGSE1*NVAPUCWKYVPEHTK31jVHPG . YCHRVDVN
9G
37
43
94
90
90
73
Aedeo aegypti
Daros AAF57140
Daros AAF57141
Droa AAF56666
Daros AXP54494
Daros AAP4 9948
Daros AAF44817
130
83
89
140
130
148
120
Aadtts aegypti
Droa AAF57140
Droa AAF57141
Daros AXF56666
Daros AAF54494
Daros AXF4 9948
Daros AAF44817
ICWLFHV8NQ. DNTHMDWLET8QD
JONP .NRIFPGLSKVMSALP
rAIVRKDNAKSTPtSRLMEAW
IFFFNLDE . DEGAGtVTINRHtH
. TPNEAIQSIIKSLGA.
GD . K8TG6YEGFTNLK8
....HHAEtDKVTSLLQ
177
130
138
180
177
194
166
Aedaa a9ypti
Droa AAF57140
Daros AAP57141
Daros AAF56666
Daros AAF54494
Daros AAF49948
Daros AAF44817
IRDAACKSAWJCCKUIPHNPLP KNRTS1
RVTKYNAKTI PGGLM^QHLGNVNpRDF
BVPIEDSNAjpVFGQSltDQLIGGVSHRDF
LIADANQEAtLNVTFK, S 8L lAGVDVD*
VKSTOSMN1CPVLVADSIAVX)DLVP8|IEI
QIDRKGKSVMMTNPLP1/3EYIS 1
FVLHKGDRVfLPQGOJPGQLLP. .1
226
180
188
236
227
243
214
Asiaa a*gyp ti
Daros AXF57140
Droa AAF57141
Droa AAF56666
Dros AXF54494
Dros AAF49948
Dros AAF44817
275
229
235
285
276
291
264
Aedaa a5YPti
Dros AAFS7140
Dros AAF57141
Dros AAF56666
Dros AAF54494
Dros AAF49948
Dros AAF44817
G8GAIPKLSLTLIVAAIAALIAK
GAYLGK
TSIATLKHEGEYLKYDWFY
SG3AGLTASVSLGI>1TLIIAGQKFLL
HNMGS X PLVDAEHAAGKWRAQAAAVLLPLWLAAL3RTSIPRGF
REIGGH
298
235
235
304
302
335
270
B
Homology (%)
100 80 60 40 20
I I I I I
Figure 3-5. Comparison of the extrapolated amino acid sequences of A-CA with six
putative dipteran CA genes identified in the D. melanogaster gene
databases. (A) An alignment of A-CA with the amino acid sequences of
the six D. melanogaster genes (accession numbers listed) identified
through bioinformatics searching. Regions of exact homology across all
species are highlighted in blue (100%); regions with less homology are
highlighted in red (>75%) and green (>50%). (B) A homology tree
comparison of these seven Dipteran CAs.

58
Figure 3-6. Polymerase chain reaction (PCR) analysis of Ae. aegypti amplified cDNA
from different gut regions. PCR was performed using exact primers for the
cloned A-CA. Anterior midgut (lane 2), gastric caeca (lane 3), posterior
midgut (lane 4). whole gut RNA control (lane 5), Malpighian tubules (lane
6) and a water template control (lane 7) are shown. Note the primary
product in gastric caeca and posterior midgut samples at the expected size
of approximately 894 nucleotides. Also note the absence of this band from
other gut regions but the appearance of bands of higher molecular masses.
Lane 1 is a 100 bp molecular mass ladder (Promega).

59
Figure 3-7. Hanssons histochemistry of whole mount Ae. aegypti gut. A. Intense dark
staining is observed in the cardia, gastric caeca (GC) and posterior midgut
(PMG), indicating CA activity. B. Higher magnification of the gastric
caeca. The distal lobes of the gastric caeca (Cap cells) exhibit relatively
low levels of reaction product, indicating lower levels of enzyme activity in
these cells relative to other cells of the gastric caeca. C. Higher
magnification of the PMG shows large, relatively unstained columnar cells
(*) contrasted with the smaller stained cuboidal cells (arrow). Scale bars
represent 150 pm in A and B, and 75 pm in C.

60
B.
Figure 3-8. Localization of CA mRNA expression in larval Ae. aegypti. A. An isolated
whole mount gut probed with DIG-labeled cRNA for A-CA. Abundant
hybridization is observed in the cardia. gastric caeca (GC), and posterior
midgut (PMG). B. The smaller cuboidal cells (arrow) display stronger
hybridization than the larger columnar cells (*). C. Isolated midgut reacted
with the sense (control) cRNA for A-CA. Scale bars represent 300 pm in
A, 75 pm in B and C.

CHAPTER 4
A GPI-LINKED CARBONIC ANHYDRASE EXPRESSED IN THE
LARVAL MOSQUITO MIDGUT
Introduction
The CA enzyme expressed in the midgut of larval mosquitoes shares some
characteristics with the mammalian CA IV isozyme, including a glycosyl-phosphatidyl-
inositol (GPI) link to the plasma membrane. Mammalian CA IV enzymes have been
found in dynamic organs such as kidney, lung, gut, brain, eye, and capillary endothelium
(Chegwidden and Carter, 2000). The human CA IV isoform was found to be as active as
the CA II isoform in carbon dioxide hydration and even more active in bicarbonate
dehydration (Baird et al., 1997). Studies of larval mosquito CAs are being pursued to
better understand the alkaline gut system. As the anterior midgut of the larval mosquito
lacks a highly active cytosolic CA II-like isozyme (previous chapter; Corena et al., 2002),
the presence of a highly active CA IV-like isozyme within the mosquito gut may be able
to provide the buffering capacity that is needed within the highly alkaline anterior
midgut. A more detailed characterization of larval Aedes aegypti CA is presented in this
study as well as the sequence of a homologous CA isoform from Anopheles gambiae.
New tools and techniques, such as the generation of a mosquito-specific CA antibody and
real time PCR, as well as improved methodology for in situ hybridization, have enabled
this further analysis.
61

62
Results
Bioinformatics of Aedes Aegypti CA
We have previously cloned a CA cDNA from the Ae. aegypti midgut (accession
number AF395662; Corena et al., 2002). Our initial structure prediction indicated that
the protein was cytosolic. However, further characterization has indicated that this CA is
actually membrane associated via a GPI-link. We have determined that the CA
propeptide sequence encodes an extracellular protein with a hydrophobic tail region. The
first 17 amino acids of the propeptide are predicted by the Simple Modular Architecture
Research Tool (SMART) program to be the signal sequence (Letunic et al., 2002). This
sequence flags the message for transport to the endoplasmic reticulum (ER). Using the
PSORT II server, the prediction of membrane topology (MTOP) indicates that the Ae.
aegypti CA sequence is GPI anchored. Amino acid G-276, is predicted by the GPI
prediction server (Eisenhaber et al., 1999) to be the site for GPI attachment. The
hydrophobic tail (L278-A289) allows translocation of the transcript through the ER
plasma membrane and is also predicted to stabilize the protein with the membrane until
the pre-formed GPI anchor is transferred to the protein. The hydrophobic tail is then
cleaved to produce a completely extracellular protein that is tethered to the cell by the
GPI link (for a review of GPI-linked proteins see Brown and Waneck, 1992).
Sequence Comparisons of CA IV-like Isoforms
I also cloned a CA IV-like cDNA from the gut of An. gambiae, an important
vector in the spread of malaria. This CA isoform (Ensembl gene ID:
ENSANGG00000018824, chromosome 2L) is partially predicted by the Ensembl CA
protein family (ENSF00000000228) as 1 of the 14 gene members found in the An.

63
gambiae genome. These cloned mosquito cDNAs from Ae. aegypti and An. gambiae are
61% identical to each other in amino acid residues and show similarities to the
mammalian CAIV isozyme. However, in contrast to the mammalian CA IV, which is
encoded by 7 exons (Sly and Hu, 1995), only 3 exons make up the An. gambiae CA
isoform. Alignment of the mosquito CA IV-like isoforms from Ae. aegypti and An.
gambiae with various mammalian CA IV isozymes reveals conserved features within this
CA isoform (Fig. 4-1). For example, the multiple leucine (L) residues within the amino
terminus of the mammalian CA IV propeptides that comprise the signal sequence are
found in the Ae. aegypti and An. gambiae CA IV-like isoforms. One important feature of
the mosquito CA IV-like sequences is the conserved alignment of G-69 (human CA IV
numbering) with the human, bovine, and rabbit CA IV sequences. This particular amino
acid residue has been changed to glutamine (Q) in rat and mouse CA IV, which results in
reduced enzyme activity (Tamai et al., 1996b). Additionally, all of the CA IV sequences,
including the mosquito isoforms, display a hydrophobic tail region. In addition to the
conserved CA IV-like features of GPI-linked proteins, there are also conserved cysteine
residues (C28 and C211, human CA IV numbering) between all of these CAs (Fig. 4-1).
It has been determined via cysteine labeling, proteolytic cleavage and sequencing that
these two cysteine residues, in the human CA IV, form a disulfide bond (Waheed et ah,
1996). A second disulfide bond is present in the mammalian CA IVs between residues
C6 and Cl 8 (human CA IV numbering; Waheed et ah, 1996). This second pair of
cysteine residues, and hence the resultant disulfide bond, is not present in either of the
mosquito isoforms.

64
Although the mosquito CA isoforms display features similar to mammalian CA
TVs, such as a 5 signal sequence, a hydrophobic 3 tail, and extracellular GPI expression,
there is one striking difference in the amino acid composition of the mosquito CAs'
active sites. The active site within all of the 14 characterized mammalian CA isoforms is
tightly conserved. Three histidine residues (His-94, His-96, and His-119) are essential
for CA activity through their coordinated binding of a required zinc molecule. The
absence of one or more of these histidine residues results in inactive proteins called CA-
related proteins (CARPs), as found in mammalian CA isoforms VIII, X, and XI (Tashian
et al., 2000). The mosquito CA IV-like isoforms contain all three of the required
histidine residues along with all of the other 13 highly conserved residues found in most
other CAs (refer to Fig 4-1; Tashian, 1992; Sly and Hu, 1995; Tamai et al., 1996a).
However, as the alignment shows in Figure 4-1, there is a conserved gap within the active
site of the mosquito CAs that is not present in any of the mammalian active sites.
Because this shortened active site was found in mosquitoes but was not found in any
mammalian CA isoform, I searched the Drosophila melanogaster genome for potential
CA homologs. The D. melanogaster genome was found to contain 14 putative CA genes
(ENSF00000000228), the same number found in An. gambiae. One out of the fourteen
D. melanogaster CA isoforms was found to contain the identical number of deleted
amino acids as the mosquito forms within the active site region. This D. melanogaster
CA sequence (accession number CG3940-PA) may also be a GPI-linked iso form due to
the presence of a lysine-rich 5 signal sequence and hydrophobic tail region. Figure 4-2
shows an alignment of the three Dipteran CAs with shortened active site regions and all

65
of the human CAs, which show several additional amino acids within the conserved
active site region.
Localization of CA IV-like Isoform in the Mosquito Midgut
In situ hybridization analyses indicate that the Ae. aegypti CA message is
expressed most heavily within the epithelial cells of the gastric caeca and posterior
midgut (Fig. 4-3). An antisense cRNA probe corresponding to the entire cDNA sequence
generated strong cytoplasmic staining of the proximal gastric caeca, while the distal Cap
cells showed no detectable hybridization (Fig. 4-3B). Rostral to the gastric caeca, a
strong localization was evident in a small subset of cardia cells that encircle the tissue,
forming a collar (Fig 4-3B). These collar cells are clearly different from the
surrounding cells in this same area. This technique also highlighted a set of specific
epithelial cells that are found only in a subset of the posterior midgut. These CA-positive
cells form a ring of about 5 cells in width that circumscribe the lower-posterior gut region
(Fig. 4-3C). The CA message was also localized to longitudinal and circular muscle
fibers of the anterior and posterior midgut (Fig. 4-4). Following the longitudinal muscle
fibers, in close association, are distinct nerve fibers that also show strong CA mRNA
expression (Fig. 4-4). Epithelial cells of the anterior midgut were clearly void of signal
beneath the labeled muscle and nerve cells. Specific staining was also evident however
within the abdominal ganglia central nervous system (CNS) and peripheral nerve tissue
(Fig. 4-5). No labeling was seen in the Malpighian tubules.
Real Time PCR Analysis of Aedes aegypti CA IV-like Transcripts
Real time PCR was used to compare the levels of Ae. aegypti CA mRNA within
specific tissue regions of the larvae. The guts of 20 fourth instar Ae. aegypti larvae were

66
dissected and the head, gastric caeca (GC), anterior midgut (AMG), posterior midgut
(PMG), and Malpighian tubules (MT) were pooled. RNA was isolated from each tissue
sample for subsequent real time PCR analysis. Ae. aegypti ribosomal RNA (Genbank
accession number M95126) was used to normalize the quantity of transcript from each
sample. The results are presented in graph format in figure 4-6. Gastric caeca contain
the greatest quantity of CA message within the gut sections (Fig. 4-6). The head tissue
contained roughly half as much message as the gastric caeca (Fig. 4-6). The localization
of CA rV-like message within the larval head is consistent with the localization of CA
message to CNS tissue by in situ hybridization. The anterior midgut, posterior midgut,
and Malpighian tubule collections showed CA message only marginally greater than zero
(Fig. 4-6).
Immunolocalization of CA IV-like Protein in the Mosquito Gut
The amino terminal peptide sequence (GVINEPERWGGQCETGRR) was chosen
from the Ae. aegypti CA sequence as an antigen for antibody production. The resultant
antiserum was used to immunolocalize the CA IV-like isoform within the mosquito gut.
The pre-immune serum was used as a control for all experiments. Immunoreactivity was
found within the gastric caeca region of the gut as well as on muscle fibers along the
anterior midgut (Fig. 4-7A). A subset of anterior muscle fibers displays the strongest and
most striking labeling on their extracellular surface, while other muscle fibers show little
or none. Immunoreactivity was also found within the CNS ganglia and immunoreactive
nerve fibers that traverse the gut (Fig. 4-8). There was no immunoreactivity detected in
the Malpighian tubules.

67
Antibody Cross-Reactivity with Other Mosquito Species
The 18 amino acid sequence from the Ae. aegypti CA, used to elicit the antibody
response, shares 14 identical residues with the homologous An. gambiae CA protein
(refer to Fig. 4-1), and therefore, the antiserum recognizes the CA IV-like isoform present
in An. gambiae as well. The immunoreactivity within the An. gambiae gut displays a
strikingly similar, yet species-distinctive pattern of CA IV-like protein expression (Fig. 4-
7B). Similar to Ae. aegypti (Fig. 4-7A), not all of the muscle fibers were localized in the
An. gambiae gut. However, in An. gambiae the immunolabeled muscle fiber network
runs down the lateral sides of the midgut, while the dorsal and ventral muscle fibers are
not immunoreactive (compare Fig. 4-7A to 4-7B).
The high sequence conservation between the chosen antigenic peptide from Ae.
aegypti and the An. gambiae CA, prompted us to check other mosquito species for
immunoreactivity. The other species tested also displayed the same strong labeling on a
subset of anterior gut muscle fibers that included both circular and longitudinal muscle
fibers. Figures 4-9 and 4-10 display the immunoreactive results obtained from Aedes
albopictus and are representative of the results from the other mosquito species including
Ae. aegypti and An. gambiae. The CA IV-like immunolabeling is clearly confined to
only a subset of actin-containing muscle fibers (Fig. 4-9D). Labeled phalloidin, which
binds to actin, labeled all of the muscle fibers within the gut, while the antibody for CA
IV-like mosquito CA only recognized a subset of the anterior muscle fibers. The CA
antibody is also specific for the extracellular plasma membrane of these muscle cells,
which is clearly shown by cross-sectional analysis (Fig. 4-10). Thus, there appear to be
two different sets of muscle fibers within the anterior mosquito midgut is an intriguing

68
discovery. This immunoreactive subset of CA-containing muscle fibers traverses the
cells that surround the highly alkaline anterior gut lumen. Determining the role of these
CA-specific muscle fibers holds promise for deciphering the necessary CA component of
mosquito gut alkalization.
Phospholipase C Treatment
In order to validate the CA IV-like isoform cloned from Ae. aegypti is indeed GPI
linked to the membrane, live fourth instarle, aegypti larvae were subjected to
phosphoinositol-specific phospholipase C (PI-PLC) treatment and subsequent
immunohistochemistry. This compound is capable of breaking the GPI-anchor and
therefore severs GPI-linked proteins from the plasma membrane. Larvae subjected to PI-
PLC treatment showed a decrease in CA immunoreactivity along the midgut muscle
fibers, as compared to the non PI-PLC treated controls (Fig. 4-11). This evidence
supports the prediction that the mosquito CA IV-like isoform is in fact GPI-linked to the
outer plasma membrane.
Discussion
In this study, we show that two GPI-linked CAs are expressed in the midguts of
two different mosquito species that rely on an alkaline digestive strategy. These
mosquito CAs share characteristics with the mammalian CA IV isozyme, including the
GPI link to the membrane. In situ hybridization localized CA message predominantly to
the gastric caeca and posterior midgut epithelial cells, as well as muscle and nerve fibers
along the anterior midgut, and CNS ventral ganglia. Real time PCR analyses confirmed
the presence of CA message within the Ae. aegypti gut and CNS. The gastric caeca were
found to contain the greatest amount of CA message in relation to the other gut samples

69
while the head sample contained roughly half of the gastric caeca concentration. CA
immunoreactivity was most striking on specific muscle fibers of the anterior midgut,
along with labeling of the gastric caeca and CNS ganglia. The localization of CA protein
in a distinct subset of muscle fibers was found in several different mosquito species. This
finding suggests that many mosquito species express a similar CA IV-like protein as well
as confirms the immunoreactivity by only a subset of muscles. Immunolocalization of
the CA IV-like isozyme within the mosquito gut and CNS also demonstrates that the CA
message is being translated into protein and agrees with the localization of CA mRNA in
the gastric caeca and muscle fibers. Interestingly, not all of the muscle fibers that show
CA mRNA expression also show CA protein. Therefore, only a fraction of the muscle
fibers that contain the CA mRNA are translating the message into protein. This may
represent a form of regulation, in which the muscle fibers not expressing the CA protein
could be turned on to translate the CA message if needed. This ability would be very
advantageous if indeed this particular CA isoform is involved in buffering the alkaline
gut.
The posterior midgut was also found by in situ hybridization to express the CA
mRNA. However, both the real time PCR analysis of CA mRNA expression and the
immunolocalization of CA protein failed to determine the presence of this particular CA
within the posterior midgut. One explanation for this may be the inability of our CA
antibody to permeate the posterior midgut tissue. However, the real time analysis, which
is very specific for a region of mRNA, also did not find this particular CA message in
that region. The most likely cause for the in situ hybridization showing a positive CA
result in the posterior midgut, is the existence of a very similar CA isoform within that

70
region. Evidence for this is supported by genome data showing 14 different CA
isoforms, all with regions of high nucleotide identity. The posterior midgut region does
display CA activity, but apparently not as a result of the GPI-linked CA isoform
presented here. The specific isoform or number of CAs contributing to the activity of the
posterior midgut is still unknown.
We have previously shown that the application of CA-specific inhibitors
dramatically decreases the alkaline gut pH, and in fact is lethal to the larval mosquitoes
(Corena et ah, 2002). We now present evidence that a CA found in the mosquito gut is
most similar to the mammalian CA IV isozyme but contains a novel active site motif
unlike any of the mammalian CA IV isoforms (Fig. 4-1). The finding of a novel CA
active site within the mosquito may facilitate the construction of a mosquito-specific CA
inhibitor for use in larval mosquito control. We are hopeful that the ongoing mosquito
CA crystallization project will yield further significant structural differences from the
mammalian CA IV structure. These differences may be useful in the design of a
mosquito-specific CA inhibitor.
Out of the 14 mammalian CAs identified thus far as cytosolic, membrane-bound,
secreted, and mitochondrial, only CA IV has a GPI link to the cell membrane. The
localization of this highly active mammalian isozyme to dynamic tissues such as the gut,
brain, kidney, and lung supports the important catalyst role of CA. It should not be
surprising that the gut of a mosquito, a highly alkaline and fluctuating system, has been
found to contain a presumably active CA IV-like isoform as well. The single amino acid
substitution of glycine-63 to glutamine is unique to rodents (rat and mouse) CA IV, and
was found to be responsible for their reduced activity rate of only 10-20% of the human

71
CAIV enzyme (Tamai et al., 1996b). Mutating glutamine-63 to glycine within the
rodent sequence resulted in almost three times greater CA activity (Tamai et al., 1996b).
Unlike the rodent sequences, both of the mosquito CA IV-like sequences display the high
activity glycine residue adjacent to histidine-69 (Human CA IV numbering, refer to Fig.
4-1).
The task ahead is to decipher if a GPI-linked CA is better equipped to function in
a highly dynamic system than other CA isoforms. Perhaps the GPI link affords the
mosquito CA enzyme a characteristic advantage in buffering such an alkaline pH through
its exclusively extracellular expression. Residing at the plasma membrane intrinsically
affords this isozyme the best location for monitoring CO2 and HCO3' flux. Indeed,
mammalian CA IVs are expressed on membrane surfaces where large fluxes of CO2 and
/or HCO3' are expected (Sly, 2000). The most compelling ability of GPI-linked proteins
is that they are known to elicit second messengers for signal transduction (Brown and
Waneck, 1992). The alkaline pH of the larval mosquito gut was found to drop within two
to three minutes after being narcotized or just simply handled (Dadd, 1975). This
handling effect lends itself to our prediction that larval mosquitoes may exert neuronal
control over the generation of the gut lumens pH. Since a GPI-linked CA was localized
within the mosquito gut and CNS tissue we propose that a GPI-linked CA may regulate
the pH of the mosquito gut by severing the GPI-link and starting a signal cascade.
Further studies are being pursued within the mosquito gut to encompass the
localization and characteristics of other CA isoforms as well as bicarbonate exchangers.

72
Aedes CA
Anoph CA
Human CA IV
Bovine CA TV
Rabbit CA IV
Murine CA TV
Rat CA IV
Aedes CA
Anoph CA
Human CA XV
Bovine CA IV
Rabbit CA IV
Murine CA IV
Rat CA IV
Aedes CA
Anoph CA
Human CA XV
Bovine CA IV
Rabbit CA IV
Murine CA XV
Rat CAIV
Aedes CA
Anoph CA
Human CA XV
Bovine CA IV
Rabbit CA IV
Murine CA XV
Rat CA IV
Aedes CA
Anoph CA
Human CA IV
Bovine CA IV
Rabbit CA IV
Murine CA XV
Rat CA IV
Aedes CA
Anoph CA
Human CA IV
Bovine CA IV
Rabbit CA XV
Murine CA IV
Rat CA IV
MIALFVATLL PSTIRADE1
MKSFTLLLCYALFVLHAARGDE1
MRMLLALLALSAARPSASAESHW
MRLLLALLVLAAAPPQAF&ASHW
MQLLFALLALGALRPLAG EELHW
MQLLLALLALAYVAPST-EDSGWi
MQLLLALLALAYVAPST-JEDSHWi
* 1
Antigenic peptide
PTPAPNfGVINEPERWGGQCETGRRSPIDLTY
¡PTPG- -TNGVMSEPERWGGQCDNGRRC£ PIDLTI
QAESSNYPCLVPVKWGGNCQKD-RC S PINIVT
IQVKPSNYTCLEPDEWEGSCQNN-RC S PVNIVT
IQA--SNYSCLGPDKWQEDCQKS-RC S PINIVT
IQTKDPRSSCLGPEKWPGACKEN-QQ S PINIVT
IQAKEPNSHCSGPEQWTGDCKKN-QQsjpiNIVT
k k k kirk t
QAAVKGDFAPFLF-SNYMNPIRNAQLTHTGHSIQIDSTDPSVTLYGGGLPGKFVLDttiHF
AAAVRGQFAPLFF-SNYMLPLKQPRVTNTGHSIQINNRDSAITMQGGGLGGRFVLD QMHF
TKAKVDKKLGRFFFSGYDK-KQTWTVQjNNGHSVMMLLEN-KASISGGGLPAPYQAKQ LHL
AKTQLDPNLGRFSFSGYNM-KHQWWQNNGHTVMVLLEN-KPSIAGGGLSTRYQAT Q LHL
TKAEVDHSLGRFHFSGYDQ-REARLVEN NGHSVMVSLGD-EISISGGGLPARYRAlj QLHL
ARTKVNPRLTPFILVGYDQ-KQQWPIK NNQHTVEMTLGG-GACIIGGDLPARYEAVQLHL
SKTKLNPSLTPFTFVGYDQ-KKKWEVKHNQHSVEMSLGE-DIYIFGGDLPTQYKAigLHL
HWG Sfi
HWG Si
HWSDLPYKGS
HWSRAMDRGS
HWSQELDRGS
HWSNGNDNGS
HHSEESNKGS
kk kkk
TIAGVRYGQBLHMVHHDSRYNS--LTEAGAVKNAVAVIGVLFHVSNQD
TLDDTRYGi]e LHLVHHDTRYAS--LEDAVQARNGVAVLGVLFHVGSQP
sldgehfaMe mhivhekekgtsrnvkeaqdpedeiavlaflveagtqv
SFDGERFAME3MHIVHEKEKGLSGNASQNQFAEDEIAVLAFMVEDG-SK
IVHQKETGTSGNEVQD--SDDSIAVLAFLVEAGPTM
SIDGRHFAMElMHIVHKKLTS SKEDSKDKFAVLAFMIEVGDKV
S IDGKHFAiMHWHKKMTTGDKVQDSDSKDKXAVLAFMVEVGNEV
CAESVIW
lACAESVlW
CDEB
CDEB
lACSEI
CDETVIW
iBTVljw
** ****** * *
NTHMDWLETSQDIRDAAGKSAPLK-GKLSPHNPLPKNRTSYFPSEGS
NMHIDTILDTATEIQNEVGKEALLR-GKLSPYNLLPSNRTSFYRyegs:
NEGFQPLVEALSNIPKPEMSTTMAE-SSLLDLLPKEEKLRHYFPiLGS:
NVNFQPLVEALSDIPRPNMNTTMKEGVSLFDLLPEEESLRHYFF V LGS:
NEGFQPLVTALSAISIPGTNTTMAP-SSLWDLLPAEEELRHYFFYMGS:
nkgfqplvealpsiskphststvre-sslqdmlppstkmytyfryngs:
NEGFQPLVEALSRLSKPBTNSTVSE-SCLQDMLPEKKKLSAYFpjYQGS:
*
TVFTESLPVSLDQVELFKT IHDPSGHELVIKfKSLQPLNARALVYHTDMDYSGs}gAI
TVFTESISVSLEQVERFKA ihdqtgrelvn netr SVQPLNTRALVYATEWDQHGNNFA
TVFREPIQLHREQILAFSQKLYYDKEQTVSMKDNVR PLQQLGQRTVIKSjGAPGRPLPHAL
TVFQKPIQLHRDQILAFSQKLFYDDQQKVNMTDN vjRPVQSLGQRQVFRS GAPGLLLAQPL
TVFQEPIRLHRDQILEFSSKLYYDQERKMNMKD N V R PLQRLGDRSVFKS QAAGQLLPLPL
TVYKQPIKIHKNQFLEFSKNLYYDEDQKLNMKDKV R PLQPLGKRQVFKSHAPGQLLSLPL
tvfeepikihkdqflefskklyydqeqklnmkdmvbplqplgnrqvfrsHASGRLLSLPL
. * .**.**.*.
PKLSLTLIVAAIAALLAK
TKMTSNWFLGAIVLLVITSRLSYH
PALLGPMLACLLAGFLR
PTLLAPVLACLTVGFLR
PTLLVPTLACVMAGLLR
PTLLVPTLTCLVANFLQ
PTLLVPTLTCLVASFLH
Figure 4-1. Alignment of several mammalian CA IV enzymes with two mosquito CA
isoforms. The leucine-rich signal sequences are found in all aligned
isoforms, along with the 3 essential zinc-binding histidines (red), and
cysteine residues (green) that form disulfide bonds. The reduced activity in
rodent CA IVs is caused by the glycine-69 mutation to glutamine (orange;
Tamai et al., 1996b), which the mosquito CAs do not display. Important
conserved residues are boxed. The position of mammalian signal sequence
cleavage is shown and the following amino acid is residue #1 in the
functional protein. Antigenic peptide sequence is also displayed.

73
AAL72625 Aedes CA
AAQ21365 Anoph CA
CG3940-PA Dros CA
P00915 Human CA I
P00918 Human CA II
P07451 Human CA III
P22748 Human CA IV
AAB47048 Human CA V
CAC42429 Human CA VI
P43166 Human CA VII
JN0576 Human CA VIII
AAH14950 Human CA IX
Q9NS85 Human CA X
AAH02662 Human CA XI
AAH23981 Human CA XII
BAA85002 Human CA XIV
FVLDQMHFHWG SEHTIAGVRYGQELHMVHHDS
FVLDQMHFHWG SEHTLDDTRYGLELHLVHHDT
FWEQIHMHWW SEHTINDIRYPLEVHIVHRNT
YRLFQFHFHWGSTNEHGSEHTVDGVKYSAELHVAHWNS
YRLIQFHFHWG--SLDGQGSEHTVDKKKYAAELHLVHWNT
YRLRQFHLHWGS SDDHG SEHTVDGVKYAAELHLVHWNP
YQAKQLHLHWSDLPYKGSEHSLDGEHFAMEMHIVHEKE
YRLKQFHFHWG--AVNEGGSEHTVDGHAYPAELHLVHWNS
YIAQQMHFHWGGASSEISGSEHTVDGIRHVIEIHIVHYNS
YRLKQFHFHWG--KKHDVGSEHTVDGKSFPSELHLVHWNA
FELYEVRFHWGRENQRGSEHTVNFKAFPMELHLIHWNS
YRALQLHLHWG--AAGRPGSEHTVEGHRFPAEIHWHLST
HRLEEIRLHFGSEDSQGSEHLLNGQAFSGEVQLIHYNH
HRLSELRLLFG--ARDGAGSEHQINHQGFSAEVQLIHFNQ
YSATQLHLHWG-NPNDPHGSEHTVSGQHFAAELHIVHYNS
YVAAQLHLHWG-QKGSPGGSEHQINSEATFAELHIVHYDS
*** *... *
Figure 4-2. Clustal alignment of CA protein sequences. All characterized human CA
isoforms are presented along with putative GPI-linked isoforms from Ae.
aegypti, An. gambiae, and D. melanogaster. Three histidine residues that are
required for the essential binding of zinc are shaded in blue. Note that 1 or
more of these histidine residues are missing from the inactive human CA-
related proteins VIH, X, and XI while all three histidines are present within
the Dipteran sequences. The three CAs from Dipterans contain a shortened
active site region (marked by red dashes) when compared to any of the human
or other mammalian CA sequences. This difference may provide a potential
target for mosquito-specific CA inhibitors, for use as larvacides.

74
Figure 4-3. Localization of CA mRNA in a wholemount preparation of early 4th instar
Ae. aegypti. A. The wholemount gut preparation localizes CA message to
specific cells of the gastric caeca (GC) and posterior midgut (PMG). B. A
subset of cardia (arrows) and gastric caeca cells display the CA message.
The distal lobes of the caeca, called Cap cells, display no staining (*). C.
There is a distinctive labeling pattern of CA message within a specific band
of posterior epithelial cells. In addition, numerous trachea (arrows) are
heavily labeled along the length of the midgut. Scale bar represents 300 pm
in A, 150 pm in B, 75 pm in C.

75
Figure 4-4. Expression of CA mRNA in Ae. aegypti anterior midgut. While the cardia,
gastric caeca and posterior midgut display heavy epithelial hybridization,
there is also specific CA mRNA expression seen in muscle and nerve cells.
A. A representative whole mount Ae. aegypti larvae displaying the strong
CA expression in epithelia, along with muscle fiber staining that can be
overlooked at low magnification. B. The beginning of the anterior midgut
shows hybridization to both muscle (arrowheads) and nerve fibers (arrows).
The labeled fibers reveal striated muscle running longitudinally down the
length of the anterior gut and circularly around the girth of the gut. C. The
anterior midgut (AMG) displayed strong hybridization in muscle
(arrowheads) and nerve fibers (arrows) while displaying no epithelial cell
labeling. The posterior midgut (PMG) shows intense fiber labeling as well
as epithelial cell labeling (*). The scale bar represents 300 pm in A, 25 pm
in B, 50 pm in C.

76
Figure 4-5. Localization of CA IV-like message within Ae. aegypti CNS tissue. A. In
situ hybridization localized the CA IV-like mRNA within all ventral ganglia
CNS clusters (arrows) as well as hair sensory cells (*) and longitudinal
nerve fibers (arrowheads). B. The sense control probe displayed no specific
hybridization. Scale bar represents 300 pm.

77
Aedes aegypti Carbonic Anhydrase
Ae. aegypti Tissue Sections
Figure 4-6. Relative quantification of CA IV-like message in Ae. aegypti larvae using
real time PCR. The gastric caeca tissue displays the greatest amount of CA
IV-like message. Data was normalized to the gastric caeca (GC) sample. The
anterior and posterior midgut along with the Malpighian tubules display very
little CA message. The head section displays roughly half the amount of
message found in the gastric caeca.

78
B
Figure 4-7. Ae. aegypti and An. gambiae CA protein labeling. The antibody generated
against the Ae. aegypti CA can also be used to localize the homologous CA
isoform within An. gambiae. The larvae were incubated with phalloidin
(red) and the CA-specific antiserum (green). Colocalization of the red and
green signals appears yellow. A. The antibody localization shows the
strongest labeling in Ae. aegypti for a subset of muscle fibers in the anterior
midgut and the proximal portions of the gastric caeca. B. Antibody
localization of An. gambiae CA is depicted by the yellow muscle fibers,
while the red muscle fibers are not recognized by the antibody. The scale
bar represents 300 pm in A, 150 pm in B.

79
Fig 4-8. The Ae. aegypti CNS ganglia express the CA IV-like isoform. A. Pre-immune
serum does not show any detectable labeling of the CNS tissue. B. Strong
immunolabeling for the mosquito CA IV-like isoform is displayed in the
ventral ganglion clusters, as displayed by the fluorescent green coloring as
compared to the yellow control (pre-immune) ganglia. The scale bars
represent 100 pm.

80
Figure 4-9. Immunolocalization of mosquito CA IV-like enzyme in Aedes albopictus.
Muscle and nerve fibers within the anterior midgut region are heavily labeled.
A. Selective labeling of particular muscle fibers (green). B. Labeled
phalloidin (red) was used to localize actin and labels all muscle fibers,
including those that were not recognized by the antibody against mosquito
CA. C. A nuclear label (blue) was used to distinguish cell numbers present.
D. Overlay of all three signals. Colocalization of the green and red signals
appear yellow. The CA antibody recognizes only a subset of anterior muscle
fibers, and is seen in several different mosquito species. The scale bar
represents 50 pm.

81
Figure 4-10. High magnification of immunoreactive muscle fibers within the Aedes
albopictus midgut. A. Labeling of muscle fibers appears to be
extracellular. B. Cross section of the same fiber demonstrates that the
localization pattern is confined to the extracellular plasma membrane of
the midgut muscle fibers. The scale bars represent 25 pm.

82
Figure 4-11. Immunoreactivity of Ae. aegypti guts for the CA IV-like isozyme. The red
labeling is specific for muscle fibers. The green labeling shows localization
of the mosquito CA IV-like protein. The yellow labeling shows co
localization of the mosquito CA IV-like isoform and actin. Prior to
immunolabeling, the guts were treated with PI-PLC to determine if the CA
IV-like isoforms are GPI-linked to the cell membrane. A. Immunolabeling
of the gastric caeca, without the PI-PLC treatment, displays heavy yellow
labeling of the GPI-linked CA isoform. B. After PI-PLC treatment there is
no CA IV-like immunolabeling of the gastric caeca. C. The anterior midgut
displays the immunolocalization of the CA IV-like isoform along a subset of
muscle fibers. D. After PI-PLC treatment the yellow immunolabeling for
the GPI-linked mosquito CA is greatly reduced. The decreased
immunolocalization within the gastric caeca and anterior midgut signifies
that the PI-PLC was successful in severing the GPI-link and releasing the
CA from the membrane association. The scale bars represent 100 pm.

CHAPTER 5
ANION EXCHANGER EXPRESSED WITHIN
THE LARVAL ANOPHELES GAMBIAE MOSQUITO
Introduction
The larval mosquito gut provides an ideal model for studying epithelial transport
due to its cellular simplicity, being only one cell layer thick. The transport of bicarbonate
within the larval mosquito gut was prompted by several studies (Boudko et al., 2001a,b;
Corena et al., 2002). Bicarbonate is the main pH buffer in most complex organisms
(Sterling and Casey, 2002) so it was predicted that de-protonated bicarbonate (ie.
carbonate) is necessary for attaining the highly alkaline lumen of the larval mosquito
midgut. However, the anterior midgut was found to apparently lack an active CA
enzyme (Corena et al., 2002). An alternative to bicarbonate being rapidly produced
within the anterior midgut, is the transport of bicarbonate into the anterior midgut. This
alternative was supported by a previous study which implicated a chloride/ bicarbonate
anion exchanger within the larval mosquito gut through the use of self-referencing ion-
selective (SERIS) microelectrodes (Boudko et al., 2001b). However, the molecular
identity of the chloride/ bicarbonate anion exchanger (AE) was not determined. We now
present the first anion exchanger (AE) to be cloned and characterized from the An.
gambiae mosquito, in an attempt to unravel the physiology of an extremely alkaline
digestive system.
Localizing an AE within the larval mosquito gut is important because intracellular
pH is known to be regulated by exchangers of bicarbonate and chloride (Phillips and
83

84
Baltz, 1999). AE localization will also distinguish whether the AE is co-expressed within
the same epithelial cells as the H+ V-ATPase and if the polarity of basal or apical
expression is also the same. The localization of a V-ATPase within the larval
mosquito gut was found to be apical in the gastric caeca, basal in the AMG, and then
apical again in the PMG (Zhuang et al., 1999).
The highly dynamic system of alkaline digestion in the larval mosquito gut does
not exist in any known mammalian system. However, mammalian organs such as the
kidney are able to perform many parallel functions of the mosquito gut, such as water
regulation, filtration, and ionic homeostasis. The CA, AE, and H+ V-ATPase proteins, in
particular, have been extensively studied and localized within the mammalian kidney due
to their dynamic roles in acid-base balance (Huber et al., 1999; Schwartz, 2002). The co
localization and polar expression of an AE with a Hf V-ATPase will define the epithelial
cells of the mosquito gut as resembling the mammalian kidney A-intercalated cell type,
the B-intercalated cell type, or the non-A non-B intercalated cell type (types as defined
by Brown and Breton, 1996 and Kim et al., 1999).
Results
%
An. gambiae AE Sequence Analysis
The full length An. gambiae AE1 (AgAEl) cDNA was cloned from midgut tissue
and contains 3309 bases (accession number AY280611) with a molecular weight of 123
kDa for the predicted protein. The NCBI conserved domain search tool (CDD)
determined that the protein sequence was part of a family of bicarbonate transporters
(BT) and cotransporters (PF00955) as well as sodium-independent chloride/bicarbonate
exchangers and related sodium/bicarbonate cotransporters (KOG1172; Geer et al., 2002).

85
This complex of BTs contains both the solute carrier 4A (SLC4) and solute carrier 26A
(SLC26A) proteins. More specifically, the EnsEMBL database places this particular AE
cDNA sequence within the anion exchange/ band3 protein family (ENSF00000000189)
as 1 of the 8 putative anion exchange band3 transcripts (ENSANGP00000010112)
encoded in the An. gambiae genome. These 8 transcripts arise from 3 different genes
(ENSANGG00000007623, ENSANGG00000004501, and ENSANG00000012483). The
gene that gives rise to the cloned AE that we are presenting (ENSANGG00000007623) is
located on chromosome 3R. The other 2 genes (ENSANGG00000004501 and
ENSANG00000012483) are located on chromosome 2L (Hubbard et ah, 2002; Clamp et
al 2003).
The 1102 amino acids comprising the An. gambiae AE form a cytosolic
framework at the amino terminus while the carboxy terminus is composed of 12
transmembrane spanning domains also with an intracellular cytosolic terminus (hmmtop
v.2; Tusnady and Simon 1998; Tusnady and Simon 2001). This hmmtop prediction was
generated based on two assumptions: 1), that the CA binding site is within the carboxy
terminus; and 2), that the C-terminus is intracellular, as is found for all known AEs. This
structure is consistent with the predicted structure of the Drosophila melanogaster
sodium dependent anion exchanger (NDAE1), which consists of a 12 membrane-
spanning pattern with intracellular carboxy and amino termini (Romero et al., 2000). The
highly conserved sequence identity of the An. gambiae AE1 with respect to the D.
melanogaster NDAE1 allows the predictive 12 transmembrane-spanning domains of the
D. melanogaster protein to be superimposed upon the An. gambiae protein (Fig. 5-1).

86
The amino terminus of the AgAEl protein contains 523 cytoplasmic residues
enriched with multiple binding sites for cytoskeletal proteins (Bairoch et al., 1997). The
carboxy terminus contains the membrane-spanning domains that are responsible for ion
transport. According to the PROSITE motif search and the hmmtop server, sites of
potential post-translational modification include 2 cAMP and cGMP-dependent protein
kinase phosphorylation sites, 16 protein kinase C phosphorylation sites, 11 casein kinase
II phosphorylation sites, 16 N-myristoylation sites, 1 prokaryotic membrane lipoprotein
lipid attachment site, and 1 leucine zipper pattern (Bairoch et al., 1997; Gupta et al.,
2002). The last 82 amino acids of the An. gambiae AE carboxy terminus are predicted to
project into the cytosol, the correct orientation that is expected for binding of a cytosolic
CA. The CAII binding site comprises a hydrophobic amino acid residue followed by at
least two acidic residues within the next four residues (Vince and Reithmeier, 2000).
Figure 5-2 displays the CA II binding sites found in several AE proteins along with the
putative CA II binding site (LDDIM) in the An. gambiae AE.
BT Sequence Comparisons
Following the sequence prediction that this An. gambiae BT is an AE, the closest
characterized protein sequence is the NDAE1 (accession number AAF98636) from D.
melanogaster. The An. gambiae AE1 shares 72% identity with NDAE1 (Fig. 5-3).
However, the greatest similarity is with an uncharacterized splice form of NDAE1
(AAF52497) that contains an inserted sequence that is also found in the An. gambiae AE
sequence (Fig. 5-4). Other BTs such as the sodium bicarbonate cotransporters (NBCs)
and AE4 show 45%-52% identity to AgAEl, AE1-3 exhibits 36% identity, and the BTs

87
that are also capable of transporting sulfate (SLC26A group) show only 11% amino acid
identity (Fig. 5-3).
Anion exchangers, specifically AE2 and AE3, were determined to be pH
sensitive. AE3 is stimulated by intracellular alkalization whereas AE1 is not. More
specifically, a region of amino acids (WRETARWIKFEE) within the carboxy terminus is
responsible for the pH sensitivity seen in AE2 (Vince et al., 2000). The An. gambiae AE
sequence in this same region contains 14 of the 16 residues found within AE2 while the
other two amino acids are conserved (Fig. 5-5). AE1 shares only 8 of the 16 amino acids
in this region.
Localization of Anion Exchanger mRNA in An. gambiae Larvae
A DIG-labeled antisense cRNA probe comprising the full length AE cDNA was
employed to localize the AgAEl mRNA. A DIG-labeled sense probe was used as a
control. The AgAEl mRNA was found in every region of the larval gut including gastric
caeca, anterior midgut, posterior midgut, Malpighian tubules, and rectum (Fig. 5-6). In
the gastric caeca and posterior midgut regions, the probe was localized to epithelial cells.
Within the gastric caeca the labeling is most intense in the area where the lobes face the
lumen. The gastric caeca labeling was confined to the proximal cells, whereas the Cap
cells displayed no label (Fig. 5-6A,B). The rectum displayed staining in a small subset of
epithelial cells along with tracheoles (Fig. 5-6D).
Muscle, nerve, and trachea cells that traverse the outer plasma membrane of the
anterior gut epithelial cells were labeled with the AE antisense probe (Fig. 5-7). Labeled
tracheal fibers are displayed in close association with the gastric caeca (Fig. 5-8). These

88
trachea extend from the gastric caecal region and become incorporated with the anterior
midgut where they are intimately associated with nerve fibers (Fig. 5-9).
The third abdominal segment marks the beginning of the posterior midgut and is
located by the fourth pair of trachea that connect to this part of the midgut. This junction
displays a marked contrast in AgAEl mRNA transcript expression. Strong staining of
the epithelial cells begins here, coinciding with the beginning of the posterior midgut and
change in epithelial cell morphology (Fig. 5-10). Along with small epithelial cell
labeling, tracheal fibers also display strong label for AgAEl mRNA within the posterior
midgut. The larger type of epithelial cell within the posterior midgut, the columnar cell,
shows extensive labeling for AgAEl mRNA expression, with signal localized near the
plasma membranes (Fig. 5-1 OB). The extreme end of the posterior midgut displays
strong labeling within a cluster of small epithelial cells known as cuboidal cells (Fig. 5-
11). Cellular processes that extend rostral and lateral from these cells are also labeled.
All cells of the Malpighian tubules label positively for AgAEl transcript expression (Fig.
5-12).
AgAEl mRNA expression is also localized to the ventral midgut ganglion. Each
ventral ganglion displays specific labeling within one or two longitudinally directed
neurons that traverse the same plane (Fig. 5-13). No other neuronal cells display signal
for AE mRNA expression. The labeling is very specific for precisely one neuronal
pathway within each ganglion (Fig. 5-13). The sense (control) DIG probes display no
hybridization (Fig. 5-14).

89
Antibody Localization of AE Protein
Two antigenic peptide sequences (EVRKRPPEKNPKEEIDEE and
KPKQQPVTTISVTKVAEQ) were chosen from the cytosolic framework of the amino
terminus and the anion exchange carboxy terminus respectively, of the translated AE
cDNA (accession number AAQ21364). Chickens were used to produce antibodies
against these antigenic peptides. The resultant chicken antisera were used to localize the
AgAEl protein within whole mounts of the An. gambiae fourth instar larvae. The
AgAEl protein was localized to the plasma membranes of the gastric caeca (Fig. 5-15)
and posterior midgut epithelial cells (Fig. 5-16). A three-dimensional reconstruction of
the cellular localization of AgAEl enabled us to discriminate the immunolabeled
basolateral membranes from the non-labeled apical membranes. Antisera for both
peptides displayed the same basolateral immunolocalization pattern. Neuronal cells
within the AMG displayed immunoreactivity (Fig. 5-17). Pre-immune sera displayed no
specific immunoreactivity.
AE Functional Expression in Oocytes
In order to ascertain the functional characteristics of the AgAEl, the protein was
expressed inXenopus oocytes. The AgAEl was subcloned into the pXOOM vector
(Jespersen et al., 2002; generous gift from Dr. T. Jespersen) for oocyte expression. The
Xenopus oocytes were injected with either AgAEl RNA or water to serve as the control.
Three to seven days post-injection the oocytes were tested for AgAEl expression using
two-electrode voltage clamp electrophysiology. Oocytes expressing AgAEl displayed a
decreased volume as compared to the water-injected controls (T. Sern, unpublished
observation), which correlates with the AE regulatory functions of cell pH and volume.

90
The AgAEl was determined to be a functional protein with the capacity to transport
chloride. No sodium ion or potassium ion dependence was determined with the voltage
clamp assay. A comparison of current versus voltage (I-V plots) for both the AgAEl
expressing oocytes and the water injected controls are compared when two different bath
solutions are applied to the oocytes. The I-V plots for the water-injected controls
displayed no transport with or without chloride (Fig. 5-18A). When the solution contains
chloride (N98), the AgAEl expressing oocytes are capable of transporting chloride, as
seen by the steep rise in current (Fig. 5-18B). When chloride is replaced by a non-ionic
equivalent (N98-C1) there is no transport (Fig. 5-18B). The transporter blockers, 4,4-
diisothiocyanodihydrostilbene-2,2 -disulfonate (DIDS) and niflumic acid (NA) both
inhibited the transport capabilities of the expressed AE1 protein (NA not shown). The
application of DIDS inhibited the transport of chloride such that the I-V plot showed an
affect similar to the removal of chloride ions from the bath solution (Fig. 5-19A). The
likeness of removing chloride and the inhibitory affect of DEDS is easily viewed by
comparing the differences in current between the DIDS inhibition and the removal of
chloride (Fig. 5-19B). When the difference between chloride and chloride removal are
compared to the blocked and chloride removal transport, a large difference can be seen
(Fig. 5-19B). Activity of the mosquito AE1 was also inhibited when the CA-specific
inhibitor acetazolamide was added to the media (data not shown). Because
acetazolamide is known to have no direct inhibitory effect on AEs, unlike other
sulfonamides, it can be inferred that the AgAEl is inhibited by acetazolamide due to its
tight coordination and regulation by endogenous CA, as was found to be the case in
mammalian systems (Sterling et al., 2001a). Endogenous CA was bound by mammalian

91
AEs when they were expressed in HEK293 cells, and was shown to increase the rate of
bicarbonate transport. Furthermore, co-expression with mutant CAII (non-active) was
shown to result in decreased bicarbonate transport due to the displacement of the active
endogenous CA. The CA II binding motif found in this mosquito AE could similarly
bind the oocytes endogenous CA, also resulting in an increased transport rate. This may
explain the inhibition seen in AE transport when acetazolamide was applied. AgAEl
expression studies in oocytes are presently ongoing to further assess the function and
inhibition of this protein.
Discussion
An AE cDNA was cloned from fourth instar, larval An. gambiae gut tissue. The
translated 123 kDa protein is predicted to consist of intracellular amino and carboxy
termini and 12 transmembrane segments. In situ hybridization and antibody
immunolocalization identified AE mRNA message expression and protein localization
within epithelial cells of cardia, gastric caeca, posterior midgut, rectum, and Malpighian
tubules as well as tracheal, nerve, and muscle cells. Expression of AgAEl in Xenopus
oocytes displayed a reversible transport of chloride.
The most similar characterized protein sequence is the NDAE1 from D.
melanogaster. Xenopus oocyte expression with pH analysis determined that this protein
was sodium ion dependent. The An. gambiae AE1 protein has 72% sequence identity to
the NDAE1 but does not display sodium ion dependence. The carboxy termini of these
proteins show little similarity and therefore the carboxy terminus of NDAE1, unlike
AgAEl, may contain the necessary domain for sodium ion dependence (refer to Fig. 5-4).

92
In Xenopus oocyte expression tests of AgAEl, activity was decreased with a CA-
specific inhibitor, acetazolamide. Bicarbonate, rapidly formed by the hydration of carbon
dioxide by CA, is a substrate for the AE. The decreased ability to exchange ions in the
presence of acetazolamide leads to the prediction that this mosquito AE is directly
regulated by CA activity. There is evidence within the mammalian system for the
tethered coordination of anion exchangers with carbonic anhydrase. The mammalian
anion exchanger (AE1/ band3) has been shown to interact with and actually bind to CA II
at its carboxy terminus (Sterling et al., 2001b). AEs and in fact all bicarbonate
transporters (except DRA) identified to date have potential CA E-binding sites at their
carboxy termini (Sterling et al., 2002b) including this An. gambiae AE. The inhibitory
effect of acetazolamide on AgAEl expressed in Xenopus oocytes suggests that this
mosquito AE is coupled with an active CA enzyme, speculatively an endogenous
Xenopus CA protein. A protein complex consisting of a membrane-spanning AE and a
cytoplasmic CA has the ability to maintain tight pH homeostasis both inside and outside
of the cell at the same time. Furthermore, this complex brings together bicarbonate
production and transport in such a way that virtually all lag time is abolished by the
tethered coordination of the system (Sterling et al., 2001b). This type of bicarbonate
transport metabolon, if found to exist within the mosquito, may explain how the mosquito
gut is capable of driving and supporting a pH greater than 10 within the lumen while
sustaining a near neutral pH within the adjacent cells. Now that an AE has been localized
to CA active regions within the mosquito gut, namely the gastric caeca and posterior
midgut, it will be necessary to determine whether they form a bicarbonate metabolon as
proposed.

Several acid-base controlling proteins have now been identified within the
mosquito gut (Zhuang et al., 1999; Corena et al., 2002). The distribution of these
93
proteins along the length of the mosquito gut is both heterogeneous and discontinuous.
Along the length of the gut, non-adjacent regions such as the gastric caeca and posterior
midgut display similar protein expression and CA activity, while the region that separates
them, the anterior midgut, displays a different pattern of protein expression. This is not
surprising as the highly alkaline pH of the anterior midgut also contrasts with the nearly
neutral pH of the flanking gut regions. The most intriguing part of the novel expression
profile of these mosquito proteins is the parallel expression profile for the same proteins
within the mosquito midgut and the well-characterized mammalian kidney. As
characteristic of most epithelia, the epithelial cells found in both the mammalian kidney
and the mosquito midgut share several specific morphological features. Both populations
are mitochondria-rich, display apical microvilli, contain active cytosolic CA activity, and
express a proton translocating FC V-ATPase on specific domains of their plasma
membranes (Clements, 1992; Sterling et al., 2001a). Similarly, cell polarity proteins can
also be compared between the A (alpha) intercalated cells of the collecting tubules of the
mammalian kidney and the mosquito epithelial cells of the gastric caeca and posterior
midgut. The A cell subpopulation of kidney intercalated cells expresses, and is defined
by, an apical H+ V-ATPase, and a basolateral AE (Matsumoto et al., 1994). The
epithelial cells of the gastric caeca and posterior midgut also express an apical FT V-
ATPase (Zhuang et al., 1999), and a basolateral AE. Furthermore, B (beta) intercalated
cells of the mammalian kidney are defined by a Ff V-ATPase expressed within the
basolateral plasma membrane and an apical AE. This apical AE is different from AE1,

94
but has yet to be cloned and characterized (Kim et al., 1999). Like B cells, anterior
mosquito midgut cells have been shown to express a basolateral bT V-ATPase whereas
expression of AgAEl was absent from this region. The closely associated A and B
intercalated cells of the mammalian kidney are known to function in acid secretion and
bicarbonate secretion, respectively. Interestingly, the bicarbonate-secreting moiety, for
which the function of the B cell is defined, is unknown.
The co-occurrence of these A and B cell types provides for the tight regulatory
control of the maintenance of near-neutral pH in the kidney. Perhaps the pH differential
of 4 units, as seen in the mosquito anterior midgut, is achieved by the decoupling of these
cell types. In the mosquito gut, the A-like cells of the gastric caeca and posterior midgut
translocate protons toward the lumen and reduce the alkaline pH to near neutral levels.
The B-like cells of the alkaline anterior midgut secrete bicarbonate and protect the cells
from the high lumenal pH. Although the presented AgAEl is not that moiety, the
parallels between cells of the mammalian kidney and the mosquito midgut are evident.
The similarities between epithelial cells of the mammalian kidney and the mosquito
midgut support the use of the mosquito midgut as a simple model in which to study cell
polarization, pH balance, protein targeting and trafficking, as well as disease states. The
decoupling of A and B intercalated cells, as they may exist in the mosquito midgut,
provides an excellent model for studying human diseases such as distal renal tubular
acidosis, which is caused by mutations in either the basolateral AE1 or different subunits
of the apical fT V-ATPase (Alper, 2002).
The simple epithelium of the mosquito midgut may continue to reveal
mechanisms and pathways that also function within the complex metabolic network

95
comprising the mammalian kidney. Many studies have sought to determine specific
amino acids responsible for pH sensitivity within the anion exchangers. The sixteen
amino acid pH sensitive region of AE2 is almost identical to the comparable region
within the AgAEl sequence (refer to Fig. 5-5). The mosquito midgut provides an
excellent model for studying precisely the pH dependent moieties and proteins due to the
large pH gradient that it supports. The one cell layer epithelium that divides the
alkalinity of the lumen (pH 11) from the neutral pH of 7-8 within the cell cytosol has yet
to reveal the cell polarity that is capable of maintaining this system.
The elucidation of such a metabolon within the mosquito gut, in which an AE is
directly tethered to one or more CAs, such as in the mammalian system (Sterling et al.,
2001a), would provide a mosquito model which could be used as a simple framework for
uncovering metabolic networks within complicated mammalian systems such as the
kidney.

96
Multiple Alignment
Figure 5-1. Structural prediction of the An. gambiae AE1. A. Hydrophobicity plot of the
DNAman-aligned D. melanogaster NDAE1 and An. gambiae AE1 sequences
suggests a nearly identical protein topology of 12 membrane-spanning
domains in both proteins. B. Illustration depicting the intracellular location of
both protein termini as well as the predicted 12 transmembrane domains.

97
AgAEl [AAQ21364]
Human AE1[P02730]
Human AE2[P04920]
Human AE3[NP005061]
Human AE4[Q96Q91]
LDYIFTKRELKILDDIMPEMTKRARADDLHQLEDGEVG
LPLIFRNVELQCLDADDAKATFDEEEGRDEYDEVAMPV
LTRIFTDREMKCLDANEAEPVFDEREGVDEYNEMPMPV
LPRLFQDRELQALDSEDAEPNFDE-DGQDEYNELHMPV
LERVFSPQELLHLDELMPEEERSIPEKGLEPEHSFSGS
* * **
Figure 5-2. Putative amino terminus CAII binding motif. The highlighted conserved
leucine (L) was shown to be necessary for the specificity of CA binding in
AE1 (Vince et al., 2000) and is also present in AgAEl. The motif consists
of at least two acidic amino acids within the four residues following the
conserved non-polar L. The ability of AEs to bind CA enzymes greatly
raises their ability to regulate ionic homeostasis. Identical residues (*) are
noted in the alignment as well as conserved residues (.).

98
100% 80% 60% 40% 20% 0%
I I I I 1 1
Human AE2
Human AE1
Human AE3
Human NBC4a
Human AE4
Dros NDAE1
Anoph AE1
Human NBC8
Human SLC26A7
60%
58%
48%
45%
72%
53 A
36%
Human DRA
Human SLC26A6
22%
31%
11%
Figure 5-3. Homology tree depicting the amino acid identity between several BTs. The
An. gambiae AE1 amino acid sequence displays the closest identity to the D.
melanogaster NDAE1 (AAF98636) and human NBC8 (NP004849) sequences
with 72% and 53% identity, respectively. The human AEs display 36% to
45% identity while the sulfate transporters (SLC26 group) display only 11%
identity to AgAEl. Accession numbers: human AE2 (P04920), human AE1
(P02730), AE3 (NP005061), NBC4a (NP067019), AE4 (Q96Q91), SLC26A7
(NP439897), DRA (P40879), and SLC26A6 (Q9BXS9).

99
AgAEl AAQ21364
Dros AAF52497
NDAE1 AAF98636
MMDHGWDEEAP ID PRLKNRTFTAD QD FEGHRAHTVFVGVHIPGSSR 47
MAEKNEyiELPWTMNSSSGDDEAPKDPRTGGEDFTQQFTENDFEGHRAHTVYVGVHVPGG-R 61
MAEKNEYIELPWTMNSSSGDDEAPKDPRTGGEDFTQQFTENDFE 24
AgAEl AAQ21364
Dros AAF52497
NDAE1 AAF98636
RHSQRRRHKHHQASKENGDKGSTG SEAERPVTPPAQRVQFILG
RHSQRRRKHHHSGPGGGGGGGGGGGSIGGSGSVGGGAGKDNVSEKQQEVERPVTPPAQRVQFILG
VTPPAQRVQFILG
90
126
57
Figure 5-4. Alignment of carboxy terminus amino acids of An. gambiae and D.
melanogaster AEs. The characterized D. melanogaster NDAE1 (AAF98636)
is 72% identical to our An. gambiae AE1 sequence. The greatest number of
amino acid differences occurs at the carboxy terminus, the regulation domain.
However, an uncharacterized splice variant of NDAE1 (AAF52497) displays
an inserted sequence at the carboxy terminus that the characterized protein
does not. This inserted sequence shows similarity to the AgAEl sequence and
therefore is the closest predicted protein to AgAEl.

100
Anoph AE1
Human AE1
Human AE2
Human AE3
Human AE4
GDEMAWKETARWVKFEEDVEEGG
NQELRWMEAARWVQLEENLGENG
NQEPdWRE TARWIKFEE DVE E|E T
soephIwretarwirfeedveeJet
SITLSTHLHHRWVLFEEKLEVAA
* *
Figure 5-5. Alignment of An. gambiae and human AEs. pH sensitivity of AE2 (P04920)
and AE3 (NP005061) was mapped to the boxed 16 amino acids shown (Vince
et al., 2000). Unlike AE3, AE1 (P02730) was not stimulated by intracellular
alkalization (Vince et al., 2000). The An. gambiae AE1 (AAQ21364)
sequence shows a strong similarity to the identified amino acid sequence and
has 14 identical residues. The human AE1 sequence has only 8 identical
residues. pH sensitivity of an AE within the mosquito gut would be an
important attribute due to the regional compartmentalization of the pH flux.
The recently identified AE4 (Q96Q91) was included in this alignment for
completeness. It displays the least conservation with only 6 identical residues.
Stars indicate identical residues within all aligned sequences.

101
Figure 5-6. Localization of AgAEl mRNA within whole mount An. gambiae larvae. A.
Gastric caeca (GC), posterior midgut (PMG), and Malpighian tubules (MT)
show extensive expression while other gut regions display more restricted
hybridization. B. The gastric caeca as well as the cardia region (arrows)
display label. C. Extensive expression of AgAEl mRNA in the Malpighian
tubules. D. Specific cells (arrows) and trachea (arrowheads) of the rectum
show expression of AgAEl. Scale bars represent 300 pm in A, 100 pm in B,
and 25 pm in C and D.

102
Figure 5-7. Localization of AgAEl mRNA in muscle, nerve, and trachea in An. gambiae.
A. The whole mount gut preparation localizes AE message to specific muscle,
nerve (long arrow), and trachea (short arrow) fibers of the anterior midgut
(AMG). B. A high magnification of the AMG region detailing the tracheal
fibers (short arrows) and neuronal cells (long arrow) that express the AgAEl
message. Scale bar represents 75 pm in A and B.

103
Figure 5-8. In situ hybridization of AgAEl in whole mount An. gambiae consistently
shows positive labeling of tracheal fibers along the midgut. A. A thick
tracheal stalk penetrates the gastric caeca while thinner branches join the
AMG. This main tracheal stalk that is closely associated with the gastric
caeca consistently displays labeled particles that may be secretory vesicles
(arrow). B. Labeled trachea (arrows) traverse the AMG and coincide with
labeled nerve fibers (arrowheads) that extend down the midgut. C and D.
Labeled trachea (arrows) are randomly associated with the area surrounding
the gastric caeca. Scale bars represent 25 pm.

104
Figure 5-9. Anion exchanger mRNA localization reveals trachea and nerve fibers along
with neuronal cell labeling. A. Trachea (large arrow) display a random
pattern of distribution on the An. gambiae midgut along with pairs of neuronal
cells (small arrows). B. A labeled tracheal stalk (large arrow) shows abundant
labeling where it joins with the midgut (*). The finer branches of the trachea
can be seen joining (small arrow) the parallel nerve fibers (arrowhead) that
also display label. C. Neuronal cells (small arrows) scattered over the AMG
display AE label along with the nerve (arrowheads) and tracheal fibers (large
arrows). D. Strong labeling is consistently seen where the thick trachea
connects to the midgut (*) and sends out smaller trachiole fibers (arrows).
Scale bars represent 25 pm in A, 50 pm in B, 50 pm in C, and 50 pm in D.

105
Figure 5-10. Localization of AgAEl mRNA to the PMG of larval An. gambiae. A. The
whole mount gut preparation displays the unlabeled AMG on the left side of
the photo as compared to the labeled PMG on the right side of the photo. The
arrows point to the tracheal stalks that join the gut at the third body segment,
corresponding to the beginning of the PMG region. B. Outer margins of large
columnar PMG cells display AE mRNA labeling (*) along with labeling of
tracheal fibers (arrows) and small cuboidal cells (arrowheads). Scale bar
represents 50 pm in A, 25 pm in B.

106
A
¡
Figure 5-11. Larval An. gambiae displays strong AgAEl expression in the hindgut, the
pylorus. A. Distal to the joining of the Malpighian tubules with the gut, the
pylorus displays pronounced labeling of small epithelial cells (arrowhead)
and closely-associated muscle fibers. B. Higher magnification of the boxed
region displays labeled epithelial cells (arrowheads) of the pylorus with
closely associated circular muscle fibers (arrows) that form a pyloric
sphincter. The pylorus, a part of the hindgut, functions in ionic and osmotic
regulation. Scale bars represent 25 pm in A and B.

107
Figure 5-12. Localization of AE mRNA in An. gambiae shows abundant labeling of the
Malpighian tubules. A. The entire length of the Malpighian tubules displayed
labeling of AE mRNA. B. Labeled tracheal fibers are also associated with the
Malpighian tubules. C. Labeled tracheal fibers also extend from the tips of
Malpighian tubules and may contain secretory vesicles (arrow). Scale bar
represents 25 pm in A, B. and C.

108
Figure 5-13. Expression of AE mRNA was found throughout the ventral midgut ganglia.
Between the ventral midgut and the ventral integument a labeled neuronal
pathway connected each ganglia to the next. A. Two neurons (arrows) label
positively for the AE mRNA, rendering them highly visible above the other
neuronal cells within the unstained ganglia. B. Between each ganglia
cluster, a tissue crossbridge (*) passes in a 90 angle between the ganglia
and the gut. The neuronal pathway showing AE mRNA expression makes
contact with this junction and continues on to the next ganglia cluster on the
other side. C. Two neuronal cells within the following ganglia show clear
expression of AE mRNA. D. This panel shows an unobstructed view of a
single triangular-shaped ganglia cluster. It is clear that the AE mRNA is
located within a distinct population of neuronal cells in each ganglion.
Scale bars represent 50 pm.

109
Figure 5-14. Sense AE probes display no specific hybridization. A. These whole mount
preparations show no AE sense (control) label in the integument, midgut or
hindgut (B). A higher magnification of the gastric caeca (C) and posterior
midgut (D) with Malpighian tubules (MT) shows no hybridization with the
sense AE probe. These experiments were performed side by side with the
antisense probes and therefore length of exposure was identical. Scale bars
represent 600 pm in A, 300 pm in B, 100 pm in C, and 100 pm in D.

110
Figure 5-15. Antibody localization of AgAEl protein to the gastric caeca in An. gambiae
larvae. A. Our AE specific antibody displays immunoreactivity within the
cardia (*) and gastric caeca. B. Phalloidin was used to label the actin-
containing muscle fibers throughout the mosquito gut. C. Draq-5 was used
to label nuclear DNA. D. Three signal overlay depicting the AE protein in
relation to muscles and cell nuclei. E-H shows higher magnification views
of the gastric caeca with the same labeling profile as in A-D. AE protein
expression can be seen on plasma membranes (arrows) of the gastric caeca
(E). Scale bars represent 25 pm in A-D and 100 pm in E-H.

Ill
Figure 5-16. Localization of AgAEl protein within the PMG of An. gambiae larvae. A.
Our AE specific antibody displayed immunoreactivity within the PMG;
most prominantly within a specific band of cells that encircle the PMG
region. B. Phalloidin was used to label muscle fibers. C. Draq-5 was used
to localize nuclear DNA. D. Overlay of AE labeled cell membranes in
relation to muscle fibers and nuclei. E. High magnification of AE protein
localization within the PMG. Cell membranes of both large and small cells
are clearly labeled with our antibody. Scale bars represent 150 pm in A-D
and 75 pm in E.

112
Figure 5-17. Neuronal cells within the AMG display immunoreactivity for our An.
gambiae AE specific antibody. These neuronal cells also displayed AE
mRNA expression (refer to Fig. 5-13A) and are most often seen in pairs (*).
Scale bar represents 50 pm.

113
AgAE1 Expressing Oocyte
Figure 5-18. Current-voltage (I-V) plots depicting ion transport by the AgAEl
expressing oocytes in contrast to the water injected control oocytes. A.
When chloride is removed from the solution bathing the control oocyte and
replaced with a nonionic equivalent there is no change in the slope of the
curve, signifying no ionic transport. B. When chloride is removed from the
media surrounding the AgAEl expressing oocyte, the ionic transport is
eliminated, as seen by the decrease in slope.

114
Expression and Inhibition of AgAE1
Voltage (V)
Figure 5-19. Inhibition of AgAEl mediated chloride transport by DIDS. A. DIDS
blocks the transport of chloride by AgAEl. The result is similar to taking
chloride out of the bath solution. B. There is almost no difference between
blocking chloride transport with DIDS and removing chloride. In contrast,
removing chloride from the uninhibited exchanger shows a large difference
in chloride transport.

CHAPTER 6
CYTOSOLIC CA EXPRESSION IN LARVAL ANOPHELES GAMBIAE
Introduction
Carbonic anhydrase (CA) represents a superfamily of enzymes that reversibly
hydrates carbon dioxide to form bicarbonate and a proton. There are three families of
CAs (a, P, and y). Mammals have fourteen different CA isoforms, all belonging to the a
CA family. The vast characterizations of the mammalian CAs have revealed cytosolic,
membrane-bound, membrane-spanning, and mitochondrial isoforms. Interestingly, it is
becoming increasingly apparent with the recent availability of genome sequences, that
less complex organisms also contain a large array of CAs. The Anopheles gamhiae
genome contains at least 14 putative CA genes (EnsEMBL protein family
ENSF00000000228). Whether the 14 well-characterized mammalian CAs serve the same
functions as the 14 mosquito genes is intriguing. This question will remain unanswered
until all of the CAs from An. gambiae or a similarly distant species are characterized. We
can, however, speculate as to their relatedness through sequence identity and
conservation.
Carbonic anhydrase is an interesting enzyme to study in the context of the
mosquito for several reasons. Unlike most animals, mosquitoes use a highly alkaline
digestive strategy instead of an acid environment. Additionally, the functions of the
mosquito gut are similar to the mammalian kidney in filtering wastes and maintaining
ionic homeostasis. Uncovering the function of CAs in an insect, such as the mosquito,
115

116
will lead to evolutionary clues within the family of a CAs. The relationship between the
three distinct CA families is believed to represent convergent evolution. The relationship
between a CAs from distantly related species is unknown, mostly due to the lack of
characterized CAs from non-mammalian species.
We report here the first cytosolic CA that has been cloned and characterized from
the An. gambiae mosquito, in an attempt to unravel the physiology of an extremely
alkaline digestive system. There are at least two different CA isoforms expressed within
the larval mosquito gut. One is a GPI-linked CA isoform expressed in a specific subset
of muscle and nerve fibers that traverse the anterior midgut and gastric caeca regions
(refer to Chapter 4). The other is a cytosolic CA isoform expressed primarily within the
gastric caeca and posterior midgut regions.
Results
Anopheles gambiae CA Sequence Analysis
A CA cDNA was cloned from An. gambiae gut tissue. The full-length CA cDNA
sequence (accession number AY280613) represents 1 of 14 putative CA genes in the An.
gambiae genome predicted by the Ensembl CA protein family (ENSF00000000228).
This CA is comprised of 257 amino acids and is predicted to be a cytosolic isoform (no
signal sequence or hydrophobic transmembrane domains; Letunic et al., 2002). The
molecular weight is predicted to be 29 kDa (DNAman software). Multiple sites of
potential post-translational modification include 5 protein kinase C phosphorylation sites,
4 casein kinase II phosphorylation sites, and 5 N-myristoylation sites (Bairoch et ah,
1997). The numerous potential sites for protein modification may contribute to
regulatory control.

117
This CA displays all of the 13 highly conserved residues found in most other
active CA proteins, including the three necessary histidine residues required for the
binding of a zinc atom (Tashian, 1992; Sly and Hu, 1995; Tamai et al., 1996a). Figure 6-
1 shows an alignment of the active sites of An. gambiae, D. melanogaster, and human
CA proteins. The active site region of the cytosolic CA from An. gambiae contains one
amino acid difference from all of the mammalian isoforms. This particular residue (C-
89; An. gambiae CA numbering), is replaced by a serine (S) residue in all of the known
human CA isoforms (including mouse CA XIII in lieu of the uncharacterized human CA
XIII). This C to S amino acid change is also found in 1 of the 14 CA isoforms predicted
from the D. melanogaster genome (accession number CGI 1284-PA; Fig. 6-1).
The fourteen CA genes of humans are apparently required to perform the many
metabolic functions that complex organisms require. The recent release of the sequenced
genomes of two insect species, An. gambiae and D. melanogaster has revealed that these
organisms also have fourteen CAs. Comparisons based on amino acid composition
revealed that these 14 dipteran CAs are probably not homologs to each of the fourteen
human CAs. A phylogenetic analysis of the CA protein sequences was performed using
the Neighbor-Joining method (Saitou and Nei, 1987) as implemented in DNAman
software. A rooted tree shows the relationship between the human, mouse, and dipteran
CAs (Fig. 6-2). The human and mouse CAs cluster together, and the dipteran CAs
cluster together, however the mammalian and the dipteran proteins cluster separately. A
bootstrapping test was performed to determine the confidence value of the phylogenetic
tree. The An. gambiae CA that has the S to C difference within the active site pairs with
the D. melanogaster CA with the same difference (Fig. 6-2). These two dipteran CAs

118
appear to be CA homologs. Further studies must be performed to determine whether any
differences in the activity or inhibitory profile of these dipteran CA proteins exists due to
this active site difference. This conserved amino acid change found in An. gambiae and
D. melanogaster, but not humans, may suggest an evolutionary importance that could be
exploited in future mosquito larvacide production. Ongoing efforts aimed at
crystallization and x-ray analyses with collaborators at the University of Florida will also
reveal the structural identity of this An. gambiae cytosolic CA.
Localization of CA Activity in Anopheles gambiae Larvae
A modified version of Hanssons CA histochemistry method (Hansson, 1967) was
used to localize CA activity within the An. gambiae larvae. Precipitated cobalt salts
marked the regions of CA activity within the larvae. The regions of CA activity include
the cardia, gastric caeca and posterior midgut (Fig. 6-3). A small subset of specific cells
within the rectum also stained positively for CA activity (Fig. 6-3E).
Localization of Cytosolic CA mRNA in Anopheles gambiae Larv ae
A DIG-labeled antisense RNA probe was utilized to localize the CA message.
The most intense labeling was consistently viewed within epithelial cells of the gastric
caeca and posterior midgut (Fig. 6-4). The localization of cytosolic CA mRNA to the
gastric caeca and posterior midgut correlates with the location of CA activity within the
An. gambiae gut, as determined by CA histochemistry (refer to Fig. 6-3). The cardia was
also labeled, as well as a subset of nerve cells and fibers that traverse the AMG
longitudinally (Fig. 6-4B,C). Within the posterior midgut, the labeling was confined to
the outer edges of a subset of large columnar cells and the small cuboidal cells (Fig.6-5).

119
Labeling was also seen within the rectum and the last distal cell of the Malpighian
tubules (MT; Fig. 6-6).
Antibody Localization of CA Protein
The antigenic peptide sequence QYIRSPDAQTEIDAD was chosen from the An.
gambiae translated CA cDNA sequence (accession number AAQ21366) to elicit the
production of antibody. This peptide was chosen for its antigenic capacity as well as its
uniqueness among the 14 putative An. gambiae CA genes. The resultant chicken
antiserum was used to localize the CA protein within whole mount preparations of fourth
instar An. gambiae larvae.
Immunoreactivity for the cytosolic CA was predominantly displayed within the
gastric caeca (Fig.6-7). The PMG also displayed immunolabeling along the periphery of
both the large and small epithelial cells. Immunoreactivity was also evident within a
small population of neuronal cells scattered along the midgut, most often seen in pairs
(Fig.6-8). The pre-immune antisera displayed no specific immunoreactivity.
Bacterial Expression and Purification oiAnopheles gambiae Cytosolic CA
The full-length CA cDNA was subcloned into a pETlOO directional expression
vector (pETlOO/D-TOPO; Invitrogen) for expression of the recombinant protein with an
N-terminal tag containing an Xpress epitope and a polyhistidine (6X His) tag. The 6X
His tag was utilized when purifying the CA protein. Antibodies against both the Xpress
epitope and the CA peptide recognized a band of the predicted molecular weight (33
kDa) for the recombinant CA protein (Fig. 6-9).
Purified CA fractions were tested for CA activity using 180 isotope exchange
experiments (Silverman and Tu, 1986). This technique showed that CA activity was

120
present within the purified, recombinant CA fractions (data not shown). Activity was
partially inhibited by the application of methazolamide, a CA-specific inhibitor (data not
shown). These analyses confirmed that this cytosolic CA, cloned from an An. gambiae
gastric caeca cDNA library, has CA activity and therefore contributes to the CA activity
in the gastric caeca and posterior midgut regions, as determined by CA histochemistry.
Discussion
The CA isoform, cloned from the An. gambiae midgut, is a predicted cytosolic
protein that is expressed in the cardia, gastric caeca and posterior midgut regions.
Carbonic anhydrase activity was localized to these same regions through CA
histochemistry. The purified recombinant CA was shown to have CA activity through
180 isotope exchange experiments. This CA isoform is therefore responsible, at least in
part, for the CA activity displayed within the gastric caeca and posterior midgut regions
of the larval mosquito gut.
The one amino acid difference (C instead of S) within the active site of this
mosquito CA is not found in any mammalian isoform. The D. melanogaster genome
however displays an identical (C instead of S) difference in one of its putative CA
isoforms. The S residue at this position is present in every mammalian CA and therefore
may represent an evolutionary divergence within the a CA family. The phylogenetic
analysis of mammalian and dipteran CAs shows a more distant relationship between the
Dipteran and mammalian proteins than within the mammalian CA family. The dipteran
CA isoforms do not cluster with the mammalian CAs despite common functional
characteristics and sequence homology in all known a CAs. Instead, the mosquito CA
sequences cluster together, apart from the mammalian CA clusters.

The mammalian anion exchanger (AE1) has been shown to physically bind a
cytosolic CA isoform (Sterling et al., 2001a). An AE has been cloned from the An.
gambiae midgut (refer to chapter 5) and protein expression has been localized to the
gastric caeca and posterior midgut regions. This AE also has a putative CA binding site
within its intracellular carboxy terminus. The membrane localization of this predicted
cytosolic CA may be caused by its interaction with a membrane protein such as an AE.
Future experiments will reveal if this cytosolic CA is indeed coupled to an AE within the
gastric caeca and posterior midgut regions of the mosquito gut, forming a bicarbonate
transport metabolon. The existence of such a metabolon within gut regions flanking the
highly alkaline anterior midgut may be the mechanism through which cellular
homeostasis is maintained.

122
AAQ21365 CAIV-like
AAQ21366 CAII-like
ENSANGPO0000001812
ENSANGP00000018999
ENSANGPO0000029518
ENSANGP00000011908
ENSANGP00000001574
ENSANGP00000011013
ENSANGPO0000012957
ENSANGP00000016412
ENSANGPO0000010017
ENSANGP00000014948
ENSANGPO0000014919
ENSANGP00000001278
CG9235-PA Dros
CG18672-PA Dros
CG10899-PA Dros
CG11284-PA Dros
CG12309-PA Dros
CG3940-PA Dros
CG32698-PA Dros
CG6074-PA Dros
CG18673-PA Dros
CG1402-PA Dros
CG3669-PA Dros
CG6906-PA Dros
CG5379-PA Dros
CG7820-PA Dros
P00915 CA-I
P00918 CA-II
P07451 CA-III
P22748 CA-IV
AAB47048 CA-V
CAC42429 CA-VI
P43166 CA-VII
JN0576 CA-VIII
AAH14950 CA-IX
Q9NS85 CA-X
AAH02662 CA-XI
AAH23981 CA-XII
AAK16672 mCA-XIII
BAA85002 CA-XIV
QMHFHWG SEHTLDDTRYGLELHLVH
QLHFHWGIGDGSG JEHTLEGSTYSMEAHAVH
QFHFHAP SENLIKGHSYPLEGHLVH
QLHFHWGLSALDGSEHTIDGYRLPLELHVIH
QFHCHWGCSDSR- -GSEHTVDGESFAGELHLVH
EIHVHYGLHDQF--GSEHSVEGYTFPAEARHIQ
EIYFHYGTDNNQGSEHHIHGYSFPGEIQLYG
QLHFHWGDNDTF- -GSEDMIDNHRFPMELHWF
GLHFHWGDKNNR--GAEHVLNDIRYPLEMHIIH
QFHCHWGCSDSR- -GSEHTVDGESFAGELHLVH
QLHFHWGPDDAV- -GSEHLLDGRAHSMEAHLVH
QLHFHWGADNGRGSEHTFDGVAWAAEAHFVF
QFHFHWGVNSTVGSEHVYDYQRYPMEIHLVF
QMHFHWGPNNSEGSEHRINGERFPLEVHLVF
EISFRWSWASSLGSEHTLDHHHSPLEMQCLH
GLHFHWGSYKSR--GSEHLINKRRFDAEIHIVH
QLHFHWGSALSK--GSEHCLDGNYYDGEVHIVH
QLHFHWSDCDESG JEHTLEGMKYSMEAHAVH
ELRFHWGWCNSE- -GSEHTINHRKFPLEMQVMH
QIHMHWW SEHTINDIRYPLEVHIVH
EIHMHYGLNDQF--GSEHSVEGYTFPAEIQIFG
GLHFHWGDKNNR--GSEHVINDIRYTMEMHIVH
SVHFHWGSREAK- -GSEHAINFQRYDVEMHIVH
EIYIHYGTENVR--GSEHFIQGYSFPGEIQIYG
AFHFHWGSPSSRGSEHSINQQRFDVEMHIVH
QFHFHWGENDTI -GSEDLINNRAYPAELHWL
AVHFHWGSPESK--GSEHLLNGRRFDLEMHIVH
QFHCHWGCTDSK--GSEHTVDGVS YSGELHLVH
QFHFHWGSTNEH--GSEHTVDGVKYSAELHVAH
QFHFHWGSLDGQ--GSEHTVDKKKYAAELHLVH
QFHLHWGSSDDH--GSEHTVDGVKYAAELHLVH
QLHLHWSDLPYK--GSEHSLDGEHFAMEMHIVH
QFHFHWGAVNEG--GSEHTVDGHAYPAELHLVH
QMHFHWGGAS SEISGSEHTVDGIRHVIEIHIVH
QFHFHWGKKHDV--GSEHTVDGKSFPSELHLVH
EVRFHWGRENQR- -GSEHTVNFKAFPMELHLIH
QLHLHWGAAGRP--GSEHTVEGHRFPAEIHWH
EIRLHFGSEDSQGSEHLLNGQAFSGEVQLIH
ELRLLFGARDGA--GSEHQINHQGFSAEVQLIH
QLHLHWGNPNDPH-GSEHTVSGQHFAAELHIVH
QFHLHWGSADDHGSEHWDGVRYAAELHWH
QLHLHWGQKGSPG-GSEHQINSEATFAELHIVH
*
Figure 6-1. Clustal alignment of active sites within An. gambiae, D. melanogaster, and
human CA proteins. All active a CAs are tethered to a zinc molecule by the
coordination with three histidine (H) residues. There are three human CAs
which do not possess H residues in this orientation and have been shown to
lack CA activity. Within the active site region, some of the An. gambiae CAs
display novel differences, as compared to the tightly-conserved human CAs.
Without exception in the human isoforms, a conserved glutamine (E; marked
with *) always follows a serine (S). In contrast, the An. gambiae alignment
shows a change from the S to a cysteine (C) in one of their 14 putative
isoforms. The An. gambiae alignment also displays a gap within the active
site regions of two of the CAs. One of these sequences (AAQ21365) was
cloned and determined to be a GPI-linked CA IV-like isoform that was
discussed in a previous chapter. These novel active site differences may be
exploited in the formulation of a specific mosquito larvacide. The same active
site differences are also displayed within the D. melanogaster genome and
therefore may also provide clues to the evolutionary mechanism of these
proteins.

123
0.05
I I
92
95
96
IOC
IOC
IOC
94
IOC
IOC
IOC
IOC
IOC
IOC
IOC
IOC
BAA85002 CA-XIV
NP035927 MCA-XIV
AAH23981 CA-XII
NP848483 MCA-XII
AAH14950 CA-IX
- NP647466 MCA-IX
CAC42429 CA-VI
NP033932 MCA-VI
P22748 CA-IV
NP031633 MCA-IV
Anoph CA II-LIKE
CG11284-PA DROS CA
10Cj
XP134293 MCA-VII
10C
ioc
P43166 CA-VI I
NP078771 MCA-XIII
XP372051 CA-XIII
P00915 CA-I
P13634 MCA-I
" P07451 CA-III
P16015 MCA-III
P00918 CA-II
NP033931 MCA-II
AAB47048 CA-V
NP031634 MCA-VA
r NP031618 MCA-VI 11
JN0576 CA-VI 11
IOC
Anoph CA IV-LIKE
Aedes CA IV-LIKE
CG3940-PA Dros CA
Figure 6-2. Phylogenetic analysis between mammalian (human and mouse) and dipteran
{An. gambiae and D. melanogaster) CAs. This phylogenetic tree was
assembled using the Neighbor-Joining method. The tree has been rooted and
the bootstrapping confidence values are shown. The mammalian sequences
are shown in blue, MCA designates the mouse, while the dipteran .sequences
are shown in green. The mammalian CAs cluster separately from the dipteran
CA clusters, signifying a high divergence between mammalian and dipteran
CAs. There is also a high divergence between the dipteran CA -like
sequences and the dipteran CA IV-like sequences.

124
Figure 6-3. Localization of An. gambiae CA activity. Using CA histochemistry. CA
activity was localized to the cardia, gastric caeca, and posterior midgut
regions. A. A whole mount view of the entire gut region. B. A higher
magnification view of the gastric caeca and cardia regions that displayed CA
activity. Trachea are indicated with arrowheads. C. A higher magnification
view of the AMG to PMG boundary. The PMG displayed epithelial CA
activity while the AMG does not. The trachea are indicated by arrowheads.
D. Specific cells of the cardia along with cells of the AMG (arrows) displayed
CA staining. E. Small cells within the rectum also displayed CA activity
(arrows). Scale bars represent 300 pm in A, 100 pm in B and C, and 150 pm
in D and E.

125
A
B
Figure 6-4. Localization of CA mRNA expression within An. gambiae whole mounts.
A. A view of the entire gut region hybridized with the CA cRNA probe.
The most prominent hybridization was found in the gastric caeca and
posterior midgut. B. A higher magnification view of the gastric caeca. Both
the proximal cells and distal cap cells of the gastric caeca show CA mRNA
expression. Rostral to the gastric caeca the cardia region also displayed
intense label (arrow). C. Top view of the cardia displays consistent labeling
around the circumference of the cardia region (arrows). Scale bars represent
300 pm in A, 75 pm in B and C.

126
Figure 6-5. Localization of CA mRNA expression within the posterior midgut o An.
gambiae. A. The PMG displays staining for CA mRNA within the
peripheral borders of both the large and small epithelial cells. B. Higher
magnification of the PMG shows the stained large columnar cells (*) and
the labeled small cuboidal cells (arrows). C. Side view of the PMG reveals
CA mRNA expression near the plasma membranes (arrows) and not
throughout the cytoplasm. Scale bars represent 150 pm in A, 25 pm in B,
and 50 pm in C.

127
Figure 6-6. Localization of CA mRNA expression within the hindgut. A. The hindgut
displays CA label within the Malpighian tubules (MT, arrow) and the
rectum. B. Higher magnification of the Malpighian tubules shows staining
for CA expression confined to the most distal one or two cells. All other
cells of the MT show no hybridization. Scale bars represent 300 pm in A
and 25 pm in B.

128
Figure 6-7. Localization of CA protein within gastric caeca of An. gambiae larvae. A.
The CA specific antibody prominently labels cells of the gastric caeca. B.
Phalloidin was used to localize muscle fibers. C. Draq-5 was used to
localize nuclear DNA. D. Overlay of the three signals shows the location of
CA in relation to muscle fibers and nuclei. Scale bars represent 100 pm.

129
Figure 6-8. Localization of CA protein within the PMG of An. gambiae. The CA-
specific antibody labels cell membranes of both large (double arrows) and
small cells (small arrows), mimicking the protein localization pattern for the
AgAEl protein. Since this particular CA isoform is predicted to be a
cytosolic isoform, its localization to cell membranes suggests an interaction
with a membrane protein. The AgAEl protein, known to be a membrane
protein, capable of binding CA II, and the localization of the AE1 protein to
the same cells supports the hypothesis of a bicarbonate metabolon within the
mosquito gut. Scale bar represents 25 pm.

130
0 1234 56 24
Figure 6-9. Protein gels and western blots of recombinantly expressed CA protein. A.
Brilliant blue staining of CA protein induction (33 kDa) from 0 to 24 hours.
B. Fast green staining of recombinant CA protein expression (33 kDa). C.
The XPRESS (XP) antibody detected a 33 kDa band. Pre-immune sera
from two chickens inoculated with a conjugated CA peptide (PI) did not
detect a band at 33 kDa. The antisera collected from one of the inoculated
chickens, detects a visible protein band at the expected 33 kDa (94; red
arrow). The antisera collected from the other chicken (93), does not
recognize the 33 kDa protein band on the western blot. The lane containing
the molecular weight standard is marked with an M.

CHAPTER 7
CONCLUSIONS AND FUTURE DIRECTIONS
Conclusions
Molecular cloning techniques using isolated mosquito guts from Aedes aegypti
and Anopheles gambiae resulted in the full-length cDNA cloning of three carbonic
anhydrase (CA) genes. Our study of CA(s) within the mosquito gut began with a simple
understanding that the anterior midgut lumen has a pH of 11. The ability of C A to
greatly enhance the production of bicarbonate (and carbonate), a strong buffer, made it a
likely candidate for buffering the mosquito gut. Mammalian CA isoforms have been
studied extensively throughout the past several decades. However, the relationship
between a CAs from mammals, and those of less complex species such as mosquitoes, is
unknown. This is partly due to the lack in characterization of multiple CA isoforms from
a single non-mammalian species.
Fourteen different a CA isoforms from mammals have been characterized. The
An. gambiae and D. melanogaster genomes also display fourteen CAs. However, a
phylogenetic analysis of amino acid sequences has shown that the fourteen mammalian
CA isoforms and the fourteen dipteran CA isoforms are not direct homologs. When more
CA isoforms are characterized from insects it will be extremely interesting to determine
which isoforms are represented or omitted from the insect divergence of CAs.
Before the fourteen CA isoforms in mammals and mosquitoes can be truly
compared, every mosquito CA isoform must be characterized to determine their
131

132
functional identities. Two different CA isoforms from larval mosquitoes were partially
characterized within this study. One GPI-linked CA isoform was found in both An.
gambiae andAe. aegypti, and the other cytosolic CA isoform was found in An. gambiae.
The additional characterization of more mosquito CAs in the future, will present a great
opportunity for studying the evolution of CAs within the same a CA family.
A GPI-linked CA isoform, similar to the mammalian CA IV isoform was
localized to the gastric caeca and a specific subset of muscle fibers in the anterior midgut
region of both Ae. aegypti and An. gambiae. This isoform is different from the other a
CA isoforms (characterized in mammals) in that the entire CA protein is located
extracellularly, with only the GPI-link maintaining an association to the plasma
membrane. These GPI-linked CA isoforms from two different mosquito species also
have a direct homolog in the D. melanogaster genome database. These three dipteran CA
IV-like isoforms all display a shortened active site region. How this novel active site
affects the activity of the CA protein is unknown. However, it is known that mammalian
CA IV proteins are oriented by the GPI attachment so that the active site is directed away
from the membrane, thereby affording the greatest accessibility of substrate to the active
site. The GPI-linked CA of the anterior midgut muscle fibers would be in the prime
location and conformation for taking up substrate from the hemolymph.
The other full-length CA cDNA isolated from An. gambiae gut tissue is a
cytosolic CA isoform. The expression of mRNA was localized to the cardia, gastric
caeca and posterior midgut epithelial cells. These regions were also shown to contain an
active CA enzyme through CA histochemistry. Recombinant expression of this An.
gambiae CA protein in bacteria, produced a purified CA-active eluate, as measured by

133
180 exchange. The activity was shown to be sensitive to the CA-specific inhibitor,
methazolamide. This cytosolic CA isoform therefore contributes to the CA activity
present within the cardia, gastric caeca and posterior midgut epithelial cells.
A full-length anion exchanger (AE) cDNA was also cloned from the gut tissue of
An. gambiae. The AEs are a small group within a large bicarbonate transporter (BT)
superfamily. Studies of mammalian BTs are ongoing and many new forms are still being
discovered and characterized to date. AEs reversibly transport chloride for bicarbonate in
a 1:1 electroneutral exchange. Cloning and localizing an AE within the mosquito gut was
therefore a direct progression from localizing the CAs, relating the production and
transport of bicarbonate within the alkaline gut.
The larval mosquito AE was expressed in Xenopus oocytes and was shown to
have electrophysiological characteristics of known AEs (i.e. chloride transport and DIDS
inhibition). The AE contains an intracellular carboxy terminus that is predicted to
moderate ion exchange and an amino terminus that performs ion exchange via the twelve
membrane-spanning domains.
The antibodies we produced, specific to carboxy and amino terminal peptides,
label the membranes of both gastric caeca and posterior midgut epithelial cells. To
maximize ion exchange, the An. gambiae AE contains an amino terminus CA binding
motif. If indeed the AE binds a cytosolic CA, the gastric caeca and posterior midgut
regions would have control over cellular pH, via both intracellular and extracellular
means. Since both of these regions contain active cytosolic CA enzyme(s), we propose
that the AE spans the membrane in the GC and PMG and binds a cytosolic CA, forming a
bicarbonate transport metabolon to transport the bicarbonate that is made by the CA(s).

134
To further regulate ion exchange, the AE has a particular region of amino acids with
similarity to mammalian AEs 2 and 3. This region was found to confer pH sensitivity
through intracellular alkalization. The ability of a mosquito AE to detect intracellular
alkalization may be the underlying mechanism through which the alkalinity remains
confined to the anterior midgut region. The AE- and CA-containing regions (gastric
caeca and posterior midgut) flank the anterior midgut and could possibly modulate their
transport rates in response to the encroaching or retreating alkaline pH. This bicarbonate
metabolon could have the ability to maintain a large pH gradient as is displayed in the
larval mosquito gut. Discovering the proteins involved in the production and
maintenance of such an alkaline pH could define a fundamental metabolon that is critical
for pH homeostasis.
New Model
The cloning and localization of CAs and an AE within the larval mosquito gut has
uncovered an unpredicted model of physiology. Our original model, based on Manduca
sexta (refer to Fig. 1-3), has evolved considerably due to our new findings within larval
mosquitoes. The most unexpected finding, was the failure to detect a CA within the
anterior midgut. Carbonic anhydrase histochemistry, l80 isotope exchange, in situ
hybridization, immunohistochemistry, and real time PCR all failed to give evidence for a
CA within the AMG region. Although these data did not support our original model of
anterior midgut alkalization, our new model describes a system in which CA is not
necessaiy within the AMG epithelial cells. Furthermore, our new data, including the
localization of AE within the mosquito gut, has provided insight as to why our new
model of AMG alkalization is more efficient in the absence of CA.

135
Although CA is a reversible enzyme, the mammalian CAIV is even faster than
CA II at bicarbonate dehydration. This ability, along with prior studies that have found a
very low concentration of bicarbonate in the hemolymph surrounding the mosquito
anterior midgut, have led to a prediction of the role of the CA IV-like enzyme in the
mosquito midgut. The high alkalinity of the AMG lumen is energized by a H+ V-
ATPase, which pumps protons from the epithelial cells into the hemolymph surrounding
the anterior midgut. Only the AMG contains a basally-oriented H+ V-ATPase, whereas
the other regions of the midgut (gastric caeca and PMG) express an apically-oriented H+
V-ATPase (Zhuang et al., 1999). The CA IV-like enzymes that are suspended in the
hemolymph are in the perfect position to provide a sink for these protons. Through the
action of the CA IV-like enzyme in the hemolymph, protons are combined with
bicarbonate to form carbon dioxide and water. Due to this coordinated effort, the H+ V-
ATPase can keep pumping protons out into the hemolymph without creating a
concentration gradient. The CA activity on the hemolymph side of the AMG therefore
allows the H+ V-ATPase to function more efficiently.
In contrast to the benefit that the CA IV-like enzyme provides to the FT V-
ATPase, the existence of CA within the epithelial cells of the AMG would actually
hinder the H+ V-ATPase. The benefits of not having a CA in the AMG are twofold, due
to the reversibility of CA. First, because there is no CA to dehydrate bicarbonate, there is
no competition with the H+ V-ATPase for protons. Secondly but most importantly,
because there is no CA to hydrate carbon dioxide to provide protons to the H+ V-ATPase,
the protons must come into the cell from the lumen to replace the protons that are being
pumped out to the hemolymph. This need for proton replacement may be the driving

136
force that pulls the proton from bicarbonate in the lumen to leave a carbonate ion. The
proton may enter the cell through a channel or be exchanged for a cation, such as
potassium. This carbonate or potassium carbonate could then drive the pH up to the high
alkalinity. The absence of an anion exchanger in the AMG would also contribute to the
alkalinity in the lumen by not providing a route for bicarbonate to leave the AMG lumen.
The localization of the AE to the basolateral membranes in both gastric caeca and
posterior midgut correlates with the localization of a cytosolic CA isoform to the same
regions. This AE brings bicarbonate into the cell in exchange for dumping chloride into
the hemolymph. This AE could account for the high levels of chloride and low levels of
bicarbonate that have been measured in the hemolymph (Boudko et al., 2001a). The
bicarbonate that enters the cell provides a substrate for the cytosolic CA. Due to the CA
binding motif at the amino terminal of the AE, the binding of these two proteins would
form a bicarbonate transport metabolon. This complex would maximize bicarbonate
transport due to the elimination of diffusion, normally relied upon to get substrate from
one protein to the next.
Our new model reflects the absence of CA within the AMG epithelial cells and
instead shows a GPI-linked CA bound to specific muscle fibers traversing the AMG and
GC, as well as a cytosolic CA within epithelial cells of the cardia, GC and PMG (Fig. 7-
1). The AE is found in regions that express CA and flank the alkaline anterior midgut.
Therefore, the cloned AE is shown in the GC and PMG. This AE has the ability to bind a
cytosolic CA, whereby the coordinated efforts of producing and transporting bicarbonate
are enhanced. The new model reflects a bicarbonate transport metabolon within the
gastric caeca and posterior midgut regions.

137
Future Directions
There are fourteen predicted CA isoforms in An. gambiae; however, this
dissertation research only dealt with the cloning and characterization of two. Although
both were found to have expression in the larval midgut, there may be more that can
influence our model. However, if any additional active CAs are localized within the
larval mosquito midgut, they most likely will also be localized to the gastric caeca and/or
PMG. All of our CA studies failed to show any evidence for a CA in the AMG epithelial
cells. An apically expressed CA in the AMG may have avoided the CA histochemical
staining but the 0 isotope exchange assay should have uncovered any CA activity
present within this gut region.
The hypothesis that a bicarbonate transport metabolon exists within the regions
flanking the alkaline anterior midgut needs to be investigated. If indeed the AE is found
to bind CA, then the putative CA binding sequence located within the intracellular
carboxy terminus needs to be examined. The amino acids necessary for binding should
be studied to determine whether the binding motif of mammalian AEs is valid for the
binding of mosquito CA(s) to the mosquito AE, as proposed.
The localization of two different CA isoforms and an AE within the larval
mosquito gut has established a new model upon which to build. Preceeding this
investigation, only a fT V-ATPase was localized within the larval mosquito gut. Other
channels and transporters need to be identified and localized, such as chloride channels
and sodium/ hydrogen exchangers.
The results of these studies may be used to formulate specific mosquito
larvacides. Both of the cloned mosquito CA isoforms display novel active site regions.

138
CA inhibitors can be manufactured to potentially block only CAs containing this novel
active site. Although these novel active sites are not found in any mammalian CA, they
are also found in other insects such as D. melanogaster. A larvacide specific for
mosquitoes can still be made due to the incorporation of an alkaline pH trigger. Species
that do not have a highly alkaline digestive strategy, such as D. melanogaster, will
therefore not be exposed to the specific CA inhibitor(s).
The larval mosquito gut provides a simple model for kidney epithelial transport.
An important finding was that the mosquito epithelial cells resemble the a and p
intercalated cells of the mammalian kidney. The addition of more ion transporters and
exchangers to our mosquito model will enable further comparisons between the mosquito
gut and the mammalian kidney. Diseases related to anion exchangers and V-ATPases
can also be studied using the mosquito gut due to the one cell layer epithelium that allows
for easier tracking of ions, as compared to the complex kidney.

139
A.
Midgut Hindgut
hco;
B.
lumen H
hco3
HCO, +H+
hco3 + co32 h<
AMG 2H>K+ PMG
co2 + h2o
hco3
.co2 +
HzO Hemolymph
H+ + HCO/ PH 78
Figure 7-1. New larval mosquito model. A. The larval mosquito gut is divided into the
foregut, midgut, and hindgut. The gastric caeca (GC) and anterior midgut
(AMG) express a GPI-linked CA isoform on muscle fibers (shown in yellow).
The cardia, GC, posterior midgut (PMG), rectum, and last distal cell of
Malpighian tubules (MT) express a cytosolic CA isoform. The GC, PMG,
MT and rectum express a chloride/ bicarbonate anion exchanger (AE). In the
GC, and PMG, the AE may bind a cytosolic CA isoform forming a metabolon.
A V-ATPase is expressed in GC, AMG, and PMG. B. Diagram of a
representative cell from GC, AMG, and PMG displaying the cell polarity.
The V-ATPase is expressed apically in GC and PMG, and basally in AMG.
The AE is expressed basally in the GC and PMG. The GPI-linked CA
isoform is expressed extracellularly on muscle fibers in the GC and AMG.

REFERENCES
Alper, S. L. (2002). Genetic diseases of acid-base transporters. Annu. Rev. Physiol. 64.
899-923.
Alper, S. L., Chernova, M. N. and Stewart, A. K. (2001). Regulation of Na+-
independent C1-/HC03- exchangers by pH. Jop. 2, 171-175.
Altschul, S. F., Gish, W., Miller, W., Myers, E. W. and Lipman, D. J. (1990). Basic
local alignment search tool. J. Mol. Biol. 215, 403-410.
Baird, T. T., Jr., Waheed, A., Okuyama, T., Sly, W. S. and Fierke, C. A. (1997).
Catalysis and inhibition of human carbonic anhydrase IV. Biochemistry 36, 2669-2678.
Bairoch, A., Bucher, P. and Hofmann, K. (1997). The PROSITE database, its status in
1997. Nucleic Acids Res. 25, 217-221.
Berenbaum, M. (1980). Adaptive significance of midgut pH in larval Lepidoptera. Am.
Nat. 115,138-146.
Boudko, D. Y., Moroz, L. L., Harvey, W. R. and Linser, P. J. (2001a). Alkalinization
by chloride/ bicarbonate pathway. Proc. Natl. Acad. Sci. USA 98,15354-15359.
Boudko, D. Y., Moroz, L. L., Linser, P. J., Trimarchi, J. R., Smith, P. J. S., Harvey,
W. R. (2001b). In situ analysis of pH gradients in mosquito larvae using non-invasive,
self-referencing, pH-sensitive microelectrodes. J. Exp. Biol. 204, 691-699.
Brown, D. and Breton, S. (1996). Mitochondria-rich, proton-secreting epithelial cells. J.
Exp. Biol. 199 (Pt 11), 2345-2358.
Brown, D. and Waneck, G. L. (1992). Glycosyl-phosphatidylinositol-anchored
membrane proteins. J. Am. Soc. Nephrol. 3, 895-906.
Chegwidden, W. R. and Carter, N. D. (2000). Introduction to the carbonic anhydrases.
Exs. 14-28.
Clamp, M., Andrews D., Barker, D., Beven, P., Cameron, G., Chen, Y., Clark, L.,
Cox, T., Cuff, J., Curwen, V., et al. (2003). Ensembl 2002: accommodating
comparative genomics. Nucleic acids res. 31 (1), 38-42.
140

141
Clark, T. M., Koch, A. and Moffett, D. F. (1999). The anterior and posterior stomach
regions of larval Aedes aegypti midgut: regional specialization of ion transport and
stimulation by 5-hydroxytryptamine. J. Exp. Biol. 202 (Pt 3), 247-252.
Clements, A. N. (1992). The biology of mosquitoes. Chapman & Hall, London, UK.
Corena, M. P., Sern, T. J., Lehman, H. K., Ochrietor, J. D., Kohn, A., Tu, C. and
Linser, P. J. (2002). Carbonic anhydrase in the midgut of larval Aedes aegypti:
cloning, localization and inhibition. J. Exp. Biol. 205, 591-602.
Dadd, R. H. (1975). Alkalinity within the midgut of mosquito larvae with alkaline-active
digestive enzymes. J. Insect Physiol. 21,1847-1853.
Darsie, R. F., Jr. and Morris, C. D. (2000). Keys to the adult females and fourth instar
larvae of the mosquitoes of Florida (Dptera, Culicidae). Bull. Fla. Mosq. Control
Assoc.
Dodgson, S. J. (1991). Why are there carbonic anhydrases in the liver? Biochem. Cell
Biol. 69, 761-763.
Donaldson, T. L., and Quinn, J. A. (1974). Kinetic constants determined from
membrane transport measurements: carbonic anhydrase activity at high concentrations.
Proc. Natl. Acad. Sci. USA 71, 4995-4999.
Dow, J. A. (1984). Extremely high pH in biological systems: a model for carbonate
transport. Am. J. Physiol. 24, (4 Pt 2), R633-636.
Eisenhaber, B., Bork, P. and Eisenhaber, F. (1999). Prediction of potential GPI-
modification sites in proprotein sequences. J. Mol. Biol. 292, 741-758.
Fernley, R. T. (1988). Non-cytoplasmic carbonic anhydrases. Trends Biochem. Sci. 13,
356-359.
Frohman, M. A. and Zhang, Y. (1997). Using rapid amplification of cDNA ends
(RACE) to obtain full-length cDNAs. Methods Mol. Biol. 69, 61-87.
Geer, L. Y., Domrachev, M., Lipman, D. J. and Bryant, S. H. (2002). CDART:
protein homology by domain architecture. Genome Res. 12, 1619-1623.
Gubler, D. J. (1997). in: Dengue and dengue hemorrhagic fever, (ed. Gubler, D. J. &
Kuno, G.), CAB International, Wallingford, U.K. pp. 1-22.
Gupta, R., Jung, E., and Brunak S. (In preparation 2002) Prediction of N-glycosylation
sites in human proteins.

142
Halstead, S. B. (1997). in: Dengue and dengue hemorrhagic fever, (ed. Gubler, D. J. &
Kuno, G.), CAB International, Wallingford, U.K. pp. 23-44.
Hansson, H. P. J. (1967). Histochemical demonstration of carbonic anhydrase activity.
Histochemie 11, 112-128.
Harvey, W. R. (1992). Physiology of V-ATPases. J. Exp. Biol. 172,1-17.
Haskell, J, A., Clemons, R. D. and Harvey, W. R, (1965). Active transport by the
Cecropia midgut. I. Inhibitors, stimulants and potassium-transport. J. Cell. Comp.
Physiol. 65,45-56.
Hewett-Emmett, D. (2000). Evolution and distribution of the carbonic anhydrase gene
families. Exs. 90, 29-76.
Hewett-Emmett, D. and Tashian, R. E. (1996). Functional diversity, conservation, and
convergence in the evolution of the alpha-, beta-, and gamma-carbonic anhydrase gene
families. Mol. Phylogenet. Evol. 5, 50-77.
Holt, R. A. Subramanian, G. M. Halpern, A. Sutton, G. G. Charlab, R. Nusskern, D.
R. Wincker, P. Clark, A. G. Ribeiro, J. M. Wides, R. et al. (2002). The genome
sequence of the malaria mosquito Anopheles gambiae. Science 298, 129-149.
Hogue, C. W. (1997). Cn3D: a new generation of three-dimensional molecular structure
viewer. Trends Biochem. Sci. 22, 314-316.
Hubbard, T., Barker, D., Birney, E., Cameron, G., Chen, Y., Clark, L., Cox, T.,
Cuff, J., Curwen, V., Down, T. et al. (2002). The ensembl genome database project.
Nucleic Acids Res. 30, 38-41.
Huber, S., Asan, E., Jons, T., Kerscher, C., Puschel, B. and Drenckhahn, D. (1999).
Expression of rat kidney anion exchanger 1 in type A intercalated cells in metabolic
acidosis and alkalosis. Am. J. Physiol. 277, F841-849.
Jespersen, T., Grunnet, M., Angelo, K., Klaerke, D. A. & Olesen, S. P. (2002). Dual
function vector for protein expression in both mammaliancells and Xenopus laevis
oocytes. Biotechniques 32, 536-538, 540.
Jungreis, A. M., Barron, D. N. and Johnston, J. W. (1981). Comparative properties of
tobacco homworm Manduca sexta, carbonic anhydrases. Am. J. Physiol. 241, R92-99.
Kim, J., Kim, Y. H., Cha, J. H., Tisher, C. C., and Madsen, K M. (1999). Intercalated
cell subtypes in collecting tubule and cortical collecting duct of rat and mouse. J. Amer.
Soc. Nephrol. 10, 1-12.

143
Letunic, I., Goodstadt, L., Dickens, N. J., Doerks, T., Schultz, J., Mott, R., Ciccarelli,
F., Copley, R. R., Ponting, C. P. and Bork, P. (2002). Recent improvements to the
SMART domain-based sequence annotation resource. Nucleic Acids Res. 30, 242-4.
Lindskog, S. (1997). Structure and mechanism of carbonic anhydrase. Pharmacol. Ther.
74,1-20.
Marchler-Bauer, A., Panchenko, A. R., Shoemaker, B. A., Thiessen, P. A., Geer, L.
Y. and Bryant, S. H. (2002). CDD: a database of conserved domain alignments with
links to domain three-dimensional structure. Nucleic Acids Res. 30,281-283.
Martin, M. M., Martin, J. S., Kukor, J. J. and Merrit, R. W. (1980). The digestion of
protein and carbohydrate by the stream detritivore, Tpula abdominalis (Dptera,
Tipulidae). Oecologia 46,360-364.
Matsumoto, T., Winkler, C. A., Brion, L. P. and Schwartz, G. J. (1994). Expression
of acid-base-related proteins in mesonephric kidney of the rabbit. Am. J. Physiol. 267,
F987-997.
Matz, M., Shagin, D., Bogdanova, E., Britanova, O., Lukyanov, S., Diatchenko, L.
and Chenchik, A. (1999). Amplification of cDNA ends based on template-switching
effect and step-out PCR. Nucleic Acids Res. 27, 1558-1560.
Meldrum, N. U. and Roughton, F. J. W. (1933). Carbonic anhydrase: Its preparation
and properties. J. Physiol. London 80, 113-142.
Osborne, W. R. A. and Tashian, R. E. (1975). An improved method for the purification
of carbonic anhydrase isozymes by affinity chromatography. Anal. Biochem. 64, 297-
303.
Parkkila, S., Rajaniemi, H., Parkkila, A. K., Kivela, J., Waheed, A., Pastorekova, S.,
Pastorek, J. and Sly, W. S. (2000). Carbonic anhydrase inhibitor suppresses invasion
of renal cancer cells in vitro. Proc. Natl. Acad. Sci. USA 97, 2220-2224.
Phillips, K. P. and Baltz, J. M. (1999). Intracellular pH regulation by HC03-/C1-
exchange is activated during early mouse zygote development. Dev. Biol. 208, 392-405.
Ridgway, R. L, and Moffett, D. F. (1986). Regional differences in the histochemical
localization of carbonic anhydrase in the midgut of tobacco homworm (Manduca
sexta). J. Exp. Zool. 237, 407-412.
Rockstein, M, (1964). The physiology of insecta, (ed. Rockstein, M.), Academic Press,
New York, NY. pp. 380-387.

144
Romero, M. F., Henry, D., Nelson, S., Harte, P. J., Dillon, A. K. and Sciortino, C. M.
(2000). Cloning and characterization of a Na+-driven anion exchanger (NDAE1). A
new bicarbonate transporter. J. Biol. Chem. 275, 24552-24559.
Saitou, N. and Nei, M. (1987). The neighbor-joining method: a new method for
reconstructing phylogenetic trees. Mol. Biol. Evol. 4, 406-25.
Schwartz, G. J. (2002). Physiology and molecular biology of renal carbonic anhydrase.
J. Nephrol. 15 Suppl 5, S61-74.
Shahabuddin, M. and Pimenta, P. F. (1998). Plasmodium gallinaceum preferentially
invades vesicular ATPase-expressing cells in Aedes aegypti midgut. Proc. Natl. Acad.
Sci. USA 95, 3385-3389.
Silverman, D. N. and Tu, C. K. (1986). Molecular basis of the oxygen exchange from
C02 catalyzed by carbonic anhydrase III from bovine skeletal muscle. Biochemistry 25,
8402-8408.
Sly, W. S. (2000). The membrane carbonic anhydrases: from C02 transport to tumor
markers. Exs. 95-104.
Sly, W. S. and Hu, P. Y. (1995). Human carbonic anhydrases and carbonic anhydrase
deficiencies. Annu. Rev. Biochem. 64, 375-401.
Spielman, A. and D'Antonio, M. (2001). Mosquito: A natural history of our most
persistent and deadly foe. Hyperion Press, New York, NY.
Sterling, D., Alvarez, B. V. and Casey, J. R. (2002a). The extracellular component of a
transport metabolon. Extracellular loop 4 of the human AE1 C1-/HC03- exchanger
binds carbonic anhydrase IV. J. Biol. Chem. 277, 25239-25246.
Sterling, D., Brown, N. J., Supuran, C. T. and Casey, J. R. (2002b). The functional
and physical relationship between the DRA bicarbonate transporter and carbonic
anhydrase II. Am. J. Physiol. Cell Physiol. 283, Cl522-1529.
Sterling, D. and Casey, J. R. (2002). Bicarbonate transport proteins. Biochem. Cell Biol.
80, 483-97.
Sterling, D., Reithmeier, R. A. and Casey, J. R. (2001a). A transport metabolon.
Functional interaction of carbonic anhydrase II and chloride/bicarbonate exchangers. J.
Biol. Chem. 276, 47886-47894.
Sterling, D., Reithmeier, R. A. and Casey, J. R. (2001b). Carbonic anhydrase: in the
driver's seat for bicarbonate transport. Jop. 2, 165-170.

145
Stobbart, R. H. (1971). Evidence ofNa+/H4 and C17HC03' exchanges during
independent sodium and chloride uptake by the larvae of the mosquito Aedes aegypti. J.
Exp. Biol. 54,19-27.
Sun, M. K. and Alkon, D. L. (2002). Carbonic anhydrase gating of attention: memory
therapy and enhancement. Trends Pharmacol. Sci. 23, 83-89.
Tamai, S., Cody, L. B. and Sly, W. S. (1996a). Molecular cloning of the mouse gene
coding for carbonic anhydrase IV. Biochem. Genet. 34, 31-43.
Tamai, S., Waheed, A., Cody, L. B. and Sly, W. S. (1996b). Gly-63-->Gln substitution
adjacent to His-64 in rodent carbonic anhydrase IVs largely explains their reduced
activity. Proc. Natl. Acad. Sci. USA 93, 13647-13652.
Tashian, R. E. (1969). in: Biochemical methods in red cell genetics, (ed. Yunis, J. J.),
Academic Press, New York, NY. pp. 307-336.
Tashian, R. E. (1992). Genetics of the mammalian carbonic anhydrases. Adv. Genet. 30,
321-356.
Tashian, R. E., Hewett-Emmett, D., Carter, N. and Bergenhem, N. C. (2000).
Carbonic anhydrase (CA)-related proteins (CA-RPs), and transmembrane proteins with
CA or CA-RP domains. Exs. 105-120.
Thompson, J. D., Higgins, D. G. and Gibson, T. J. (1994). CLUSTAL W: Improving
the sensitivity of progressive multiple sequence alignment through sequence weighting,
position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22, 4673-
4680.
Tong, C. K., Brion, L. P., Suarez, C. and Chesler, M. (2000). Interstitial carbonic
anhydrase (CA) activity in brain is attributable to membrane-bound CA type IV. J.
Neurosci. 20, 8247-8253.
Turbeck, O. B. and Foder, B. (1970). Studies on a carbonic anhydrase from the midgut
epithelium of larvae of lepidoptera. Biochim. Biophys. Acta. 212, 139-149.
Tusnady, G. E. and Simon, I. (1998). Principles governing amino acid composition of
integral membrane proteins: application to topology prediction. J. Mol. Biol. 283. 489-
506.
Tusnady, G. E. and Simon, I. (2001). The HMMTOP transmembrane topology
prediction server. Bioinformatics 17, 849-50.
Vince, J. W. and Reithmeier, R. A. (2000). Identification of the carbonic anhydrase II
binding site in the Cl(-)/HCO(3)(-) anion exchanger AE1. Biochemistry 39, 5527-5533.

146
Vince, J. W., Carlsson, U. and Reithmeier, R. A. (2000). Localization of the Cl-
/HC03- anion exchanger binding site to the amino-terminal region of carbonic
anhydrase II. Biochemistry 39, 13344-13349.
Volkmann, A. and Peters, W. (1989a). Investigations on the midgut caeca of mosquito
larvae I. Fine structure. Tissue & Cell 21,243-251.
Volkmann, A. and Peters, W. (1989b). Investigations on the midgut caeca of mosquito
larvae II. Functional aspects. Tissue & Cell 21, 253-261.
Waheed, A., Okuyama, T., Heyduk, T. and Sly, W. S. (1996). Carbonic anhydrase IV:
purification of a secretory form of the recombinant human enzyme and identification of
the positions and importance of its disulfide bonds. Arch. Biochem. Biophys. 333,432-
438.
Westerfield, M. (1994). The zebrafish book: A guide for the laboratory use of zebrafish
(Brachydanio rerio). (ed. Westerfield, M.), University of Oregon Press, Eugene, OR.
pp. 9.16-9.21.
Wieczorek, H., Gruber, G., Harvey, W.R., Huss, M. and Merzendorfer, H. (1999).
The plasma membrane H+-V-ATPase from tobacco homworm midgut. J. Bioenerg.
Biomembr. 31, 67-14.
Wieczorek, H., Gruber, G., Harvey, W. R., Huss, M., Merzendorfer, H. and Zeiske,
W. (2000). Structure and regulation of insect plasma membrane H(+)V-ATPase. J. Exp.
Biol. 203 (Pt 1), 127-135.
Zhang, Y. and Frohman, M. A. (1997). Using rapid amplification of cDNA ends
(RACE) to obtain full-length cDNAs. Methods Mol. Biol. 69, 61-87.
Zhuang, Z., Linser, P. J. and Harvey, W. R. (1999). Antibody to H(+) V-ATPase
subunit E colocalizes with portasomes in alkaline larval midgut of a freshwater
mosquito (Aedes aegypti). J. Exp. Biol. 202 (Pt 18), 2449-2460.

BIOGRAPHICAL SKETCH
Theresa J. Sern was bom and raised in Connecticut with her older sister and
younger brother. She enjoyed the outdoors and wildlife from a very young age and
continues to pursue activities such as scuba diving and kayaking. She received her
Bachelor of Science degree from the University of Connecticut in 1995. She then
ventured off to Saint Thomas in the U.S. Virgin Islands, where she lived for twelve
months to satisfy her enthusiasm for traveling and diving. Upon returning to the United
States, she was employed by Boehringer-Ingelheim Pharmaceuticals in Danbury,
Connecticut, where she was introduced to the world of scientific research and discovery.
This path was strengthened by a relocation to Miami, Florida, and a second research
position at Noven Pharmaceuticals. In 1997, she moved to Gainesville, Florida, to pursue
an advanced degree at the University of Florida. She became acquainted with The
Whitney Laboratory in Saint Augustine, Florida, when she was chosen to participate in a
summer research program. The following fall she was admitted to the University of
Florida graduate school in the Department of Fisheries and Aquatic Sciences. Here she
was able to combine her research background with her enthusiasm for aquatic animals.
Her research project was carried out at The Whitney Laboratory under the direction of
Dr. Paul Linser. The project focused on the enzyme carbonic anhydrase and its unknown
role in larval mosquito physiology. Upon completion of her doctorate degree she plans to
continue scientific research within the aquatic realm.
147

I certify that I have read this study and that in my opinion it conforms to acceptable
standards of scholarly presentation and is fully adequate, in scope and quality, as a
dissertation for the degree of Doctor of Philosophy.
Edward J. Phlips, Chair
Professor of Fisheries and Aquatic Sciences
I certify that I have read this study and that in my opinion it conforms to acceptable
standards of scholarly presentation and is fully adequate, in scope and quality, as a
dissertation for the degree of Doctor of Philosophy.,*
Linser, Coer
1 Professor of Anatomy and Cell Biology
I certify that I have read this study and that in my opinion it confor
standards of scholarly presentation and is fully adequate, in scoj
dissertation for the degree of Doctor of Philosophy.
ceptable
, as a
L. Moroz
Assistant Professor of Neuroscience
I certify that I have read this study and that in my opinion it conforms to acceptable
standards of scholarly presentation and is fully adequate, in scope and quality, as a
dissertation for the degree of Doctor of Philosophy.
Assistant Professor of Fisheries and Aquatic
Sciences
I certify that I have read this study and that in my opinion it conforms to acceptable
standards of scholarly presentation and is fully adequate, in scope and quality, as a
dissertation for the degree of Doctor of Philosophy
Robert M.
Associate Professor of Neuroscience

This dissertation was submitted to the Graduate Faculty of the College of
Agricultural and Life Sciences and to the Graduate School and was accepted as partial
fulfillment of the requirements for the degree of Doctor nf Philosophy
El '
May 2004
Dean, Collegof Agricul
Sciences
Dean, Graduate School



117
This CA displays all of the 13 highly conserved residues found in most other
active CA proteins, including the three necessary histidine residues required for the
binding of a zinc atom (Tashian, 1992; Sly and Hu, 1995; Tamai et al., 1996a). Figure 6-
1 shows an alignment of the active sites of An. gambiae, D. melanogaster, and human
CA proteins. The active site region of the cytosolic CA from An. gambiae contains one
amino acid difference from all of the mammalian isoforms. This particular residue (C-
89; An. gambiae CA numbering), is replaced by a serine (S) residue in all of the known
human CA isoforms (including mouse CA XIII in lieu of the uncharacterized human CA
XIII). This C to S amino acid change is also found in 1 of the 14 CA isoforms predicted
from the D. melanogaster genome (accession number CGI 1284-PA; Fig. 6-1).
The fourteen CA genes of humans are apparently required to perform the many
metabolic functions that complex organisms require. The recent release of the sequenced
genomes of two insect species, An. gambiae and D. melanogaster has revealed that these
organisms also have fourteen CAs. Comparisons based on amino acid composition
revealed that these 14 dipteran CAs are probably not homologs to each of the fourteen
human CAs. A phylogenetic analysis of the CA protein sequences was performed using
the Neighbor-Joining method (Saitou and Nei, 1987) as implemented in DNAman
software. A rooted tree shows the relationship between the human, mouse, and dipteran
CAs (Fig. 6-2). The human and mouse CAs cluster together, and the dipteran CAs
cluster together, however the mammalian and the dipteran proteins cluster separately. A
bootstrapping test was performed to determine the confidence value of the phylogenetic
tree. The An. gambiae CA that has the S to C difference within the active site pairs with
the D. melanogaster CA with the same difference (Fig. 6-2). These two dipteran CAs


18
methazolamide to a final concentration of 10'6 M. Recombinantly expressed and purified
mosquito CAs were also tested for activity using this assay.
Isolation of RNA and Synthesis of cDNA
Total RNA was isolated from freshly dissected fourth instar mosquito larval
midguts using TRI Reagent (Molecular Research Center Inc., Cincinnati, Ohio)
according to the manufacturer's instructions. Briefly, 100 Ae. aegypti gut epithelial
organs, including fore-, mid-, and hindgut (approximately 20 mg) were dissected in HSS
and transferred to a sterile microcentrifuge tube containing TRI Reagent (600 pi). The
tissue was homogenized and incubated for 5 min at room temperature. The homogenate
was then extracted with chloroform (40 pL) and precipitated with isopropanol (100 pL).
The RNA pellet was washed with 75% ethanol (200 pL), air-dried and resuspended in 50
pL diethylpyrocarbonate (DEPC; Sigma-Aldrich)-treated H2O. RNA concentrations
were calculated from the absorbance at 260 nm. Total RNA (10 pg) was reverse-
transcribed for 2 hours at 42C in a 20 pi reaction mixture using 5 pmol of oligo(dT)12-
18, RNasin (1:40 dilution), IX first strand buffer, 1 mM dNTPs, and 200 units (U) of
Superscript II reverse transcriptase (Invitrogen Inc., Carlsbad, California). This cDNA
was used to clone the first fragment of Ae. aegypti CA.
Bioinformatics
The National Center for Biotechnology Information (NCBI) website
(www.ncbi.nlm.nih.gov) was used for the majority of the bioinformatical data presented
in this study. The first mosquito genome, An. gambiae, was released in 2002 (Holt et al.,
2002), and made accessible to the public on the NCBI website. The basic local alignment
search tool (BLAST; Altschul et ah, 1990) was employed for primer construction as well


124
Figure 6-3. Localization of An. gambiae CA activity. Using CA histochemistry. CA
activity was localized to the cardia, gastric caeca, and posterior midgut
regions. A. A whole mount view of the entire gut region. B. A higher
magnification view of the gastric caeca and cardia regions that displayed CA
activity. Trachea are indicated with arrowheads. C. A higher magnification
view of the AMG to PMG boundary. The PMG displayed epithelial CA
activity while the AMG does not. The trachea are indicated by arrowheads.
D. Specific cells of the cardia along with cells of the AMG (arrows) displayed
CA staining. E. Small cells within the rectum also displayed CA activity
(arrows). Scale bars represent 300 pm in A, 100 pm in B and C, and 150 pm
in D and E.


CARBONIC ANHYDRASES AND BICARBONATE TRANSPORT
IN LARVAL MOSQUITOES
By
THERESA J. SERON
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
2004


CHAPTER 6
CYTOSOLIC CA EXPRESSION IN LARVAL ANOPHELES GAMBIAE
Introduction
Carbonic anhydrase (CA) represents a superfamily of enzymes that reversibly
hydrates carbon dioxide to form bicarbonate and a proton. There are three families of
CAs (a, P, and y). Mammals have fourteen different CA isoforms, all belonging to the a
CA family. The vast characterizations of the mammalian CAs have revealed cytosolic,
membrane-bound, membrane-spanning, and mitochondrial isoforms. Interestingly, it is
becoming increasingly apparent with the recent availability of genome sequences, that
less complex organisms also contain a large array of CAs. The Anopheles gamhiae
genome contains at least 14 putative CA genes (EnsEMBL protein family
ENSF00000000228). Whether the 14 well-characterized mammalian CAs serve the same
functions as the 14 mosquito genes is intriguing. This question will remain unanswered
until all of the CAs from An. gambiae or a similarly distant species are characterized. We
can, however, speculate as to their relatedness through sequence identity and
conservation.
Carbonic anhydrase is an interesting enzyme to study in the context of the
mosquito for several reasons. Unlike most animals, mosquitoes use a highly alkaline
digestive strategy instead of an acid environment. Additionally, the functions of the
mosquito gut are similar to the mammalian kidney in filtering wastes and maintaining
ionic homeostasis. Uncovering the function of CAs in an insect, such as the mosquito,
115


6 CYTOSOLIC CA EXPRESSION IN LARVAL ANOPHELES GAMBIAE 115
Introduction 115
Results 116
Anopheles gambiae CA Sequence Analysis 116
Localization of CA Activity in Anopheles gambiae Larvae 118
Localization of Cytosolic CA mRNA in Anopheles gambiae Larvae 118
Antibody Localization of CA Protein 119
Bacterial Expression and Purification of Anopheles gambiae
Cytosolic CA 119
Discussion 120
7 CONCLUSIONS AND FUTURE DIRECTIONS 131
Conclusions 131
New Model 134
Future Directions 137
REFERENCES 140
BIOGRAPHICAL SKETCH 147
vii


114
Expression and Inhibition of AgAE1
Voltage (V)
Figure 5-19. Inhibition of AgAEl mediated chloride transport by DIDS. A. DIDS
blocks the transport of chloride by AgAEl. The result is similar to taking
chloride out of the bath solution. B. There is almost no difference between
blocking chloride transport with DIDS and removing chloride. In contrast,
removing chloride from the uninhibited exchanger shows a large difference
in chloride transport.


CHAPTER 3
CARBONIC ANHYDRASE IN THE MIDGUT OF LARVAL AEDES AEGYPTI:
CLONING, LOCALIZATION, AND INHIBITION1
Introduction
Bicarbonate (and ultimately carbonate) ions are produced in vivo primarily by the
enzymatic action of carbonic anhydrase (CA). Its activity contributes to the transfer and
accumulation of FT1' or HCO3' in bacteria, plants, vertebrates and invertebrates. Although
there are innumerable reports related to the isolation of CA from vertebrates, studies
involving CA from invertebrates are very rare and there are no reports of the isolation of
CA from adult or larval mosquitoes.
There is strong immunohistochemical (Zhuang et al., 1999) and physiological
(Clark et al., 1999; Boudko et al., 2001b) evidence that an electrogenic, basal H+ V-
ATPase energizes luminal alkalinization in the anterior midgut of the larval mosquito by
producing a net extrusion of protons out of the lumen and a hyperpolarization of the basal
membrane. In contrast, ET V-ATPase appears to be localized in the apical membrane of
the posterior midgut and gastric caeca providing a reversed FT*"- pumping capacity relative
to the anterior midgut (Zhuang et al., 1999). A system capable of generating a high
luminal pH is likely to be buffered by carbonate (CO32), which has a pKa of
approximately 10.5.
This chapter was slightly modified and reprinted with permission from The Company of
Biologists LTD. Corena, M. P., Sern, T. J., Lehman, H. K., Ochrietor, J. D., Kohn,
A., Tu, C. and Linser, P. J. (2002). Carbonic anhydrase in the midgut of larval Aedes
aegypti: cloning, localization and inhibition. J. Exp. Biol. 205, 591-602.
39


34
Anion Exchanger Oocyte Expression
The full-length anion exchanger (AE) sequence was subcloned into the pXOOM
vector, which is optimized for both oocyte and mammalian expression (Jespersen et al.,
2002; a generous gift from Dr. T. Jespersen). In addition to a T7 RNA polymerase
promoter, this vector contains Xenopus-sptciTic 5 and 3 UTR sequences flanking the
insert in both directions. cRNA synthesis was performed using the T7 mMessage
mMachine kit (Ambion, Austin, Texas), after the cDNA was linearized using PMEI.
One day after surgical removal of the eggs from the frog, the eggs were injected
with either AE cRNA or water (control). After injection the eggs were incubated at 16C
for 4 days, long enough for measurable protein production and expression. The oocytes
were maintained in ND96 (96 mM NaCl, 2 mM KC1, 1 mM MgC^, 10 mM HEPES, pH
7.4 with NaOH). The medium was changed daily and dead oocytes were removed.
Anion Exchanger Physiology
Expression of the An. gambiae AE was examined using 2-electrode voltage clamp
electrodes. The voltage electrodes were pulled using 1.2 mm glass (M1B120F-3, World
Precision Instruments), and showed resistances between 1 -2 Mil Oocytes were clamped
to -50 mV and stepped from -90 mV to +70 mV in 10 mV increments. The wTater
injected eggs served as the control in evaluating any activity exerted by endogenous
proteins found in the Xenopus oocytes. Several different solutions were used to
determine the exchangers functional activities (refer to table 2-2). The transporter
blockers, 4,4-diisothiocyanodihydrostilbene-2,2-disulfonate (DIDS, Calbiochem, La
Jolla, California) and niflumic acid ( Sigma-Aldrich) were used to inhibit the transporter
capabilities of the expressed AE1 protein.


37
Aedes CA Primer Linearization
BrCA equations:
y = -3.2655k 39.958
R2 = 0.9776
18s equations:
y -3.1312x + 27.781
R2 = 0.9931
BrCA primers GC
18s primers WG
Figure 2-1. Efficiency plots for real-time PCR primers. Serially diluted cDNA samples
were tested with each primer set to determine the efficiency of
amplification. A linear regression was performed to determine the slope and
intercept for each primer set. These values were then used in an algorithm
to compare cDNA concentrations within the samples.


19
as analyzing PCR products. The NCBI Blast Flies database
(www.ncbi.nlm.nih.gov/BLAST/Genome/FlyBlast.html), together with the Ensembl
database (www.ensembl.org/Anopheles gambiae/) were used to predict the number of CA
genes in the Drosophila melanogaster and An. gambiae genomes by inputting the Ae.
aegypti CA as the search sequence. These partial sequence results were then annotated to
reflect the 2 full-length CA sequences that we have cloned from An. gambiae and
presented within this manuscript.
Ensembl is a joint project between the European Bioinformatics Institute and the
Sanger Institute to bring together genome sequences with annotated structural and
functional information. The NCBI protein database (pdb) and the BLAST were used in
conjunction with the 3-dimensional structure viewer (Cn3D; Hogue, 1997) for the
prediction of antibody accessible peptide regions in mosquito proteins. BLAST analyses
also confirmed that the chosen antigenic peptides were unique. The conserved domain
database (CDD; Marchler-Bauer et al., 2002) and the conserved domain architecture
retrieval tool (CDART; Geer et al., 2002) were used to predict the function of our newly
cloned mosquito proteins. Alignments were produced using Clustal W (Thompson et al.,
1994), as implemented in DNAman software (Lynnon Biosoft, Vaudreuil, Quebec,
Canada).
Cloning of CA from Aedes aegypti Larval Midgut
Degenerate oligonucleotides were designed against the regions of conserved
amino acids among CA proteins as determined by the BLAST analysis of several
vertebrate and two putative, but annotated, CA proteins from the D. melanogaster
sequence database.


68
discovery. This immunoreactive subset of CA-containing muscle fibers traverses the
cells that surround the highly alkaline anterior gut lumen. Determining the role of these
CA-specific muscle fibers holds promise for deciphering the necessary CA component of
mosquito gut alkalization.
Phospholipase C Treatment
In order to validate the CA IV-like isoform cloned from Ae. aegypti is indeed GPI
linked to the membrane, live fourth instarle, aegypti larvae were subjected to
phosphoinositol-specific phospholipase C (PI-PLC) treatment and subsequent
immunohistochemistry. This compound is capable of breaking the GPI-anchor and
therefore severs GPI-linked proteins from the plasma membrane. Larvae subjected to PI-
PLC treatment showed a decrease in CA immunoreactivity along the midgut muscle
fibers, as compared to the non PI-PLC treated controls (Fig. 4-11). This evidence
supports the prediction that the mosquito CA IV-like isoform is in fact GPI-linked to the
outer plasma membrane.
Discussion
In this study, we show that two GPI-linked CAs are expressed in the midguts of
two different mosquito species that rely on an alkaline digestive strategy. These
mosquito CAs share characteristics with the mammalian CA IV isozyme, including the
GPI link to the membrane. In situ hybridization localized CA message predominantly to
the gastric caeca and posterior midgut epithelial cells, as well as muscle and nerve fibers
along the anterior midgut, and CNS ventral ganglia. Real time PCR analyses confirmed
the presence of CA message within the Ae. aegypti gut and CNS. The gastric caeca were
found to contain the greatest amount of CA message in relation to the other gut samples


42
Carbonic Anhydrase Activity and Alkalization
A classic CA inhibitor methazolamide, was tested in live fourth instar larvae to
examine the influence of CA on the maintenance of the pH extremes inside the midgut,
and the effect of the enzyme on the net alkalinization of the growth medium by the intact
animals. Previous investigations have shown that living mosquito larvae excrete
bicarbonate, which results in the net alkalization of their surrounding aqueous medium
(Stobbart, 1971). Equal numbers of living larvae of equivalent age and size were placed
in culture plate wells containing lightly buffered medium and the pH indicator BTB. The
tissue culture plates used in this assay were scanned before and after addition of various
concentrations of methazolamide. In the absence of methazolamide, the blue color of the
medium, indicating a pH of at least 7.6, was maintained (Stobbart, 1971). Actual
measurement of the pH in each well showed a slow increase over time (data not shown).
Upon addition of methazolamide, the culture medium slowly became acidic, with a
resulting change in color to yellow as the pH dropped below 7.6 (Figure 3-1). All of the
controls that did not contain methazolamide remained blue. Addition of methazolamide,
at various concentrations, to wells containing only medium with BTB (no mosquito larva
control) remained blue. These data show that CA activity is present in the living larvae
and that it plays some role in acid/base excretion.
Moreover, fourth instar larvae cultured in BTB-containing medium ingest the dye,
which can then be used as a visible indicator of the pH in the gut lumen. Treatment of
the cultured larvae with methazolamide showed a direct impact of inhibited CA activity
on gut luminal pH. Figure 3-2 compares the luminal pH of dissected larval midguts with
and without a 5 hour exposure to methazolamide. The micrographs reveal that


110
Figure 5-15. Antibody localization of AgAEl protein to the gastric caeca in An. gambiae
larvae. A. Our AE specific antibody displays immunoreactivity within the
cardia (*) and gastric caeca. B. Phalloidin was used to label the actin-
containing muscle fibers throughout the mosquito gut. C. Draq-5 was used
to label nuclear DNA. D. Three signal overlay depicting the AE protein in
relation to muscles and cell nuclei. E-H shows higher magnification views
of the gastric caeca with the same labeling profile as in A-D. AE protein
expression can be seen on plasma membranes (arrows) of the gastric caeca
(E). Scale bars represent 25 pm in A-D and 100 pm in E-H.


Blood
(Acidic)
pH 6.5
Caterpillar gut call
(Neutral)
pH* 7.0
Lumen
(Alkaline)
pH* 11.0
13
H*V-ATPaee Q Anion exchanger
Cation exchanger Carbonic anhydraae
Amino add: K+ cotranaportor
Channel
Figure 1-3. Preliminary mosquito anterior midgut model based on M. sexta. This
theoretical model places a CA II-like isoform within the cell cytosol where
it combines carbon dioxide and water to form bicarbonate and a proton.
Alkalization is driven by a proton pumping V-ATPase that resides in the
apical membrane and pumps protons into the lumen. A chloride/
bicarbonate exchanger, that is also located in the apical membrane,
exchanges bicarbonate from the CA, for chloride from the lumen. A cation
exchanger transfers potassium to the lumen while stripping protons from the
bicarbonate for the exchange. The potassium ion combines with the de-
protonated carbonate ion to form potassium carbonate, which brings the pH
to highly alkaline levels.
K*pump


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EESN3FXSQ_J3NTFC INGEST_TIME 2015-04-01T19:43:44Z PACKAGE AA00029856_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


100
Anoph AE1
Human AE1
Human AE2
Human AE3
Human AE4
GDEMAWKETARWVKFEEDVEEGG
NQELRWMEAARWVQLEENLGENG
NQEPdWRE TARWIKFEE DVE E|E T
soephIwretarwirfeedveeJet
SITLSTHLHHRWVLFEEKLEVAA
* *
Figure 5-5. Alignment of An. gambiae and human AEs. pH sensitivity of AE2 (P04920)
and AE3 (NP005061) was mapped to the boxed 16 amino acids shown (Vince
et al., 2000). Unlike AE3, AE1 (P02730) was not stimulated by intracellular
alkalization (Vince et al., 2000). The An. gambiae AE1 (AAQ21364)
sequence shows a strong similarity to the identified amino acid sequence and
has 14 identical residues. The human AE1 sequence has only 8 identical
residues. pH sensitivity of an AE within the mosquito gut would be an
important attribute due to the regional compartmentalization of the pH flux.
The recently identified AE4 (Q96Q91) was included in this alignment for
completeness. It displays the least conservation with only 6 identical residues.
Stars indicate identical residues within all aligned sequences.


123
0.05
I I
92
95
96
IOC
IOC
IOC
94
IOC
IOC
IOC
IOC
IOC
IOC
IOC
IOC
BAA85002 CA-XIV
NP035927 MCA-XIV
AAH23981 CA-XII
NP848483 MCA-XII
AAH14950 CA-IX
- NP647466 MCA-IX
CAC42429 CA-VI
NP033932 MCA-VI
P22748 CA-IV
NP031633 MCA-IV
Anoph CA II-LIKE
CG11284-PA DROS CA
10Cj
XP134293 MCA-VII
10C
ioc
P43166 CA-VI I
NP078771 MCA-XIII
XP372051 CA-XIII
P00915 CA-I
P13634 MCA-I
" P07451 CA-III
P16015 MCA-III
P00918 CA-II
NP033931 MCA-II
AAB47048 CA-V
NP031634 MCA-VA
r NP031618 MCA-VI 11
JN0576 CA-VI 11
IOC
Anoph CA IV-LIKE
Aedes CA IV-LIKE
CG3940-PA Dros CA
Figure 6-2. Phylogenetic analysis between mammalian (human and mouse) and dipteran
{An. gambiae and D. melanogaster) CAs. This phylogenetic tree was
assembled using the Neighbor-Joining method. The tree has been rooted and
the bootstrapping confidence values are shown. The mammalian sequences
are shown in blue, MCA designates the mouse, while the dipteran .sequences
are shown in green. The mammalian CAs cluster separately from the dipteran
CA clusters, signifying a high divergence between mammalian and dipteran
CAs. There is also a high divergence between the dipteran CA -like
sequences and the dipteran CA IV-like sequences.


59
Figure 3-7. Hanssons histochemistry of whole mount Ae. aegypti gut. A. Intense dark
staining is observed in the cardia, gastric caeca (GC) and posterior midgut
(PMG), indicating CA activity. B. Higher magnification of the gastric
caeca. The distal lobes of the gastric caeca (Cap cells) exhibit relatively
low levels of reaction product, indicating lower levels of enzyme activity in
these cells relative to other cells of the gastric caeca. C. Higher
magnification of the PMG shows large, relatively unstained columnar cells
(*) contrasted with the smaller stained cuboidal cells (arrow). Scale bars
represent 150 pm in A and B, and 75 pm in C.


79
Fig 4-8. The Ae. aegypti CNS ganglia express the CA IV-like isoform. A. Pre-immune
serum does not show any detectable labeling of the CNS tissue. B. Strong
immunolabeling for the mosquito CA IV-like isoform is displayed in the
ventral ganglion clusters, as displayed by the fluorescent green coloring as
compared to the yellow control (pre-immune) ganglia. The scale bars
represent 100 pm.


88
trachea extend from the gastric caecal region and become incorporated with the anterior
midgut where they are intimately associated with nerve fibers (Fig. 5-9).
The third abdominal segment marks the beginning of the posterior midgut and is
located by the fourth pair of trachea that connect to this part of the midgut. This junction
displays a marked contrast in AgAEl mRNA transcript expression. Strong staining of
the epithelial cells begins here, coinciding with the beginning of the posterior midgut and
change in epithelial cell morphology (Fig. 5-10). Along with small epithelial cell
labeling, tracheal fibers also display strong label for AgAEl mRNA within the posterior
midgut. The larger type of epithelial cell within the posterior midgut, the columnar cell,
shows extensive labeling for AgAEl mRNA expression, with signal localized near the
plasma membranes (Fig. 5-1 OB). The extreme end of the posterior midgut displays
strong labeling within a cluster of small epithelial cells known as cuboidal cells (Fig. 5-
11). Cellular processes that extend rostral and lateral from these cells are also labeled.
All cells of the Malpighian tubules label positively for AgAEl transcript expression (Fig.
5-12).
AgAEl mRNA expression is also localized to the ventral midgut ganglion. Each
ventral ganglion displays specific labeling within one or two longitudinally directed
neurons that traverse the same plane (Fig. 5-13). No other neuronal cells display signal
for AE mRNA expression. The labeling is very specific for precisely one neuronal
pathway within each ganglion (Fig. 5-13). The sense (control) DIG probes display no
hybridization (Fig. 5-14).


CHAPTER 4
A GPI-LINKED CARBONIC ANHYDRASE EXPRESSED IN THE
LARVAL MOSQUITO MIDGUT
Introduction
The CA enzyme expressed in the midgut of larval mosquitoes shares some
characteristics with the mammalian CA IV isozyme, including a glycosyl-phosphatidyl-
inositol (GPI) link to the plasma membrane. Mammalian CA IV enzymes have been
found in dynamic organs such as kidney, lung, gut, brain, eye, and capillary endothelium
(Chegwidden and Carter, 2000). The human CA IV isoform was found to be as active as
the CA II isoform in carbon dioxide hydration and even more active in bicarbonate
dehydration (Baird et al., 1997). Studies of larval mosquito CAs are being pursued to
better understand the alkaline gut system. As the anterior midgut of the larval mosquito
lacks a highly active cytosolic CA II-like isozyme (previous chapter; Corena et al., 2002),
the presence of a highly active CA IV-like isozyme within the mosquito gut may be able
to provide the buffering capacity that is needed within the highly alkaline anterior
midgut. A more detailed characterization of larval Aedes aegypti CA is presented in this
study as well as the sequence of a homologous CA isoform from Anopheles gambiae.
New tools and techniques, such as the generation of a mosquito-specific CA antibody and
real time PCR, as well as improved methodology for in situ hybridization, have enabled
this further analysis.
61


106
A
¡
Figure 5-11. Larval An. gambiae displays strong AgAEl expression in the hindgut, the
pylorus. A. Distal to the joining of the Malpighian tubules with the gut, the
pylorus displays pronounced labeling of small epithelial cells (arrowhead)
and closely-associated muscle fibers. B. Higher magnification of the boxed
region displays labeled epithelial cells (arrowheads) of the pylorus with
closely associated circular muscle fibers (arrows) that form a pyloric
sphincter. The pylorus, a part of the hindgut, functions in ionic and osmotic
regulation. Scale bars represent 25 pm in A and B.


11
Midgut
I
GC
AMG
PMG
Hindgut
1 1
MT
Figure 1-1. Illustration showing the regions of the larval mosquito gut. The midgut is
composed of the cardia, gastric caeca (GC), anterior midgut (AMG), and
the posterior midgut (PMG). The hindgut is composed of the Malpighian
tubules (MT) and the rectum.


104
Figure 5-9. Anion exchanger mRNA localization reveals trachea and nerve fibers along
with neuronal cell labeling. A. Trachea (large arrow) display a random
pattern of distribution on the An. gambiae midgut along with pairs of neuronal
cells (small arrows). B. A labeled tracheal stalk (large arrow) shows abundant
labeling where it joins with the midgut (*). The finer branches of the trachea
can be seen joining (small arrow) the parallel nerve fibers (arrowhead) that
also display label. C. Neuronal cells (small arrows) scattered over the AMG
display AE label along with the nerve (arrowheads) and tracheal fibers (large
arrows). D. Strong labeling is consistently seen where the thick trachea
connects to the midgut (*) and sends out smaller trachiole fibers (arrows).
Scale bars represent 25 pm in A, 50 pm in B, 50 pm in C, and 50 pm in D.


TABLE OF CONTENTS
Eige
DEDICATION iii
ACKNOWLEDGEMENTS iv
LIST OF TABLES viii
LIST OF FIGURES ix
ABSTRACT xii
CHAPTER
1 INTRODUCTION 1
Alkaline Gut 2
Carbonic Anhydrase 4
Mosquito Development and Control 5
Carbonic Anhydrase Inhibition 6
Bicarbonate Transport 7
Gut Alkalization Model 8
Specific Aims 9
2 MATERIALS AND METHODS 14
Experimental Insects 14
Preparation and Fixation of Tissue 15
Bromothymol Blue Qualitative Assay 16
Effect of Methazolamide on the Alkalization of the Midgut of Live Larvae 16
ID # # #
0 Exchange Method to Measure Carbonic Anhydrase Activity 17
Isolation of RNA and Synthesis of cDNA 18
Bioinformatics 18
Cloning of CA from Aedes aegypti Larval Midgut 19
Construction of Amplified cDNA Pools 20
3' and 5' Rapid Amplification of cDNA Ends and Sequencing 23
Construction of In Situ Hybridization Probes 23
In Situ Hybridization 26
CA Histochemistry 27
Real Time PCR 27
v


43
alkalinization of the midgut was inhibited by methazolamide as shown by the color
change of the BTB indicator. Interestingly, the effect was most pronounced in the
anterior midgut, where the pH indicator changed from blue in the midgut of larvae reared
in the absence of inhibitor to yellow in as little as 30 minutes when methazolamide (10"6
M) was added to the culture. The indicator also changed color progressively from blue
through green to yellow in the gastric caeca (Figure 3-2). No apparent change was
observed in the posterior midgut. The color of the midgut in this region was yellow both
in the untreated larvae and in the larvae treated with methazolamide. Since the pH of the
posterior midgut has been associated with values close to 7.6, no change in color was
evident using this qualitative method.
lsO Isotope-Exchange Experiments
The relative activity of CA, normalized to total protein content, was calculated as
described by Silverman and Tu (1986). The relative activity of CA was highest in the
gastric caeca, followed by the posterior midgut and Malpighian tubules (Figure 3-3). The
relative activity of CA in the anterior midgut was either extremely low or non-existent,
falling at or below that of the buffer blank. The specificity of the reaction was confirmed
by complete inhibition with the addition of 1 O'6 M methazolamide (results not shown).
Cloning of Carbonic Anhydrase from Aedes Aegypti Larvae
We utilized a cDNA cloning strategy to obtain a specific carbonic anhydrase
cDNA from the midgut epithelial cells of the larval Ae. aegypti. A comparison of twelve
CA sequences, including two putative CA sequences that had been annotated but not
characterized in the Drosophila melanogaster databases, was made. We then produced
degenerate PCR primers from consensus regions of the CA gene family. The initial 297


62
Results
Bioinformatics of Aedes Aegypti CA
We have previously cloned a CA cDNA from the Ae. aegypti midgut (accession
number AF395662; Corena et al., 2002). Our initial structure prediction indicated that
the protein was cytosolic. However, further characterization has indicated that this CA is
actually membrane associated via a GPI-link. We have determined that the CA
propeptide sequence encodes an extracellular protein with a hydrophobic tail region. The
first 17 amino acids of the propeptide are predicted by the Simple Modular Architecture
Research Tool (SMART) program to be the signal sequence (Letunic et al., 2002). This
sequence flags the message for transport to the endoplasmic reticulum (ER). Using the
PSORT II server, the prediction of membrane topology (MTOP) indicates that the Ae.
aegypti CA sequence is GPI anchored. Amino acid G-276, is predicted by the GPI
prediction server (Eisenhaber et al., 1999) to be the site for GPI attachment. The
hydrophobic tail (L278-A289) allows translocation of the transcript through the ER
plasma membrane and is also predicted to stabilize the protein with the membrane until
the pre-formed GPI anchor is transferred to the protein. The hydrophobic tail is then
cleaved to produce a completely extracellular protein that is tethered to the cell by the
GPI link (for a review of GPI-linked proteins see Brown and Waneck, 1992).
Sequence Comparisons of CA IV-like Isoforms
I also cloned a CA IV-like cDNA from the gut of An. gambiae, an important
vector in the spread of malaria. This CA isoform (Ensembl gene ID:
ENSANGG00000018824, chromosome 2L) is partially predicted by the Ensembl CA
protein family (ENSF00000000228) as 1 of the 14 gene members found in the An.


80
Figure 4-9. Immunolocalization of mosquito CA IV-like enzyme in Aedes albopictus.
Muscle and nerve fibers within the anterior midgut region are heavily labeled.
A. Selective labeling of particular muscle fibers (green). B. Labeled
phalloidin (red) was used to localize actin and labels all muscle fibers,
including those that were not recognized by the antibody against mosquito
CA. C. A nuclear label (blue) was used to distinguish cell numbers present.
D. Overlay of all three signals. Colocalization of the green and red signals
appear yellow. The CA antibody recognizes only a subset of anterior muscle
fibers, and is seen in several different mosquito species. The scale bar
represents 50 pm.


55
12
GC AM PM MT
Figure 3-3. Relative activity of CA in different pooled segments of the midgut of larval
Ae. aegypti. Midguts were dissected from early fourth-instar larvae and
separated into gastric caeca (GC), anterior midgut (AM), posterior midgut
(PM) and Malpighian tubules (MT). The relative activity of CA was
measured using the l80 mass spectrometry method (Silverman and Tu,
1986), normalized to total protein content. The activity of the anterior
midgut was lower than that of the water blank and, thus, is set as zero
activity.


BIOGRAPHICAL SKETCH
Theresa J. Sern was bom and raised in Connecticut with her older sister and
younger brother. She enjoyed the outdoors and wildlife from a very young age and
continues to pursue activities such as scuba diving and kayaking. She received her
Bachelor of Science degree from the University of Connecticut in 1995. She then
ventured off to Saint Thomas in the U.S. Virgin Islands, where she lived for twelve
months to satisfy her enthusiasm for traveling and diving. Upon returning to the United
States, she was employed by Boehringer-Ingelheim Pharmaceuticals in Danbury,
Connecticut, where she was introduced to the world of scientific research and discovery.
This path was strengthened by a relocation to Miami, Florida, and a second research
position at Noven Pharmaceuticals. In 1997, she moved to Gainesville, Florida, to pursue
an advanced degree at the University of Florida. She became acquainted with The
Whitney Laboratory in Saint Augustine, Florida, when she was chosen to participate in a
summer research program. The following fall she was admitted to the University of
Florida graduate school in the Department of Fisheries and Aquatic Sciences. Here she
was able to combine her research background with her enthusiasm for aquatic animals.
Her research project was carried out at The Whitney Laboratory under the direction of
Dr. Paul Linser. The project focused on the enzyme carbonic anhydrase and its unknown
role in larval mosquito physiology. Upon completion of her doctorate degree she plans to
continue scientific research within the aquatic realm.
147


35
Aedes degenerate CA primers:
CA5F:
5'
GAR
CAR
TTY
CAY
TKY
CAY
TGG
GG
CA3R:
5'
GTI
ARI
SWN
CCY
TCR
TA
N=G,A,T,C; K=G,T; S=G,C; VI
/=A,T;
Y=C,
T; R=,
A,G
Amplified cDNA adaptor primers:
DAP:
5'
CGA
CGT
GGA
CTA
TCC
ATG
AAC
GCA
TRsa:
5'
CGC
AGT
CGG
TAC
TTT
TTT
TTT
TTT
T
Anopheles exai
ct
ZA prim
lers:
Ag1CA2F:
5'
CAG
TCA
CCT
ATC
GAC
CTA
AC
Ag1CA4R:
5'
CTC
GCG
TGT
TCA
ATG
GTT
G
Ag4CA11F:
5'
GGA
GGC
GTC
CTT
GGC
AAC
Ag4CA12R:
5'
CTG
CAC
TGA
CCG
GAA
GTT
G
Anopheles exa(
:ti
\E prim
ers:
AgAEIF:
5
CCT
GGA
AGG
AAA
CGG
CAC
G
AgAE4R:
5'
CCT
CGA
GCT
GGT
GCA
GAT
C
Aedes CA Real
tin
fie PCR
prime
rs:
5SPCAF1:
5
GCA
ACA
CTG
CTT
CCG
TCT
ACA
A
5SPCAR1:
5'
CCG
GTT
CGT
TAA
TAA
CTC
CAT
TG
18s RIBF:
5'
CGC
TAC
TAC
CGA
TGG
ATT
ATT
TAG
TG
18s RIBR:
5'
GTC
AAC
TTC
AGC
GAT
TCA
AAT
GTA
A
Aedes CA expn
ISl
lion pri
mers:
ExCAshortF:
5'
CACC
ATG
GAC
GAA
TGG
CAC
T
ExCAshortR:
5'
TTA
GTA
ATC
CAT
ATC
GGT
GTG
GT
Anopheles CA
ex
>ressioi
n prim
ers:
ExCA4F:
5'
CACC
ATG
GCA
TCA
AAA
ACA
ACA
AAG
CA4end:
5'
TTA
CAG
CTT
CGA
AAG
CAC
AAC
GG
Table 2-1. PCR primer sequences.


3
cells are capable of maintaining physiological homeostasis while facing a pH range of 7-
11 along the length of the mosquito gut lumen (Dadd, 1975). This range in pH, along the
length of the mosquito gut, is presumed to support digestive and assimilation functions
(Clements, 1992). The epithelial cells of the anterior midgut (AMG) surround a highly
alkaline lumen (pH 11) while those of the gastric caeca (GC) and posterior midgut
(PMG) surround a neutral to mildly alkaline lumen (pH 7-8; Clements, 1992; Zhuang et
al., 1999). The different pH values found along the midgut may support the various
metabolic functions that are active in each gut region. The gastric caeca perform ion and
water transport, the anterior midgut performs alkaline digestion, the posterior midgut
performs nutrient absorption, and the Malpighian tubules (part of the hindgut) actively
transport potassium and fluid (Clements, 1992).
The role of the alkaline pH in the anterior midgut is a point of some controversy.
It has been suggested that the high pH contributes to the digestion of plant detritus and, in
particular, to the dissociation of tannin-protein complexes (Martin et al., 1980). The high
pH restricts the conglomeration of proteins within the anterior midgut that could interfere
with the insects normal physiology. These complexes could also interfere with insect
digestion by blocking the active sites of many different digestive enzymes. Therefore,
the alkaline gut serves as a proposed benefit to the insects by allowing ingested food to
remain soluble. The alkalinity therefore keeps the gut free from attachable tannin-protein
complexes and enhances the assimilation of proteins. Berenbaums review (1980) of
Lepidopteran insects correlated gut pH (range from 7.0-10.3) with diet. Caterpillars
feeding on leaves containing tannins were found to display a more alkaline pH (average
pH 8.76) than those feeding on low tannin diets (average pH 8.25; Berenbaum, 1980).


96
Multiple Alignment
Figure 5-1. Structural prediction of the An. gambiae AE1. A. Hydrophobicity plot of the
DNAman-aligned D. melanogaster NDAE1 and An. gambiae AE1 sequences
suggests a nearly identical protein topology of 12 membrane-spanning
domains in both proteins. B. Illustration depicting the intracellular location of
both protein termini as well as the predicted 12 transmembrane domains.


97
AgAEl [AAQ21364]
Human AE1[P02730]
Human AE2[P04920]
Human AE3[NP005061]
Human AE4[Q96Q91]
LDYIFTKRELKILDDIMPEMTKRARADDLHQLEDGEVG
LPLIFRNVELQCLDADDAKATFDEEEGRDEYDEVAMPV
LTRIFTDREMKCLDANEAEPVFDEREGVDEYNEMPMPV
LPRLFQDRELQALDSEDAEPNFDE-DGQDEYNELHMPV
LERVFSPQELLHLDELMPEEERSIPEKGLEPEHSFSGS
* * **
Figure 5-2. Putative amino terminus CAII binding motif. The highlighted conserved
leucine (L) was shown to be necessary for the specificity of CA binding in
AE1 (Vince et al., 2000) and is also present in AgAEl. The motif consists
of at least two acidic amino acids within the four residues following the
conserved non-polar L. The ability of AEs to bind CA enzymes greatly
raises their ability to regulate ionic homeostasis. Identical residues (*) are
noted in the alignment as well as conserved residues (.).


107
Figure 5-12. Localization of AE mRNA in An. gambiae shows abundant labeling of the
Malpighian tubules. A. The entire length of the Malpighian tubules displayed
labeling of AE mRNA. B. Labeled tracheal fibers are also associated with the
Malpighian tubules. C. Labeled tracheal fibers also extend from the tips of
Malpighian tubules and may contain secretory vesicles (arrow). Scale bar
represents 25 pm in A, B. and C.


Antibody Production 29
Immunohistochemistry 30
CA Protein Expression 32
Anion Exchanger Oocyte Expression 34
Anion Exchanger Physiology 34
3 CARBONIC ANHYDRASE IN THE MIDGUT OF LARVAL AEDES
AEGYPTI: CLONING, LOCALIZATION, AND INHIBITION 39
Introduction 39
Results 40
Bromothymol Blue Qualitative Assay 40
Carbonic Anhydrase Activity and Alkalization 42
l80 Isotope-Exchange Experiments 43
Cloning of Carbonic Anhydrase from Aedes aegypti Larvae 43
Localization of the Enzyme in the Midgut Epithelium: Carbonic
Anhydrase Enzyme Histochemistry 45
In Situ Hybridization 46
Discussion 46
4 A GPI-LINKED CARBONIC ANHYDRASE EXPRESSED IN THE
LARVAL MOSQUITO MIDGUT 61
Introduction 61
Results 62
Bioinformatics of Aedes aegypti CA 62
Sequence Comparisons of CA IV-like Isoforms 62
Localization of CA IV-like Isoform in the Mosquito Midgut 65
Real Time PCR Analysis of Aedes aegypti CA IV-like Transcripts 65
Immunolocalization of CA IV-like Protein in the Mosquito Gut 66
Antibody Cross-Reactivity with Other Mosquito Species 67
Phospholipase C Treatment 68
Discussion 68
5 ANION EXCHANGER EXPRESSED WITHIN THE LARVAL
ANOPHELES GAMBIAE MOSQUITO 83
Introduction 83
Results 84
An. gambiae AE Sequence Analysis 84
BT Sequence Comparisons 86
Localization of Anion Exchanger mRNA in An. gambiae Larvae 87
Antibody Localization of AE Protein 89
AE Functional Expression in Oocytes 89
Discussion 91
vi


CHAPTER 2
MATERIALS AND METHODS
Experimental Insects
Ae. aegypti eggs were obtained from a colony maintained by the United States
Department of Agriculture (USDA) laboratory in Gainesville, Florida. The eggs were
allowed to hatch in 20 ml of 2% artificial seawater (ASW; 8.4 mM NaCl, 1.7 mM KC1,
0.1 mM CaCl2,0.46 mM MgCl2, 0.51 mM MgS04, and 0.04 mM NaHC03). The
mosquito larvae were reared in 2% ASW at room temperature. The Ae. aegypti larvae
were fed a mixture of yeast and liver powder (1:1.5 g respective dry weight; ICN
Biomedicals Inc., Aurora, Ohio). Eight to ten days were required for this species to reach
the early fourth instar.
An. gambiae eggs were obtained from the Centers for Disease Control and
Prevention (CDC) in Atlanta, Georgia. Strict handling guidelines were followed with this
particular species, which does not currently inhabit Florida, due to its inherent ability to
acquire and transmit the Plasmodium protozoan, which causes malaria. This Anopheles
species was therefore reared in deionized water inside of a locked incubator set at 30C.
A mesh screen served as a second barrier within the incubator while the sealed (but not
airtight) containers harboring the An. gambiae larvae served as the third barrier against
escape. The An. gambiae larvae were fed a Wardley tropical fish flake food (The Hartz
Mountain Corp., Secaucus, New Jersey). Early fourth instar larvae were chosen for all
experiments. Ten to twelve days from the hatch day were required for this species to
14


54
Figure 3-2. Effect of methazolamide on the alkalization of the midgut using
Bromothymol Blue (BTB) assay of pH within living, but isolated, gut
tissue. Gut tubes were dissected after pre-loading with BTB and then
incubated for 5 hours in hemolymph substitute (Clark et al., 1999)
in the absence (A) or presence (B) of 1 O'6 M methazolamide. The loss of
blue coloration in B shows that the internal pH of the gut lumen has
dropped below 7.6. Scale bar represents 300 pm.


52
mM in the midgut lumen in larval Hyalophora cecropia (Turbeck and Foder, 1970).
Recent measurements using capillary zone electrophoresis of larval Ae. aegypti fluids
have revealed a bicarbonate/carbonate level as high as 50.84.21 mM in the midgut
lumen compared with 3.962.89 mM in the hemolymph (Boudko et al., 2001a). These
values correlate with those observed by Turbeck and Foder (1970). This combined
evidence suggests that the CO2 that reaches the midgut lumen in the larvae of
lepidopterans is rapidly converted to a mixture of bicarbonate and carbonate. The role of
CA in the alkalization process would be of great significance. The generation of
antibodies against A-CA will facilitate a detailed analysis of the cellular and subcellular
distribution of this key enzyme in this system.


66
dissected and the head, gastric caeca (GC), anterior midgut (AMG), posterior midgut
(PMG), and Malpighian tubules (MT) were pooled. RNA was isolated from each tissue
sample for subsequent real time PCR analysis. Ae. aegypti ribosomal RNA (Genbank
accession number M95126) was used to normalize the quantity of transcript from each
sample. The results are presented in graph format in figure 4-6. Gastric caeca contain
the greatest quantity of CA message within the gut sections (Fig. 4-6). The head tissue
contained roughly half as much message as the gastric caeca (Fig. 4-6). The localization
of CA rV-like message within the larval head is consistent with the localization of CA
message to CNS tissue by in situ hybridization. The anterior midgut, posterior midgut,
and Malpighian tubule collections showed CA message only marginally greater than zero
(Fig. 4-6).
Immunolocalization of CA IV-like Protein in the Mosquito Gut
The amino terminal peptide sequence (GVINEPERWGGQCETGRR) was chosen
from the Ae. aegypti CA sequence as an antigen for antibody production. The resultant
antiserum was used to immunolocalize the CA IV-like isoform within the mosquito gut.
The pre-immune serum was used as a control for all experiments. Immunoreactivity was
found within the gastric caeca region of the gut as well as on muscle fibers along the
anterior midgut (Fig. 4-7A). A subset of anterior muscle fibers displays the strongest and
most striking labeling on their extracellular surface, while other muscle fibers show little
or none. Immunoreactivity was also found within the CNS ganglia and immunoreactive
nerve fibers that traverse the gut (Fig. 4-8). There was no immunoreactivity detected in
the Malpighian tubules.


2
(Gubler, 1997). Another mosquito example, Anopheles gambiae kills millions of people
each year in Africa by infecting them with the deadly Plasmodium parasite that causes
malaria. Many studies have therefore been undertaken to understand the life cycle and
physiology of these insects that exert such a large socio-economic impact.
The mosquitos ability to acquire, harbor and transmit deadly pathogens has
spurred research into the workings of the mosquito gut. Specific cells of the midgut,
which express a proton pumping V-ATPase, have been found to be preferentially invaded
by pathogens (Shahabuddin and Pimenta, 1998). Studies have also shown that the
mosquito gut is not a static organ but is comprised of several different regions. Each
region displays different characteristics and is made up of different cell types.
Alkaline Gut
Larval mosquitoes, as well as some caterpillars, are known to possess a highly
alkaline digestive system (Dadd, 1975). The tobacco homworm, M. sexta, has a gut
lumen pH that can exceed 11, while the larval mosquito, Aedes aegypti, displays a pH
greater than 10 in its anterior midgut region (Zhuang et al., 1999). These insects are not
only unharmed by this caustic pH, but are able to generate this property' while
maintaining cellular homeostasis.
The larval midgut is involved in ionic and osmotic regulation as well as digestion,
absorption, and excretion (Clements, 1992). It is subdivided into four structurally
distinguishable regions: cardia, gastric caeca, anterior stomach, and posterior stomach
(Fig. 1-1). Each of these regions consists of one cell layer of epithelial cells, composed
of large columnar cells and much smaller cuboidal cells, which vary in character
somewhat from region to region. Belying this simple architecture however, the epithelial


85
This complex of BTs contains both the solute carrier 4A (SLC4) and solute carrier 26A
(SLC26A) proteins. More specifically, the EnsEMBL database places this particular AE
cDNA sequence within the anion exchange/ band3 protein family (ENSF00000000189)
as 1 of the 8 putative anion exchange band3 transcripts (ENSANGP00000010112)
encoded in the An. gambiae genome. These 8 transcripts arise from 3 different genes
(ENSANGG00000007623, ENSANGG00000004501, and ENSANG00000012483). The
gene that gives rise to the cloned AE that we are presenting (ENSANGG00000007623) is
located on chromosome 3R. The other 2 genes (ENSANGG00000004501 and
ENSANG00000012483) are located on chromosome 2L (Hubbard et ah, 2002; Clamp et
al 2003).
The 1102 amino acids comprising the An. gambiae AE form a cytosolic
framework at the amino terminus while the carboxy terminus is composed of 12
transmembrane spanning domains also with an intracellular cytosolic terminus (hmmtop
v.2; Tusnady and Simon 1998; Tusnady and Simon 2001). This hmmtop prediction was
generated based on two assumptions: 1), that the CA binding site is within the carboxy
terminus; and 2), that the C-terminus is intracellular, as is found for all known AEs. This
structure is consistent with the predicted structure of the Drosophila melanogaster
sodium dependent anion exchanger (NDAE1), which consists of a 12 membrane-
spanning pattern with intracellular carboxy and amino termini (Romero et al., 2000). The
highly conserved sequence identity of the An. gambiae AE1 with respect to the D.
melanogaster NDAE1 allows the predictive 12 transmembrane-spanning domains of the
D. melanogaster protein to be superimposed upon the An. gambiae protein (Fig. 5-1).


98
100% 80% 60% 40% 20% 0%
I I I I 1 1
Human AE2
Human AE1
Human AE3
Human NBC4a
Human AE4
Dros NDAE1
Anoph AE1
Human NBC8
Human SLC26A7
60%
58%
48%
45%
72%
53 A
36%
Human DRA
Human SLC26A6
22%
31%
11%
Figure 5-3. Homology tree depicting the amino acid identity between several BTs. The
An. gambiae AE1 amino acid sequence displays the closest identity to the D.
melanogaster NDAE1 (AAF98636) and human NBC8 (NP004849) sequences
with 72% and 53% identity, respectively. The human AEs display 36% to
45% identity while the sulfate transporters (SLC26 group) display only 11%
identity to AgAEl. Accession numbers: human AE2 (P04920), human AE1
(P02730), AE3 (NP005061), NBC4a (NP067019), AE4 (Q96Q91), SLC26A7
(NP439897), DRA (P40879), and SLC26A6 (Q9BXS9).


109
Figure 5-14. Sense AE probes display no specific hybridization. A. These whole mount
preparations show no AE sense (control) label in the integument, midgut or
hindgut (B). A higher magnification of the gastric caeca (C) and posterior
midgut (D) with Malpighian tubules (MT) shows no hybridization with the
sense AE probe. These experiments were performed side by side with the
antisense probes and therefore length of exposure was identical. Scale bars
represent 600 pm in A, 300 pm in B, 100 pm in C, and 100 pm in D.


53
Figure 3-1. Effect of CA inhibition on culture medium pH with fourth-instar Ae.
aegypti larvae. Mosquito larvae typically alkalize the medium in which
they are reared (Stobbart, 1971). (A) Six culture wells each containing five
fourth-instar larvae incubated for 5 hours in medium containing 0.003%
Bromothymol blue (BTB). The blue color is retained, indicating a pH
greater than 7.6. (B) The same as A, except that each well also contains a
different concentration of the CA inhibitor methazolamide ranging from
10"6 to 10'3 M from left to right. A yellow color indicates a pH below 7.6.


70
region. Evidence for this is supported by genome data showing 14 different CA
isoforms, all with regions of high nucleotide identity. The posterior midgut region does
display CA activity, but apparently not as a result of the GPI-linked CA isoform
presented here. The specific isoform or number of CAs contributing to the activity of the
posterior midgut is still unknown.
We have previously shown that the application of CA-specific inhibitors
dramatically decreases the alkaline gut pH, and in fact is lethal to the larval mosquitoes
(Corena et ah, 2002). We now present evidence that a CA found in the mosquito gut is
most similar to the mammalian CA IV isozyme but contains a novel active site motif
unlike any of the mammalian CA IV isoforms (Fig. 4-1). The finding of a novel CA
active site within the mosquito may facilitate the construction of a mosquito-specific CA
inhibitor for use in larval mosquito control. We are hopeful that the ongoing mosquito
CA crystallization project will yield further significant structural differences from the
mammalian CA IV structure. These differences may be useful in the design of a
mosquito-specific CA inhibitor.
Out of the 14 mammalian CAs identified thus far as cytosolic, membrane-bound,
secreted, and mitochondrial, only CA IV has a GPI link to the cell membrane. The
localization of this highly active mammalian isozyme to dynamic tissues such as the gut,
brain, kidney, and lung supports the important catalyst role of CA. It should not be
surprising that the gut of a mosquito, a highly alkaline and fluctuating system, has been
found to contain a presumably active CA IV-like isoform as well. The single amino acid
substitution of glycine-63 to glutamine is unique to rodents (rat and mouse) CA IV, and
was found to be responsible for their reduced activity rate of only 10-20% of the human


DEDICATION
I wish to dedicate this dissertation to my incredible family, Mom, Dad,
Grandparents, Tracey, and George. My family has witnessed my struggles and triumphs
and has helped me through it all. I am so proud to call them my family. I cannot thank
them enough for all of their support. This achievement is really a reflection of all of us.
I also want to dedicate this milestone to my soon to be husband, Dr. Peter Lovell.
Coming home to his love, humor, and music, has given me true joy. My family would be
incomplete if I did not mention our furry companions, Frodo and Princess, who remind us
that a nap can solve most problems.
m


21
was vortexed and centrifuged at 14,000 g for 30 seconds at 4C. The upper, aqueous
phase was transferred to a clean tube and 5 pL glycogen solution (Pharmacia Quick Prep
Micro RNA purification kit, Piscataway, New Jersey). The RNA was precipitated by the
addition of 100% ice-cold ethanol (550 pL) followed by centrifugation at 14,000 g for 6
minutes at room temperature. The supernatant was removed and 1 mL of ice-cold
ethanol (80%) was added. The mixture was centrifuged at 14,000 g for 10 minutes at
room temperature, the supernatant was removed, and the pellet was air-dried.
For first strand synthesis, the pellet was resuspended in DEPC-treated water (5
pL) and combined with the TRsa primer (1 pM; Table 2-1). This mixture was incubated
at 50C for 3 minutes and immediately placed on ice. Then IX ligation buffer (Marathon
cDNA Amplification kit, BD Biosciences, Palo Alto, California), 0.01 M DDT, 1 U
Superscript II (Life Technologies; Rockville, Maryland), and 0.5 pL dNTP mix (10 mM
each dNTP, Marathon cDNA Amplification kit) were added to a total volume of 10.5 pL.
This reaction mixture was incubated at 42C for 1 hour and immediately put on ice.
For second strand synthesis, DEPC-treated water (49 pL) was added to the first
strand reaction mix. The mixture was then combined with 1.6 pL dNTP mix (10 mM
each, Marathon cDNA Amplification kit), IX reaction buffer (Marathon cDNA
Amplification kit), and 4 pL second strand synthesis enzyme mix (Marathon cDNA
Amplification kit) in 80 pL total volume. The reaction mix was then incubated at 16C
for 1.5 hours. T4 DNA polymerase (1 U; Marathon cDNA Amplification kit) was added
to the reaction mixture and the entire mixture was incubated at 16C for an additional 0.5
hour. The reaction was stopped by incubation at 65C for 5 minutes.


136
force that pulls the proton from bicarbonate in the lumen to leave a carbonate ion. The
proton may enter the cell through a channel or be exchanged for a cation, such as
potassium. This carbonate or potassium carbonate could then drive the pH up to the high
alkalinity. The absence of an anion exchanger in the AMG would also contribute to the
alkalinity in the lumen by not providing a route for bicarbonate to leave the AMG lumen.
The localization of the AE to the basolateral membranes in both gastric caeca and
posterior midgut correlates with the localization of a cytosolic CA isoform to the same
regions. This AE brings bicarbonate into the cell in exchange for dumping chloride into
the hemolymph. This AE could account for the high levels of chloride and low levels of
bicarbonate that have been measured in the hemolymph (Boudko et al., 2001a). The
bicarbonate that enters the cell provides a substrate for the cytosolic CA. Due to the CA
binding motif at the amino terminal of the AE, the binding of these two proteins would
form a bicarbonate transport metabolon. This complex would maximize bicarbonate
transport due to the elimination of diffusion, normally relied upon to get substrate from
one protein to the next.
Our new model reflects the absence of CA within the AMG epithelial cells and
instead shows a GPI-linked CA bound to specific muscle fibers traversing the AMG and
GC, as well as a cytosolic CA within epithelial cells of the cardia, GC and PMG (Fig. 7-
1). The AE is found in regions that express CA and flank the alkaline anterior midgut.
Therefore, the cloned AE is shown in the GC and PMG. This AE has the ability to bind a
cytosolic CA, whereby the coordinated efforts of producing and transporting bicarbonate
are enhanced. The new model reflects a bicarbonate transport metabolon within the
gastric caeca and posterior midgut regions.


87
that are also capable of transporting sulfate (SLC26A group) show only 11% amino acid
identity (Fig. 5-3).
Anion exchangers, specifically AE2 and AE3, were determined to be pH
sensitive. AE3 is stimulated by intracellular alkalization whereas AE1 is not. More
specifically, a region of amino acids (WRETARWIKFEE) within the carboxy terminus is
responsible for the pH sensitivity seen in AE2 (Vince et al., 2000). The An. gambiae AE
sequence in this same region contains 14 of the 16 residues found within AE2 while the
other two amino acids are conserved (Fig. 5-5). AE1 shares only 8 of the 16 amino acids
in this region.
Localization of Anion Exchanger mRNA in An. gambiae Larvae
A DIG-labeled antisense cRNA probe comprising the full length AE cDNA was
employed to localize the AgAEl mRNA. A DIG-labeled sense probe was used as a
control. The AgAEl mRNA was found in every region of the larval gut including gastric
caeca, anterior midgut, posterior midgut, Malpighian tubules, and rectum (Fig. 5-6). In
the gastric caeca and posterior midgut regions, the probe was localized to epithelial cells.
Within the gastric caeca the labeling is most intense in the area where the lobes face the
lumen. The gastric caeca labeling was confined to the proximal cells, whereas the Cap
cells displayed no label (Fig. 5-6A,B). The rectum displayed staining in a small subset of
epithelial cells along with tracheoles (Fig. 5-6D).
Muscle, nerve, and trachea cells that traverse the outer plasma membrane of the
anterior gut epithelial cells were labeled with the AE antisense probe (Fig. 5-7). Labeled
tracheal fibers are displayed in close association with the gastric caeca (Fig. 5-8). These


144
Romero, M. F., Henry, D., Nelson, S., Harte, P. J., Dillon, A. K. and Sciortino, C. M.
(2000). Cloning and characterization of a Na+-driven anion exchanger (NDAE1). A
new bicarbonate transporter. J. Biol. Chem. 275, 24552-24559.
Saitou, N. and Nei, M. (1987). The neighbor-joining method: a new method for
reconstructing phylogenetic trees. Mol. Biol. Evol. 4, 406-25.
Schwartz, G. J. (2002). Physiology and molecular biology of renal carbonic anhydrase.
J. Nephrol. 15 Suppl 5, S61-74.
Shahabuddin, M. and Pimenta, P. F. (1998). Plasmodium gallinaceum preferentially
invades vesicular ATPase-expressing cells in Aedes aegypti midgut. Proc. Natl. Acad.
Sci. USA 95, 3385-3389.
Silverman, D. N. and Tu, C. K. (1986). Molecular basis of the oxygen exchange from
C02 catalyzed by carbonic anhydrase III from bovine skeletal muscle. Biochemistry 25,
8402-8408.
Sly, W. S. (2000). The membrane carbonic anhydrases: from C02 transport to tumor
markers. Exs. 95-104.
Sly, W. S. and Hu, P. Y. (1995). Human carbonic anhydrases and carbonic anhydrase
deficiencies. Annu. Rev. Biochem. 64, 375-401.
Spielman, A. and D'Antonio, M. (2001). Mosquito: A natural history of our most
persistent and deadly foe. Hyperion Press, New York, NY.
Sterling, D., Alvarez, B. V. and Casey, J. R. (2002a). The extracellular component of a
transport metabolon. Extracellular loop 4 of the human AE1 C1-/HC03- exchanger
binds carbonic anhydrase IV. J. Biol. Chem. 277, 25239-25246.
Sterling, D., Brown, N. J., Supuran, C. T. and Casey, J. R. (2002b). The functional
and physical relationship between the DRA bicarbonate transporter and carbonic
anhydrase II. Am. J. Physiol. Cell Physiol. 283, Cl522-1529.
Sterling, D. and Casey, J. R. (2002). Bicarbonate transport proteins. Biochem. Cell Biol.
80, 483-97.
Sterling, D., Reithmeier, R. A. and Casey, J. R. (2001a). A transport metabolon.
Functional interaction of carbonic anhydrase II and chloride/bicarbonate exchangers. J.
Biol. Chem. 276, 47886-47894.
Sterling, D., Reithmeier, R. A. and Casey, J. R. (2001b). Carbonic anhydrase: in the
driver's seat for bicarbonate transport. Jop. 2, 165-170.


I certify that I have read this study and that in my opinion it conforms to acceptable
standards of scholarly presentation and is fully adequate, in scope and quality, as a
dissertation for the degree of Doctor of Philosophy.
Edward J. Phlips, Chair
Professor of Fisheries and Aquatic Sciences
I certify that I have read this study and that in my opinion it conforms to acceptable
standards of scholarly presentation and is fully adequate, in scope and quality, as a
dissertation for the degree of Doctor of Philosophy.,*
Linser, Coer
1 Professor of Anatomy and Cell Biology
I certify that I have read this study and that in my opinion it confor
standards of scholarly presentation and is fully adequate, in scoj
dissertation for the degree of Doctor of Philosophy.
ceptable
, as a
L. Moroz
Assistant Professor of Neuroscience
I certify that I have read this study and that in my opinion it conforms to acceptable
standards of scholarly presentation and is fully adequate, in scope and quality, as a
dissertation for the degree of Doctor of Philosophy.
Assistant Professor of Fisheries and Aquatic
Sciences
I certify that I have read this study and that in my opinion it conforms to acceptable
standards of scholarly presentation and is fully adequate, in scope and quality, as a
dissertation for the degree of Doctor of Philosophy
Robert M.
Associate Professor of Neuroscience


84
Baltz, 1999). AE localization will also distinguish whether the AE is co-expressed within
the same epithelial cells as the H+ V-ATPase and if the polarity of basal or apical
expression is also the same. The localization of a V-ATPase within the larval
mosquito gut was found to be apical in the gastric caeca, basal in the AMG, and then
apical again in the PMG (Zhuang et al., 1999).
The highly dynamic system of alkaline digestion in the larval mosquito gut does
not exist in any known mammalian system. However, mammalian organs such as the
kidney are able to perform many parallel functions of the mosquito gut, such as water
regulation, filtration, and ionic homeostasis. The CA, AE, and H+ V-ATPase proteins, in
particular, have been extensively studied and localized within the mammalian kidney due
to their dynamic roles in acid-base balance (Huber et al., 1999; Schwartz, 2002). The co
localization and polar expression of an AE with a Hf V-ATPase will define the epithelial
cells of the mosquito gut as resembling the mammalian kidney A-intercalated cell type,
the B-intercalated cell type, or the non-A non-B intercalated cell type (types as defined
by Brown and Breton, 1996 and Kim et al., 1999).
Results
%
An. gambiae AE Sequence Analysis
The full length An. gambiae AE1 (AgAEl) cDNA was cloned from midgut tissue
and contains 3309 bases (accession number AY280611) with a molecular weight of 123
kDa for the predicted protein. The NCBI conserved domain search tool (CDD)
determined that the protein sequence was part of a family of bicarbonate transporters
(BT) and cotransporters (PF00955) as well as sodium-independent chloride/bicarbonate
exchangers and related sodium/bicarbonate cotransporters (KOG1172; Geer et al., 2002).


28
minutes at 75C. The lysed tissues were treated with 2 U of DNase I for 30 minutes at
37C. The DNase I was then inactivated by heating to 75C for 5 minutes. For the
reverse transcription reaction, 10 pL of cell lysate was combined with 4 pL dNTP mix
(contains 2.5 mM each dNTP) and 5 pM random decamer first strand primer in 16 pL
total volume. The mixture was incubated at 70C for 3 minutes and then chilled on ice
for 1 minute. This mixture was then combined with IX RT buffer, 1 U M-MLV reverse
transcriptase, and 10 U RNase inhibitor, and incubated at 42C for 1 hour. The reverse
transcriptase was then inactivated by incubation at 95C for 10 minutes. Primers (Table
2.1) were designed using Primer Express software (Applied Biosystems; Foster City,
California). The SYBR Green PCR Master mix, which includes SYBR Green I dye,
Amplitaq Gold DNA Polymerase, dNTPs, and buffer, was used for all real time PCR
investigations. Each cycle of PCR was detected by measuring the increase in
fluorescence caused by the binding of the SYBR Green dye to double-stranded DNA
using an ABI Prism 7000 Sequence Detection System (Applied Biosystems). Initially,
each primer set, including the control 18s ribosomal RNA (Genbank accession M95126),
was assessed to determine the optimal concentration of primer to be used. All real time
experiments used the same 2-step cycling profile: 50C for 2 minutes followed by 95C
for 10 minutes and 40 cycles of 95C for 15 seconds and 60C for 1 minute. Whole gut
cDNA (100 nM) was used as template with 500 nM, 300 nM, 100 nM, or 50 nM of each
primer set and IX SYBR green I master mix in 25 pL total volume. Each reaction was
done in triplicate. The optimal concentration was then chosen based on the amplification
plots and the dissociation curves generated. Once a concentration was chosen for each
primer set, the efficiency of amplification of that set was determined. Serial dilutions of


128
Figure 6-7. Localization of CA protein within gastric caeca of An. gambiae larvae. A.
The CA specific antibody prominently labels cells of the gastric caeca. B.
Phalloidin was used to localize muscle fibers. C. Draq-5 was used to
localize nuclear DNA. D. Overlay of the three signals shows the location of
CA in relation to muscle fibers and nuclei. Scale bars represent 100 pm.


127
Figure 6-6. Localization of CA mRNA expression within the hindgut. A. The hindgut
displays CA label within the Malpighian tubules (MT, arrow) and the
rectum. B. Higher magnification of the Malpighian tubules shows staining
for CA expression confined to the most distal one or two cells. All other
cells of the MT show no hybridization. Scale bars represent 300 pm in A
and 25 pm in B.


102
Figure 5-7. Localization of AgAEl mRNA in muscle, nerve, and trachea in An. gambiae.
A. The whole mount gut preparation localizes AE message to specific muscle,
nerve (long arrow), and trachea (short arrow) fibers of the anterior midgut
(AMG). B. A high magnification of the AMG region detailing the tracheal
fibers (short arrows) and neuronal cells (long arrow) that express the AgAEl
message. Scale bar represents 75 pm in A and B.


116
will lead to evolutionary clues within the family of a CAs. The relationship between the
three distinct CA families is believed to represent convergent evolution. The relationship
between a CAs from distantly related species is unknown, mostly due to the lack of
characterized CAs from non-mammalian species.
We report here the first cytosolic CA that has been cloned and characterized from
the An. gambiae mosquito, in an attempt to unravel the physiology of an extremely
alkaline digestive system. There are at least two different CA isoforms expressed within
the larval mosquito gut. One is a GPI-linked CA isoform expressed in a specific subset
of muscle and nerve fibers that traverse the anterior midgut and gastric caeca regions
(refer to Chapter 4). The other is a cytosolic CA isoform expressed primarily within the
gastric caeca and posterior midgut regions.
Results
Anopheles gambiae CA Sequence Analysis
A CA cDNA was cloned from An. gambiae gut tissue. The full-length CA cDNA
sequence (accession number AY280613) represents 1 of 14 putative CA genes in the An.
gambiae genome predicted by the Ensembl CA protein family (ENSF00000000228).
This CA is comprised of 257 amino acids and is predicted to be a cytosolic isoform (no
signal sequence or hydrophobic transmembrane domains; Letunic et al., 2002). The
molecular weight is predicted to be 29 kDa (DNAman software). Multiple sites of
potential post-translational modification include 5 protein kinase C phosphorylation sites,
4 casein kinase II phosphorylation sites, and 5 N-myristoylation sites (Bairoch et ah,
1997). The numerous potential sites for protein modification may contribute to
regulatory control.


17
mM Tris HC1, 0.1 M Na2S04 buffer, pH 8.5. BTB solution was added to each well until
a 0.003% solution was achieved. Five live early fourth-instar larvae that had been placed
in BTB indicator solution for 2 hours were added to each of the wells, and the larvae
were allowed to adjust to their new environment for 30 minutes. Methazolamide
dissolved in Dimethyl Sulfoxide (DMSO; Sigma-Aldrich) at concentrations ranging from
10"6 M to 8x10'3 M was added to the wells. Controls included wells containing DMSO
with BTB indicator but no inhibitor and wells containing BTB indicator but no DMSO.
The plates were scanned using a Hewlett Packard ScanJet 6100C scanner before addition
of the inhibitor and 5 hours later. In addition, the midguts were dissected and
photographed to record the pH within the gut lumen as revealed by the color of ingested
BTB.
180 Exchange Method to Measure Carbonic Anhydrase Activity
Tissue homogenate carbonic anhydrase activity was measured using the O
exchange method (Silverman and Tu, 1986). Midguts were dissected, and the peritrophic
membrane was removed together with its contents. Individual measurements of CA
activity were performed with pooled samples of gastric caeca, anterior midgut, posterior
midgut and Malpighian tubules. The method involved adding O-labeled NaHCCb to
0.1 M Hepes buffer, pH 7.6, at 9.5C. The disappearance of 180 isotopes from C02
and/or HCO3' upon addition of the enzyme preparations was monitored. Measurements
10
of O in C02 were accomplished with a mass spectrometer, using a C02-permeable inlet
that allowed very rapid, continuous measurement of the isotopic content of C02 in
solution. All samples were centrifuged at 14,000 rpm at room temperature prior to the
assay to remove food and insoluble material. Inhibition was accomplished by adding


basal side of the anterior midgut. This CA resembles the mammalian CA IV isozyme in
that a glycosylphosphatidylinositol (GPI)-link tethers the enzyme to the extracellular
membrane. The other CA isoform, an active cytosolic enzyme, was localized to the
gastric caeca and posterior midgut regions. It was also determined that the AE transports
chloride and is expressed in the gastric caeca, posterior midgut, and Malpighian tubules.
We were unable to detect CA within the anterior midgut epithelial cells using a variety of
assays. My studies have therefore led to an alternative hypothesis that one or more CAs
within the mosquito gut, but located outside of the anterior midgut epithelial cells,
contribute to buffering the alkaline pH of 11 within the anterior midgut lumen. The
localization of two CA isoforms, one extracellular and the other cytosolic, and an AE
possessing a putative CA binding sequence, to the regions flanking the anterior midgut,
supports the prediction of a bicarbonate transport metabolon within the gastric caeca and
posterior midgut regions. Such a metabolon has only been studied in mammals, however,
the colocalization of CA and AE within the mosquito gut suggests a similar network of
bicarbonate production and transport.


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
CARBONIC ANHYDRASES AND BICARBONATE TRANSPORT IN
LARVAL MOSQUITOES
By
Theresa J. Sern
May 2004
Chair: Edward J. Phlips
Cochair: Paul J. Linser
Major Department: Fisheries and Aquatic Sciences
Carbonic anhydrase (CA) is an important enzyme due to its involvement in many
pH-dependent, physiological processes. CA reversibly converts CO2 and H2O into
bicarbonate and a proton. The anterior midgut lumen of the larval mosquito has an
extremely alkaline pH; therefore, we hypothesized that an active CA within the epithelial
cells surrounding this region would rapidly produce bicarbonate to buffer the high pH. A
cDNA cloning strategy, followed by in situ hybridization, was employed to isolate and
localize CA and anion exchanger (AE) transcripts within the mosquito gut. Localization
of CA enzymatic activity was assessed via histochemical analyses. Enzymatic and
electrophysiological analyses of recombinant CAs and AE were also performed. In this
dissertation, cDNAs encoding three CA genes and an AE were cloned and localized
within the larval mosquito gut. One isoform of mosquito CA, which was cloned from
two different mosquito species, was localized to a specific subset of muscle fibers on the
Xll


91
AEs when they were expressed in HEK293 cells, and was shown to increase the rate of
bicarbonate transport. Furthermore, co-expression with mutant CAII (non-active) was
shown to result in decreased bicarbonate transport due to the displacement of the active
endogenous CA. The CA II binding motif found in this mosquito AE could similarly
bind the oocytes endogenous CA, also resulting in an increased transport rate. This may
explain the inhibition seen in AE transport when acetazolamide was applied. AgAEl
expression studies in oocytes are presently ongoing to further assess the function and
inhibition of this protein.
Discussion
An AE cDNA was cloned from fourth instar, larval An. gambiae gut tissue. The
translated 123 kDa protein is predicted to consist of intracellular amino and carboxy
termini and 12 transmembrane segments. In situ hybridization and antibody
immunolocalization identified AE mRNA message expression and protein localization
within epithelial cells of cardia, gastric caeca, posterior midgut, rectum, and Malpighian
tubules as well as tracheal, nerve, and muscle cells. Expression of AgAEl in Xenopus
oocytes displayed a reversible transport of chloride.
The most similar characterized protein sequence is the NDAE1 from D.
melanogaster. Xenopus oocyte expression with pH analysis determined that this protein
was sodium ion dependent. The An. gambiae AE1 protein has 72% sequence identity to
the NDAE1 but does not display sodium ion dependence. The carboxy termini of these
proteins show little similarity and therefore the carboxy terminus of NDAE1, unlike
AgAEl, may contain the necessary domain for sodium ion dependence (refer to Fig. 5-4).


139
A.
Midgut Hindgut
hco;
B.
lumen H
hco3
HCO, +H+
hco3 + co32 h<
AMG 2H>K+ PMG
co2 + h2o
hco3
.co2 +
HzO Hemolymph
H+ + HCO/ PH 78
Figure 7-1. New larval mosquito model. A. The larval mosquito gut is divided into the
foregut, midgut, and hindgut. The gastric caeca (GC) and anterior midgut
(AMG) express a GPI-linked CA isoform on muscle fibers (shown in yellow).
The cardia, GC, posterior midgut (PMG), rectum, and last distal cell of
Malpighian tubules (MT) express a cytosolic CA isoform. The GC, PMG,
MT and rectum express a chloride/ bicarbonate anion exchanger (AE). In the
GC, and PMG, the AE may bind a cytosolic CA isoform forming a metabolon.
A V-ATPase is expressed in GC, AMG, and PMG. B. Diagram of a
representative cell from GC, AMG, and PMG displaying the cell polarity.
The V-ATPase is expressed apically in GC and PMG, and basally in AMG.
The AE is expressed basally in the GC and PMG. The GPI-linked CA
isoform is expressed extracellularly on muscle fibers in the GC and AMG.


LIST OF FIGURES
Figure E§££
1-1. Illustration showing the regions of the larval mosquito gut 11
1 -2. Illustration of the mosquito life cycle 12
1 -3. Preliminary mosquito anterior midgut model based on M. Sexta 13
2-1. Efficiency plots for real-time PCR primers 37
2-2. Three-dimensional (Cn3D) depiction of human CAIV (1ZNC) 38
3-1. Effect of CA inhibition on culture medium pH with fourth-instar Ae.
aegypti. larvae 53
3-2. Effect of methazolamide on the alkalization of the midgut using
Bromothymol Blue (BTB) assay of pH within living, but isolated, gut tissue 54
3-3. Relative activity of CA in different pooled segments of the midgut of larval
Ae. aegypti 55
3-4. Carbonic anhydrase from the midgut of larval Ae. aegypti 56
3-5. Comparison of the extrapolated amino acid sequences of A-CA with
six putative dipteran CA genes identified in the D. melanogaster
gene databases 57
3-6. Polymerase chain reaction (PCR) analysis of Ae. aegypti amplified cDNA
from different gut regions 58
3-7. Hanssons histochemistry of whole mount Ae. aegypti gut 59
3-8. Localization of CA mRNA expression in larval Ae. aegypti 60
4-1. Alignment of several mammalian CA IV enzymes with two mosquito CA
isoforms 72
4-2. Clustal alignment of CA protein sequences 73
IX


REFERENCES
Alper, S. L. (2002). Genetic diseases of acid-base transporters. Annu. Rev. Physiol. 64.
899-923.
Alper, S. L., Chernova, M. N. and Stewart, A. K. (2001). Regulation of Na+-
independent C1-/HC03- exchangers by pH. Jop. 2, 171-175.
Altschul, S. F., Gish, W., Miller, W., Myers, E. W. and Lipman, D. J. (1990). Basic
local alignment search tool. J. Mol. Biol. 215, 403-410.
Baird, T. T., Jr., Waheed, A., Okuyama, T., Sly, W. S. and Fierke, C. A. (1997).
Catalysis and inhibition of human carbonic anhydrase IV. Biochemistry 36, 2669-2678.
Bairoch, A., Bucher, P. and Hofmann, K. (1997). The PROSITE database, its status in
1997. Nucleic Acids Res. 25, 217-221.
Berenbaum, M. (1980). Adaptive significance of midgut pH in larval Lepidoptera. Am.
Nat. 115,138-146.
Boudko, D. Y., Moroz, L. L., Harvey, W. R. and Linser, P. J. (2001a). Alkalinization
by chloride/ bicarbonate pathway. Proc. Natl. Acad. Sci. USA 98,15354-15359.
Boudko, D. Y., Moroz, L. L., Linser, P. J., Trimarchi, J. R., Smith, P. J. S., Harvey,
W. R. (2001b). In situ analysis of pH gradients in mosquito larvae using non-invasive,
self-referencing, pH-sensitive microelectrodes. J. Exp. Biol. 204, 691-699.
Brown, D. and Breton, S. (1996). Mitochondria-rich, proton-secreting epithelial cells. J.
Exp. Biol. 199 (Pt 11), 2345-2358.
Brown, D. and Waneck, G. L. (1992). Glycosyl-phosphatidylinositol-anchored
membrane proteins. J. Am. Soc. Nephrol. 3, 895-906.
Chegwidden, W. R. and Carter, N. D. (2000). Introduction to the carbonic anhydrases.
Exs. 14-28.
Clamp, M., Andrews D., Barker, D., Beven, P., Cameron, G., Chen, Y., Clark, L.,
Cox, T., Cuff, J., Curwen, V., et al. (2003). Ensembl 2002: accommodating
comparative genomics. Nucleic acids res. 31 (1), 38-42.
140


146
Vince, J. W., Carlsson, U. and Reithmeier, R. A. (2000). Localization of the Cl-
/HC03- anion exchanger binding site to the amino-terminal region of carbonic
anhydrase II. Biochemistry 39, 13344-13349.
Volkmann, A. and Peters, W. (1989a). Investigations on the midgut caeca of mosquito
larvae I. Fine structure. Tissue & Cell 21,243-251.
Volkmann, A. and Peters, W. (1989b). Investigations on the midgut caeca of mosquito
larvae II. Functional aspects. Tissue & Cell 21, 253-261.
Waheed, A., Okuyama, T., Heyduk, T. and Sly, W. S. (1996). Carbonic anhydrase IV:
purification of a secretory form of the recombinant human enzyme and identification of
the positions and importance of its disulfide bonds. Arch. Biochem. Biophys. 333,432-
438.
Westerfield, M. (1994). The zebrafish book: A guide for the laboratory use of zebrafish
(Brachydanio rerio). (ed. Westerfield, M.), University of Oregon Press, Eugene, OR.
pp. 9.16-9.21.
Wieczorek, H., Gruber, G., Harvey, W.R., Huss, M. and Merzendorfer, H. (1999).
The plasma membrane H+-V-ATPase from tobacco homworm midgut. J. Bioenerg.
Biomembr. 31, 67-14.
Wieczorek, H., Gruber, G., Harvey, W. R., Huss, M., Merzendorfer, H. and Zeiske,
W. (2000). Structure and regulation of insect plasma membrane H(+)V-ATPase. J. Exp.
Biol. 203 (Pt 1), 127-135.
Zhang, Y. and Frohman, M. A. (1997). Using rapid amplification of cDNA ends
(RACE) to obtain full-length cDNAs. Methods Mol. Biol. 69, 61-87.
Zhuang, Z., Linser, P. J. and Harvey, W. R. (1999). Antibody to H(+) V-ATPase
subunit E colocalizes with portasomes in alkaline larval midgut of a freshwater
mosquito (Aedes aegypti). J. Exp. Biol. 202 (Pt 18), 2449-2460.


20
The primer sequences used initially for Ae. aegypti CA were CA5F and CA3R
(see Table 2-1). PCRs were performed in a total volume of 20 pi, and the reaction
mixture contained 0.1 pg of cDNA as template, 0.2 pM of each primer, 200 pM each of
dNTPs, IX PCR buffer and 1 U of Taq polymerase (Promega; Madison, Wisconsin).
The PCR cycling profile was: 94C for 5 min, 55C for 2 min and 72C for 3 min,
followed by six cycles of 94C for 0.5 min, 53C (in increments of 2C/cycle) for 1 min
and 72C for 1 min and 35 cycles of 94C for 0.5 min, 45C for 1 min and 72C for 2
min followed by a final extension at 72C for 15 min. The PCR products were visualized
on 1 % agarose gels and specific products were isolated using a QIAquick gel extraction
kit (Qiagen, Inc, Valencia, California), diluted 1:100 in water, and used as template for a
second, identical PCR. The resulting 297 base-pair (bp) product was gel-purified, ligated
into pGem-T (Promega) and transformed into JM109 Escherichia coli (Promega) for
subcloning. This partial Ae. aegypti CA cDNA was completed using amplified cDNA
pools from gastric caeca and posterior midgut.
Construction of Amplified cDNA Pools
Adapter-ligated, amplified cDNA pools (libraries) were constructed from
different regions of the fourth instar larval gut of both Ae. aegypti and An. gambiae using
a technique optimized for invertebrate tissues (Matz et al., 1999). The gastric caeca,
anterior midgut, posterior midgut, rectal salt gland, Malpighian tubules, and anal papillae
of ten larvae were dissected in HSS and collected separately, resulting in six discreet
tissue pools. The tissue was dissolved in Buffer D (500 pL; 4 M guanidine thiocyanate,
30 mM sodium citrate, and 30 mM beta-mercaptoethanol). The mixture was placed on
ice and combined with phenol (500 pL, pH 7.0) and chloroform (100 pL). The mixture


4-3. Localization of CA mRNA in a whole mount preparation of early 4th
instarle, aegypti 74
4-4. Expression of CA mRNA in Ae. aegypti anterior midgut 75
4-5. Localization of CA IV-like message within Ae. aegypti CNS tissue 76
4-6. Relative quantification of CA IV-like message in Ae. aegypti larvae
using real time PCR 77
4-7. Ae. aegypti and An. gambiae CA protein labeling 78
4-8. The Ae. aegypti CNS ganglia express the CA IV-like isoform 79
4-9. Immunolocalization of mosquito CA IV-like enzyme in Aedes albopictus 80
4-10. High magnification of immunoreactive muscle fibers within the Aedes
albopictus midgut 81
4-11. Immunoreactivity of Ae. aegypti guts for the CA IV-like isozyme 82
5-1. Structural prediction of the An. gambiae AE1 96
5-2. Putative amino terminus CA II binding motif 97
5-3. Homology tree depicting the amino acid identity between several BTs 98
5-4. Alignment of carboxy terminus amino acids of An. gambiae and D.
melanogaster AEs 99
5-5. Alignment of An. gambiae and human AEs 100
5-6. Localization of AgAEl mRNA within whole mount An. gambiae
larvae 101
5-7. Localization of AgAEl mRNA in muscle, nerve, and trachea in
An. gambiae 102
5-8. In situ hybridization of AgAEl in whole mount An. gambiae consistently
shows positive labeling of tracheal fibers along the midgut 103
5-9. Anion exchanger mRNA localization reveals trachea and nerve fibers
along with neuronal cell labeling 104
5-10. Localization of AgAEl mRNA to the PMG of larval An. gambiae 105
x


106
5-11. Larval An. gambiae displays strong AgAEl expression in the hindgut,
the pylorus
5-12. Localization of AE mRNA in An. gambiae shows abundant labeling of
the Malpighian tubules 107
5-13. Expression of AE mRNA was found throughout the ventral midgut
ganglia 108
5-14. Sense AE probes display no specific hybridization 109
5-15. Antibody localization of AgAEl protein to the gastric caeca in An.
gambiae larvae 110
5-16. Localization of AgAEl protein within the PMG of An. gambiae larvae 111
5-17. Neuronal cells within the AMG display immunoreactivity for our
An. gambiae AE specific antibody 112
5-18. Current-voltage (I-V) plots depicting ion transport by the AgAEl expressing
oocytes in contrast to the water injected control oocytes 113
5-19. Inhibition of AgAEl mediated chloride transport by DIDS 114
6-1. Clustal alignment of active sites within An. gambiae, D. melanogaster,
and human CA proteins 122
6-2. Phylogenetic analysis between mammalian (human and mouse) and
dipteran (An. gambiae and D. melanogaster) CAs 123
6-3. Localization of An. gambiae CA activity 124
6-4. Localization of CA mRNA expression within An. gambiae whole mounts 125
6-5. Localization of CA mRNA expression within the posterior midgut of An.
gambiae 126
6-6. Localization of CA mRNA expression within the hindgut 127
6-7. Localization of CA protein within gastric caeca of An. gambiae larvae 128
6-8. Localization of CA protein within the PMG of An. gambiae 129
6-9. Protein gels and western blots of recombinantly expressed CA protein 130
7-1. New larval mosquito model 139
xi