A Phylogenetic and Evolutionary Study of Endogenous Cellulose Digestion in Higher Termites (isoptera

MISSING IMAGE

Material Information

Title:
A Phylogenetic and Evolutionary Study of Endogenous Cellulose Digestion in Higher Termites (isoptera termitidae)
Physical Description:
1 online resource (78 p.)
Language:
english
Creator:
Bujang,Nurmastini
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Entomology and Nematology
Committee Chair:
Su, Nan-Yao
Committee Members:
Harrison, Nigel A
Kern, William H
Giblin-Davis, Robin M
Scheffrahn, Rudolf H
Elliott, Monica L

Subjects

Subjects / Keywords:
cellulase -- termites -- termitidae
Entomology and Nematology -- Dissertations, Academic -- UF
Genre:
Entomology and Nematology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
Cellulose is the most abundant biopolymer in the world whereas termites are the most important metazoan cellulose processors. Termites are divided into lower and higher termites, with the latter being the most derived and most specious. Although termites are known for their ability to digest wood, members of the family Termitidae (higher termites) are nutritionally diverse in their use of cellulose. There have been numerous studies on the evolution of termites, but the evolution of endogenous cellulose digestion in termites, especially in higher termites, is poorly understood. Endogenously-produced termite cellulases consist of endo-beta-1,4-glucanases and beta-glucosidases only. Hence, using phylogenetic inferences from mitochondrial (16S) ribosomal RNA, nuclear (28S), endo-beta-1,4-glucanase and beta-glucosidase sequences, I attempt to explain the evolution of endogenous cellulose digestion in higher termites. The translated endo-beta-1,4-glucanase amino acid sequences obtained during this study showed high similarity to endo-beta-1,4-glucanases in the glycosyl hydrolase family 9 (GHF9). The inferred endo-beta-1,4-glucanase phylogenetic tree showed congruency with the mitochondrial/nuclear tree, with the fungus-growers being the most basal group and the soil/litter- and wood/lichen/grass/litter-feeders being the most distal diphyletic feeding groups. The phylogenetic placement of the bacterial comb-grower was determined as the ?missing link? between the fungus-growers and the soil/litter- and wood/lichen/grass/litter-feeders. There was also a strong diphyletic relationship between endo-beta-1,4-glucanases of upper layer soil-feeders and the other soil-feeders. Within the monophyletic wood/lichen/grass/litter-feeding termites? clade, the nasutitermitines were polyphyletic and a strong diphyletic relationship was also observed in the most distal groups, the lichen- and the grass/litter-feeders. In some species, I was able to obtain up to four paralogous copies with high degrees of substitutions among them, suggesting different alleles, and subsequently resulting in different gene function. For beta-glucosidase, the deduced amino acid sequences showed that they were similar to beta-glucosidases in the glycosyl hydrolase family 1 (GHF1). However, phylogenetic incongruity was observed between the mitochondrial/nuclear and beta-glucosidase trees. Instead of being the most basal feeding group, the fungus-growers formed a strong diphyletic relationship with the wood/grass- and soil/litter-feeders. Furthermore, instead of being an intermediary between the basal and derived groups as initially hypothesized, data suggested that bacterial comb-grower beta-glucosidases were probably derived from fungus-growers beta-glucosidases. Two different sequences from the bacterial comb-feeder suggest the involvement of two different alleles, resulting in different gene functionality. Lastly, while there is a high level of evidence to support the vertical gene transfer hypothesis in GHF9, the evolutionary origin of GHF1 could not be deduced at present.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Nurmastini Bujang.
Thesis:
Thesis (Ph.D.)--University of Florida, 2011.
Local:
Adviser: Su, Nan-Yao.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2011
System ID:
UFE0043228:00001


This item is only available as the following downloads:


Full Text

PAGE 1

A PHYLOGENETIC AND EVOLUTIONARY STUDY OF ENDOGENOUS CELLULOSE DIGESTION IN HIGHER TERM ITES (ISOPTERA:TERMITIDAE) By NURMASTINI SUFINA BINTI BUJANG 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 2011 1

PAGE 2

2011 Nurmastini Sufina binti Bujang 2

PAGE 3

To my family, for their endless love, even when Im at my most unlovable Family is a haven in a heartless world. Christopher Lasch 3

PAGE 4

ACKNOWLEDGMENTS In my journey to become a better scientist, there are three indi viduals whom I am forever indebted to: Doctor Nan-Yao Su, for his guidance, limitless form of support and for allowing me the fre edom to find my own niche in te rmitology; Dr. Nigel Harrison, for his patience, advice, and selflessness to help overcome numerous failed PCRs and termite-related assays; and Dr. Lee Chow Yang, for his unwavering support from afar. I will strive to make them proud in my future research endeavors. I thank my committee member s: Dr. Robin Giblin-Davis, Dr. Rudolf Scheffrahn, Dr. William Kern and Dr. Monica Elliott for their advice, guidance and help. Having six professors on the committee is a bit of a crowd, but I deem all advice and help from every individual crucial throughout this jour ney. A special thanks goes to Dr. Gaku Tokuda for his advice, Dr. Natsumi Kanzaki, Dr. Alain Robert and Aaron Mullins for their effort in getting the specimens, Paul Madei ra for his experimental advice and help, and Ericka Helmick, the laboratory-techniques-teacher-extraordina ire. To the late Paul M. Ban, thank you for your help and friendship. You will be missed. I also thank all my colleagues and fr iends at Fort Lauderdale Research and Education Center for making life interesti ng and fulfilling, and Debbi e Hall, Sarah Kern, Veronica Woodard, Joanne Korvick and Mike Ry abin for making life easier. I thank all my friends near and far for sharing my tear s and laughter, and for keeping me firmly rooted and sane. I am forever grateful to my family who gave me the fr eedom to carve my own path in life. They do not always understand the choi ces I make, but they have continued to provide me with infinite love, undivided s upport, and constant guidance, nonetheless. They are my meaning of life. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDG MENTS .................................................................................................. 4LIST OF TABLES ............................................................................................................ 7LIST OF FI GURES .......................................................................................................... 8ABSTRACT ..................................................................................................................... 9 CHAPTER 1 TERMITE ENDOGENOUS CELLULAS ES ............................................................. 11General Intr oduction ............................................................................................... 11Cellulos e ................................................................................................................. 11Cellulase and its Functions ..................................................................................... 12Cellulose Digestion in Lower Te rmites .................................................................... 13Cellulose Digestion in Higher Te rmites ................................................................... 15Problem Stat ement ................................................................................................. 16Objectiv es ............................................................................................................... 172 THE PHYLOGENY OF HIGHER TERMITES BASED ON MITOCHONDRIAL AND NUCLEAR MARKERS ................................................................................... 18Materials and Methods ............................................................................................ 19Termite S pecies ............................................................................................... 19DNA Extrac tion ................................................................................................. 20Polymerase Chai n Reacti on ............................................................................. 20Visualization and Sequencin g .......................................................................... 20Sequence Analysis ........................................................................................... 21Results and Discussion ........................................................................................... 223 A PHYLOGENETIC AND EVOLUTION ARY STUDY OF ENDO-BETA-1,4GLUCANASE IN HIGH ER TERM ITES ................................................................... 28Materials and Methods ............................................................................................ 31mRNA Extraction and c DNA Synthes is ............................................................ 31Polymerase Chai n Reacti on ............................................................................. 31Cloning ............................................................................................................. 31Sequence Analysis ........................................................................................... 33Results and Discussion ........................................................................................... 334 A PHYLOGENETIC AND EVOLUTION ARY STUDY OF BETA-GLUCOSIDASE IN HIGHER TE RMITES .......................................................................................... 49 5

PAGE 6

Materials and Methods ............................................................................................ 51Polymerase Chai n Reacti on ............................................................................. 51Cloning and Sequenc e Analysi s ....................................................................... 52Results and Discussion ........................................................................................... 525 CONCLUDING REMARKS AND FUTURE DIRE CTIONS ...................................... 63LIST OF RE FERENCES ............................................................................................... 66BIOGRAPHICAL SKETCH ............................................................................................ 78 6

PAGE 7

LIST OF TABLES Table page 2-1 List of 25 termitid spec ies used in th is study ...................................................... 242-2 List of nuclear and mitochondrial ma rker primers used in this study .................. 253-1 List of endo--1,4-glucanase primers used in this study ..................................... 404-1 List of -glucosidase primers used in this study ................................................. 58 7

PAGE 8

LIST OF FIGURES Figure page 2-1 Agarose gel showing PCR amplification from 25 species of higher termites ...... 262-2 Consensus Bayesian tree inferred fr om combined 16s and 28s sequences ...... 273-1 Agarose gel showing PCR amplification of endo-1,4-glucanase from four representative species of higher te rmites ........................................................... 413-2 Restriction fragment length pr ofiles of four representative endo-1,4glucanase clones (ca. 1.35 kb) from f our species of higher termites .................. 423-3 Multiple alignments of endo-1,4-glucanase amino acid sequences from higher term ites .................................................................................................... 433-4 Consensus Bayesian tree inferred from endo-1,4-glucanase sequences ....... 484-1 Agarose gel showing PCR amplification of -glucosidase from four species of higher te rmites ................................................................................................ 584-2 Restriction fragment length profiles of five -glucosidase clones (ca. 1.6 kb) from four species of higher te rmites ................................................................... 594-3 Multiple alignments of -glucosidase amino acid sequences from higher termite s ............................................................................................................... 604-4 Consensus Bayesian tree inferred from -glucosidase sequences. ................... 62 8

PAGE 9

Abstract of Dissertation Pr esented to the Graduate School of the University of Florida in Partial Fulf illment of the Requirements for t he Degree of Doctor of Philosophy A PHYLOGENETIC AND EVOLUTIONARY STUDY OF ENDOGENOUS CELLULOSE DIGESTION IN HIGHER TERM ITES (ISOPTERA:TERMITIDAE) By Nurmastini Sufina binti Bujang August 2011 Chair: Nan-Yao Su Major: Entomology and Nematology Cellulose is the most abundant biopolymer in the world whereas termites are the most important metazoan cellulose processors. Termites are divided into lower and higher termites, with the latter being the mo st derived and most specious. Although termites are known for their ability to diges t wood, members of the family Termitidae (higher termites) are nutritionally diverse in their use of cellulose. There have been numerous studies on the evolution of te rmites, but the ev olution of endogenous cellulose digestion in termites, especially in higher termites, is poorly understood. Endogenously-produced termite cellulases consist of endo-1,4-glucanases and glucosidases only. Hence, using phylogenet ic inferences from mitochondrial (16S) ribosomal RNA, nuclear (28S), endo--1,4-glucanase and -glucosidase sequences, I attempt to explain the evolut ion of endogenous cellulose diges tion in higher termites. The translated endo-1,4-glucanase amino acid sequences obtained during this study showed high similarity to endo-1,4-glucanases in the glycosyl hydrolase family 9 (GHF9). The inferred endo-1,4-glucanase phylogenetic tree showed congruency with the mitochondrial/nuclear tree, with the fungus-growers being the most basal group and 9

PAGE 10

10 the soil/litterand wood/lichen/grass/litter-feeder s being the most distal diphyletic feeding groups. The phylogenetic placement of the bacterial comb-grower was determined as the missing link between the fungus-growers and the soil/litterand wood/lichen/grass/litter-feeders. There was also a strong diphyletic relationship between endo-1,4-glucanases of upper layer soil-feeders and the other soil-feeders. Within the monophyletic wood/lichen/grass/litter-feeding termites clade, the nasutitermitines were polyphyletic and a strong diphyletic relationshi p was also observed in the most distal groups, the lichenand the grass/litter-feeders. In some species, I was able to obtain up to four paralogous copies with high degr ees of substitutions among them, suggesting different alleles, and subsequently re sulting in different gene function. For -glucosidase, the deduced amino acid sequences showed that they were similar to -glucosidases in the glycosyl hydr olase family 1 (GHF1). However, phylogenetic incongruity was observed between the mitochondrial/nuclear and glucosidase trees. Instead of being the most basal feeding group, the fungus-growers formed a strong diphyletic relationship with the wood/grassand soil/litter-feeders. Furthermore, instead of being an intermediary between the basal and derived groups as initially hypothesized, data sugge sted that bacterial comb-grower -glucosidases were probably derived from fungus-growers -glucosidases. Two different sequences from the bacterial comb-feeder suggest the involvement of two different alleles, resulting in different gene functionality. Lastly, while there is a high level of evidence to support the vertical gene transfer hypothesis in GHF9, the evolutionary origin of GHF1 could not be deduced at present.

PAGE 11

CHAPTER 1 TERMITE ENDOGENOUS CELLULASES General Introduction Termites are eusocial insects comprising over 2,600 described species (Kambhampati and Eggleton, 2000). They are phylogenetica lly classified into lower (families Mastotermitidae, Kalotermiti dae, Hodotermitidae, Termopsidae, and Rhinotermitidae) and higher termites (family Termitidae). The family Termitidae is recognized as being the most recently evol ved and derived family (Miura et al., 1994). While it was suggested that primitive termite s originated from the Upper Jurassic period (~161 Ma), the earliest recorded termitid foss il was from the Eoc ene period (~55.8 Ma) (Thorne et al., 2000). However, E ngel et al. (2009) estimated that the family Termitidae arose from the family Rhin otermitidae in Early Paleogen e and began radiating in the Late Eocene period (~40 Ma). The family Termitidae comprises 84% of a ll termite species and is divided into seven subfamilies: Macrotermitinae, Sphaerotermitinae, Foraminitermitinae, Syntermitinae, Nasutitermiti nae, Termitinae and Apicotermitinae (Engel et al., 2009). Although termites are known for their ability to digest wood, members of this family actually exploit a wide variety of feeding substrates, ranging from fungi, bacterial comb, wood, lichen, grass and soil/litter. Cellulose The three major components of lignoce llulose are cellulose (28-50%), hemicelluloses (20-30%) and lig nin (18-30%) (Thompson, 1983). Cellulose is composed of unbranched anhydro-1,4-glucose chains linked together by a -1,4-D-glycosidic bond, which is cleaved by cellulases during cellulose degradation (H an et al., 1995). 11

PAGE 12

According to Brune (2006), the pathway for cellu lose digestion is: wood polysaccharides mono-, diand o ligosaccharides lactate (ethanol) acetate/formate CO2/CH4. The penultimate end products of these reactions are acetate (which is used by termites as their energy source) and formate (H2 and CO2). Termites mainly utilize the glucose from the breakdown of cellulose and the acetate from th e fermentation in the hindgut (Slaytor, 2000). Both H2 and CO2 are utilized by endosymbiotic : i) spirochetes to produce acetate; ii) acetogenic bacteria to produce acetate; and iii) methanogenic bacteria to produce methane (Adams and Boopathy, 2005). Cellulose is the most abundant biopol ymer in the biosphere and one of the cheapest resources to solve the problem of chemical and energy production (Sakka et al., 2000). Hence, cellulase can be utilized to manage cellulosic industrial and municipal waste by converting them into useful subs tances such as ethanol, or acetic acid (Sukhumavasi et al., 1989). Lately, there ha s been an effort to use termites and/or their endosymbionts as a potential resource of fu nctional genes for industrial applications, where useful cellulolytic, lignolytic and aromatic hydrocarbon degradation genes are proposed for use in environmental solutions, biomass utilization and fine chemicals production (Matsui et al., 2009). Cellulase and its Functions There are three types of cellulases: i) endo-1,4-glucanases (EG) (1,4-D-glucan 4-glucohydrolase, EC 3.2.1.4); ii) exoglucanase ( -1,4-D-glucan cellobiohydrolase, EC 3.2.1.91) and finally iii) -glucosidases (BG) ( -D-glucoside glucohydrolase, EC 3.2.1.21) (Han et al., 1995). However, endogenous termite cellulases only consist of endo-1,4-glucanases and -glucosidases. According to Robson and Chambliss (1989), endo-1,4-glucanase works by cleaving the internal -1,4-D-glycosidic bonds at 12

PAGE 13

random. Exoglucanases remove the cellobios e unit from the non-reducing end, while glucosidases acts by cutting cellobiose and cello-oligosaccharides to convert them into glucose. Cellulases are mem bers of the glycosyl hydrolas e superfamily (GHF), which comprise 125 classified families based on amino acid sequence comparison (CAZy, Carbohydrate-Active enZYmes Database, website: http://www.cazy.org). Cellulose Digestion in Lower Termites In lower termites, endosymbiotic protozoans play an important role in the digestion of cellulose (Cleveland, 1923; 1924; 1925). According to Konig et al. (2006), the intestinal microbiota of lower termites c onsists of a mixture of protozoans, fungi, archaea and bacteria. Termites have a long ev olved symbiotic relationship with these gut microbiota, which play important roles in the degradation of cell ulose, hemicellulose, and aromatic compounds, as well as in nitrogen fixation (Breznak and Brune, 1994; Brune and Ohkuma, 2011). Using Coptotermes formosanus Shiraki as an example, its symbiotic protozoan Pseudotrichonympha grassii Koidzumi decomposes highly polymerized cellulose while Holomastigotoides sp. and Spirotrichonympha leidyi Koidzumi utilizes low molecular weight cellulose only (Brugerolle and Radek 2006). Watanabe et al. (2002) proved that protozoan symbionts of C. formosanus produced an endo-1,4-glucanase homologous to GHF7. Nakashima et al. (2002a) found that crude extracts from both the midgut and hindgut produced sugar and reducing sugar from crystalline cellulose and that GHF7 in the hindgut must have origi nated from the protozoans bec ause secreting cells were absent there. Nakashima et al. (2002b) also isolated and characterized cellulase genes from P. grassii in C. formosanus and found that the nuc leotide sequences ( PgCBHhomos ) showed similarity to GHF7 and the prim ary structure was similar to that of 13

PAGE 14

cellulase Cel7A from the aerobic fungus Trichoderma reesei Simmons. Later, Inoue et al. (2005) found that S. leidyi in C. formosanus produced endo--1,4-glucanase similar to GHF5. The site of cellulase secretion for lower termites is the salivary glands (Slaytor, 2000). About 80% of endo-1,4-glucanase activity is found in the salivary gland of C. formosanus with its N-terminal amino acid sequence similar to fungal endo-1,4glucanase and cellobiohydrolases from the GHF7, but not fr om GHF9 (Nakashima and Azuma, 2000). Nakashima et al. (2002a) reported that endo-1,4-glucanase much like GHF9 occured in the salivary gland, foregut and midgut of C. formosanus which transformed cellulose into cellobiose. Mo et al. (2004) also found high -glucosidase activity in the C. formosanus midgut, which transformed cellobiose into glucose. According to Nakashima et al. (2002a), an independent dual cellulose-digesting system (endogenous and exogenous) occurs in C. formosanus This was supported by Tokuda et al. (2002) in a drywood termite, Neotermes koshunensis (Shiraki) and Scharf et al. (2010) in the Eastern subterranean termite, Reticulitermes flavipes (Kollar). Nakashima et al. (2002a) also proposed that cellulose was partly ingested through the termitederived system (endogenous) first, and then the remaining undigested cellulose moved to the hindgut to be further digested by the protozoans (exogenous). According to Nakashima and Azuma (2000), the total localiz ed cellulase activity in the digestive system of C. formosanus are 80.8%, 2.4%, 8.9% and 7. 9% in the salivary glands, foregut, midgut, and hindgut, respectively. Th is two-step cellulose degradation method is highly efficient and these termites were able to assimilate >90% of the wood (Ohkuma, 2003). 14

PAGE 15

Cellulose Digestion in Higher Termites According to Breznak (1984), the intestinal microbiota of higher termites consists of prokaryotes alone. Warnecke et al (2007) who conducted a major metagenomic study of the microflora in the hindgut paunch of Nasutitermes sp. and found 1,750 bacterial 16S rRNA gene sequences that represent 12 phyla and 216 phylotypes supported this. Heterogenous bact erial populations reside wit hin the hindgut of woodeating higher termites (Anklin-Muhlemann et al., 1995). According to Slaytor (1992), because higher termites do not harbor protozoans the role of bacteria in cellulose digestion was unclear as both the lo wer and higher termites produce endogenous cellulases. However, Lenoir-Labe and Ro uland (1993) proved the presence of cellulolytic activity of bacteria in Cephalotermes rectangularis (Sjoestedt). Tokuda and Watanabe (2007) who showed the presence of endocellulases from symbiotic bacteria in the hindgut of Nasutitermes takasagoensis (Shiraki) and Nasutitermes walkeri (Hill) confirmed this. Warnecke et al. (2007) r eported the presence of multiple sets of bacterial genes responsible for cellu lose digestion in the hindgut of Nasutitermes sp.. Nevertheless, according to Bignell (2000), it is unproven if symbi onts are exclusively responsible for cellulose digestion in termi tes. Furthermore, Sl aytor (1992) found no evidence that exocellulases are necessary for cellulose digestion in termites. Earlier, Rouland et al. ( 1988a) purified cellulase IT, II, 1,4-glucan glucanohydrolase and 1,4-glucan cellobiohydrolase from Macrotermes muelleri (Sjostedt) and its symbiotic fungus Termitomyces sp.. According to Rouland et al. (1990), the subfamily Macrotermitinae degra de plant material using double symbiosis with the basidiomycete of the genus Termitomyces sp.; exosymbiosis when the termite workers consume pre-digested inferior fungus comb and endosymbiosis when the 15

PAGE 16

Termitomyces sp. within the termite gut further di gests cellulose together with termitederived cellulase and endosymbiont -derived cellulase. Hyodo et al. (2000) confirmed that the role of mutualistic fungus Termitomyces sp. in the fungus-growing termite Macrotermes gilvus (Hagen) was to degrade lignin while increasing cellulose digestibility for the host. K ouame et al. (2005) purified -fucosidases (-glycosidase A and B) from Macrotermes subhyalinus (Rambur) and proposed that these degrade diand oligosaccharides from hemic elluloses and celluloses. According to Brune and Ohkuma (2011), a dual cellulose digestion system also occurs in higher termites, with both the hosts endogenous cellulase and hindgut bacteria engaging in cellulose degradation. T he site of cellulase secretion for higher termites is the midgut epithelium (Slaytor 2000), although evidence from later studies by Tokuda et al. (2004; 2009) provided some variat ions to the initial finding. More recent developments in the higher termites endogenous cellulases are discussed in detail in Chapters 3 and 4. Problem Statement A major occurrence in termite evolution is a single event loss of flagellates from the hindgut of higher termites (Breznak, 2000; Inoue et al., 2000; Tokuda et al., 2004), in parallel with significant changes in gut structure and nutrition (Bignell, 1994; Donovan et al., 2000). While there have been numerous st udies on the evolution of termites, the evolution of cellulose digestion in termites, especially in higher termites, is still poorly understood. Inward et al. (2007a) found that it was impossible to elucidate the evolution of feeding group within the highe r termites using nuclear and mitochondrial markers. To date, endogenous cellulase sequences have been available only from four termitid species; Odontotermes formosanus (Shiraki), Na. takasagoensis Na. walkeri 16

PAGE 17

17 and Sinocapritermes mushae (Oshima and Maki); in the GenBank database. Due to the large number of termitid species and their diverse feeding habits, there is a substantial gap in our knowledge on endogenous cellulases in the molecular database to help us understand more about the evolution of endogenous cellulose digestion in higher termites. Hence, the targeted enzymes for this study are the endogenous cellulases (endo-1,4-glucanase and -glucosidase) because they are endogenously produced by the termites to digest cellulose. The termitids diverse diets drive selective pressure, which cause changes in the coding sequence of these endogenous cellulases to encode for the appropriate enzyme to match a particular type of food. By examining the sequences that encode for these enzymes, I might begin to understand how termite endogenous digestion evolves at the molecular level. Objectives The main question I wished to answer was How did endogenous cellulose digestion evolve in higher termites? Therefore, my specific objectives were: I. to elucidate the phylogeny of a represent ative selection of higher termites in this study with mitochondrial (16S) an d nuclear (28S) markers, II. to purify, clone, and sequence endo-1,4-glucanases, and to determine its evolution across higher termites of different feeding guilds, III. to purify, clone, and sequence -glucosidases, and to elucidate its evolution across nutritionally diverse Termitidae, and finally, IV. to determine the phylogenetic placement of the bacterial comb-grower in the evolution of endogenous cellulose digestion in termitids and to posit how feeding behavior evolved in higher termites

PAGE 18

CHAPTER 2 THE PHYLOGENY OF HIGHER TERM ITES BASED ON MITOCHONDRIAL AND NUCLEAR MARKERS Termites have been closely linked with cockroaches through various morphological structures (Walker, 1922; McKittrick, 1965; Thorne and Carpenter, 1992; Klass, 1995), as well as the presence of endosymbionts (Cleveland et al., 1934; Koch, 1938). According to Kambhampati (1995; 1996) termites are sister-groups with the cockroach-mantid clade. However, th rough 18S ribosomal RNA, mitochondrial cytochrome oxidase subunit II (COII) and endo-1,4-glucanase gene sequences analyses, Lo et al. (2000) showed that eusocial termites probably shared a stem ancestor with the cockroach, Cryptocercus More recently, Inward et al. (2007b) showed that termites form a cl ade within the cockroaches with Cryptocercus as their sister group, thus corroborating the hypothesis that termites are actually eusocial cockroaches. While these previous studies have deal t with a broader question, the phylogeny within the family Termitidae has received little consensus, especially due to poor taxon sampling (Eggleton, 2001) and an inadequate num ber of genetic loci (Inward et al., 2007a). Similar problems exist not only in mo lecular-based studies, but also in gut anatomy-based studies (Lo and Eggleton, 2011). More comprehensive molecular studies were done by Inward et al. (2007a) and Legendre et al. (2008) using various molecular markers. Later, Engel et al. (2009) published the first morphology-based termite phylogeny that combined fossil and recent data and this has provided insight into the evolution of the family Termitidae. In phylogenetic determinations, mitochondrial DNA (mtDNA) is widely used because it exhibits various genotypic char acters and evolves rapidly (Moore, 1995). 18

PAGE 19

However, because the gene is inherited as a single linkage group ( haplotype), it does not provide independent estima tes of the species tree (Moor e, 1995). Nuclear genes, on the other hand, provide an i ndependent estimate of the s pecies tree because they can be selected from distinct chromosomes (Moore, 1995). Consequen tly, Pamilo and Nei (1988) proposed that one should use sequences of many different and independentlyevolving (unlinked) loci to construct a s pecies tree and Wu (1991) suggested the use of more than five non-orthologous loci to determine species phylogeny. In this study, my goal was to elucidate the phylogeny of a selection of 25 species of higher termites with mitochondrial (16S) ribosomal RNA and nuclear (28S) gene sequences. Both gene markers have been used with success to infer the phylogeny of different species within the family Termi tidae (Inward et al., 2007a, Legendre et al., 2008). I hypothesized that these gene segment s will provide a clear separation among the 25 termitid species (Table 2-1), and sh ow that fungus-growers are the most phylogenetically basal, while woodand soil/ litter-feeders are the most derived groups. I predict that phylogenetic trees inferred from these sequences will provide a clear delineation among different f eeding groups of the higher te rmites and show congruency with the results from endogenous cellulase analyses. Materials and Methods Termite Species The 25 higher termite species used in this study and their feeding habits are listed in Table 2-1. These termitids were selected because of their differing nutritional biology. Nasutitermes corniger (Motschulsky) was the only s pecimen obtained fresh from a laboratory population. Others were pr eserved in ethanol or on Whatman FTA Plantsaver cards. 19

PAGE 20

DNA Extraction DNA from fresh and ethanol-preserved spec imens were extracted using Qiagen DNeasy Blood & Tissue Kit. Those preserved on Whatman FTA Plantsaver cards were processed according to the methodology of Bujang et al. (2011). On ly the termite heads were used to prevent contaminat ion from the hi ndgut microbiota. Polymerase Chain Reaction PCR was performed with 16S mitoc hondrial and 28S (D4 ex pansion segment) nuclear markers as listed in Table 2-2. Am plifications were conducted in 50 L final reaction volumes, each containing 2 L DNA template, 33.8 L dH2O, 5 L PCR Buffer (1.675 L dH2O, 1.25 L 1 M KCl, 1 L 1 M Tris, 0.5 L 5% Tween 20, 0.075 L 1 M MgCl2 and 0.5 L 1% gelatin), 0.04 mM of each dNTP, 50 ng of each primer and 1 U EconoTaq DNA polymerase (Lucigen Corp., Mi ddleton, WI). Depending on the DNA extraction method, either wa ter or a strip of unused FTA card was incorporated into PCR reaction mixtures to serve as negative controls. For the 16S gene, after initial denaturat ion at 94C for 45 s, the thermocycling profile for 40 cycles was 94C for 1 min, 50 C for 1 min, and 72C for 2 min, followed by final extension at 72C for 10 min (Ye et al., 2004) before c ooling to 4C. For the 28S gene segment, after precycle denaturation at 94 C for 2 min, the thermocycler profile for 40 cycles was 94C for 1 min, 50C for 1 min, and 72C for 2 min, followed by a postcycle extension at 72C for 10 min before cooling to 4C. Visualization and Sequencing The amplified PCR products were elec trophoresed through 1% Agarose Low EEO electrophoresis grade agar (Fisher Scientific Pittsburg, PA) using TAE (40 mM Trisacetate and 1 mM EDTA) as running buffer and stained with ethidium bromide (EtBr) 20

PAGE 21

before they were visualized by UV transillumination. Successful amplicons were purified using Wizard PCR Preps DNA Purification System (Promega Corp., Madison, WI) and quantified by comparison with serial diluti ons of uncut lambda DNA (Promega Corp., Madison, WI) by 1% Agar ose (Low EEO) (Fisher Sc ientific, Pittsburg, PA) electrophoresis using TAE. Fi nally, purified PCR products were sent for sequencing using their respective primer pairs to t he BioAnalytical Services Laboratory (BASLab), University of Maryland, Baltimore, MD. Sequence Analysis Consensus sequences for each PCR product were obtained using DNA Baser Sequence Assembler (Heracle BioSoft Pitesti, Romania) and Lasergene SeqMan Pro v7.2.0 (DNASTAR, Inc, Madison, WI) software. Cons ensus sequences were then aligned using Mega 4.1 software (Tamura et al., 2007). Sequenc e similarity or putative sister groups of nucleotide sequences we re searched using BLAST (Basic Local Alignment Search Tool) in Na tional Center for Biotechnology Information, USA (website: http://blast.ncbi.nlm.nih.gov/Blast.cgi). Sequence variation analysis was performed with Mega 4.1 software (Tamura et al., 2007). T he homogeneity test of base frequencies was conducted using PAUP* 4.0 (Swofford, 2002, Sinauer Associates, Inc. Publishers, Sunderland, MA). Phylogenetic analyses were done under Baye sian criteria. The appropriate model of DNA substitution for each of the 16S, 28S, and their combined dataset was chosen using MODELTEST 3.0 (Posada and Crandall, 1998). The parameters obtained under Akaike's Information Criterion (AIC) were subsequently used in PAUP* 4.0 (Swofford, 2002, Sinauer Associates, Inc. Publishers, Sunderland, MA). Tr ee inferences and posterior probabilities estimation with Markov Chain Monte Carlo sampling were carried 21

PAGE 22

out using MrBayes v3.1.2 software (Ronqu ist and Huelsenbeck, 2003) for 1,000,000 generations and burnin setting at 1,000. All si tes with missing data were regarded as missing characters. Results and Discussion I was able to sequence partial fragment s of about ~600 bp and ~700 bp for the 16S (GenBank Accession No. xxxxxxx to xxxxxxx) and 28S (GenBank Accession No. xxxxxxx to xxxxxxx) genes, respectively, which prov ided a total of ~1,300 bp characters (Fig. 2-1) (GenBank Accession No. xxxxxxx to xxxxxxx ). The results from the inferred tree supported my earlier hypothe sis that fungus-growers ar e the most phylogenetically basal group, while woodand soil/litter-f eeders are the most der ived groups, with the bacterial comb-feeder intermediate betw een the other subgroups (Fig. 2-2). Results also supported earlier findings by Inward et al. (2007a), Legendre et al. (2008) and Engel et al. (2009), which posited that t he family Termitidae is monophyletic. Macrotermitines formed a separate basal cla de with strong posterior probability, distal from the other subgroups. The next major br anch is the subfamily Sphaerotermitinae, which corroborated the molecular findings by Inward et al. (2007a). However, based on the morphological data of Engel and Krishna (2004), they placed Sphaerotermitinae as a sister group to the Macrotermitinae because of a combination of plesiomorphic and apomorphic traits. Moving apically, the monophyly of subfamily Apicotermitinae is consistent with the findings of Inward et al. (2007a). The inferr ed tree also showed that the apicotermitines form a separate clade away from other soil/ litter feeders. A paraphyletic relationship between the soil/litter feeders and the w ood/grass/lichen/litter-feeders was also revealed. This outcome supported the study of Donovan et al. (2001), who outlined the 22

PAGE 23

23 evolution of feeding groups based on gut cont ent analysis and morphological characters of worker termites. They suggested that t he ancestor of apicotermi tines, termitines and nasutitermitines might have been the soil-feeders, which acquired the hindgut bacterial community from ingested soil. Hence, the presence of wo od-feeding termitines in the same clade as soil interface-feeding termiti nes agreed with the trend of feeding group evolution proposed by Donovan et al. (2001). As it is more conceivable to think that the soil/litter-feeders are the most phylogenetic ally apical group, the presence of woodfeeding termitids amongst them shows the independent progression of this group (Donovan et al., 2001). The termitines were paraphyletic with the syntermitines and nasutitermitines, consistent with the findings of Donovan et al. (2001) and Inward et al. (2007a). Instead of nesting within the nasutitermitines as init ially expected, the syn termitines were found nesting within the termitines, as was show n by Inward et al. (2007a). Finally, my findings departed in some respects from Engel et al. (2009) and Legendre et al. (2008) regarding the nasutitermitines. Data from th is study showed that Nasutitermitinae is polyphyletic, with Subulitermes baileyi (Emerson) falling outside of the otherwise monophyletic Nasutitermitinae group. Thus, this outcome is in agreement with Inward et al. (2007a).

PAGE 24

Table 2-1. List of 25 termitid species used in this study Species Subfamily Feeding habit Locality Macrotermes carbonarius (Hagen) Macrotermitinae Wood/Grass/Litter, Fungus-grower Malaysia, Pulau Pinang Macrotermes gilvus (Hagen) Macrotermitinae Wood/Grass/Litter, Fungus-grower Malaysia, Pulau Pinang Macrotermes subhyalinus (Rambur) Macrotermitinae Wood/Grass/Litter, Fungus-grower Tanzania, -4.67710/29.62260 Microtermes pallidus (Haviland) Macrotermitinae Wood/Grass/Litter, Fungus-grower Malaysia, Pulau Pinang Odontotermes formosanus (Shiraki) Macrotermitinae Wood/Grass/Litter, Fungus-grower Taiwan, Pingtung County Odontotermes hainanensis (Light) Macrotermitinae Wood/Grass/Litter, Fungus-grower Malaysia, Pulau Pinang Sphaerotermes sphaerothorax (Sjoestedt) Sphaerotermitinae Wood, Bacterial comb-grower Congo, Pointe Noire Syntermes grandis (Rambur) Syntermitinae Grass/Litter French Guyana, 5.67540/-53.59198 Rhynchotermes bulbinasus Scheffrahn Syntermitinae Grass/Litter Co lombia, 9.31634/-74.90097 Amitermes dentatus (Haviland) Termitinae Wood Malaysia, Pulau Pinang Amitermes foreli Wasmann Termitinae Grass Colombia, 8.92399/-75.8381 Microcerotermes crassus Snyder Termitinae Wood Malaysia, Pulau Pinang Globitermes sulphureus (Haviland) Termitinae Wood/Litter Malaysia, Pulau Pinang Hospitalitermes bicolor (Haviland) Nasutitermitinae Lichen Malaysia, Pulau Pinang Constrictotermes cavifrons (Holmgren), Nasutitermitinae Lichen French Guyana, 5.02389/-53.0249 Constrictotermes guantanamensis Krecek, Scheffrahn and Roisin Nasutitermitinae Lichen Cuba, 19.934/-75.098 Nasutitermes corniger (Motschulsky) Nasutitermitinae Wood USA, Dania Beach 24

PAGE 25

25 Table 2-1. Continued Species Subfamily Feeding habit Locality Nasutitermes takasagoensis (Shiraki) Nasutitermitinae Wood Japan, Iriomate Island Nasutitermes sp. Nasutitermitinae Wood Malaysia, Pulau Pinang Subulitermes baileyi (Emerson) Nasutitermitinae Soil/Li tter Venezuela, 10.18533/-65.82158 Pericapritermes nitobei (Shiraki) Termitinae Upper soil/Litter Taiwan, Taitung County Pericapritermes sp. Termitinae Upper soil/Litter Malaysia, Pulau Pinang Sinocapritermes mushae (Oshima and Maki) Termitinae Upper soilLitter Taiwan, I-Lan County Grigiotermes metoecus Mathews Apicotermitinae Soil/Litte r Venezuela, 10.40245/-68.00039 Anoplotermes schwarzi Banks Apicotermitinae Soil/Litte r Guatemala, 14.69649/-89.62552 Table 2-2. List of nuclear and mitochondr ial marker primers used in this study Name Gene Orientation Sequence (5 to 3) References Mitochondrial 16Sar 16S Forward CCGGTCTGAACTC AGATCACGT Simon et al., 1994, Marini and Mantovani, 2002 16Sbr 16S Reverse CGCCTGTTTAAC AAAAACAT Simon et al., 1994, Marini and Mantovani, 2002 Nuclear Hux 28S Forward ACACGGACCAAGGAGT CTAAC Inward et al., 2007a Win 28S Reverse GTCCTGCTGTCTTAAGCAACC Inward et al., 2007a

PAGE 26

A B Figure 2-1. Agarose gel show ing PCR amplificati on from 25 species of higher termites used in this study with positive an d negative (water) control. A) PCR amplification of 16S mitochondrial (16S) ribosomal RNA gene and B) PCR amplification of nuclear (28S) gene. 26

PAGE 27

27 Figure 2-2. Consensus Bayesian tree inferred from combined 16s and 28s gene sequences (total ~1300 bp). Numbers abov e branch nodes indicate posterior probabilities recovered by the Bayesian analysis. Branch lengths are proportional to the number of changes.

PAGE 28

CHAPTER 3 A PHYLOGENETIC AND EVOLUTIONARY ST UDY OF ENDO-BETA-1,4-GLUCANASE IN HIGHER TERMITES Endo-1,4-glucanases are members of GHF9, which comprise four known representatives and 732 components, as listed in CAZy (Carbohydrate-Active enZYmes Database, website: http://www.cazy.org ). They occur in various insects, such as beetles, flies, cockroaches, and termites (Willis et al., 2010) and are an important component of termite digestion because t hey randomly hydrolyze the internal -1,4-Dglycosidic bonds on cellulose and convert it into cellobiose and cello-oligosaccharides (Robson and Chambliss, 1989). Slay tor (1992) reported that endo-1,4-glucanase is active against both crystalline cellulose and ca rboxymethylcellulose, hence proving that exocellulases are not crucial for cellulose digestion in termites. In lower termites, Watanabe et al. ( 1998) were the first to sequence an endogenous cellulase, RsEG (GenBank Acce ssion No. AB008778) from a termite, Reticulitermes speratus (Kolbe), after which endo-1,4-glucanases from other lower termite species were also sequenced (Tokuda et al., 2004; Zhou et al. 2007; Zhang et al. 2011). Zhang et al. (2009) cloned and overexpressed CfEG3a from C formosanus in Escherichia coli and found that the hydrolytic activity of the recombinant native form (nCfEG) was higher than that of the C-terminal His-tagged form (tCfEG). Zhang et al. (2010) later conducted a functional analysis on recombinants of endo--1,4-glucanase and -glucosidase obtained from the cDNA library of C. formosanus and found successful conversion from cellulose to glucos e. Fujita et al. ( 2008) reported that in Hodotermopsis sjoestedti Holmgren, endo--1,4-glucanase activity was highest in the salivary gland, and was significantly higher in termite worker s than soldiers. 28

PAGE 29

Due to the loss of flagellates from their hindgut, higher temites depend heavily on endogenous cellulases for cellulose digestion (Slaytor et al., 1997; Ohkuma, 2003; Tokuda et al., 2004). Kovoor (1970) was the first to suggest that Microcerotermes edentatus Wasmann produce their own cellulase. According to Lo et al. (2011), endogenously-produced cellulases play a majo r role in termitid metabolism. Endo-1,4glucanase activity has been quantified in numerous species such as Trinervitermes trinervoides (Sjostedt) (Potts and Hewitt, 1973), Macrotermes natalensis (Haviland) (Martin and Martin, 1978), Speculitermes cyclops Wasmann (Mishra and Sen-Sarma, 1985a), Na. walkeri Nasutitermes extiosus (Hill) (Hogan et al., 1988), Nasutitermes lujae (Wasmann) (Chararas and Noirot 1988), Crenetermes albotarsalis (Sjostedt) (Rouland et al., 1989a), M. subhyalinus, Macrotermes michaelseni (Sjostedt) (Veivers et al., 1991), Na. takasagoensis (Tokuda et al., 1997; 2005; Tokuda and Watanabe, 2007; Fujita et al., 2008) and O. formosanus (Yang et al., 2004; Tokuda et al., 2005). Slaytor (1992) reported that the cellulolytic activity in the hindg ut of higher termites was either undetectable or very low. Later, Slaytor ( 2000) reported that in areas where the gut microflora was absent, or present in trace amounts, such as the salivary glands and midgut, cellulolytic activity was found to be high. The entire sequence of the coding region of Na. takasagoensis NtEG (GenBank Accession No. AB019146) was first determined by Tokuda et al. (1999) using a PCRbased strategy and found to consist of 10 exons and interrupted by nine introns. Khademi et al. (2002) later r eported the structure of an endo-1,4-glucanase from Na. takasagoensis Based on endo-1,4-glucanase sequences from Na. takasagoensis a 29

PAGE 30

sea squirt, Ciona intestinalis (Linnaeus) and an abalone, Haliotis discus hannai Ino, Lo et al. (2003) suggested that GHF9 was present in the ancestor of all bilaterian animals. From endo-1,4-glucanase sequences of only three species of higher termites ( O. formosanus Na. takasagoensis and S. mushae ), Tokuda et al. (2004) concluded that the more phylogenetically basal group is Ma crotermitinae (fungus-g rowers), while the more apical groups are Termitinae (soilfeeders) and Nasutitermitinae (wood-feeders). They also found that the expression site s of endogenous cellulases in lower termites and O. formosanus were in the salivary glands while in those more distal expression sites occured in the midgut. Termitids hav e a wide range of feeding substrates, but there is a paucity of molecular information on endo-1,4-glucanases to help us understand the evolution of endogenous cellulose digestion in higher termites. Furthermore, Inward et al. (2007a) found t hat it was impossible to elucidate the evolution of feeding groups wit hin the higher termites using nuclear and mitochondrial markers only. In this chapter, my goal was to purify, clone, and sequence endo-1,4glucanases, and to compare its evolution across higher termites of different feeding guilds. I hypothesized that the evolution of endo--1,4-glucanase should be congruent with the evolution of termites according to the mitochondrial and nuclear markers. I predict that the endo-1,4-glucanase phylogenetic tr ee will provide clues to understand the evolution of feeding groups within the higher termites. The bacterial comb-grower, S. sphaerothorax is of particular interest because it is the only species of higher termite known to cultivate bacterial combs and consum e the bacterial pellets following bacterial action. Despite their unique feeding habit, how their feeding behavior was derived and 30

PAGE 31

where this feeding group falls along the ev olutionary line has yet to be investigated. Hence, the phylogenetic placement of the bac terial comb-grower in the evolution of endo-1,4-glucanase was also determined. Materials and Methods mRNA Extraction and cDNA Synthesis Only the termite heads and salivary glands were used to prevent contamination from the hindgut microbiota. Messenger RNA was extracted using Aurum Total RNA Mini Kit (Bio-Rad Laboratories, Hercules CA). cDNA synthesis was performed with iScript cDNA Synthesis Kit (Bio-Rad Laborat ories, Hercules, CA) according to the manufacturers protocol. Polymerase Chain Reaction PCR was performed with specifically-designed endo-1,4-glucanase primers (Clone Manager 9 Professional Edit ion, Scientific & Educati onal Software, Cary, NC) as listed in Table 3-1. The procedure used was the same as described in Chapter 2, but used a different thermal cyclin g program. The temperature pr ofile for the first cycle was 94C for 2 min, 52C for 2 min, and 72C fo r 3 min. For the remaining 44 cycles, the temperature profile was 94 C for 1 min, 52C for 2 mi n, and 72C for 3 min before cooling to 4C. Cloning After electrophoresis of PCR products thro ugh a 1% Agarose Low EEO gel (Fisher Scientific, Pittsburg, PA) followed by EtBr staining and visualization by UV transillumination, amplified DNA products were gel-purified with Lonza SeaPlaque GTG Agarose (Lonza Rockland, Inc., Rockland, ME) and excised bands were purified using Wizard PCR Preps DNA Purification System (Promega Corp., Madison, WI). The 31

PAGE 32

purified fragments were ligated overnight with pGEM-T Vector System I (Promega Corp., Madison, WI) at 4C and then used to transform into One Shot TOP10 Chemically-competent E. coli (Invitrogen Corp., Carlsbad, CA). The transformed bacterial cultures were grown overnight at 37C on Luria-Bertani (LB) medium (15g Bacto Agar, 10g Bacto Tryptone, 5g Bacto Yeast Extract, 10g NaCl in 1L volume) containing 100 g/ml ampicillin (A100) wit h isopropyl beta-D-thiogalactopyranoside (IPTG) and 5-bromo-4-chloro-3-indolyl-beta-D-galactopyranoside (Xgal) as an overlay on the agar to enable blue/white colony screening. Fifty isolated white colonies were carefully selected and grown overnight at 37C on LBA100 media patch plate. PCR was carried out again on each transfo rmed bacterial colony as described earlier, but using dH2O-suspended bacterial cells of each clone as the DNA template. Restriction Fragment Length Polymorphism (RFLP) with restriction enzymes Hinf 1 and Mse 1 (New England BioLabs, Waverley, MA, USA) was conducted on each amplified product from successfully-transformed products to assess polymorphisms. These restriction enzymes were selected after virtual electrophoresis screening of endo-1,4glucanase sequences from N. takasagoensis (GenBank Accession No. AB013272) and N. walkeri (GenBank Accession No. AB013273) with pDRAW32 1.0 (AcaClone software, http://www.acaclone.com). Diges ted products were electrophoresed through an 8% nondenaturing polyacrylamide gel using TBE (90mM Tris-borate, 2mM EDTA) as running buffer, stained with EtBr and then vi sualized by UV transillumination. Forty-one clones were selected and the recombinant plasmids were grown overnight at 37C in LB broth (10g Bact o Tryptone, 5g Bacto Yeast Extract, 10g NaCl in 1L volume). After purification with Wizard Plus Minipreps DNA Purification 32

PAGE 33

Systems (Promega Corp., Madison, WI), the plasmids were each quantified by comparison with serial dilutions of uncut lambda DNA (Promega Corp., Madison, WI) in 1% Agarose Low EEO electrophoresis grade agar (Fisher Scientific, Pittsburg, PA). Finally, they were sent for insert sequenci ng with the M13F and M13R primer pair at the Interdisciplinary Center for Biotechnology Research DNA Sequencing Core Laboratory, University of Florida, Gainesville, FL. Sequence Analysis Consensus sequences were assembled and aligned as described in Chapter 2. Sequence similarity or putative sister gr oups of nucleotide and amino acid sequences were searched using BLAST (Basic Local Alignment Search Tool) at the National Center for Biotechnology In formation, USA (website: http://blast.ncbi.nlm.nih.gov/Blast.cgi). Nucleotide and amino acid sequence variation analysis was performed with Mega 4.1 softwa re (Tamura et al., 2007). The homogeneity test of base frequencies was conducted using PAUP* 4.0 (Swofford, 2002, Sinauer Associates, Inc. Publishers, Sunderland, MA). Phylogenetic analyses followed the outline described in Chapter 2. Amino acid sequences were subsequently aligned with Mega 4.1 (Tamura et al., 2007). Catalysis and substrate-binding sites were inferred from Sakon et al. (1997). Finally, N-glyco sylation site search was performed with NGlycoSite HCV sequence database ( http://hcv.lanl.gov/content/s equence/GLYCOSITE/glycosite.html) (Zhang et al., 2004). Results and Discussion Single PCR fragments of about 1.35 kb in length from 23 species of higher termites were amplified (Fig. 3-1). Based on RFLP profiles of the clones, 41 different clones from 23 termitid species were obtained (Fig. 3-2) (GenBank Accession No. 33

PAGE 34

xxxxxxx to xxxxxxx). Collectively, the sequences showed at least 70% amino acid similarity with Mastotermes darwiniensis Froggatt endo--1,4-glucanase and at least 61% identity with Panesthia cribrata Saussure endo--1,4-glucanase. Protein BLAST searches also showed similarity to me mbers of GHF9 family, which contain endo-1,4glucanases from plants, bacteria and slime molds. However, plants and bacteria GHF9 lacks the linker and cellulose -binding domains sometimes found in members of this family (Tomme et al., 1995). Each of the sequences contain the pr oton donor glutamate and the nucleophile aspartate, which are typi cal characteristics of GHF9 (Watanabe and Tokuda, 2001; Zhou et al. 2007). Thes e findings were consistent with endo-1,4glucanases from other termitids (Tok uda et al., 2004) and lower termites (Watanabe and Tokuda, 2001). The sequences obtained were consistent with the findings of Nakashima et al. (2002a) and Zhou et al. (2007) in as mu ch as each sequence possesses conserved motifs involved in substrate binding and catalysis as described by Khademi et al. (2002). These motifs are NEVA with E (Glu, glutamic acid) as the proton donor and DAGD with both Ds (Asp, aspartic ac id) as the nucleophiles (Fig. 3-3). The substrate-binding cleft structure allows random binding on the cellulose chain (Khademi et al. 2002; Zhou et al. 2007). The N-linked glycosylation, putative catalytic glutamic acid and aspartic acid residues (proton donor and nucleophile, respectively) and stop codon positions are shown in Fig. 3-3. Only a few insertion/deletion events were evident among all the sequences obtained. The inferred endo-1,4-glucanase phylogenetic tree (Fig. 3-4) showed congruency with the mitochondrial/ nuclear tree (Fig. 2-2). Generally, the fungus-growers 34

PAGE 35

were the most basal group and the most di stal diphyletic feeding groups were the soil/litterand wood/lichen/grass/litter-feeders. Results of endo--1,4-glucanase sequences suggest that the bacterial comb-g rower is phylogenetically placed as the missing link between the fungus-growers and the soil/litterand wood/lichen/grass/litter-feeders. Up to four paralogous copies from each of the 23 termitid species were obtained, with high degrees of substitutions among so me paralogs. The high divergence among these paralogs suggests the involvement of different alleles. As predicted, endo-1,4glucanase paralogs from the same species clustered with one anot her except in the case of O. formosanus Although Lo et al. (2011) stated th at the roles of multiple endo-1,4-glucanase gene copies remained unclear I support the neofunctionalization hypothesis by Ohno (1970), Force et al. (1999) and Hahn (2009), which suggests a different function from the original gene. In terms of endo-1,4-glucanase evolution, fungus-growers were the most phylogenetically basal group, concordant with the mitochondrial/nucl ear data. This was also in agreement with earlier studies by To kuda et al. (2004), who further showed that in O. formosanus endo-1,4-glucanase is mostly expressed in the salivary glands, as observed in lower termites. There was a st rong posterior probability for the monophyly of macrotermitines, although there was poor phylogenetic resolution for endo-1,4glucanase sequences between specie s, except in the case of M. carbonarius and M. gilvus According to Lo et al. (2011), although endogenously-produced cellulases are less important in fungus-growers, they are still preserve d in the genome. 35

PAGE 36

The low divergence of endo-1,4-glucanases from this feeding group was due to their high dependency on fungi to digest their f ood. With the loss of flagellates in the ancestors of the family Termitidae [through abr asion, as proposed by Rouland-Lefevre and Bignell (2001)], the macrot ermitines close association with cellulolytic fungi has allowed them to continue to use wood and litter as a raw product in their food production. By relying on fungal symbionts to process their food (Darlington, 1994), the fungus-growers have some form of an external gut that partially digests cellulose for them. The fungus grows on fungus combs, which were constructed from primary feces (Grasse, 1978). According to Sands (1969), this method of fecal reuse substituted the role of proctodeal feeding. From a nutrition point of view Sands (1956) found that the exclusion of fungi from Odontotermes badius (Haviland) diet caused an effect similar to that when starved. Because both the comb and fungi are major sources of food for the fungus-growers (Arshad et al., 1 987), I suspect that the type s of fungi with which the termites are associated drives selective pre ssure to code for the different substitution patterns to match a particular type of fungus. In the monophyletic sphaerotermitine, mole cular evidence suggests that bacterial comb-feeding termites bridge the transition fr om being fungus-growers to soil/litterand wood-feeders. Donovan et al. (2001) sugge sted that the soil-f eeding ancestor of Apicotermitinae, Termitines and Nasutitermitinae acquired the bacterial community from ingested soil. However, I pr opose that their hindgut bacterial community was acquired through bacterial comb-feeding. I was able to obtain four paralogous copies from S. sphaerothorax The high divergence among t hese four paralogs suggests the involvement of different alleles. In fungus-growing termites, the fungal symbiont, 36

PAGE 37

basidiomycete Termitomyces spp. grows on termite-constructed fungus-combs (made from mylosphere or primary faeces) as mycelium or seasonal basidiocarps (Heim, 1977). According to Garnier-Sillam (1989) Sphaerotermitinae builds two types of bacterial combs within its nest; the first occurs as a result of primary feces accumulation, while the second is by final feces accumulation (Garnier-Sillam, 1989). Both types of combs contain nitrogen-fixi ng bacteria, with the older (lighter) comb containing a significantly higher number of bacteria than the younger (darker) comb (Garnier-Sillam, 1989). Just as fungus-gro wers consume the comb following fungal action, bacterial comb-feeder eats up the li ght-colored pellets formed by bacterial action. Garnier-Sillam (1989) also reported that one bacterial type is more abundant than the others within the sphaerotermitine nest. I suspect that different bacterial strains present within the nest might produce bacterial pellets of different chemical/nutritional properties. In support of the i dea of different functionality, I propose that the various bacterial strains present in bacterial combs might have caused different endo-1,4glucanase paralogs to occur. I suspect t hat different paralogs may be used to degrade pellets of different physical and chemical properties that resulted from different bacterial strains. A shift has occured from having an ext ernal gut in the fungusand bacterial comb-grower to having an internal gut in the wood/lichen/grass/litterand soil/litterfeeders. The inferred tree suggested a diphyle tic relationship between the soil/litterfeeders and the wood/lichen/grass/ litter-feeders. According to Donovan et al. (2001), the ancestor of woodand soil/litter-feeders may have been the soil-feeders, where the acquisition of bacterial comm unity within their hindgut was presumably achieved from 37

PAGE 38

ingested soil. Earlier, Noirot (1992) suggested that the majo r source of nutrient for soilfeeding termites is the bacterial-fermented ar omatic humus compound, but later, Ji and Brune (2001) provided evidence that soil-feeding termites, Cubitermes orthognathus (Emerson) utilize plant and bacterial polysacc harides as well as microbial biomass as their nutrient source. Although Tokuda et al. (2004) predict ed the next major branch to be the apicotermitines, sequences from this st udy showed a strong separation between Pericapritermes and the other soil-feeders, indicati ng high nucleotide substitutions between these groups. According to Brauman et al. (2000), genuine soil-feeders feed widely in the soil profile. Eggleton et al. (1995) and Egglet on and Bignell (1995) classified the wood/soil interface feeders as those that feed on hi ghly-humified but still recognizable organic matter. Donovan et al. (2001) have categorized Pericapritermes under group III feeders that feed on the organic rich upper layers of the soil, while Grigiotermes was placed under group IV feeders (tr ue soil feeders) that ingest mineral soil. Tokuda et al. (2004) s howed that the majority of endo-1,4-glucanase activity in S. mushae occurs in the midgut, although its overall endo-1,4-glucanase activity was almost imperceptible when compared with that of lower te rmites. Regardless, it is retained in the genome despite its insignificance for soil/li tter-feeders (Lo et al., 2011). Within the wood/lichen/grass/litter-feeding clade, it was interesting to note the polyphyly of the nasutitermitines and the str ong diphyletic relationship between the most distal groups, the lichen-feeder s and the grass/litter feeders. The high divergence among Globitermes and Amitermes paralogs also suggests the involvement of different alleles. I speculate that this high level of substitution was because the selection 38

PAGE 39

39 pressure on wood-feeders was more relax ed, thus allowing for the occurrence of multiple endo-1,4-glucanase paralogs to achieve increased resource utilization efficiency. Nonetheless, Tokuda et al. (2004) reported that the overall endo-1,4glucanase activity in Na. takasagoensis was only 10% than that of lower termites. The evolutionary origin of the endo-1,4-glucanase obtained here remains uncertain because of the unavailability of complete coding sequences. However, Zhou et al. (2007) found evidence to support the i dea of vertical transfer of GHF9 from a cockroach ancestor in R. flavipes Also, in Na. takasagoensis endo--1,4-glucanase NtEG (GenBank Accession No. AB019146), i dentical intron positions between GHF9 genes from Na. takasagoensis with those from two marine organisms suggested vertical transfer of this gene from a common ancestor. Nevertheless, the gene transfer status prior to that common ance stor remains unanswered.

PAGE 40

Table 3-1. List of endo-1,4-glucanase primers used in this study Name Gene Orientation Sequence (5 to 3) EG1f EG Forward GCGGACCTGAAGGTAACTTG EG1r EG Reverse AGTACGCGCTGAGTTCCATC EG2f EG Forward CGCTTTGCCAAGGTGCTTAC EG2r EG Reverse GGCGAGAGCTGATTGGAAAC EG3f EG Forward CATGCTGCTTGCGACTAC EG3r EG Reverse AGCGACGAGAGCTGATTG EG4f EG Forward ATGATAGCGGCCAGAACG EG4r EG Reverse TAACCCAGCGCTACGAGAAC EG5f EG Forward GCTGCCGACTACAAGAAAG EG5r EG Reverse GGCGGATCAATGACCCAAC EG6f EG Forward CTTGGCGGAAAGATTCAG EG6r EG Reverse GTTGAGTGCCATCAAGAG EG7f EG Forward TTTGCCAAGCTGCGTATG EG7r EG Reverse ATAATCGCAGGCCACTTC EG8f EG Forward AAGAACGGACTGGACCTTAC EG8r EG Reverse TGGGCCACTAATAGCCTAAC EG9f EG Forward AAGGACTCCGCCTTAAACG EG9r EG Reverse ATACGAAACGGCAGGACAG EG10f EG Forward TTTGCCAAACTGCTTACC EG10r EG Reverse AGCCTGCGTTATAATCTG EG11f EG Forward GGAAAGATTCAGCCCTGAAC EG11r EG Reverse CCATCAAGTGGGCATGAAC EG12f EG Forward AAGGATTCCGCCCTCAATG EG12r EG Reverse CGTTACGAGAACGGAGATAG EG13f EG Forward AGCTGCTTACGACTATAACC EG13r EG Reverse TGGAAGCCTGCGTTATAATC 40

PAGE 41

Figure 3-1. Agarose gel show ing PCR amplification of endo-1,4-glucanase from four representative species of higher termites used in this study with positive and negative (water) control. 41

PAGE 42

A B Figure 3-2. Restriction fragment length profiles of four representative endo--1,4glucanase clones (ca. 1.35 kb) from four species of higher termites used in this study. A) Digestion with Hinf 1 and B) Digestion with Mse 1. 42

PAGE 43

Figure 3-3. Multiple alignments of endo-1,4-glucanase amino acid sequences from higher termites. The number of encoded amino acids was listed next to the sequence names. N-linked glycosylation site s are highlighted in red. Blue dot indicates putative proton donor. Red dot indicates putative nucleophile. Black dot indicates the position of stop codon. 43

PAGE 44

Figure 3-3. Continued. 44

PAGE 45

Figure 3-3. Continued. 45

PAGE 46

Figure 3-3. Continued. 46

PAGE 47

Figure 3-3. Continued. 47

PAGE 48

48 Figure 3-4. Consensus Bayesian tree inferred from endo-1,4-glucanase sequences. Numbers above branch nodes indicate posterior probabilities recovered by the Bayesian analysis. Branch lengths are proportional to the number of changes.

PAGE 49

CHAPTER 4 A PHYLOGENETIC AND EVOLUTIONARY STUDY OF BETA-GLUCOSIDASE IN HIGHER TERMITES -glucosidases (or -D-glucoside glucohydrolase) are Class 1 and 2 glycosidases which degrade cellobiose and ot her disaccharides (Marana et al., 2001). -glucosidases are members of GHF1, whic h comprise 19 known representatives and 2,889 components, as listed in CAZy (C arbohydrate-Active enZYmes Database, website: http://www.cazy.org ). -glucosidase occurs in various insects, including termites (reviewed by Willis et al., 2010). It is a crucial component of termite digestion because it completes cellulose digesti on by cleaving cellobiose and cellooligosaccharides converting them into glucose (Robson and Chambliss, 1989). According to Lo et al. (2011), endogenously-produced cellulolytic enzymes are especially important in higher termites due to the lack of flagellates in their hindgut. In higher termites, Potts and Hewitt (1972) reported -glucosidase activity in the head and gut of T. trinervoides However, at the time its function was unknown because their findings showed that the -glucosidase was incapable of hydrolyzing cellobiose. -glucosidase activity was later r eported in the foregut and midgut of C. albotarsalis (Rouland et al., 1989a) and shown as being most active in the salivary glands of M. muelleri (Rouland et al., 1989b). More recently, Binate et al. (2008) purified and characterized two -glucosidases ( -Glc A and B) from Macrotermes bellicosus (Smeathman) intended fo r glycobiotechnology. Tokuda et al. (2002) were the first to molecularly characterize -glucosidase from termites, which was obtained from N. koshunensis (NkBG) (GenBank Accession No. AB073638). Following that, Ni et al. (2007) successfully overexpressed this NkBG cDNA in E. coli which showed a 3-fold increase in the recombinant enzymes specific 49

PAGE 50

activity. Later, Tokuda et al. (2009) successfully sequenced -glucosidases from the salivary gland and midgut of a higher termite, Na. takasagoensis Scharf et al. (2010) later sequenced two -glucosidases, RfBGluc-1 and RfBGluc-2, which were expressed in the salivary glands and foregut of R. flavipes Zhang et al. (2010) recovered and cloned an endo-1,4-glucanase and a -glucosidase from the cDNA library of C. formosanus and found successful cellulose to glucose conversion using the recombinant enzymes. From an earlier study, Tokuda et al. (1997) found that the majority of glucosidase activity occurs in the salivary glands of Na. takasagoensis although they were uncertain of the function at the time (Tokuda et al., 2009), The recent discovery of egg-mimicry by the cuckoo fungus, Matsuura et al. (2009) demonstrated that glucosidase and lysozyme constitute the termite-egg recognition pheromone in Reticulitermes termites. Termite eggs were administered with -glucosidase during transportation into the nursery chamber and egg-grooming. La ter, Matsuura and Yashiro (2010) reported a similar type of te rmite egg-mimicry by a different fungus with Na. takasagoensis This is not surprising, because according to Ketudat Cairns and Esen (2010), -glucosidases occur universally and serve many functions, including defense and plant-insect interactions. Among different termitid castes, Deng et al. (2008) showed that -glucosidase activity was highest in O. formosanus workers but that the level was not significantly different between soldiers and the T. albuminosus fungus with which the termites were associated. Fujita et al. (2008) reported that -glucosidase activity was highest in the midgut of all castes of Na. takasagoensis Low titers of -glucosidase were also found in 50

PAGE 51

the salivary glands of major and minor workers of Na. takasagoensis (Fujita et al., 2008), suggesting a similar egg-marking function to what was observed in R. speratus. As mentioned previously, it is impossibl e to elucidate the evolution of feeding group within the higher termites using nuclear and mitochondrial markers alone (Inward et al., 2007a). To date, there have only been eight -glucosidase sequence accessions from two species of higher termites [ Na. takasagoensis a wood-feeder (GenBank Accession No. AB 508954-AB508960) and O. formosanus a fungus-grower (GenBank Accession No. GU591172)] deposited in the GenB ank database. The family Termitidae constitutes the majority of all termite spec ies and the scarcity of molecular information on -glucosidase makes it impossible to cu rrently understand how termite endogenous digestion evolves at the molecular level. Hence, in this study, my goal was to purify, clone, and sequence -glucosidases, and elucidate its evolution across nutritionally diverse Termitidae. I hypothesized that the evolution of -glucosidase will be the same as endo--1,4-glucanase, and congruent with the results from mitochondrial and nuclear markers. In addition, I aimed to determine the phylogenetic placement of S. sphaerothorax in the evolution of glucosidases among the higher termites. Materials and Methods Polymerase Chain Reaction PCR was performed with specifically-designed -glucosidase primers (Clone Manager 9 Professional Edition, Scientific & Educational Software, Cary, NC) as listed in Table 4-1. Amplifications were conduct ed using Terra PCR Direct Polymerase Mix (Clontech Laboratories, Inc., CA) in 50 L fi nal reaction volumes, each containing 12 L dH2O, 25 L 2X Terra PCR Direct Buffer (with Mg2+ and dNTP), 2 L cDNA template 51

PAGE 52

(as was used in Chapter 3), 100 ng of each primer and 1.25 U Terra PCR Direct Polymerase Mix. After the init ial denaturation at 98C for 2 min, the temper ature profile for 35 cycles was 98C for 10 s, 60C for 15 s, and 68C for 2 min before cooling to 4C. Cloning and Sequence Analysis The cloning procedures followed the method described in Chapter 3. However, the plasmids were sent for insert sequencin g with M13F and M13R primer pairs to BioAnalytical Services Laboratory (BASLab), Un iversity of Maryland, Baltimore, MD. Sequence assembly, alignment and analyses of nucleotide and amino acid sequences also followed the procedures outlined in Chapter 3. Results and Discussion Of 25 termitid species, I was only able to amplify single PCR fragments of about 1.6 kb in length from four species of higher termites (Fig. 4-1). Based on RFLP profiles of the clones, I was able to obtain five different clones from M. carbonarius S. sphaerothorax Anoplotermes schwarzi Banks and R. bulbinasus (Fig. 4-2) (GenBank Accession No. xxxxxxx to xxxxxxx). Sequences from this study showed at least 63% amino acid similarity with N. koshunensis -glucosidase and at least 52% identity with Tenebrio molitor Linnaeus -glucosidase (GenBank Accession No. AAG26008). Protein BLAST search also showed that the amino acid sequences were similar to glucosidases in GHF1. This was consistent with -glucosidases from Na. takasagoensis (Tokuda et al., 2009), although a -glucosidase from GHF3 has been found in the salivary glands of a lower termite, H. sjoestedti (Yuki et al. 2008). The sequences obtained also showed consistency with the findings of Tokuda et al. (2002; 2009) and Scharf et al. (2010), in t hat they possess conserved motifs involved 52

PAGE 53

in substrate binding and catalysi s, which are NEPL, with E (Glu, glutamic acid) as the proton donor and TENG with E (Glu, glutam ic acid) as the nucleophile. The N-linked glycosylation, putative catalytic glutamic acid residues (proton donor and nucleophile) and stop codon positions are s hown in Fig. 4-3. The inferred -glucosidase phylogenetic tree (F ig. 4-4) showed discrepancy with the mitochondrial/nuclear tree (Fig. 2-2). Instead of being the most basal group, the fungus-growers formed a strong diphyletic re lationship with the woodand soil/litterfeeders. The result further suggests that bacterial comb-grower -glucosidases were derived from fungus-growers -glucosidases. As predicted, -glucosidase paralogs from the same species clustered with one another, as seen with Na. takasagoensis and S. sphaerothorax The -glucosidase sequence from R. bulbinasus, a grass-/litter-feeder clus tered within the wood-feeders. Both woodand grass-feeders formed a paraphyletic group from the soil/litter-feeders. In the fungus-growers/bacterial comb -grower clade, I suspect that the glucosidases of fungus-growers are least divergent because of the high dependency on fungi to digest their food. High levels of -glucosidase are present in the fungal nodules with which M. bellicosus, Odontotermes pauperans (Silvestri), Ancistrotermes cavithorax (Sjoestedt) and Pseudocanthotermes militaris (Hagen) are associated (Sengupta and Sengupta, 1990). As stated by Darlington (1994) and Rouland-Lefvre (2000), fungus-growers depend on fungal symbi onts to process their food. When their ancestors lost their flagella tes, these termites were st ill able to feed on wood and litter because of their close association with cellulo lytic fungi, which function as an external gut by partially digesting the wood for termi tes. According to Mishra and Sen-Sarma 53

PAGE 54

(1985b), T albuminosus contain glucosidases as well as laccase, chitinase and esterase, which are all essential in lignoc ellulose degradation. Rouland et al. (1988b) later reported that fun gus-derived cellulases, -glucosidase and another termite-derived cellulase work in synergy to digest cellulo se. Earlier on, Abo-Khatwa (1978) showed that -glucosidase activities in the Termitomyces conidiophores and midgut and hindgut of M. subhyalinus are almost equal, thus proving that the fungal nodules were able to replace the role of the missing flagellates. Moreover, in macrotermitines, Martin and Martin (1978; 1979) suggested that Termitomyces -acquired digestive enzymes are required for digestion by M. natalensis. This was contested by Bignell et al. (1994), who suggested t hat the termites endogenous cellulase activity alone is sufficient for resource utilization. According to a study by Hyodo et al. (2000), fungus-feeders consume the mature portion of the comb because the cellulose degr adation in the old comb is thr ee times higher than that of the fresh comb. Rouland et al. (1991) suggested that some fungus species produce fungal cellulases to match a substrate while others did not. Rouland-Lefvre et al. (2006) later divided Termitomyces into a relatively generalist fungal genus (which contained various degradation enzymes depending on the substrate and grown by several termite species) and relatively specialist fungal species (which produced degradation enzymes for specific substrates and only associated with a single termite species). In the case of endogenous -glucosidase evolution, I speculate that because the fungus itself is a major carbon source for these termites (Ars had et al., 1987), the types of fungi with which the termites are associated might have driven the selective pressure to code for the most adaptive -glucosidase to occur. 54

PAGE 55

In sphaerotermitines, I provide molecular evidence to suggest that bacterial combgrower -glucosidases probably evolved from fungus-growers -glucosidases. Two glucosidase sequences, SSBG1 and SSBG2 were obtained from the bacterial combgrower, suggesting two different alleles. Although Lo et al. (2011) stated that the role of different gene copies was unclear, in the case of S. sphaerotermes the occurrence of two distinctly different -glucosidases probably supports the neofunctionalization hypothesis by Ohno (1970), Force et al. (1999) and Hahn (2009), which resulted in a different function from the original gene. I su spect that different paralogs may be used to degrade cellobiose and cello-oligosaccharides of different physical and chemical properties, hence increasing cell ulose digestion efficiency. Because of the presence of different bacterial strains within the S. sphaerothorax nest, (Garnier-Sillam, 1989), I speculate that this may have been the factor that drove the selection for different glucosidases to occur. The nutritional dependency on fungus and bacteria has been lost in woodand soil/litter-feeders. Henceforth, a shift has occured from having an external gut in fungusand bacterial comb-grower to having an internal gut in woodand soil/litterfeeding termites. Rouland et al. (1986; 1989a) reported a lo w endogenous digestive enzyme activity in the midgut and hindgut of soil/litter-feeding te rmites. Even though endogenously-produced cellulases ar e less important in the digestion of soil/litterfeeders, however, the gene is still maintained in the genome (Lo et al., 2011). I suspect that soil/litter-feeders developed an increased dependency on their hindgut microbiota to digest food with low cellulose content because of their highly humified nutritional requirements. Brauman et al. (2000) proposed that some compounds such as 55

PAGE 56

polysaccharides are digested to a certain ext ent by a generalist gut flora after alkaline pretreatment and selected reduced substrates such as polyaromatic compounds are digested by a specialized symbiont population. There were some speculation that cellulolytic amoebae could play a role in cellulos e utilization in some termitids (cited by Brauman, 2000; Slaytor, 2000; Eggleton, 2006). The inferred tree shows that R. bulbinasus -glucosidase, RbBG, falls within the wood-feeders clade. In wood-feeding termi tids, Hogan et al. (1988) showed that glucosidase is secreted in the midgut of Na. walkeri While the expression site has shifted exclusively to the midgut for endo-1,4-glucanase, -glucosidase was still secreted in the salivary glands, as well as the midgut of Na. takasagoensis (Tokuda et al., 2009). This explains the considerabl e differences between salivary and midgut glucosidase paralogs, as observed in the inferred tree. According to Tokuda et al. (1999) and Slaytor (2000), the shift of expre ssion sites from salivary glands in lower termites to the midgut in hi gher termites (which lack hi ndgut flagellates) enhanced cellulose digestion ability. Because I obtained only five different sequences from four termitid species, I could not support the idea of Tokuda et al. (2009) and Lo et al. (2011) that -glucosidase gene copy numbers increased with more derived species. Nevertheless, I speculate that their high divergence was because selection pressure on wood-feeders was more relaxed, thus selecting for multiple types of -glucosidases to occur, resulting in an increased e fficiency in resource utilization. The evolutionary origin of termitid -glucosidases remains to be answered because there has yet to be a complete codi ng sequence of this gene available in the molecular database. While there is a high level of evidence to reject the horizontal gene 56

PAGE 57

hypothesis in GHF9 (Lo et al., 2003; Daviso n and Blaxter, 2005), the evolutionary origin of GHF1, whether it was from a metaz oan ancestor or horizontally-acquired more recently, is still debatable. Tokuda et al (2002) supported vertical gene transfer because the N. koshunensis -glucosidase (NkBG) that they obtained was closely related to those from various insects in the GenBank. However, in contrast, Ketudat Cairns and Esen (2010) have suggested that insect -glycosidases have diverged from plants because some insects have adapted glycosides and glycoside hydrolases from the plants that they consumed. Unfortunately, such speculation could not be confirmed at present with endogenous termite -glucosidases because of the very limited amount of available information. 57

PAGE 58

Table 4-1. List of -glucosidase primers used in this study Name Gene Orientation Sequence (5 to 3) BG1f BG Forward GGCAGAGCAACGAAATG BG1r BG Reverse AAGCGCCAGGGATATG BG2f BG Forward TTCCCGATGGATTTCTG BG2r BG Reverse ACAAACCGCTAGATGAAG BG3f BG Forward GACAGTTTGCTTCGTTATC BG3r BG Reverse ACTCGTAATCAGCAGTATG BG4f BG Forward ATTCCCCGATGGATTTC BG4r BG Reverse AACAAACCGCGTTTTC Figure 4-1. Agarose gel show ing PCR amplification of -glucosidase from four species of higher termites used in this study wit h positive and negative (water) control. 58

PAGE 59

59 A B Figure 4-2. Restriction fragment length profiles of five -glucosidase clones (ca. 1.6 kb) from four species of higher termites used in this study. A) Digestion with Hinf 1 and B) Digestion with Mse 1.

PAGE 60

Figure 4-3. Multiple alignments of -glucosidase amino acid sequences from higher termites. The number of encoded amino acids was listed next to the sequence names. N-linked glycosylation sites are high lighted in red. Blue dot indicates putative proton donor. Red dot indicates putative nucleophile. Black dot indica tes the position of stop codon. 60

PAGE 61

61 Figure 4-3. Continued.

PAGE 62

Figure 4-4. Consensus Bayesian tree inferred from -glucosidase sequences. Numbers above branch nodes indicate posterior probabilities recovered by the Bayesian analysis. Branch lengths are proportional to the number of changes. 62

PAGE 63

CHAPTER 5 CONCLUDING REMARKS A ND FUTURE DIRECTIONS The family Termitidae comprise a majority of all termite species and boasts an ability to exploit a wide range of feeding substrates, ranging from fungi, bacterial comb, plant materials of various stages of humificat ion and soil/litter. In this study, I attempted to explain how endogenous cellulase digestion evolved in higher termites, with respect to the evolution of endo-1,4-glucanases and -glucosidases. Firstly, however, the phylogeny of a select ion of 25 species of higher termites was delineated using mitochondrial (16S) ribosomal RNA and nuc lear (28S) gene markers. Inferences from the sequenc es of fragments of the 16S and 28S genes corroborated with the outcome of studies by Inward et al. (2007a) and Legendre et al. (2008) regarding the monophyletic status of the family Termitidae re lative to the lower termites. The fungus-growing termites were the most phylogenetically basal group, while the wood/lichen/grass/litterand soil/litter-feeding termites were the most derived groups. The bacterial comb-growing termites were phylogenetically placed between fungusgrowing and wood/lichen/ grass/litterand soil/litter-feeding termitids. The macrotermitines and apicotermitines were monophyletic, the termitines were paraphyletic and the nasutitermitines were polyphyletic. Finally, there was a paraphyletic relationship between the soil/li tter feeders and the wood/grass/lichen/litterfeeders. For endo-1,4-glucanase, I hypothesized that the evolution of endo-1,4glucanase will be congruent with the evolution of termites a ccording to the mitochondrial and nuclear markers. Forty-one endo-1,4-glucanase sequences were obtained from 23 species of higher termites. The deduced amino acid sequences showed that they 63

PAGE 64

were similar to endo-1,4-glucanase in GHF9. The inferred tree suggested that the fungus-growing termites were the most phylogenetically basal group, while the woodand soil/litter-feeding termites were the most distal groups. The bacterial comb-grower was phylogenetically placed as the intermediar y that bridged the transition from being a fungus-growing termite to being a woodan d soil/litter-feeding termite. There were strong diphyletic relationships between endo-1,4-glucanases of upper layer soilfeeders and the other so il-feeders, and between the lichenand the grass/litter-feeders within the wood/lichen/grass/litter-feeding termi tes clade. Sequences from the bacterial comb-feeder also showed that they were signi ficantly different from each other, thus suggesting different alleles, which subsequently resulted in different functions from the original gene. As with endo-1,4-glucanase, I hypothesiz ed that the evolution of -glucosidase would also be congruent with t he evolution of termites acco rding to the mitochondrial and nuclear markers. Five -glucosidase sequences were obtai ned from four species of higher termites. The deduced amino acid sequenc es show that they are similar to glucosidases in GHF1. The inferred -glucosidase phylogenetic tree conflicts with the mitochondrial and nuclear data, with the f ungus-growers forming a strong diphyletic relationship with the woodand soil/litter-feeders. Furthermore, it suggests that bacterial comb-grower -glucosidases were probably derived from fungus-growers glucosidases. Two different sequences obtained from the bac terial comb-feeder suggest the involvement of two different allele s, resulting in different gene functionality. While there was a high level of evi dence to support vertical gene transfer hypothesis in GHF9, the evolutionary origin of GHF1 remains unanswered because of 64

PAGE 65

65 the limited amount of av ailable information. Further res earch should focus on providing the complete coding sequences to under stand the evolutionary origin of -glucosidases. Molecular characterization of termite endoge nous cellulases serves as a critical initial step towards using termites as bioresour ces for industrial applications. While this study has focused solely on termite endogenous cellulases, termites are also a rich source of xylanases, amylase, pectinase, lignin peroxidases, manganese peroxidases and laccases, regardless of whether they are termite-derived or of microbiotic origin. These recombinant enzymes or genes can be mass-produced for use in industrial application and biof uel production.

PAGE 66

LIST OF REFERENCES Abo-Khatwa, N., 1978. Cellulase of fungus-g rowing termites: A new hypothesis on its origin. Experentia. 34, 559-560. Arshad, M.A., Schnitzer, M., Preston, C., 1987. Characteristi cs of a termite fungus comb studied by chemical and spectroscopic methods. Soil Biol. Biochem. 19, 227230. Adams, L., Boopathy, R., 2005. Isolation and characterizati on of enteric bacteria from the hindgut of Formosan termite. Bioresource Technol. 96, 1592-1598. Anklin-Muhlemann, R., Bignell, D. E., Veivers, P. C., Leuthold, R. H., Slaytor, M., 1995. Morphological, microbiological and biochemical studies of the gut flora in the fungus-growing termite Macrotermes subhyalinus J. Insect Physiol. 41, 929-940. Bignell, D.E., 1994. Soil-feeding and gut morphology in higher termites, in: Hunt, J.H., Nalepa, C.A. (Eds.), Nourishment and Evol ution in Insect Societies. Westview Press, Boulder, Colo rado, pp. 131-158. Bignell, D.E., 2000. Introduction to symbiosis in: Abe, T., Bignell, D.E., Higashi, M. (Eds.), Termites: evolution, sociality, symbioses, ecology. Kluwer Academic Publisher, Dordrecht, The Netherlands, pp. 189-208. Bignell, D.E., Slaytor, M., Vievers, P.C., Muhlemann, R., Leuthold, R.H., 1994. Functions of symbiotic fungus gardens in higher termites of the genus Macrotermes : evidence against the acquired enzym e hypothesis. Acta Microbiol. Hung. 41, 391-401. Binate, S., Ndri, D., Toka, M., Kouame, L.P., 2008. Purifi cation and characterisation of two beta-glucosidases from termite workers Macrotermes bellicosus (Termitidae: Macrotermitinae). J. Appl. Biosci. 10, 461-470. Brauman, A., Bignell, D.E., Tayasu, I., 2000. Soil-feeding termites: Biology, microbial associations and digestive mechanisms, in : Abe, T., Bignell, D.E., Higashi, M. (Eds.), Termites: evolution, sociality, symbioses, ecology. Kluwer Academic Publisher, Dordrecht, The Netherlands, pp. 233-259. Breznak, J.A., 1984. Biochemical aspects of symbiosis between termites and their intestinal microbiota, in: Anderson, J.M. Rayner, A.D.M., Walt on, D.W.H. (Eds.), Invertebrate-microbial interactions. Ca mbridge Univ. Press, Cambridge, UK. pp. 173-203. Breznak, J.A., 2000. Ecology of prokaryotic microbes in the guts of woodand litterfeeding termites, in: Abe, T., Bignell, D.E., Higashi, M. (Eds.), Termites: evolution, sociality, symbioses, ecology Kluwer Academic Publisher, Dordrecht, The Netherlands, pp. 209-232. 66

PAGE 67

Breznak, J.A., Brune, A., 1994. Role of micr oorganisms in the digestion of lignocellulose by termites. Annu. Rev. Entomol. 39, 453-467. Brugerolle, G., Radek, R., 2006. Symbiotic prot ozoa of termites, in: Konig H., Varma, A. (Eds.), Intestinal Microorganisms of soil invertebrates. Springer-Verlag, Berlin, Germany, pp. 234-269. Brune, A., 2006. Symbiotic associations between termites and prokaryotes. Prokaryotes. 1, 439-474. Brune, A., Ohkuma, M., 2011. Role of the termite gut microbi ota in symbiotic digestion, in: Bignell, D.E., Roisin, Y., Lo, N. (Eds.), Biology of termites: A modern synthesis. Springer, New York, pp. 439-475. Bujang, N.S., Harrison, N.A., Su, N.-Y., 2011. An improved method for extraction and purification of termite endo-1,4-glucanase from FTA cards. Fla. Entomol. 94, 356-358. Carbohydrate-Active enZYmes Database (CAZY), http://www.cazy.org/Home.html Last accessed May 02, 2011. Chararas, C., Noirot, C., 1988. Les osidases du Nasutitermes lujae (Termitidae). Bull. Soc. Zool. Fr. 113, 175-180. Cleveland L.R., 1923. Symbiosis between termite s and their intestinal protozoa. Proc. Natl. Acad. Sci. 9, 424-428. Cleveland L.R., 1924. The physiology and symbiotic relationships between the intestinal protozoa of termites and their host, with special reference to Reticulitermes flavipes Kollar. Biol. Bull. 46, 117-227. Cleveland L.R., 1925. The method by which Trichonympha campanula a protozon in the intestine of termites, ingests solid particles of wood for f ood. Biol. Bull. 48, 282-289. Cleveland L.R., Hall, S.R., Saunders, E.P., Collier, J ., 1934. The wood-feeding roach Cryptocercus its protozoa, and the symbiosis between protozoa and roach. Mem. Am. Acad. Sci. 17, 185-342. Darlington, J.P.E.C., 1994. Nu trition and evolution in fungus -growing termites, in: Hunt, J.H., Nalepa, C.A. (Eds.) Nourishment and Evolution in Insect Societies. Westview Press, Boulder Colorado, pp. 105-130. Davison, A., Blaxter, M., 2005. Ancient origin of glycosyl hydrolase family 9 cellulase genes. Mol. Biol. Evol. 22, 1273-1284. 67

PAGE 68

Deng, T., Chen, C., Cheng, M., Pan, C., Zhou, Y., Mo, J., 2008. Differences in cellulase activity among different castes of Odontotermes formosanus (Isoptera: Termitidae) and the symbiotic fungus Termitomyces albuminosus Sociobiology. 51, 697-704. Donovan, S.E., Jones, D.T., Sands, W.A., Eggleton, P. 2000. The morphological phylogenetics of termites (Isoptera). Biol. J. Linn. Soc. 70, 467-513. Donovan, S.E., Eggleton, P., Bignell, D.E., 2001. Gut content analysis and a new feeding group classification of term ites. Ecol. Entomol. 26, 356-366. Eggleton, P., 2001. Termites and trees: a review of recent advances in termite phylogenetics. Insect Soc. 48, 187-193. Eggleton, P., 2006. The termite gut habitat: its evolution and co-evolution, in: Konig H., Varma, A. (Eds.), Intestinal Microor ganisms of soil invertebrates. SpringerVerlag, Berlin, Germany, pp. 373-404. Eggleton, P., Bignell, D.E., 1995. Monitoring the response of tropical insects to changes in the environment, in: Harrington, R., Stor k, N.E. (Eds.), Insects in a changing environment. Academic Press, London, pp. 473-497. Eggleton, P., Bignell, D.E., Sands, W.A., Wa ite, B., Wood, T.G. Lawton, J.H., 1995. The species richness of termites (Isopt era) under differing levels of forest disturbance in the Mbalmayo Forest Rese rve, southern Cameroon. J. Trop. Ecol. 11, 85-98. Engel, M.S., Krishna, K., 2004. Family-group names for termites (Isoptera). Am. Mus. Novit. 3432, 1-9. Engel, M.S., Grimaldi, D.A ., Krishna, K., 2009. Termites (Isoptera): Their phylogeny, classification, and rise to ecological dominance. Am. Mus. Novit. 3650, 1-27. Force, A., Lynch, M., Picke tt, F.B., Amores, A., Yan, Y.-l., Postlethwait, J., 1999. Preservation of duplicate genes by co mplementary, degenerative mutations. Genetics. 151,1531-1545. Fujita, A., Miura, T., Matsum oto, T., 2008. Differences in cellulose digestive systems among castes in two termite lineages. Physiol. Entomol. 33, 73-82. Garnier-Sillam, E., Toutain, F., Villemin, G., Renoux, J., 1989. tudes prliminaires des meules originales du termite xylophage Sphaerotermes sphaerothorax (Sjostedt). Insect. Soc. 36, 293-312. 68

PAGE 69

Grass, P.P., 1978. Physiologie des insects. Sur la vritable nature et le rle des meules champignons construites par les termites Macrotermitinae (Isoptera: Termitidae). C. R. Acad. Sci. Paris. 287, 1223-1226. Hahn, M.W., 2009. Distinguishing among evolut ionary models for the maintenance of gene duplicates. J. Hered. 100, 605-617. Han, S.J., Yoo, Y.J., Kang, H.S., 1995. Characterization of a bifunctional cellulase and its structural gene J. Biol. Chem. 270, 26012-26019. Heim, R., 1977. Termites et Champignons. Boube, Paris. Hogan, M., Veivers, P.C., Slaytor, M., Cz olij, R.T., 1988. The site of cellulose breakdown in higher termites ( Nasutitermes walkeri and Nasutitermes exitiosus ). J. Insect Physiol. 34, 891-899. Hyodo, F., Inoue, T., Azuma, J. -I., Tayasu, I., Abe, T., 2000. Role of the mutualistic fungus in lignin degradation in the fungus-growing termite Macrotermes gilvus (Isoptera; Mactotermitinae). Soil Biol. Biochem. 32, 653-658. Inoue, T., Kitade, O., Yoshim ura, T., Yamaoka, I., 2000. Symbiotic associations with protists, in: Abe, T., Bignel l, D.E., Higashi, M. (Eds .), Termites: evolution, sociality, symbioses, ecology. Kluwer Academic Publisher, Dordrecht, The Netherlands, pp. 275-288. Inoue, T., Moriya, S., Ohkuma, M., K udo, T., 2005. Molecular cloning and characterization of a cellulose gene from a symbiotic protist of the lower termite, Coptotermes formosanus Gene. 349, 67-75. Inward, D., Vogler, A.P., E ggleton, P., 2007a. A comprehens ive phylogenetic analysis of termites (Isoptera) illuminates key aspects of their evolutionary biology. Mol. Phylogenet. Evol. 44, 953-967. Inward, D., Beccaloni,G., Eggleton, P. 2007b. Death of an order: a comprehensive molecular phylogenetic study confirms that termites are eusocial cockroaches. Biol. Lett. 3, 331-335. Ji, R., Brune, A., 2001. Transformation and mineralization of 14C -labeled cellulose, peptidoglycan, and protein by the soil-feeding termite Cubitermes orthognathus Biol. Fertil. Soils. 33, 166-174. Ketudat Cairns, J.R. Esen, A., 2010. -Glucosidases. Cell. Mol. Life Sci. 67, 33893405. 69

PAGE 70

Kambhampati, S., 1995. A phylogeny of cockr oaches and related insects based on DNA sequence of mitochondrial ribosomal RNA genes Proc. Natl. Acad. Sci. USA. 92, 2017-2020. Kambhampati, S., 1996. Phylogenetic relati onship among cockroach families inferred from mitochondrial 12S rRNA gene sequence. Syst. Entomol. 21, 89-98. Kambhampati, S., Eggleton, P., 2000. Taxonomy and phylogeny of termites, in: Abe, T., Bignell, D.E., Higashi, M. (Eds.), Termites: evoluti on, sociality, symbioses, ecology. Kluwer Academic Publisher, Dordrecht, The Netherlands, pp. 1-24. Khademi, S., Guarino, L.A., Wa tanabe, H., Tokuda, G., Meyer, E.F., 2002. Structure of an endoglucanase from termite, Nasutitermes takasagoensis Acta Cryst. D58, 653-659. Klass, K.-D., 1995. Die Phylogeny der Dicty optera. PhD thesis, Ludwig Maximilians Universitt, Munich, Germany. Koch, A., 1938. Symbiosestudien, 3: Die intrazellulare symbiose von Mastotermes darwiniensis Froggatt. Z.Morph. Ok ol. Tiere. 34, 384-609. Konig, H., Frohlich, J., Hertel, H., 2006. Div ersity and lignocellulolytic activities of cultured microorganisms, in: Konig H ., Varma, A. (Eds .), Intestinal Microorganisms of soil invertebrates. Sp ringer-Verlag, Berlin, Germany, pp. 271301. Kouame, L.P., Kouame, F.A., Niamke, S.L., Faulet, B.M., Kamenan, A., 2005. Biochemical and catalytic properties of two -glycosidases purified from workers of the termite Macrotermes subhyalinus (Isoptera: Termitidae). J. Trop. Insect Sci. 25, 103-113. Kovoor, J., 1970. Presence denzymes cellu lolytiques dans lintestin dun termite suprrieur, Microcerotermes edentatus (Was.). Ann. Sci. Nat. Zool. Biol. Anim. 12, 65-71. Legendre, F., Whiting, M.F., Bordereau, C., 2008. The phylogeny of termites (Dictyoptera: Isoptera) based on mitochondr ial and nuclear markers: implications for the evolution of the worker and pseudergate castes, and foraging behaviors. Mol. Phylogenet. Evol. 48, 615-627. Lenoir-Labe, F., Rouland, C., 1993. Puri fication et proprietes de deux isodases produites par des bacteries isolees a partit du tractus digestif de Cephalotermes rectangularis (Isoptera, Termitidae). Acte s Coll. UIEIS. 8, 71-78. 70

PAGE 71

Lo, N., Eggleton, P., 2011. Termite phylogenetics and co-cladogenesis with symbionts, in: Bignell, D.E., Roisin, Y., Lo, N. (Eds.), Biology of termites: A modern synthesis. Springer, New York, pp. 27-50. Lo, N., Tokuda, G., Watanabe, H., Rose, H., Sl aytor, M., Maekawa, K., Bandi, C., Noda, H., 2000. Evidence from multiple gene sequences indicates that termites evolved from wood-feeding cockroaches Curr. Biol. 10, 801-804. Lo, N., Watanabe, H., Sugimu ra, M., 2003. Evidence for the presence of a cellulase gene in the last common ancestor of bila terian animals. Proc. R. Soc. Lond. B 270, S69-S72. Lo, N., Tokuda, G., Watanabe, H., 2011. Evol ution and function of endogenous termite celulases, in: Bignell, D.E., Roisin, Y., Lo, N. (Eds.), Biology of termites: A modern synthesis. Springer, New York, pp. 51-67. Marana, S.R., Jacobs-Lorena, M., Terra, W.R ., Ferreira, C., 2001. Amino acid residues involved in substrate binding and ca talysis in an insect digestive -glycosidase. Biochim. Biophys. Acta. 1545, 41-52. Marini, M., Mantovani, B., 2002. Molecular relationships among European samples of Reticulitermes. Mol. Phylogenet. Evol. 22, 454-459. Martin, M.M., Martin, J.S., 1978. Cellulose digestion in the midgut of the fungus-growing termite Macrotermes natalensis : the role of acquired di gestive enzymes. Science. 199, 1453-1455. Martin, M.M., Martin, J.S., 1979. The distribution and origins of the cellulolytic enzymes of the higher termite, Macrotermes natalensis Physiol. Zool. 52, 11-21. Matsui, T., Tokuda, G., Shin zato, N., 2009. Termites as f unctional gene resources. Recent Pat. Biotechnol. 3, 10-19. Matsuura, K., Yashiro, T., 2010. Parallel evolution of termi te-egg mimicry by sclerotiumforming fungi in distant termite groups Biol. J. Linn. Soc. 100, 531-537. Matsuura, K., Yashiro, T., Shimizu, K., Tatsumi, S., Tamura, T., 2009. Cuckoo fungus mimics termite eggs by producing the cellulose-digesting enzyme -glucosidase. Curr. Biol. 19, 30-36. McKittrick, F.A., 1965. A contribution to the understanding of cockroach termite affinities. Ann. Entomol. Soc. Am. 58, 18-22. Mishra, S.C., Sen-Sarma, P.K., 1985a. Orig in and distribution of carbohydrases (cellulase and non-cellulases) in higher termites Odontotermes distans (Holmg.) and Speculitermes cyclops (Wasm.). Indian J. Entomol. 47, 83-91 71

PAGE 72

Mishra, S.C., Sen-Sarma, P. K., 1985b. Nutritional significance of fungus comb and Termitomyces albuminosus (Berk.) Heim in Odontotermes obesus Rambur (Isoptera: Termitidae). Mate r. Organismen. 20, 205-213. Miura, T., Maekawa, K., Ki tade, O., Abe, T., Matsum oto, T., 1994. Phylogenetic relationships among subfamilies in higher termites (Isoptera: Termitidae) based on Mitochondrial COII gene sequences. Ann. Entomol. Soc. Am. 91, 515-523. Mo, J.-C, Yang,T.-C., Song, X.-G ., Cheng, J.-A., 2004. Cellulase activity in five species of important termites in China. Appl. Entomol. Zool. 39, 635-641. Moore, W.S., 1995. Inferring phylogenies fr om mtDNA variation: mitochondrial-gene trees versus nuclear-gene tr ees. Evolution. 49, 718-726. Nakashima, K., Azuma,J.-I., 2000. Distribution and properties of endo-1,4-glucanase from a lower termite, Coptotermes formosanus (Shiraki). Biosci. Biotechnol. Biochem. 64, 1500-1506. Nakashima, K., Watanabe, H., Saitoh, H ., Tokuda, G., Azuma, J.-I., 2002a. Dual cellulose-digesting system of the wood-feeding termite, Coptotermes formosanus Shiraki. Insect Biochem. Molec. Biol. 32, 777-784. Nakashima, K., Watanabe, H., Azuma, J.-I., 2002b. Cellulase genes from the parabasalian symbiont Pseudotrichonympha grassii in the hindgut of the woodfeeding termite Coptotermes formosanus Cell. Mol. Life Sci. 59, 1554-1560. Ni, J, Tokuda, G., Takehara, M., Watanabe, H., 2007. Heterologous expression and enzymatic characterization of b-glucos idase from the dr ywood-eating termite, Neotermes koshunensis Appl. Entomol. Zool. 42, 457-463. Noirot, C., 1992. From woodto humus-feeding: an important trend in termite evolution, in: Billen, J. (Ed.), Biology and evoluti on of social insects. Leuven University Press, Leuven. pp 107-119. Ohkuma, M., 2003. Termite symbiotic systems: efficient bio-recycling of lignocellulase. Appl. Microbiol. Biotechnol. 61, 1-9. Ohno, S., 1970. Evolution by gene duplicat ion. Springer-Verlag, Berlin, Germany. Pamilo, P., Nei, M., 1988. Relationship s between gene trees and species trees. Mol. Biol. Evol. 5, 568-583. Posada, D., Crandall, K.A., 1998. Modeltest: testing the model of DNA substitution. Bioinformatics. 14, 817-818. 72

PAGE 73

Potts, R.C., Hewitt, P.H., 1972. Some properties of an aryl-glucosidase from the harvester termite, Trinervitermes trinervoides Insect Biochem. 2, 400-408. Potts, R.C., Hewitt, P.H., 1973. The distribut ion of intestinal bacteria and cellulase activity in the harvester termite, Trinervitermes trinervoides Insect. Soc. 20, 215220. Robson, L.M., Chambliss, G.H., 1989. Celu lases of bacterial origin. Enzyme Microb. Tech. 11, 626-644. Ronquist, F., Huelsenbeck, J.P., 2003. MrB ayes 3: Bayesian phylogenetic inference under mixed models. Bioi nformatics. 19, 1572-1574. Rouland, C., Chararas, C., Renoux, J., 1986. Etude compares des osidases de trois espces de termites Africain rgime alim entaire diffrent. C.R. Acad. Sci. Paris. 9, 341-345. Rouland, C., Civas, C., Renoux, J., Petek, F., 1988a. Purification and properties of cellulases from the termite Macrotermes muelleri (Termitidae, Macrotermitinae) and its symbiotic fungus Termitomyces sp.. Comp. Biochem. Physiol. 91, 449458. Rouland, C., Civas, C., Renoux, J., Petek, F., 1988b. Synergistic activities of the enzymes involed in cellulos e degradation, purified from Macrotermes muelleri and from its symbiotic fungus Termitomyces sp.. Comp. Biochem. Physiol. 91, 459-465. Rouland, C., Chararas, C., Renoux, J., 1989a. Les osidases digestives prsentes dans l'intestin moyen, l'intestin postrieur et les glandes salivaires du termite humivore Crenetermes albotarsalis C.R. Acad. Sci. Paris. 10, 281-285. Rouland, C., Lenoir-Rousseaux, J., Mora P., Renoux, J., 1989b. Origin of glucosidase purified from the digestive tract of the f ungus-growing termite Macrotermes muelleri Sociobiology. 15, 237-245. Rouland, C., Brauman, A., Keleke, S., Labat, M., Mora P., Renoux, J., 1990. Endosymbiosis and exosymbiosis in the fungus-growing termites, in: Lesel, R. (Ed.), Microbiology in Poecilotherms Elsevier Science, Amsterdam, The Netherlands. pp. 79-82. Rouland, C., Lenoir F., Lepage, M., 1991. The ro le of the symbiotic fungus in the digestive metabolism of several specie s of fungus growing termites. Comp. Biochem. Physiol. 99, 657-663. 73

PAGE 74

Rouland-Lefvre, C., 2000. Symbio sis with fungi, in: Abe, T., Bignell, D.E., Higashi, M. (Eds.), Termites: evolution, sociality, symbioses, ecology. Kluwer Academic Publisher, Dordrecht, The Netherlands, pp. 289-306. Rouland-Lefevre, C., Bign ell, D.E., 2001. Cultivation of sym biotic fungi by termites of the subfamily Macrotermitinae, in: Seckbach, J. (Ed), Symbiosis: Mechanisms and Model Systems. Kluwer Academic Publishe r, Dordrecht, The Netherlands, pp. 731-756. Rouland-Lefvre, C., Inoue, T., Johjim a, T., 2006. Termitomyces /termite interactions, in: Knig, H., Varma, A. (Eds.) Intestinal microorganism s of termites and other invertebrates. Springer, Berlin, pp. 335-350. Sakka, K., Kimura, T., Karita S., Ohmiya, K., 2000. Molecu lar breeding of cellulytic microbes, plants and animals for biomass utilization. J. Biosci. Bioeng. 90, 227233. Sakon, J., Irwin, D., Wilson, D.B., Karplu s, P.A., 1997. Structure and mechanism of endo/exocellulase E4 from Thermomonospora fusca Nat. Struct. Biol. 4, 810818. Sands, W.A., 1956. Some factors affecting the survival of Odontotermes badius. Insect. Soc. 3, 531-536. Sands, W.A., 1969. The association of te rmites and fungi, in: Krishna, K., Weesner, F.M. (Eds.), Biology of termites, Volu me I. Academic Press, New York, pp. 495524. Scharf, M.E., Kovaleva, E.S., Jadhao, S., Ca mpbell, J.H., Buchman, G.W., Boucias, D.G., 2010. Functional and translational analyses of a beta-glucosidase gene (glycosyl hydrolase family 1) isolat ed from the gut of the lower termite Reticulitermes flavipes Insect Biochem. Mole c. Biol. 40, 611-620. Sengupta, S., Sengupta, S., 1990. -glucosidase production by the mycelial culture of the mushroom Termitomyces clypeatus Enzyme Microb. Tech. 12, 309-314. Simon, C., Frati, F., Beckenbach, A., Crespi B., Liu, H., Flook, P., 1994. Evolution, weighting, and phylogenetic utility of mitochondrial gene sequences and a compilation of conserved polymerase chain reaction primers. Ann. Entomol. Soc. Am. 87, 651-701. Slaytor, M., 1992. Cellulase digestion in te rmites and cockroaches; do symbionts play a role? Comp. Biochem. Physiol. 103, 775-784. 74

PAGE 75

Slaytor, M., 2000. Energy metabolism in the termite and its gut micr obiota, in: Abe, T., Bignell, D.E., Higashi, M. (Eds.), Termites: evoluti on, sociality, symbioses, ecology. Kluwer Academic Publisher, Dordrecht, The Netherlands, pp. 307-332. Slaytor, M., Veivers, P.C., Lo, N., 1997. Aerobic and anaerobic metabolism in the higher termite Nasutitermes walkeri (Hill). Insect Biochem. Mol. Biol. 27, 291-303. Swofford, D.L., 2002. PAUP*. Phylogenetic analysis using parsimony (*and other methods). Sinauer Associates, Sunderland, MA. Sukhumavasi, J., Ohmiya, K., Swwana-adt h, M., Shimizu, S., 1989. Conversion of agricultural wastes to ethanol and acetic acid by cellulolytic anaerobic bacteria. J. Sci. Soc. Thailand. 15, 109-120. Tamura, K., Dudley, J., Nei, M., Kumar, S., 2007. MEGA 4: Molecular Evolutionary Genetics Analysis (MEGA) software versi on 4.0. Mol. Biol. Evol. 24, 1596-1599. Thompson, N.S., 1983. Hemicellulose as a biom ass resource, in: So ltes, E.J. (Ed.), Wood and agricultural residues. Research on use for feed, fuels, and chemicals. Academic Press, New York, pp. 101-119. Thorne, B.L., Carpenter, J.M., 1992. Phylog eny of Dictyoptera. Sys. Entomol. 17, 253268. Thorne, B.L., Grimaldi, D.A., Krishna, K., 2000. Early fossil history of the termites, in: Abe, T., Bignell, D.E., Higashi, M. (Eds .), Termites: evolution, sociality, symbioses, ecology. Kluwer Academic Publisher, Dordrecht, The Netherlands, pp. 77-93. Tokuda, G., Watanabe, H., 2007. Hidden cellulases in termites: revision of an old hypothesis. Biol. Lett. 3, 336-339. Tokuda, G., Watanabe, H., Matsumoto, T., N oda, H., 1997. Cellulose digestion in the wood-eating higher termite, Nasutitermes takasagoensis (Shiraki): distribution of cellulases and properties of endo-1,4-glucanase. Zool. Sci. 14, 83-93. Tokuda, G., Lo, N., Watanabe, H., Slaytor, M., Matsumoto, T., Noda, H., 1999. Metazoan cellulase genes from termites: intron/exon structures and sites of expression. Biochim. Biophys. Acta. 1447, 146-159. Tokuda, G., Saito, H., Wa tanabe, H., 2002. A digestive -glucosidase from the salivary glands of the termite, Neotermes koshunensis (Shiraki): distribution, characterization and isolation of its precursor cDNA by 5 and 3-RACE amplifications with degenerate primers. Insect Biochem. Molec. 32, 1681-1689. 75

PAGE 76

Tokuda, G., Lo, N., Watanabe, H., Arakawa, G., Matsumoto, T., N oda, H., 2004. Major alteration of the expression site of endogenous cellulases in members of an apical termite lineage. Molec. Ecol. 13, 3219-3228. Tokuda, G., Lo, N., Watanabe, H., 2005. Mark ed variations in patterns of cellulase activity against crystallineversus car boxymethyl-cellulose in the digestive systems of diverse, wood-feeding termites. Physiol. Entomol. 30, 372-380. Tokuda, G., Miyagi, M., Ma kiya, H., Watanabe, H., Arakawa, G., 2009. Digestive bglucosidase from the wood-feeding higher termite, Nasutitermes takasagoensis: Intestinal distribution, molecular charac terization, and alteration in sites of expression. Insect Biochem. Molec. 39, 931-937. Tomme P.R., Warren, A.J., Gilkes, N.R., 1995. Cellulose hydrolysis by bacteria and fungi. Adv. Microb. Physiol. 37, 1-81. Veivers, P.C., M hlemann R., Slaytor, M., Leuthol d, R.H., Bignell, D.E., 1991. Digestion, diet and polyethism in two fungus-growing termites: Macrotermes subhyalinus Rambur and M. michaelseni Sjstedt. J. Insect. Physiol. 37, 675682. Walker, E.M., 1922. The terminal structures of orthopteroid inse cts: a phylogenetic study II The terminal abdominal structures of the male. Ann. Entomol. Soc. Am. 15, 1-87. Warnecke, F., Luginbuhl, P., Iv anova, N., Ghassemian, M., Richardson,T.H., Stege, J. T., Cayouette, M., McHardy, A.C., Djor djevic, G., Aboushadi, N., Sorek, R., Tringe, S.G., Podar, M., Martin H.G., Kunin, V., Dalevi, D., Madejska, J., Kirton, E., Platt, D., Szeto, E., Salamov, A., Ba rry, K., Mikhailova, N. Kyrpides, N.C., Matson, E. G., Ottesen, E.A., Zhang, X., Hernandez, M., Murillo, C., Acosta, L.G., Rigoutsos, I., Tamayo G., Green, B.D., Chang, C ., Rubin, E.M., Mathur, E.J., Robertson, D.E., Hugenholtz, P. Leadbetter, J.R., 2007. Metagenomic and functional analysis of hindgut microbi ota of a wood-feeding higher termite. Nature. 450, 560-565. Watanabe, H., Tokuda, G., 2001. Animal cellulases. Cell. Mo l. Life Sci. 58, 1167-1178. Watanabe, H., Noda, H., Tokuda, G., Lo, N., 1998. A cellulase gene of termite origin. Nature. 394, 330-331. Watanabe, H., Nakashima, K., Saito, H. Slaytor, M., 2002. New endo-beta-1,4glucanases from the parabasalian symbionts, Pseudotrichonympha grassii and Holomastigotoides mirabile of Coptotermes termites. Cell. Mol. Life Sci. 59, 11223-11292. 76

PAGE 77

77 Willis, J.D., Oppert, C., Jurat-Fuentes, J.L., 2010. Methods for discovery and characterization of cellulolytic enzymes from insects. Inse ct Sci. 17, 184-198. Wu, C.I., 1991. Inferences of species phylog eny in relation to segregation of ancient polymorphisms. Genet ics. 127, 429-435. Yang, T., Mo, J.-C, Cheng, J. -A., 2004. Purification and some properties of cellulase from Odontotermes formosanus (Isoptera:Termitidae). Entomol. Sinica. 11, 1-10. Ye, W., Lee, C.-Y., Scheffrahn, R.H., Aleong, J.M., Su, N.-Y., B ennett, G.W., Scharf, M.E., 2004. Phylogenetic re lationships of nearctic Reticulitermes species (Isoptera: Rhinotermitidae) wit h particular reference to Reticulitermes arenincola Goellner. Mol. Phylogenet. Evol. 30, 815-822. Yuki, M., Moriya, S., Inoue, T., Kudo, T., 2008. Transcripto me analysis of the digestive organs of Hodotermopsis sjostedti a lower termite that hosts mutualistic microorganisms in its hindgu t. Zool. Sci. 25, 401-406. Zhang, M., Gaschen, B., Blay, W., Foley, B ., Haigwood, N., Kuiken, C., Korber, B., 2004. Tracking global patterns of N-linked glycosylation site variation in highly variable viral glycoproteins: HIV, SIV, and HCV envelopes and influenza hemagglutinin. Glycobiology. 14, 1229-1246. Zhang, D., Lax, A.R., Raina, A. K., Bland, J.M., 2009. Differentia l cellulolytic activity of native-form and C-terminal tagg ed-form cellulase derived from Coptotermes formosanus and expressed in E. coli Insect Biochem. Molec. 39, 516-522. Zhang, D., Lax, A.R., Bland, J.M., Yu, J., Fedorova, N., Nierman, W.C., 2010. Hydrolysis of filter-paper cellulose to glucose by two recombinant endogenous glycosyl hydrolases of Coptotermes formosanus. Insect. Sci. 17, 245-252. Zhang, D., Lax, A.R., Bland, J.M., Allen, A.B., 2011. Ch aracterization of a new endogenous endo-b-1,4-gluc anase of Formosan subterranean termite ( Coptotermes formosanus). Insect Biochem. Molec. 41, 211-218. Zhou, X., Smith, J.A., Oi, F.M., Koehler, P.G., Bennett, G.W., Scharf, M.E., 2007. Correlation of cellulase gene expression and cellulolytic activity throughout the gut of the termite Reticulitermes flavipes Gene. 395, 29-39.

PAGE 78

BIOGRAPHICAL SKETCH Nurmastini Sufina binti Bujang was born in Kuching, Sarawak in 1977. Growing up as a child in Borneo, she had often embar ked on her own exploration of nature, taking nature walks, climbing trees, catc hing or simply observing small mammals, reptiles, amphibians, fish and insects. To her, nature is indeed, the best teacher. After obtaining a Bachelor of Applied Scienc e in parasitology from Universiti Sains Malaysia, Pulau Pinang in 2000, she develo ped a much deeper interest in insects. Under the guidance of Professor Lee Chow Y ang, she studied the bi ological parameters and control of the smooth cockroach, Symploce pallens (Stephens). Nurmastini graduated with a Master of Science in entomol ogy from Universiti Sains Malaysia in 2005. In the Fall of 2006, she enrolled in a Doctor of Philosoph y program at the University of Florida under Professor Nan-Y ao Su. After working closely with Dr Nigel Harrison, she has now developed a deep interest in molecular bi ology. Her ultimate dream one day is to return home to set up her own independent molecular laboratory to study termites from the deep rainforest of Borneo. Nurmastini graduated with a Doctor of Philosophy in entomology from the Univ ersity of Florida in the Summer of 2011. In youth we learn; in age we understand. Marie von Ebner-Eschenbach 78