Citation
Functional Analysis of Genes Involved in cAMP-Mediated Signaling in the Wide Host Range Necrotroph Sclerotinia sclerotiorum (Lib.) de Bary

Material Information

Title:
Functional Analysis of Genes Involved in cAMP-Mediated Signaling in the Wide Host Range Necrotroph Sclerotinia sclerotiorum (Lib.) de Bary
Creator:
JURICK II, WAYNE MICHAEL ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Ascomycota ( jstor )
Diseases ( jstor )
DNA ( jstor )
Enzymes ( jstor )
Fungi ( jstor )
Genes ( jstor )
Genomics ( jstor )
Infections ( jstor )
Oxalates ( jstor )
Sclerotia ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Wayne Michael Jurick, II. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
5/31/2009
Resource Identifier:
658216357 ( OCLC )

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

FUNCTIONAL ANALYSIS OF GENES INVOLVED IN CAMP-MEDIATED SIGNALING IN THE WIDE HOST RANGE NECROTROPH Sclerotinia sclerotiorum (Lib.) de Bary By WAYNE MICHAEL JURICK, II A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2006

PAGE 2

Copyright 2006 By Wayne Michael Jurick, II

PAGE 3

Dedicated to my wife, Dr. Angela C. Vin cent Jurick, for her unending love, support and generous amount of help in all areas of my life.

PAGE 4

iv ACKNOWLEDGMENTS I sincerely thank my supervisory committee chair (Dr. Jeffrey A. Rollins) for his unending patience, encouragement, guidance and friendship. I al so would like to acknowledge the members of my supervisory committee, Drs. Kuang-Ren Chung, Charles L. Guy and Wen-Yuan Song, for their helpful suggestions a nd critical evaluation of this dissertation. I would like to express my gratitude to the members of the plant pathology department for their gracious suppor t during my tenure as an undergraduate and graduate student. I especially would like to acknowledge my undergraduate mentor, Dr. F. William Zettler, and my masters mentor Dr. Prem S. Chourey for their guidance and nurturing during the early st ages of my development as a scientist. I also want to thank the members of the Rollins lab, past and present, for generously sharing their knowledge and skills. Special thanks to Ulla Benny for her unending desire to help in all areas of scientific endeavor. I thank my parents, Wayne a nd Ronda C. Jurick for their encouragement, love and financial support th at made my undergraduate and graduate experience possible. Most importantly, I th ank God for the knowledge, wisdom and gifts that I have been given which have enable d me to achieve one of my life’s dreams.

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv TABLE OF CONTENTS.....................................................................................................v LIST OF FIGURES..........................................................................................................vii ABSTRACT.......................................................................................................................ix CHAPTER 1 LITERATURE REVIEW.................................................................................................1 Sclerotinia sclerotiorum (Lib.) de Bary........................................................................1 Biology..................................................................................................................1 Epidemiology, Host Range and Control of S. sclerotiorum ..................................3 Physiological Mechanisms of Pathogenicity.........................................................6 Genome Sequence.................................................................................................9 cAMP-Mediated Signal Transduction........................................................................10 Discovery of cAMP.............................................................................................10 cAMP Signal Transduction Pathway...................................................................11 Other Targets of cAMP.......................................................................................15 Biochemical Functions of cAMP in Fungi..........................................................16 2 CHARACTERIZATION AND FUNCTIONAL ANALYSIS OF A CAMPDEPENDENT PROTEIN KINASE A CATALYTIC SUBUNIT GENE ( pka1 ) IN Sclerotinia sclerotiorum ..................................................................................................22 Introduction.................................................................................................................22 Materials and Methods...............................................................................................24 Fungal Isolates, Growth Conditions, Media and Cultural Manipulations...........24 Methods for Nucleic Acid Is olations and Manipulations....................................25 Molecular Cloning and Seque nce Identification of pka1 ....................................26 Construction of pka1 Gene Replacement and Complementation Vectors..........27 Fungal Protoplast Transformation and Evaluation of Transformants.................28 Protein Extraction and Analysis of Pka Enzyme Activity...................................28 Multiple Sequence Alignment, Phe nogram Construction and Bootstrap Analysis............................................................................................................29 Results........................................................................................................................ .30

PAGE 6

vi Molecular Cloning and Characterization of pka1 ...............................................30 Disruption of the pka1 Locus..............................................................................30 RNA Blot Analysis..............................................................................................31 Growth on 10 mM cAMP Amended PDA..........................................................32 Analysis of PKA Enzyme Activity......................................................................32 Phylogenetic Analyses of pka Catalytic Subunit Genes from Various Fungi.....33 Discussion...................................................................................................................42 Investigating the Biological Role of pka1 in S. sclerotiorum .............................42 Phylogenetic Evidence for Two pka Catalytic Subunit Genes in S. sclerotiorum .....................................................................................................44 3 ADENYLATE CYCLASE DELETION MUTANTS IN Sclerotinia sclerotiorum DISPLAYED ALTERED MYCELIAL BR ANCHING, PRODUCED ABERRANT SCLEROTIA AND WERE NON-PATHOGENIC........................................................46 Introduction.................................................................................................................46 Materials and Methods...............................................................................................49 Fungal Strains, Growth Conditions and Media...................................................49 Basic Procedures for Nucl eic Acid Manipulation...............................................50 Cloning and Identification of the sac1 Adenylate Cyclase Gene........................51 Construction of sac1 Gene Replacement and Complementation vectors...........51 Transformation of Fungal Protoplasts a nd Evaluation of Transformant Strains.54 Radial Growth Analysis, Determination of Oxalate, and Pathogenicity Assays55 Extraction and Determination of Total cAMP Levels.........................................55 Results........................................................................................................................ .56 Cloning and Characterization of sac1 .................................................................56 Deletion of the sac1 Locus..................................................................................56 Evaluation of Total Cellular cAMP Levels.........................................................57 Adenylate Cyclase Deletion Mutants E xhibit Many Morphological Defects.....58 Discussion...................................................................................................................70 Molecular Cloning and Characterization of sac1 in S. sclerotiorum ..................70 Multiple Morphological Aberrations Exist in sac 1 Deletion Mutants................70 cAMP is Detectable in sac1 KO mutants............................................................72 The Role of Adenylate Cyclase and cA MP in Pathogencity and Virulence.......72 APPENDIX IDENTIFICATION OF GE NES INVOLVED IN cAMP SIGNALING FROM THE DRAFT GENOME SEQUENCE OF Sclerotinia sclerotiorum (Lib. de Bary)..........................................................................75 LIST OF REFERENCES...................................................................................................79 BIOGRAPHICAL SKETCH.............................................................................................89

PAGE 7

vii LIST OF FIGURES Figure page 2-1 Multiple sequence alignment of am ino acid residues encompassing the Pka catalytic core from seven differe nt organisms using ClustalX................................35 2-2 Diagram of the pka1 locus illustrating the double re combination event necessary for replacement of wild type with the disrupted pka1 locus....................................36 2-3 Genomic DNA blot analysis of Wt, disrupt ant, ectopic and complemented strains...37 2-4 RNA blot analysis of Wt, disruptant, ectopic, and complemented strains..................38 2-5 Growth of wild type, disruptant, ectopi c, and complemented strains on PDA and PDA supplemented with 10 mM cAMP..................................................................39 2-6 Pka enzyme activity assay us ing crude protein extracts from Wt, disruptant, ectopic and complemented strains . ...........................................................................40 2-7 A phylogenetic tree represen ting the amino acid residues contained in the highly conserved serine/threonine protein kinase catalytic core from seven different organisms.................................................................................................................41 3-1 Genomic organization of the sac1 locus illustrating the location and relative size of exons and introns and putat ive stop and start codons..........................................60 3-2 Construction and analysis of sac1 gene replacement strains.......................................61 3-3 PCR and Northern Blot analysis of sac1 deletion and control strains.........................62 3-4 Analysis of total cellular cAMP levels in wild type (Wt) and sac1 deletion (KO1)...63 3-5 Cultural morphology of wild type (Wt), sac1 deletion (KO1), ectopic (E1), complemented (C1), and non-complemented strains (Nc1) grown on potato dextrose agar for 14 days.........................................................................................64 3-6 Growth of wild type (Wt), sac1 deletion (KO1), ectopic (E1), complemented (C1), and non-complemented (Nc1) strains on 10mM cAMP-amended potato dextrose agar for 7 days..........................................................................................................65 3-7 Radial growth analysis of wild type (Wt), sac1 deletion, (KO1), ectopic (E1), complemented (C1), and non-complemented (Nc1) strains.....................................66

PAGE 8

viii 3-8 Pathogenicity assay with wild type and transformant strains on detached tomato leaflets......................................................................................................................67 3-9 Pathogenicity assay of wild type a nd transformant strains on wounded detached tomato leaflets..........................................................................................................68 3-10 Oxalic acid accumulation kinetics fr om 0.5M MOPS-buffered YPSU cultures pH 7.0 of wild type (Wt), sac1 deletion (KO1), ectopic (E1), complemented (C1), and non-complemented (Nc1) strains......................................................................69

PAGE 9

ix Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy FUNCTIONAL ANALYSIS OF GENES INVOLVED IN CAMP-MEDIATED SIGNALING IN THE WIDE HOST RANGE NECROTROPH Sclerotinia sclerotiorum (Lib.) de Bary By WAYNE MICHAEL JURICK, II May 2006 Chair: Jeffrey A. Rollins Major Department: Plant Pathology The filamentous ascomycete phytopathogen, Sclerotinia sclerotiorum (Lib.) de Bary, is a broad host range necrotroph with global distribu tion. Dissemination and survival of this fungus is attr ibuted to resting stru ctures termed sclerotia that can remain viable in the soil for years. A sclerotium can give rise to multiple mushroom-shaped apothecia that are the source of forcibly-disch arged ascospores that serve as the primary source of inoculum for Sclerotinia diseases. Genetic regulators of sc lerotial development and pathogenesis were determined by func tionally characterizing, a cAMP-dependent protein kinase A catalytic subunit gene ( pka1 ) and an adenylate cyclase gene ( sac1 ) , that are involved in cAMP-mediate d signal transduction. Contrary to our original hypothesis, pka1 loss-of-function mutants were cAMP-responsi ve, produced wild type like sclerotia, and were pathogenic. This finding prompt ed the making of a new hypothesis that a second PKA catalytic subunit gene, pka 2, contributes the majority of PKA activity in the cell. Mining recently available data from the S. sclerotiorum genome project uncovered

PAGE 10

x the second hypothesized pka catalytic subunit gene, pka 2. In a second approach to determine the biological role of cAMP in S. sclerotiorum, an adenylate cyclase gene, sac1 , was mutated by targeted deletion. Adenyl ate cyclase knock-out (AC-KO) mutants revealed several morphological defects includi ng a reduced rate of mycelial expansion and the production of aberrant sclerotia in concentric rings in culture. Sclerotia were competent for myceliogenic germination, but di d not carpogenically germinate and form apothecia under standard conditions. AC-KO mutants pr oduced 90-branched hyphae in contrast to acutely branched hyphae observed in wild type. Cyclic AMP levels were greatly reduced in the AC-KO and it was nonpathogenic on detached tomato leaflets. The AC loss-of-function mutant was capab le of colonizing mechanically wounded leaves, but the rate of lesion expansion wa s much slower than wild type. Loss of pathogenicity may be attributed to the lack of infection cushion formation in the AC-KO strain as determined by in vitro morphogenesis assays. Slower growth rate observed in culture and in planta appears to account for the reduced virulence in wounded plants . Taken together, these results demonstrat e that cAMP plays important roles in pathogenicity, growth and vegetative development in S. sclerotiorum.

PAGE 11

1 CHAPTER 1 LITERATURE REVIEW Sclerotinia sclerotiorum (Lib.) de Bary Biology The taxonomic placement of S. sclerotiorum has been revised several times since its first description. I will not give a full acc ount of this history how ever; it is noteworthy to mention that the first Latin binomial given to S. sclerotiorum was Peziza sclerotiorum by Libert (1837). In 1884, Anton de Bary changed the name of P. sclerotiorum to its present designation. S. sclerotiorum resides in the Sclerotiniaceae family which includes stromatal-forming inoperculate discomycetes that produce stipitate apothecia with asci containing ellipsoid ascospores and globose spermatia (Whetzel 1945). The S clerotinia genus contains two other type species: S. trifoliorum Eriks, and S. minor Jagger. The Sclerotiniaceae family is based on the production of tuberoid sclerotia that do not incorporate host tissue within the sclerotia l medulla, develop an apothecial ectal excipulum composed of globose cells, and lack a disseminative conidial state (Kohn 1979a, 1979b). S. sclerotiorum possess simple, septate hyphae containing membrane-bound vesicles, mitochondria, lipid bodies, r ough and smooth endoplasmic reticulum and ribosomes (Maxwell et al . 1972). Vegetative hyphae are haploid and multinucleate, sometimes having up to 100 nuclei per cell (Willets and Wong 1979). Spermatia or microconidia are uninucleate and produced in chains from phialides on branched

PAGE 12

2 microconidiophores (Willets and Wong 1980). Spermatia have been observed on aerial mycelia in culture, on the surface of sclero tia, and on the apothecial hymenium (Kohn 1979b and Le Tourneau 1970). No biological function is known for spermatia as S. sclerotiorum is homothallic and fertilization by microconidia has not been observed (Kohn 1979b). A discernable feature of the S. sclerotiorum life cycle in culture and on infected plants is the production of highly melanized, multihyphal resting structures termed sclerotia. Three distinct stages of sclero tial development have been characterized by Townsend (1957): 1) initiation, 2) devel opment, and 3) maturation. The mature sclerotium is composed of three distinct la yers; a rind, a cortex and a medulla (Mordue and Holliday 1976). Melanin is found in high concentration in the outer rind and is responsible for the black coloration of sclerotia (J ones 1970). Under appropriate physiological conditions (exposure to optimal light, temperatur e and moisture), sclerotia can undergo carpogenic germination giving rise to multiple apothecia (Huang and Kozub 1989; Morrall 1977; Huang and Kozub 1993; Le Tourneau 1979). A mature apothecium is composed of an unbranched stalk (or stipe) and a single terminally produced cupshaped disc. Histological studies by Kosasih an d Willets (1975) warranted subdivision of the apothecial disc into 4 di stinct parts including: 1) an ectal excipulum, 2) medullary excipulum, 3) hymenium, and 4) subhymenium. The hymenial layer of the apothecium is the site of meiosis and asynchronous ascus production where mature asci forcibly discharge binucleate ascospores (de Bary 1887). Sclerotia are also capable of myceliogenic germination to produce vegetativ e hyphae that can directly infect plant

PAGE 13

3 tissues (Bardin and Huang 2001; Le Tourneau 1979), however is not a major route of infection in most pathosystems. Epidemiology, Host Range and Control of S. sclerotiorum Sclerotinia diseases favor cool, moist condi tions from 16-20C. High humidity and dew formation increase the intensity and spread of the disease (Pohronezny and Purdy, Plant Pathology Fact Sheet). Adams a nd Ayers (1980) showed that the primary source of inoculum in the field is from windbor ne ascospores that are forcibly discharged from sclerotia-derived apothecia and that scle rotia may remain viable for up to 8 years in the soil. Schwartz and Steadman (1979) repor ted that a single a pothecium can produce approximately 3 X 107 ascospores and that a single sclerotium producing multiple apothecia could produce approximately 2.3 X 108 ascospores. Ascospores must germinate and grow saprophytically to be compet ent for infection. Th is distinguishes S. sclerotiorum from most appressoria-forming fungi th at are capable of di rect infection via spore-derived germ tubes (Abawi and Grogan 1979; Lumsden 1979 and McLean 1959). In the field, requisite saprophytic growth takes place on senescent flower petals or detached leaves in which the fungal hyph ae evade and contact intact, healthy tissue (Inglis and Boland 1990; Turkington a nd Morrall 1993). Multicel lular, melanized infection cushions terminally produce infection pegs that are capable of direct penetration of the plant cuticle (Lumsden and Dow 1973). Colonization of the host tissues proceeds via ramifying hyphae that extensively colonize the dead tissue forming a mass of cottonywhite mycelia followed by sclerotial devel opment. Maceration of plant tissues and development of watery soft rot are th e most obvious symptoms of infection by S. sclerotiorum (Willets and Wong 1980).

PAGE 14

4 S. sclerotiorum is a ubiquitous phytopathogen with a wide host range. Boland and Hall (1994) compiled a list of 408 plant species from 278 genera in 75 families, most of which were dicots. Agriculturally important ve getable crops include: beans, cucurbits, carrots, celery, lettuce, peas, radish, rutabaga, turnips, rhubarb, potatoes and tomatoes. Field crops such as citrus, canola, mustard, peas, beans, lentils, sunflowers and legumes are also susceptible to S. sclerotiorum . Florida crops affected by S. sclerotiorum include: pepper, tomato, potato (stem rot) celery (pi nk rot), lettuce (drop), beans (white mold), cabbage (watery soft rot) and peanut. S. sclerotiorum has also been shown to infect the model plant Arabidopsis thaliana (Morgan 1971; Dickman and Mitra 1992). Crop loss estimates are available for diseases caused by S. sclerotiorum on a select number of agronomic and vegetable cr ops. For example, peanut production can be greatly affected by Sclerotinia blight, now pr esent in most peanut-producing countries of the world, especially in the United States. Yield losses commonly reach 10% in states like Virginia, North Carolina, Georgia and Fl orida (Compendium of Peanut Diseases 2nd edition). Soybean production is limited by Sclero tinia stem rot, which has been touted as the second most important disease of s oybean based on 1994 data. Soybean yields inversely correlate with the percent inciden ce of Sclerotinia stem rot (Compendium of Soybean Diseases 2nd edition). Lettuce production can be greatly reduced by Sclerotinia drop. Drop refers to the wilted appearance of the outer lettuce leaves affected by S. sclerotiorum where disease incidence can be as severe as 70% (Compendium of Lettuce Diseases 1st edition). South Florida was faced with an epidemic of S. sclerotiorum in 2002, as entire pepper fields were abandoned fo r fear of harvesting fruit and suffering

PAGE 15

5 post harvest losses. Crop losses for pepper th at year were estimated at about 30% for Palm Beach County alone (K.L. Pernezny personal communication). Sclerotinia diseases can be managed in a variet y of ways including: Chemical and biological control, use of transgenic and naturally derived resistance, and by implementation of cultural practices. Adequate means of chemical control have been achieved with: Actigard, Flint, Ronalin, Rovr al, Switch, Topsin, Benlate, Botran and Qaudris (Kucharek 2003). However, Benlat e and Ronalin are being fazed out of production, and limited labeling of these produc ts on a small number of crops greatly decreases their usefulness . Biological control strategies have focused on the use of mycoparasites, Coniothyrium minitans and Sporidesmium sclerotivorum which are capable of degrading the sclerotium (Ayers and Adams 1981). Presently, two biological control agents: Intercept WG and Contans WG have been approved for agricultural use to control S. sclerotiorum and S. minor in the field. Simply inherited resistance to Sclerotinia diseases has not been documented in any crop to date. However, use of partially re sistant lines in some instances has provided economically useful control for gr owers (Boland and Hall 1987; Grau et al. 1982; Kim and Diers 2000; Nelson et al. 1991). Use of transgenic resistance has focused on the degradation of oxalic acid using oxalate oxi dase and other oxalat e-degrading enzymes. Transgenic resistance has been implemented in sunflower, soybean and peanut and these crops have shown increased resistance to Sclerotinia diseases (Donaldson et al . 2001; Hu et al . 2003; Kesarwani et al . 2000; Livingstone et al . 2005). Cultural practices such as field flooding, crop rotation with non-hosts, co ntrol of alternate weedy hosts, removal of crop residues post harvest, deep plowing of fields, maintaining a well spaced plant

PAGE 16

6 density, avoiding excessive irrigation and sele cting crop varieties with open canopies are practices that may decrease the incidence of disease and can be used in conjunction with other control methods as part of an integr ated pest management program (Pernezny and Purdy; Moore 1949; Steadman 1979) Physiological Mechanisms of Pathogenicity S. sclerotiorum is a necrotrophic fungal phytopat hogen with an extremely wide host range that must kill living tissue in or der to colonize its host and complete its lifecycle. The biochemical basis for its diverse host range has been i nvestigated in several studies. Anton de Bary (1884) was the first to describe the physiological factors that were involved in S. sclerotiorum pathogenicity. He observed that host cells were killed and the middle lamella was destroyed in advance of fungal colonization. He also described the production of oxalic acid by S. sclerotiorum in planta and in vitro . When he applied oxalate alone or infected plan t sap to healthy plants, only the later facilitated symptom development. Boiling infected plant sap prior to treatment eliminated disease symptoms. Taken together, he hypothesized that a “f erment” (enzyme) was responsible for the expression of disease symptoms. Since the time of de Bary, others have investigated the physiological mechanisms of necrotrophic parasitism. Bateman and B eer (1964) showed that polygalacturonase from Sclerotium rolfsii required oxalate ions for hydrolysis of calcium pectate. They also observed a positive correlation between fungal growth and oxalate accumulation. Treatment of beans with oxalic acid caused in jury and bleaching at the application site while treatment of oxalate and an enzyme mi xture lead to tissue maceration and collapse of hypocotyls. The authors concluded th at a synergism between oxalate and polygalacturonase existed whereby oxalic aci d lowers the pH that is optimal for

PAGE 17

7 polygalacturonase activity while simultaneousl y chelating calcium fr om pectate in the middle lamella making it readily availabl e for enzyme-mediated degradation. Maxwell and Lumsden (1970) investigat ed the influence of carbon source on oxalate accumulation in culture and oxalate pr oduction and polygalacturonase activity in S. sclerotioruminfected bean hypocotyls . They found that liquid cultures of S. sclerotiorum supplemented with 73.7 mM glucose and 56 mM sodium succinate produced the highest levels of oxalate after 7 days. They also observed an increase in oxalate production in planta while the pH decreased a nd polygalacturonase activity increased with time. The authors concluded th at the production of oxa late lowered the pH of infected bean hypocotyls which would e nhance the activities of several cell wall degrading enzymes (including polyg alacturonase) that have previously been implicated in S. sclerotiorum diseases. It was also concluded that since oxalate accumulated early in pathogenesis and the existence of a positive co rrelation between disease severity and acid accumulation occurred in infected bean th at oxalate plays an important role in S. sclerotiorum pathogenesis. Marciano et al . (1983) studied the physiol ogical factors that govern S. sclerotiorum pathogenicity in sunflower. This work investigated the relationship between oxalic acid, cell wall degrading en zymes, pH and their signif icance in pathogenicity and virulence of two different S. sclerotiorum isolates. These two isolates differed in virulence and produced xylanase, polygalacturon ase, and cellulase in infected sunflower stems. However, striking differences between the two interactions were present in pH, oxalate concentrations and polyphenoloxidase activity. The strongly virulent strain produced 4 times the amount of oxalate, re duced pH to 4.0, and lacked polyphenol

PAGE 18

8 oxidase activity. The authors concluded that cell wall degrading en zymes alone were not sufficient for infection and that oxalate may inhibit plant polyphenoloxidase. The occurrence of oxalate (ethanedioic acid) in nature is widespread and has been found in animals, plants, fungi, bacteria , rocks and soil. It is produced in large quantities in some species bel onging to all classes of fungi. It can be found in a free acid form, as a soluble salt of potassium or sodi um or as insoluble calcium oxalate (Dutton and Evans 1996). Oxalic acid is a dicarboxylic acid that can be derived from oxaloacetate via the TCA (tricarboxylic acid) cycle or by action of oxaloac tetase and glyoxylate oxidase respectively. It can also be derive d from pyruvate in a two-step enzymatic synthesis via pyruvate carboxylase and oxalace tate acetyl hydrolase (Dutton and Evans 1996). Chemical preparation of oxalate ha s been achieved by passing carbon monoxide into concentrated sodium hydroxide (Wallace 1926) or by heating sodium formate in the presence of sodium hydroxide or sodium carbonate (Beckman 1951). S. sclerotiorum has been shown to produce copious amounts of oxalate and other dicarboxylic acids in vitro like fumaric, succinic, glyc olic, and malic acids (Vega et al . 1970). However, only oxalate has been shown to be a pathogenicity factor as S. sclerotiorum oxalate-minus mutants were unable to cause disease on dry beans despite possessing an arsenal of cell wall degrading enzymes (Godoy et al . 1990). Besides playing a role in pathogenicity, acting as a calcium reserve and playing a role in heavy metal detoxification (Dutton and Evans 1996), oxalate can also manipulate physiological processes in the plant host. Cessna et al . (2000) showed that a non-pathogenic oxalatedeficient mutant induced oxidative burst in susp ension-cultured plant cells , whereas inoculation of an oxalate produc ing strain did not produce measur able levels of oxidant in

PAGE 19

9 tobacco leaves and soybean cultured cells. Fr om this study it was concluded that oxalate aids in pathogen compatibility by inhibiti ng the oxidative burst reaction (a defense response) in host plants (Cessna et al . 2000). Oxalate has also been shown to regulate guard cell function in bean leaves. Guimaraes and Stotz (2004) showed that oxalate facilitates plant wilting by inducing stomatal opening and inhibits ABA-induced stomatal closure which increases plant transpiration th at in turn causes water loss and wilting. Multiple lines of physiological and biochemical evidence have revealed that Sclerotinia pathogenicity is a multi-component process involving the secretion of oxalate and the production of many cell wall degradi ng enzymes. Most studies have focused on the role of oxalate as a pathogenicity fact or. Therefore, the use of oxalate-degrading enzymes as a source of transgenic resistance w ould seem to be an attractive and effective means of control. However, this strategy has not revealed complete resistance to Sclerotinia diseases. To date, no magic bulle t has been discovered which can stop this successful ominovorous plant pathogen. Therefor e, additional studies aimed at unraveling the signaling pathways that control oxalat e production, cell wall degrading enzymes, and infection structure formation, long term, ma y aid the discovery of novel targets for disease control. Genome Sequence A draft genome sequence assembly of S. sclerotiorum was publicly released on April 13, 2005. This project was carried out by the Broad Institute for Biomedical Research in partnership with academic labs at the University of Florida (Dr. Jeffrey A Rollins) the University of Nebraska (Dr. Ma rtin B. Dickman) and the University of Toronto (Dr. Linda M. Kohn) ( S. sclerotiorum Sequencing Project. Broad Institute of Harvard and MIT. http://www.broad.mit.edu ). Genome sequencing information will be

PAGE 20

10 used to promote identification of genes for ge netic and functional studies as well as serve as a template for comparative genomics. Sequence information may also aid in the identification of potential anti-fungal targets and facilitate the el ucidation of signal transduction pathways. Whole genome shotgun sequencing provided an average of 8X coverage. It is estimated that the S. sclerotiorum genome is 38 Mb in size containing 14,522 genes where 41.3% of the genome contains coding sequence. Automated calling of genome sequence data was released on October 1, 2005 and was accomplished by analyzing genome sequence data using FGENESH and GeneID programs combined with the analysis of various expressed sequence tags (ESTs). Analysis of genome data revealed that there is ~1 gene every 2.6 kb of nucleot ide sequence with an average intron length of 140 bp. The shortest intron designated wa s 24 bp and the longest was 1494 bp. The average exon length was estimated at 389 bp with the longest being 17,212 bp and the shortest as 1 bp ( S. sclerotiorum Sequencing Project. Broad Ins titute of Harvard and MIT. http://www.broad.mit.edu ) cAMP-Mediated Signal Transduction Discovery of cAMP Rall and Sutherland (1958) published the fi rst description of a cyclic nucleotide produced in living cells which de tailed the synthesis, degrada tion and structure of 3’, 5’ cyclic adenosine monophosphate (cAMP). This landmark discovery led to a Nobel Prize awarded to Earl Sutherland in medicine a nd physiology in 1971. This and other studies carried out by Southerland led to the discovery that adrenaline stim ulated gluconeogensis via cAMP. In light of these st udies, Southerland is credited with initiating the “second messenger” concept. In eukaryotes the cAMP second messenger induces physiological

PAGE 21

11 responses ranging from growth, different iation, gene expres sion, secretion and neurotransmission (Kawasaki et al . 1998). Since the discovery of cAMP, other secondary messengers’ i.e cyclic GMP, Ca+2, diacylglycerol, and inosit ol phosphate have been discovered but their functions have not been fully elucidated (Dickman and Yarden 1999). Over the past 50 years, research into second messengers has provided a framework for understanding transmembran e signal transduction, receptor-effector coupling, protein kinase cascades, and down regulation of drug responsiveness (Beavo and Brunton 2002). cAMP Signal Transduction Pathway A general cAMP signal transduction pathwa y for animals can be described in the following manner. Binding of a hormone to its cognate receptor lead s to coupling of the receptor-hormone complex to a cellular guani ne nucleotide-binding protein (G-protein), which interacts with and activates adenylyl cyclase (AC). Activation of AC promotes synthesis of the second messenger cAMP from ATP. Cyclic AMP then binds to cAMPdependent protein kinase (PKA) which cause s the catalytic subunit to dissociate and phosphorylate protein substrates that regulate a number of biological processes (Chin et al . 2002). At the heart of cAMP-mediated signal transduction lies cAMP-dependent protein kinase A (PKA). PKA was one of the first pr otein kinases to be purified and have its crystal structure determined (Knighton et al . 1991; and Walsh et al . 1968). PKA has been found in all eukaryotes studied and remains the primary, but not the only known receptor for cAMP (Dickman and Yarden 1999). PKA is composed of two genetically distinct regulatory and catalytic subunits that form a tetramer. In the presence of cAMP, 2 regulatory subunits each bind two molecules of cAMP that allow the release of two

PAGE 22

12 catalytically active s ubunits that can phosphorylate a va riety of nuclear and cytosolic targets (Chin et al . 2002). Transcriptional regulation in the nucleus is achieved through cAMP-responsive transcription factors that bind to and regulate the expression of genes containing a cAMP-responsive element (CRE) in their promoter. These cAMPresponsive binding proteins contain a basic domain/leucine zipper motif, and translocate to the nucleus upon phosphorylation vi a the PKA catalytic subunit (Chin et al . 2002). The PKA catalytic subunit is a seri ne/threonine protein ki nase that utilizes gamma phosphate from ATP or GTP to genera te phosphate esters using alcohol groups on serine and threonine residues as acceptors (Dickman and Yarden 1999). PKAs contain 12 highly conserved subdomains of seri ne/threonine kinase s among other highly conserved hallmarks of PKA catalytic subunits . Biochemical functions for some of the subdomains have been elucidated and or have a hypothesized function. Subdomain I contains a glycine-rich motif that is thought to contact the ribose moiety of ATP. Subdomain II contains a lysine residue involved in the phos photransfer reaction, possibly mediating proton transfer. Subdomain VI cont ains asparagine, phenylalanine, and glycine residues which are implicated in ATP binding. Subdomain VIII contains alanine, proline, and glutamate which are involve d in general base catalysis by removing a proton from serine via carbonyl carbon ionization (Taylor et al . 1990). Hallmarks that differentiate PKA from other serine/thre onine protein kinases include : histidine, leucine, and tryptophan amino acids involved in regul atory subunit binding, a pentapeptide PKI inhibitor binding site and a threonine resi due capable of autophosphorylation located between two of the residues involved in regulatory subunit binding (Durrenberger et al . 1998).

PAGE 23

13 PKA regulatory subunits have been de signated type I and II. However, filamentous fungi possess only type II regulatory subunits as they contain an autophosphorylation site located in the dimer interaction motif while type I do not (Taylor et al .1990 and Dickman and Yarden 1999). Both types I and II contain the following 4 functional domains. The amino term inus contains the dimer interaction site composed of hydrophilic and charged amino aci ds that are responsible for interaction with other cellular proteins and membrane s. Two cAMP-binding sites, A and B, are located near the carboxy terminus. Thes e two domains contain a phosphate binding cassette with the following conserved residue s: F-G-E (L-I-V) A-L-(L-I-M-V)-3X-(P-V)R-(A-N-Q-V)-A (Canaves and Taylor 2002). Bi nding of cAMP occurs in a cooperative fashion. Binding domain B occurs first allowing domain A to become more accessible for cAMP. The peptide inhibitory region is lo cated between the dimer interaction and the cAMP-binding sites and serves as the area of interaction between the catalytic and regulatory subunits (Taylor et al . 1990) Cyclic AMP is synthesized by adenylat e cyclase (AC) and degraded by cAMPspecific phosphodiesterases (PDEs). These two enzymes work in concert to regulate intracellular levels of cAMP. Phosphodiesterase s are represented by a large super family of enzymes and possess a highly conserved cata lytic domain that is C-terminal and an amino terminal regulatory domain (Soderl ing and Beavo 2000). There are two types of PDEs, low affinity (Pde1) and high affinity (Pde2), which have been shown to regulate distinct biological processes and are structurally different as they share virtually no sequence similarity (Kicks et al. 2005). In mammalian systems, PDEs have been shown to regulate insulin secretion, T-cell activation, fertility and growth and penile erectile

PAGE 24

14 function (Soderling and Beavo 2000). PDEs in y east have been implicated in controlling stress tolerance, filamentation, nutrient sensi ng, entry in to S-phase of the cell cycle and maintaining cell wall and membrane integrity (Bahn et al 2003; Jung et al 2005). Many fungal phosphodiesterases require the divalent cations magnesium and manganese and can be inhibited with ethylenediaminetetr aacetic acid, theophylline, caffeine, and 1methyl-3-isobutylxanthine (Pall 1981). Adenylate cyclase is the biosynthetic en zyme that produces 3’, 5’ cAMP and pyrophosphate from ATP using base catalyzed nucleophillic elimination (Zubay 1998). Mammalian adenylate cyclases are membrane bound, except for cytosolic forms found in human sperm and Dictyostelium discoideum. Mammalian ACs are composed of two serpentine-like transmembrane domains and two cytosolic catalytic motifs that are separated by a highly conserved domain of unknown function (DUF). All mammalian AC’s are activated by G alpha(s) and the diterp enoid forskolin. Inter action of AC with Gproteins is dependent on pre nylation of the G-protein ga mma subunit (Tang and Gilman 1992). Fungal adenylate cyclases are membrane associated and contain an N-terminally located RAS-association domain, followed by mu ltiple leucine rich repeats thought to be involved in protein-protein interactions. A PP2C-phosphatase domain located near the Cterminus may mediate Ras-GTP activation of adenylate cyclase and is not known to possess phosphatase activity in yeast (Barford et al . 1998). The catalytic domain (CycC) is highly conserved among AC and guanylate cyclases (GC) containing the consensus residues (V/I)KTXGDA) where K is thought to contact the N2 atom of ATP (Roelofs et al . 2001a).

PAGE 25

15 Heterotrimeric G-proteins are associated with the inner surf ace of the plasma membrane and with cytosolic domains of seven transmembrane receptors. They are composed of 3 subunits: alpha, beta and gamma which serve distinct functions in signal transduction. The alpha subunit possesses GTPa se activity and is also the site of prenylation which anchors the complex to the plasma membrane. G alpha is active in the GTP bound state and inactive in the GDP -bound state. Upon GTP binding, G alpha dissociates from beta/gamma and is able to stimulate effector molecules like adenylate cyclase in mammalian systems (Gilman 1984). G-proteins are regulated by guanine nucleotide exchange factors (GEF’s) that cata lyze the release of GDP thus activating G alpha, whereas GTPase activating proteins (GAPs) promote inactivation of small Gproteins through activation of GTPase act ivity (Wittinghofer 1998; Bar-Sagi and Hall 2000). Other Targets of cAMP For approximately 40 years it has been wi dely accepted that the main target for cAMP is the PKA regulatory subunit (Dickman and Yarden 1999; Chin et al . 2002). However, this dogma is being re-written with the discovery of cAMP-binding effector proteins that are independent of PKA signaling (Chin et al . 2002). Studies involving cardiac cells have shown that cAMP modulates ion channel activity by dire ctly binding the carboxy terminus of HERG K+ channels where cyclic nucleotide-binding domains are located (Cui et al . 2000). Kawasaki et al . (1998) showed that families of cAMPbinding proteins were differe ntially distributed in mamm alian brain and that they possessed cAMP-binding and guanine-nucleotide exchange factor domains. cAMPregulated GEFs (cAMP-GEF’s) were show n to bind cAMP and activate RAP1 in a cAMP-dependent yet PKA-independent fashion. cAMP-binding proteins have also been

PAGE 26

16 found in the slime mold D. discoideum during the early stages of development. D. discoideum expresses cAMP-receptors at the cell surface that are important in the perception of cAMP gradients (acrasin) that signal aggreg ation of amoebae leading to sporocarp formation. Klein et al . (1988) cloned a cDNA enco ding a cAMP-receptor and the deduced amino acid sequence revealed 7 clusters of hydrophobic residues characteristic of 7-transmembrane receptors kno wn to interact with G-proteins. A serinerich cytoplasmically located carboxyl termi nus is the proposed si te of ligand-induced receptor phosphorylation (Klein et al . 1988). Despite the identification and characterization of novel cAMP -binding proteins that act independently of PKA, the signal transduction path ways in which these proteins operate have yet to be fully elucidated. Biochemical Functions of cAMP in Fungi Biochemical functions for cAMP in filame ntous fungi and yeast include nutrient sensing and mobilization of nutrient reserves. More specifically cAMP has been shown to be involved in the catabolism of carbon rese rves glycogen and tr ehalose, in yeast mitochondrial functions, and in the utiliza tion of exogenous carbon sources (Pall 1981). Catabolism of endogenous carbon has focuse d on the utilization of glycogen via activation of glycogen phosphorylase by cAMP in Neurospora crassa (Tellez-Inon, M.T. and Torres, H.N. 1970; Gold et al. 1974) , Coprinus macrorhizus (Uno and Ishikawa 1976) , and by induction of trehalas e in trehalose metabolism in Saccharomyces cereviseae (van der Platt and van Solingen 1974 and van Solingen and van der Platt 1975) . In S. cereviseae, mitochondrial synthesis is repre ssed by high glucose levels (Pall 1981). Cyclic AMP has also been shown to influence mitochondrial transcription and translation (Chandraskaran and Jayarman 1978) and the biosynthesis of ubiquinone-6

PAGE 27

17 (Sippel et al. 1979). Control and utilization of exogenous carbon sources via cAMP has been shown in the crisp mutant as it grows as well as wild type on glucose or acetate (Terenzi et al . 1979). Exogenous application of cAMP has also been reported to overcome glucose repression in S. cereviseae (Wiseman and Lim 1975 and Wiseman and Lim 1977). While an abundance of information has b een obtained in model filamentous fungi and yeast as to the biochemical roles of cA MP, research over the past few decades has focused on the biological effects of cAMP in filamentous phytopathogenic fungi. In S. sclerotiorum, high cAMP levels were shown to inhibit sclerotial development and concomitantly raise oxalic acid levels (Ro llins and Dickman 1998). Cyclic AMP has also been demonstrated to regulate in fection structure formation in Magnaporthe grisea (Lee and Dean 1993), modulate hyphal branching in Fusarium graminearum (Robson et al . 1991) and control appressorial germ tube differentiation in Blumeria graminis (Hall and Gurr 2000). These pharmacological studies have provided insight into the biological processes and pathways influenced by cAMP , and serve as an important foundation to further dissect cAMP-mediated signal tran sduction via targeted gene analysis. Functional Analyses of cA MP Signaling Components Functional analyses of PKA catalytic s ubunit genes have been carried out in a number of yeasts incl uding, but not limited to: S. cereviseae , Crytpococcus neoformans , Candida albicans, and Schizosaccharomyces pombe (Pan and Heitman 1999; Pan and Heitman 2002; Robertson and Fink 1998; D'Souza et al . 2001; Sonneborn et al . 2000; Maeda et al . 1994) and in filamentous fungi: Colletotrichum trifolii, Ustilago maydis, and M. grisea (Yang and Dickman 1999; Durrenberger et al . 1998; Xu et al . 1997; and

PAGE 28

18 Mitchell and Dean 1995). The focus of this di scussion will be limited to investigations regarding filamentous fungal phytopathogens. The Ct-PKAC catalytic subunit gene from C. trifoli was cloned, characterized and inactivated by gene replacement (Yang and Dickman 1999). Phenotypic analyses revealed that there was a small reduction in growth compared to wild type and conidiation patterns were altered. The mutant strain was also unable to infect intact, healthy alfalfa leaves despite a small delay in conidial germination and appressorium formation. However, Ct-PKAC disruption mutants were able to cause disease on artificially wounded plants. Ther efore, it was concluded that the defect in pathogenicity was due to a failure in penetration, and that PKA regulates the transition between vegetative growth and conidiation (Yang and Dickman 1999). In the dimorphic basidiomycete plant pathogen U. maydis , two PKA catalytic subunit genes were cloned and functionally characterized. Disruption of PKA catalytic subunit gene adr1 resulted in constitutive fila mentous growth, was required for pathogenicity, and found to be responsible for the majority of activity in the cell. In contrast, the second PKA catalytic subunit gene, uka1 , was disrupted and did not affect cell morphology and showed little influe nce on pathogenicity (Durrenberger et al . 1998). Taken together, these results demonstrate th at PKA activity is critical for dimporhic switch to pathogenic filamentous growth and that two PKA catalytic subunit genes exist and play different roles in the U. maydis biology. Functional analysis of a PKA catalytic subunit gene was carried out by two groups with M. grisea (Xu et al . 1997; Mitchell and Dean 1995). Mitchell and Dean cloned and disrupted a PKA catal ytic subunit gene designated cpkA. Disruption mutants

PAGE 29

19 were unable to produce appressoria even in the presence of cAMP. They reported that cpkA mutants were unable to cause disease on susceptible intact and abraded rice cultivars. The authors concluded that cAMP-dependent protein kinase ( cpkA ) is necessary for infection-related morphogenesis and pathogenesis in M. grisea . Contrasting data for cpkA mutants was reported by Xu et al . (1997). They carried out a more rigorous and detailed phenotypic ch aracterization. The cpkA mutants were severely reduced in pathogenicity on healthy plants. However, this defect was not due to the lack of infection structure formation as appressoria formed by cpkA mutants were fully melanized, but smaller in size than wild type. Appressorial formation was greatly delayed as compared to wild type, yet could be restored by a pplication of exogenous cAMP. Mutant strains were shown to produce infectious hyphae a nd cause lesions when inoculated through wounds. The authors indicated that cAMP signaling was res ponsible for appressorium penetration and that there maybe additional PKA catalytic subunits involved in the process of surface-sensing in M. grisea . Adenylate cyclase is the biosynthetic en zyme responsible for the synthesis of cAMP in the cell. In order to understand th e biological role th at cAMP plays in eukaryotic organisms, researchers have targ eted the gene encoding adenylate cyclase in mutational studies. Adenylate cy clases have been functionall y analyzed in model yeasts like S. cereviseae and S. pombe (Matsumoto et al . 1982; Maeda et al . 1990), model filamentous fungi like N. crassa (Terenzi et al . 1974) , and filamentous phytopathogenic fungi like M. grisea and Botrytis cinerea (Choi and Dean 1997; Klimpel et al . 2002). The first adenylate cyclase mutant, crisp ( cr-1 ), was also one of the first morphological mutants reported in N. crassa described by Lindegren (1936 ). The crisp mutant was

PAGE 30

20 named for it’s “crisp” morphological phenotype in culture due to production of short aerial hyphae with tight clusters of more br ightly colored conidia than wild type (Lindegren 1936) and early un iform conidiation over the ag ar surface (Perkins 1959). Almost 60 years later the identity of the muta ted gene responsible fo r the crisp phenotype was confirmed to be adenylate cyclase (Kore-eda et al . 1991). Recent studies have focused their attenti on on filamentous fungal plant pathogens. Choi and Dean (1997) cloned and functionally analyzed an adenylate cyclase gene, MAC1 from the ascomycete rice blast pathogen M. grisea . Deletion of MAC1 resulted in reduced vegetative growth, conidi ation and conidial germination. MAC1 mutants were also unable to form appressoria on inductive surfaces, but could be restored with the exogenous application of cAMP. Deletion muta nts were unable to penetrate susceptible rice leaves and perithecial form ation was abolished rendering MAC1 mutants sterile. The authors concluded that cell signaling involvi ng cAMP is central to the development and pathogenicity of M. grisea . Klimpel et al . (2002) cloned and functionally an alyzed the BAC gene encoding an adenylate cyclase in the necrotrophi c filamentous phytopathogenic fungus B. cinerea . Gene replacement resulted in reduced vege tative growth, which wa s partially restored with the addition of exogenous cAMP in culture. BAC mutants produced low levels of cAMP despite the deletion of the entire catalytic domain. Lesions caused by BAC mutants developed more slowly than w ild type and sporulation did not occur in planta . Together, these results show that cAMP signaling controls aspects of vegetative growth and pathogenicity in B. cinerea.

PAGE 31

21 S. sclerotiorum has proven itself as a highly succe ssful, widely distributed and destructive filamentous funga l plant pathogen. Its ability to survive in the soil for extended periods of time coupled with its “t hug-like” approach to pathogenicity makes it an attractive organism for studyi ng the necrotrophic form of pa rasitism. Detailed analysis of sclerotial development, apothecial ge rmination, regulation of cell wall degrading enzymes and physiological processes regulating pathogenicity are ideal areas to explore in efforts to control Sclerotinia diseases. Id entification and detailed investigation of the signaling pathways that control these biolog ical phenomena offer the greatest potential for identifying novel targets for disease control for S. sclerotiorum .

PAGE 32

22 CHAPTER 2 CHARACTERIZATION AND FUNCTIONAL ANALYSIS OF A cAMP-DEPENDENT PROTEIN KINASE A CATALYTIC SUBUNIT GENE ( pka1 ) IN Sclerotinia sclerotiorum1 Introduction S clerotinia sclerotiorum (Lib.) de Bary is a fi lamentous ascomycete plant pathogen that affects more than 400 sp ecies of plants (Boland and Hall 1994). S. sclerotiorum lacks mitotic spores, but produces sc lerotia which are compact, multihyphal, highly melanized structures capable of l ong-term quiescent survival for many years. Sclerotia germinate mycelioge nically to produce saprophytic and/or infectious hyphae, and carpogenically to form apothecia. Individual sclerotia can produce multiple apothecia each with the capacity to produce millions of ascospores (Steadman 1979). Apothecia forcibly discharge ascospores which serve as the primary source of inoculum for disease development. The enormous reproductive poten tial of sclerotia and capacity for long term survival define key characteristics of sc lerotia important in the epidemiology of Sclerotinia diseases. The structural makeup of sclerotia has b een extensively studied (Chet and Henis 1975; Le Tourneau 1979; Willets and Bullock 1992; Willets and Wong 1980). Townsend (1957) described three distinct stages of sclerotial formati on: initiation, development, and maturation. However, the molecular mechanis ms that regulate a nd signal sclerotial 1 Reprinted from Physiological and Molecular Plant Pathology, vol 64, Wayne M. Jurick II, Martin B. Dickman and Jeffrey A. Rollins, “Characterization an d functional analysis of a cAMP-dependent protein kianse A catalytic subunit gene ( pka1 ) in Sclerotinia sclerotiorum ”, 155-163, 2004, with permission from Elsevier.

PAGE 33

23 biogenesis have yet to be elucidat ed. Rollins and Dickman (1998) used a pharmacological approach to identify signali ng components responsible for sclerotial development. Of the compounds tested, onl y those that affected cAMP metabolism (caffeine, 3-isobutyl-1-methylxanthine, and Na F) and cAMP itself blocked the formation of sclerotial initials. In the presence of high cAMP levels mycelia grew normally, but failed to form multi-hyphal aggregates that are indicative of scle rotial initiation. If sclerotia were initiated in the absence of high exogenous cAMP, the normal course of development was not affected by subseque nt increases of exogenous or endogenous cAMP levels. These results suggest that si gnaling via PKA is involved in the initial event(s) of sclerotial formation. High intracellular cAMP levels not only i nhibit sclerotial development, but also increase oxalic acid levels in S. sclerotiorum (Rollins and Dickman 1998). Oxalic acid is known to play synergistic physiological role s in Sclerotinia pathogenesis (Maxwell and Lumsden 1970) as oxalic acid deficient mutant s are nonpathogenic and unable to produce sclerotia (Godoy et al. 1990). Several pathways of oxalate biosynthesis have been proposed in fungi. Oxalate can be derived fr om oxaloacetate by enzymatic removal of acetate via oxaloacetate acetyl hydrolase. Sour ces of oxaloacetate can be from the TCA cycle (mitochondria); the glyoxylate cycle (g lyoxysome) or cytoplasmically from conversion of pyruvate (derived from glycolysis) to oxaloacetate by pyruvate decarboxylase (Dutton and Evans 1996). Ambien t pH is an important factor in the regulation of oxalic acid accumulation (Maxwell and Lumsden 1973; Rollins 2003) and also influences sclerotial development (Rol lins and Dickman 1998). The enzymatic route

PAGE 34

24 of oxalic acid biosynthesis in S. sclerotiorum and the molecular roles of cAMP and pH in oxalate synthesis have yet to be established. In various phytopathogenic fungi, PKA has been determined to play a role in cellular processes centr al to pathogenicity including a ppressorial formation, vegetative dimorphism, and hyphal growth (Mitchell and Dean 1995; Xu and Hamer 1997; Yang and Dickman 1999; Durrenberger et al. 1998). PKA is a heterotetramer composed of two catalytic and two regulatory s ubunits that are bound together in an inactive form (Taylor et al. 1990). Binding of cAMP allows the ca talytic subunit to dissociate from the regulatory subunit and then the free catalytic subunit is able to function as an active kinase. The central role of the PKA catal ytic subunit in cAMP-mediated signaling suggests that it is involved in cAMP-mediated effects on sc lerotial initiation and oxalic acid production. I have begun to test this hypothesis by isolating and functionally characterizing a pka -encoded catalytic subunit gene from S. sclerotiorum . Materials and Methods Fungal Isolates, Growth Conditions, Media and Cultural Manipulations The wild type Sclerotinia sclerotiorum 1980 isolate (Godoy et al. 1990) was used to derive all strains in this study. Cultures were routinely maintained on PDA (potato dextrose agar) (Difco, Detr oit, MI) and propagated by ma ss hyphal tip transfer. Liquid shake cultures inYPSuc medium (4 g/l yeast ex tract (Difco, Detroit, MI), 15 g/l sucrose, 1 g/l K2HPO4 and 0.5 g/l MgSO4 pH 6.5) were grown as prev iously described (Rollins 2003). Pathogenicity, growth on citrate-phosphate buffered PDA, oxalic acid analysis and radial growth analysis was carried out as described by Rollins (2003). Growth and analysis of cultures on cAMP-amended PDA wa s carried out as previously described by

PAGE 35

25 Rollins and Dickman (1998). To induce heat stress, PDA cultures were placed at 37C for 24 hours prior to scleroti al initiation, and subsequen tly visually analyzed for aberrations in sclerotial development. Water agar (dd H2O + 15 g/l noble agar) was used to analyze mutant strains under nutrient star vation. Water agar medium was used as a base for the following media: nitrogen st arvation (+ 10 g/l glucose + 10 g/l K2HPO4 + 25 g/l MgSO4 + 15 g/l NaCl, carbon star vation (+ 10% w/v CaNO3 + 20 g/l K2HPO4 + 25 g/l MgSO4 + 15 g/l NaCl), high salt (4% NaCl), high osmotic potential (1M sorbitol), and sole carbon sources (10 g/l sucrose, glucose, fructose, arabinose, galactose, maltose, raffinose, sorbitol, mannitol, myoinositol, po tassium acetate, citric acid, oxalic acid, glucosamine, ascorbic acid, p -aminobenzoic acid, panthothenic acid, or nicotinic acid). Apothecia were induced from PDA culture-der ived sclerotia as previously described by Russo and Van Etten (1982). Visual asse ssment of spore production, viability and morphology was carried out using both compound and stereomicroscopes. Methods for Nucleic Acid Isolations and Manipulations Mycelia from liquid shake cultures were flash frozen in liquid nitrogen, lyophilized and stored at -80C. Lyophilized mycelia were used to isolate genomic DNA as described by Yelton et al. (1984). Total RNA was extrac ted from lyophilized mycelia using Trizol reagent (Gibco BRL, Rockvi lle, MD) according to the manufacturer’s instructions. RNA electrophoresis was conducte d as previously described (Rollins and Dickman 2001). E. coli strain DH5 was used to propagate all plasmids in this study. Plasmid isolations, agarose gel electrophor esis, DNA restriction digests, ligation reactions, and transformations of E. coli were conducted using standard procedures (Sambrook and Russell 2001). For Southern hyb ridization analyses, digested genomic

PAGE 36

26 DNAs were transferred to MagnaGraph Nyl on Membrane (Micron Separations Inc. Westborough, MA) by downward alkaline transf er (Chomczynski 1992) then UV-crosslinked. For Northern hybridization analyses , RNAs were transferred to MagnaGraph Nylon membrane by standard procedures (Ausubel et al. 1991). RNA and DNA hybridization analyses were carried out at high stringency as defined by Ausubel et al. (1991) and analyzed by autoradiography. Radioactive probes for all hybridizati ons were generated using the Random Primers DNA Labeling System (Invitrogen Life Technologies, Carlsbad, CA) as per manufacturer’s instruct ions. A full length 2.1 kb hph (hygromycin phosphotransferase) gene cassette (Redman and Rodriguez 1994) , and a 620 bp fragment corresponding to the 5’ end of the pka1 gene was used to probe both genom ic DNA and RNA blots in this study. Molecular Cloning and Sequence Identification of pka1 One hundred nanograms of genomic DNA from S. sclerotiorum were used as a template for PCR amplification. Degenerate primers: PKA1F 5’-TA(CT) (AC)G (ATGC) GA(CT) (CT)T(ACG) AA(AG) CC(ACG) GA-3’ & PKA1R 5’-CA(ATGC) (CT)(ACG)(CT) TT(GT) GC(AG) AA(ATGC) CC(AG) AA(AG) TC-3’were used. PCR cycling conditions were as follows: 94C for five min followed by 30 cycles of 94C for one min, 45C for one min, 72C for one mi n and a final cycle of 72C for 7 min. Standard PCR reaction conditions were used (Ausubel et al. 1991), except that the final magnesium concentration was kept at 1.25 mM. The resulting 77 bp amplicon was gene cleaned (Q.BIOgene, USA) and reamplifie d using the above conditions except the annealing temperature was changed to 50C. This product was separated on a two percent agarose gel, gene cleaned and cloned us ing the TOPO TA cl oning kit (Invitrogen,

PAGE 37

27 Carlsbad, CA). The clone was sequenced, and BLAST (Altschul et al. 1997) was used to determine nucleotide sequence homology to other DNA sequences. The 77 bp insert was labeled with 32P using a PCR-based technique pr eviously described by Mertz and Rashtchian (1994) and used to screen a S. sclerotiorum lambda EMBL3 genomic library (Rollins and Dickman 2001). Hybridizing sequences localized to a 3.2 kb Hin d III band were cloned into the pBluescrip t plasmid vector (Stratagene, La Jolla, California). The insert of the resulting plasmid pBSPKA 15.2, was sequenced by primer walking. This pka1 sequence was deposited in Genbank (accession AY545583). Construction of pka1 Gene Replacement and Complementation Vectors Plasmid vector pBSPKA15.2, containing the genomic pka1 gene and flanking sequences, was used to construct a gene re placement vector. The plasmid was digested with Msc I and filled in with Klenow. An hph gene cassette contained on plasmid pHA1.3-H3 (Redman and Rodri guez 1994) was digested with Hin d III, filled in with Klenow, and inserted by bl unt-end ligation into the Msc I site of the linearized pBSPKA15.2 plasmid. This new vector, PKA Hph, contained the 2.1 kb hph cassette inserted between sub domains III and IV of the protein kinase catalytic core of pka1 . To genetically complement the pka1 disruption strains, the pBSPKA15.2 vector was digested with Xho I and Not I which liberated a 3.6 kb insert. The pBARKS1 vector (Pall and Brunelli 1993) was digested with Xho I and Not I which liberated a 4.5 kb fragment containing th e bialophos resistance ge ne cassette. The 4.5 kb bar cassette was then ligated to the 3.6 kb insert containing the pka1 gene which became the 8.1 kb plasmid pPKA1BAR.

PAGE 38

28 Fungal Protoplast Transformation and Evaluation of Transformants PKA Hph was linearized with Sac II, gel purified and in troduced into the wild type genome of S. sclerotiorum strain 1980 via PEG-mediated protoplast transformation as described by Rollins (2003). Southern hybridization was used to screen DNA integration events in the transformant s. For this analysis, genomic DNA from hygromycin-resistant transformants was digested with Hin d III and hybridized with two probes. Sequences for probe one were deri ved by PCR amplification of the pBSPKA15.2 template using primers 5’-TGGTTAGAT ACAGGATCATG-3’ (SPKAC-F2) and 5’GTACCTGAAGGTATCCTCC-3’ (SPKAC-R4) . The resulting 650 bp fragment represents nucleotides 440 to 1090 of the 5’ pka1 genomic sequence (accession AY545583). Sequences for probe two were deri ved by digesting the pHA1.3-H3 vector with Hin d III and gel purifying the resulting 2. 1 kb fragment corresponding to the hph gene cassette. The pka1 disruption strain 2 was comp lemented with pPKA1BAR by PEG-mediated protoplast transformation. Tran sformants were selected on RM medium containing 10 g/ml Bialophos (Shinyo Sangyo Co., LTD. Japan). Protein Extraction and Analys is of PKA Enzyme Activity Tissue used for protein extraction was de rived from freshly frozen liquid shake culture mycelia. Approximately 100 mg of mycelial mat was ground in a mortar and pestle with liquid nitrogen. One l of PKA extraction buffer [200 mM MOPS (3morpholinopropanesulfonic acid) (pH 8.0), 100 mM PMSF (phenylmethylsulfonyl fluoride), 10 mM EGTA (ethylenebis (oxye thylenenitrolo) tetraacetic acid) (pH 8.0), 5 mM EDTA (ethylenediaminetetraacetic ac id) (pH 8.0), 10 g/ml Leupeptin, 10g/ml Aproteinin, 10 mM sodium flouride] was added per mg of mycelia. The tissue and extraction buffer were mixed thoroughly and placed on ice for 30 minutes. Samples were

PAGE 39

29 spun in a centrifuge for 30 minutes at 14K rpm at 4C and the supernatants recovered. Bio-Rad protein assay reagent (Bio-Rad, US A) was used to quantify total soluble proteins as described by Brad ford (1976). Protein extracts were normalized and used immediately for PKA enzyme analysis. Analysis of PKA enzyme activity was carried out using the PepTag Nonradioactive cAMP-dependent protein kinase assay kit as per the manufacturer’s instructions (Promega, Madison, WI). Fluor escently-labeled kemptide products were separated on a (0.8% agarose gel 10mM Tris, pH 8.0, in 10mM Tris, pH 8.0) and visualized by UV light illumination using a Gel-Doc 2000 imaging system (Bio-Rad, USA). Multiple Sequence Alignment, Phenogram Construction and Bootstrap Analysis Amino acid sequences corresponding to the catalytic subunits from seven entries ( Homo sapiens, Drosophila melanogaster, Ustilago maydis, Colletotrichum trifolii, Magnaporthe grisea, Sclero tinia sclerotiorum, and Neurospora crassa) were derived from nucleotide sequences obtained from public databases. Amino acid sequences were aligned using ClustalX (Thompson et al. 1997) with the following parameters: Pairwise align (slow-accurate), gap opening 10, gap extension 0.10. An alignment file was then transferred to GeneDoc and the sequences we re annotated. The PAUP program was used to construct a phylogenetic tree using parsim ony, and two taxa were designated as out groups. Stepwise and random addition of each taxon was used to construct the most parsimoniest tree. The data set was analy zed by bootstrap analysis, and one thousand replicates were resampled 10 times each.

PAGE 40

30 Results Molecular Cloning and Characterization of pka1 A pair of degenerate primers, based on the conserved regions of pka catalytic subunit genes from various filamentous f ungi, was used in PCR reactions with S. sclerotiorum genomic DNA. A 77 bp amplicon was isolated, sequenced and determined by BLAST (Altschul et al. 1997) analysis to have high homology with other protein kinase A catalytic subunit genes. Th e amplicon was used to screen a S. sclerotiorum genomic DNA library, and a la mbda-clone containing a 3.2 kb Hin d III digestion fragment was obtained and cloned into a pB luescript plasmid vector. The insert was sequenced and determined by BLAST to cont ain a 1.6 kb region of sequence with highest identity to other fungal pka catalytic subunit genes. The pr edicted protein possesses all 12 highly conserved sub-domains of the serine/th reonine kinase catalytic core. Amino acid residues that are specific to the PKA cla ss of serine/threonine kinases include: three conserved residues for the regul atory subunit binding site, a putative autophosphorylation site, and the quatrapeptide PKI inhibitor binding site. All of these highly conserved residues are encoded on the pka1 gene obtained from S. sclerotiorum (Fig. 2-1). Disruption of the pka1 Locus The strategy to disrupt the pka1 locus with the pka1-hph gene disruption construct described in the Materials and Methods is shown in Fig. 2-2. Using this construct, approximately 50 hygromycin-resistant transf ormants were recovered and purified by several rounds of hyphal-tip tr ansfer. Genomic DNA from 15 of these transformants was digested with Hin d III and screened by Southern blot an alysis using the 5’ portion of the pka1 coding sequence as a probe (data not shown).

PAGE 41

31 Southern blot analysis of three putat ive gene-disruptant transformants (D1, D2, and D3), the wild type strain (Wt), and a randomly chosen ectopic transformant (E1) are shown in Fig. 2-3. Hybridization with probe-1 containing the 5’ pka1 sequence (Materials and Methods) produced an expected sized ba nd of 3.2 kb in wild type. Two bands of 5.4 and 3.2 kb were detected in E1 and D1 and a 5.4 kb band was detected in D2 and D3 (Fig. 2-3). To confirm that the 5.4 kb band represented the disrupted pka1 locus containing the hph cassette, genomic DNA was hybridized with the hph sequence. As expected, probe 2 hybridized to a 5.4 kb band in all strains except the wild type (Fig.2-3). In the eleven other independent tran sformants examined, two patterns of hybridization were observed (d ata not shown). Ten of thes e transformants produced a hybridization pattern simila r to the E1 strain. Genomic DNA from three other transformants hybridized strongly to a 5.4 kb band and weakly to a 3.2 kb band. This pattern is indicative of hetero karyosis, in which wild type and gene-replacement nuclei are present in a common cytoplasm. D2 was chosen as a representative pka1 disruptant strain and utilized as a recipient for complementation via tran sformation with pPKA1BAR vector. The pPKA1BAR vector contains ~3.2 kb of the intact genomic pka1 coding and flanking sequences and the bar cassette conferring resistance to bialaphos. Bialaphos-resistant transformants were recovered and one strain, C1, was analyzed by genomic DNA hybridization (Fig. 2-3). Geno mic DNA from the C1 strain hybridized to 3.2 and 5.4 kb bands using the 5’ pka1 probe. RNA Blot Analysis To determine if pka1disruptant strains possessed wild type-sized pka1 transcript, total RNA was isolated and subjected to agarose gel electrophor esis. Fig. 2-4 shows

PAGE 42

32 Northern blot analysis afte r hybridization with the 5’ pka1 probe. The wild type transcript appears as a ~ 1.6 kb band, whereas the disrupta nt strains lack the Wt band but possess a smaller ~ 1.0 kb band. The ectopic and comp lemented strains possess both the 1.0 and 1.6 kb bands. Growth on 10 mM cAMP Amended PDA Wild type S. sclerotiorum is unable to produce sclerotia on PDA supplemented with 10 mM cAMP (Rollins and Dickman 1998) . All strains produced wild type-like sclerotia when grown on PDA, and were una ble to produce sclerotia when inoculated onto media amended with 10 mM cAMP. No difference among strains (Wt, D1, D2, D3, E1, and C1) were observed with regards to cultural morphology, mycelial growth, or sclerotial initia tion, development, and maturation (Fig. 2-5). Lack of a morphological phenotype prompted analysis of the pka1 mutants under a variety of growth conditions. Pathogenic ity, nutrient starvation, carbon and nitrogen stress, apothecial development, ascospor e production, ascospore morphology, ascospore viability, growth on high salt, growth on 1M sorbitol, growth on various sole carbon sources, temperature shift prior to scleroti al development, oxalate production, cultural growth and growth on buffered PDA medium we re examined in the mutant strains and found to be equivalent to wild type (dat a not shown). See the Materials and Methods section for specific media composition and parame ters used to screen the mutant strains. Analysis of PKA En zyme Activity Crude protein extracts from all strains were subjected to non-radioactive PKA enzyme activity assays. These extracts we re incubated with a fluorescently-labeled kemptide substrate and the reaction mixture wa s separated on a 0.8% 10 mM Tris-acetate agarose gel pH 8.0. Fig. 2-6 shows the separation of phosphorylated and non-

PAGE 43

33 phosphorylated kemptide products. Purified PKA was used as a positive control (P) and PKA enzyme was omitted from the reaction as a negative control (N). Both controls gave expected results with the positive control s howing most of the kemptide product in the phosphorylated form and the negative control exhibiting kemptide in the nonphosphorylated form. When PKA inhibitor pe ptide (8 M PKI) was added to purified PKA (P+I) the majority of the product was detected in the non-phosphorylated form. Wild type protein extract from S. sclerotiorum was also incubated with 8 M PKI (Wt + I) as a control, and the bulk of the ke mptide product was detected in the nonphosphorylated form. Crude protei n extracts from all strains (Wt, D1, D2, D3, E1, and C1) showed approximately equal levels of phosphorylated kemptide product, indicating that PKA activity was not qualitatively differe nt from the wild type strain. Phylogenetic Analyses of pka Catalytic Subunit Genes from Various Fungi The annotations of Magnaporthe grisea and Neurospora crassa genome sequence databases enabled the analysis of all the annotated serine/th reonine kinases (kinome) in the M. grisea and N. crassa genomes. A phylogenetic tree of all 85 annotated serine/ threonine kinases from M. grisea was assembled using the neighbor joining method (data not shown). Based on phylogeneti c relationships and direct examination of the encoded amino acid sequences for highly conserved P KA residues, it was determined that only two cAMP-dependent protein kinase s exist in the entire annotated M. grisea kinome (MG06821( CPK2 ) and MG02382.1 ( CPKA )). The pka1 gene from S. sclerotiorum forms a clade with the M. grisea cpk2 gene. The M. grisea cpka gene affects appressorial function (Xu and Hamer 1997), and is pres ent in a different clade from the pka1 & cpk2 genes . A cladogram of all the serine/threonine ki nases was also constructed in the same fashion from the annotated N. crassa genome database (data not shown). Only two pka

PAGE 44

34 catalytic subunit genes were identified in the N. crassa serine/threonine kinome. The S. sclerotiorum pka1 gene was located on a branch with an uncharacterized pka gene from N. crassa (data not shown). The N. crassa pka gene (accession AF264760) that affects growth polarity and development was located on a separate clade than the one containing pka1 (data not shown) . A detailed phylogenetic analysis of the amino acid sequences for eight fungalencoded PKA catalytic subunits was conducted by parsimony analysis. The resulting tree was subjected to bootstrap analysis. Two pka catalytic subunit genes, one from H. sapiens and one from D. melanogaster, were chosen as outgroups in the analysis. Two major clades of fungal PKAs were generate d by the analysis and are shown (Fig. 2-7) One clade contains Ct-PKAC from C. trifolii, accession AF264760 from N. crassa , CPKA from M. grisea, and adr1 from U. maydis . This clade is supported by a bootstrap value of 100. A second major clade, also supporte d by a bootstrap value of 100, contains pka catalytic subunit genes from N. crassa (Locus ID #NCU06821.1), M. grisea cpk2 , and S. sclerotiorum pka1 . The PKA catalytic subunit, Uka1 from U. maydis , cannot be unequivocally assigned to either of the two major clades.

PAGE 45

35 Figure 2-1.Multiple sequence alignment of amino acid residues encompassing the Pka catalytic core from seven different organisms using ClustalX. GeneDoc was used to shade the amino acid residues in the following manner: black = 100%, gray = 80% and white = 60% identit y. Roman numerals denote each sub domain of the highly conserved kinase catalytic core. A filled black circle indicates the amino acids involved in binding the regulatory subunit and a black square marks the threonine resi due involved in the autophosphorylation reaction. The quatrapeptide PKI binding site is labeled with a solid black bar. Each amino acid subunit is annotated w ith a two-letter abbreviation for the genus and species name followed by the corresponding gene name except for N. crassa genes, which are labeled with either a Genbank accession number or a locus identification number. Hs = Homo sapiens, Dm = Drosophila melanogaster , Ct = Colleotorichum trifolii , Nc = Neurospora crassa , Mg = Magnaporthe grisea , Um = Ustilago maydis , and Ss = Sclerotinia sclerotiorum. H s P K X 1 : D m D C 2 : C t P K A : N c A F 2 6 4 7 6 0 : M g C P K A : U m A d r 1 : U m U k a 1 : S s P K A 1 : M g C P K 2 : N c N C U 0 6 8 2 1 : F D T L A T V G T G T F G R V H L V K E K T A K H F F A L K V M S I P D V I R L K Q E Q H V H N E K S V L K Y Q I I K T V G T G T F G R V C L C R D R I S E K Y C A M K I L A M T E V I R L K Q I E H V K N E R N I L R F D I L R T L G T G S F G R V H L V Q S K H N Q R F Y A V K V L K K A Q V V K M K Q V E H T N D E R R M L G F E I L R T L G T G S F G R V H L V Q S R H N S R F Y A V K V L K K A Q V V K M K Q V E H T N D E R R M L A F E I L R T L G T G S F G R V H L V Q S R H N Q R F Y A V K V L K K A Q V V K M K Q V E H T N D E R K M L G F A V E R T L G T G S F G R V H L V R S R H N H R F Y A I K V L R K E Q V V K M K Q V E H T N S E R A I L S F E V V E T L G T G T F G R V L L V R L K D R D V A D R S A Y F A L K V L A K T D V I K L K Q V S H I N S E R C I L T F E L V R T L G T G T F A R V W L A R L A N P A E E D R D K V F A L K V L R K V E V I K L K Q V D H V N H E R S V L A F K K V R T L G T G T F A R V C L V R P S N P Q N E T E R N K V F A L K I L R K S E V V K L K Q I D H V R H E R A I L A F H R I R T L G T G T F A R V V L V R P A N G T E I D R Q K V Y A L K I L R K T E V I R L K Q I D H V R H E R Q I L Q H s P K X 1 : D m D C 2 : C t P K A : N c A F 2 6 4 7 6 0 : M g C P K A : U m A d r 1 : U m U k a 1 : S s P K A 1 : M g C P K 2 : N c N C U 0 6 8 2 1 : E V S H P F L I R L F W T W H D E R F L Y M L M E Y V P G G E L F S Y L R N R G R F S S T T G L F Y S A E I I C A I E E I R H P F V I S L E W S T K D D S N L Y M I F D Y V C G G E L F T Y L R N A G K F T S Q T S N F Y A A E I V S A L E E V K H P F L I T L W G T F Q D S K N L Y M V M D F V E G G E L F S L L R K S G R F P N P V A K F Y A A E V T L A L E E V K H P F L I T L W G T F Q D A K N L Y M V M D F V E G G E L F S L L R K S G R F P N P V A K F Y A A E V T L A L E E V K N P F L I T L W G T F Q D C R N L Y M V M D F V E G G E L F S L L R K S G R F P N P V A K F Y A A E V T L A L E I V R H P F L V N L W G T F K D S T F L Y M V M D Y V P G G E L F T L L R K S Q R F P H P V A K F Y A A E V A L A I D K V D H P F L V N M I A S F Q D K N C Y M L M E Y V V G G E I F S Y L R R A G H F S A D A R F Y I S T I V L A I E D V A G H P F I T T L I T S F A D H D S L Y M L L D Y C P G G E V F S Y L R K A K R F D E N T A R F Y A A E I V L I L E D V S G F P F I T N M L A S F S D H D F L Y I V L D Y V P G G E L F S Y L R K Y R R F D E D M A R F Y A A E I V L V L E D V T G H P F I T S L Q A S F S D H D F L Y L L L D Y I P G G E L F T Y L R K Y R R F D E E M A R F Y A A E I V L V L E H s P K X 1 : D m D C 2 : C t P K A : N c A F 2 6 4 7 6 0 : M g C P K A : U m A d r 1 : U m U k a 1 : S s P K A 1 : M g C P K 2 : N c N C U 0 6 8 2 1 : Y L H S K E I V Y R D L K P E N I L L D R D G H I K L T D F G F A K K L V D R T W T L C G T P E Y L H S L Q I V Y R D L K P E N L L I N R D G H L K I T D F G F A K K L R D R T W T L C G T P E Y L H S R D I I Y R D L K P E N L L L D R H G H L K I T D F G F A K R V P D K T W T L C G T P D Y L H S R D I I Y R D L K P E N L L L D R H G H L K I T D F G F A K R V P D K T W T L C G T P D Y L H A K N I I Y R D L K P E N L L L D R H G H L K I T D F G F A K R V P D K T W T L C G T P D Y L H Q N N I I Y R D L K P E N I L L S A D G H L K I T D F G F A K Y V P D V T W T L C G T P D Y L H N K V V Y R D L K P E N L L I D S N G Y T K I T D F G F A K E V E D R T W T L C G T P E F L H E R E G V A Y R D M K P E N L L L D A E G H I K L V D F G F A K R L G N R E T Y T L C G T P E Y L H E A Q D G V A Y R D L K P E N L L L D G Q G H I K L V D F G F A K R L G G R R D G D N S G T Q E T Y T L C G T P E Y L H E E Q G G I A Y R D M K P E N L L L D A D G H I K L V D F G F A K R L G Y N D V E R P V E T Y T L C G T P E H s P K X 1 : D m D C 2 : C t P K A : N c A F 2 6 4 7 6 0 : M g C P K A : U m A d r 1 : U m U k a 1 : S s P K A 1 : M g C P K 2 : N c N C U 0 6 8 2 1 : Y L A P E V I Q S K G H G R A V D W W A L G I L I F E M L S G F P P F F D D N P F G I Y Q K I L A G K I D F P R Y I A P E I I Q S K G H N K A V D W W A L G V L I Y E M L V G Y P P F Y D E Q P F G I Y E K I L S G K I E W E R Y L A P E V V S N K G Y N K S V D W W S L G I L I Y E M L C G Y T P F W D S G S P L K I Y E N I L K G K V K Y P A Y L A P E V V S N K G Y N K S V D W W S L G I L I Y E M L C G Y T P F W D G S S P M K I Y E N I L K G K V R Y P Q Y L A P E V V S N K G Y N K S V D W W S L G I L I Y E M L C G Y P P F W D S G S P M K I Y E N I L K G K V R Y P A Y L A P E I V S S K G Y N K S V D W W A L G V L L Y E M L A G H P P F F T E D G N P I K L Y E K I I A C K V R Y P P Y L A P E I I Q C S G H G A V D W W S L G I L L F E M P G Y P P F Y D P N P I L I Y E K I L A G N L V F P E Y L A P E V I Q S K G H T T A V D W W A L G I L I Y E F L T G Y P P F W H S N P I E I Y K Q I V T K P V S F P A E P Y L A P E V I H N K G H T T A V D W W A L G I L I Y E F L T G Y P P F W H Q N P I E I Y K Q I V E K P V V F P Q D P Y L A P E V I Q N K G H T T A V D W W A L G I L I Y E F L T G Y P P R K S T V T K T Q V R I V E K P V L F P S S T H s P K X 1 : D m D C 2 : C t P K A : N c A F 2 6 4 7 6 0 : M g C P K A : U m A d r 1 : U m U k a 1 : S s P K A 1 : M g C P K 2 : N c N C U 0 6 8 2 1 : H L D F H V K D L I K K L L V V D R T R R L G N M K N G A N D V K H H R W F H M D P I A K D L I K K L L V N D R T K R L G N M K N G A D D V K R H R W F Y I N P D A Q D L L S K L I T A D L T K R L G N L Y G G P N D V K T H P W F W V N P D A Q D L L E R L I T A D L S K R L G N L Y G G P Q D V K S H P W F Y I N P D A Q D L L Q R L I T A D L T K R L G N L Y G G S Q D V R N H P W F Y F E T G V K D L L K N L L T A D L S K R Y G N L H R G S K D I F G H L W F E I D P L S R D L I S S L L T A D R R R L G N L R G G A N D V K N H P W F A I S S A A K D I I R Q F C T V D R S H R L G N I S G G A A R V K D H P F F P I S P N A Q D I I R Q F C T V D R S R R L G N I S G G A A R V K E H P F F E I S E E A K D I I R S F C T V D R T M R L G N M S G G A A R V K A H P W F I II • III IV V VIa VIb VII • • VIII IX X XI

PAGE 46

36 Figure 2-2.Diagram of the pka1 locus illustrating the double recombination event necessary for replacement of wild type with the disrupted pka1 locus. A hygromycin gene cassette ( hph ) was inserted between sub domains III and IV of the kinase catalytic domain. The hph cassette is in the opposite orientation with respect to the pka1 gene. Locations of sequences used for hybridization probes (1 and 2) are indicated with a thin solid line on the disruption locus diagram. 1 kb hph III IV Hind III Hind III p PKA1 hph pk a1 pka1 replacement locus hph III IV Hind III Hind III hph pk a1 Hind III Hind III pka1 locus Msc I III IV pka1 1 2

PAGE 47

37 Figure 2-3. Genomic DNA blot analysis of Wt, disruptant, ectopic and complemented strains. The blot (upper panel) was hybridized with the 5’ region of the pka1 gene as a probe (probe sequence 1) and shows a single 3.2 kb band in wild type (Wt) and a 5.4 kb band in disruptant strains 2 and 3 (D2, D3). Disruptant strain 1 (D1) is a putative heterokary on due to the presence of a 5.4 kb band and a weakly hybridizing 3.2 kb band. Ge nomic DNA from ectopic (E1) and complemented (C1) strains hybridized to 3.2 and 5.4 kb bands as expected. The blot (lower panel) was hybridized using the hph gene cassette as a probe (probe sequence 2), which detected a single 5.4 kb band in all strains except wild type. Wt D1 D2 D3 E1 C1 5.4 kb 3.2 kb 5.4 kb pka 1 hph

PAGE 48

38 Figure 2-4. RNA blot analysis of Wt, disruptant, ectopic, and complemented strains. The RNA blot was hybridized with the 5’ pka1 probe showing a single band of ~1.6 kb in Wt (wt).The disruptant st rains (D1, D2, and D3) showed only a single band that was distinctly smaller in size than Wt, whereas ectopic (E1) and complemented (C1) strains both s how a wild type and a smaller-sized disruptant band. RNA blot analysis wa s carried out three times, each time using different fungal tissue as th e source of RNA for each experiment. Results from the three experiments were very similar to the one shown in this figure. Ethidium bromide stained 28S rRNA is shown to assess loading and overall quality of the RNA. Wt D1 D2 D3 E1 C1 1.6 kb 1.0 kb 28S rRNA pka 1

PAGE 49

39 Figure 2-5. Growth of wild type, disruptant, ectopic, and complemented strains on PDA and PDA supplemented with 10 mM cAMP. None of the strains produced sclerotia in the presence of 10 mM cA MP. The experiment was executed three times with three replications per strai n. The results of all experiments with their respective replic ations looked very similar to the figure shown below. Wt = wild type, D1 = disruptant 1, D2 = di sruptant 2, D3 = disruptant 3, E1 = ectopic 1, and C1 = complemented 1. Wt D1 D2 D3 E1 C1 PDA PDA + 10mM cAMP

PAGE 50

40 Figure 2-6. PKA enzyme activity assay using crude protei n extracts from Wt, disruptant, ectopic and complemented strains . The figure shows the separation of phosphorylated and non-phosphorylated kemp tide products on a 0.8% Tris pH 8.0 gel. Purified PKA from bovine hear t (P) was used as a positive control which phosphorylated most of the kemptide product and was completely inhibited by the addition of 8 M PKI (P +I). Crude extracts from all strains showed relatively equal amounts of phosphorylated and non-phosphorylated product. Phosphorylation of kemptide usi ng crude Wt extract was completely inhibited with the addition of 8 M PKI (Wt + I). This experiment was repeated a minimum of three times, each time using extracts derived from different liquid shake cultures. The resu lts from all three experiments look very similar to the figure shown. N = ne gative control (enzyme), Wt = wild type, D1 = disruptant 1, D2 = disruptan t 2, D3 = disruptant 3, E1 = ectopic 1, C1 = complemented 1. Wt+I Wt D1 D2 D3 E1 C1 P N P+I P

PAGE 51

41 Figure 2-7. A phylogenetic tree representing the amino acid residues contained in the highly conserved serine/thr eonine protein kinase ca talytic core from seven different organisms. The tree was co nstructed using parsimony, and bootstrap values were indicated at each branch. Br anches were labeled with a two-letter abbreviation for the genus and species in italics followed by the corresponding gene name except for N. crassa genes where either the accession and or locus ID number are indicated. Hs = Homo sapiens, Dm = Drosophila melanogaster , Ct = Colleotorichum trifolii , Nc = Neurospora crassa , Mg = Magnaporthe grisea , Um = Ustilago maydis , and Ss = Sclerotinia sclerotiorum. Hs PKX1 Ct-PPKAC Dm DC2 Nc #AF264670 Mg CPKA Um A dr1 Um Uka1 Ss PKA1 M g CPK 2 100 66 77 100 100 85 Nc #NCU06821.1 Clade 1 Clade 2

PAGE 52

42 Discussion Investigating the Biological Role of pka1 in S. sclerotiorum Previous pharmacological studies (R ollins and Dickman 1998) suggested a primary role for cAMP in scle rotial initiation and oxalate production. Based on this work, I hypothesized that pka1 loss-of-function mutants would be aberrant in sclerotial biogenesis and oxalate accumulation. Several pka1 disruption strains were created and analyzed in culture for defects in sclero tial development and oxalate production. All strains grew and developed sclerotia indistingu ishably from wild type and control strains. The pka1 mutants were further characterized by cu lturing them in the presence of 10 mM cAMP. All pka1 disruptant strains reta ined wild type responsiveness to cAMP as evidenced by their inability to produce scleroti a. Additionally, oxalate levels in all strains were found to be near wild type. Other phenotypic consequences of pka1 disruptants were sought through the evaluation of growth and development unde r various conditions. These evaluations included apothecial development, ascos pore production and morphology, and numerous physiological and nutrient limiting conditions. No discernable phenotype could be attributed to the disruption of the pka1 locus in the mutant strains. These results suggested that a second pka catalytic subunit gene or a cAMP-responsive protein other than PKA functions in S. sclerotiorum . Recent investigations in various eukaryotes have revealed that proteins other than PKA bind and are affected by cAMP. Some examples of cAMP-binding signaling proteins include: Msn2 and Msn4 transcripti on factors and Rim15 pr otein kinase from yeast (Thevelein and de Winde 1999), cA MP-regulated GEF’s (guanine nucleotide exchange factor) in mammalian cells (Kawasaki et al. 1998), cAMP-binding HERG K+

PAGE 53

43 channels (Cui et al. 2000), and a cAMP-binding G-pr otein coupled receptor in Dictyostelium discoideum (Klein et al. 1988). Evidence from pharmacological data (Rollins and Dickman 1998) and the lack of a pka1 disruptant phenotype reported here suggest that other cAMP-responsive proteins may play a role in cAMP-mediated signal transduction in S. sclerotiorum. Despite the possibility of other cAMP-responsive fact ors, many investigations involving cAMP-mediated signal transduction in filamentous fungi have reinforced the hypothesis that PKA is the main target of cAMP. In plant pathogenic fungi, mutational analyses of pka catalytic subunit genes have been carri ed out in a number of biotrophic or hemibiotrophic host parasites e.g. M. grisea, C. trifolii and U. maydis (Mitchell and Dean 1995; Xu and Hamer 1997; Durrenberger et al. 1998). These studies have revealed that PKA functions in various developmental pro cesses that are important for infectious growth and pathogenicity. Our observations that S. sclerotiorum pka1 mutants are not altered in pathogenicity or other relate d phenotypes is consistent with data obtained from the functional analysis of orthologous genes in hemi-biotrophic/biotrophic phytopathogenic fungi. In M. grisea, loss of function cpka mutants produced appressoria that were fully melanized but were smaller and developed mo re slowly than wild type (Xu and Hamer 1997). These appressoria failed to penetrate intact leaf ti ssue resulting in a loss of pathogenicity on healthy host plants (Mit chell and Dean, 1995; Xu and Hamer 1997). A second pka catalytic subunit ge ne orthologous to the S. sclerotiorum pka1 was isolated from M. grisea, CPK2 (J.-R. Xu, personal communication). Deletion of CPK2 failed to manifest an obvious phenotype (Jin-Rong X u, personal communication); however, the

PAGE 54

44 inability to create a cpka /cpk2 double mutant suggests that CPK2 functions in the absence of CPKA ( J.-R. Xu, personal communication). Mutations in the Ct-PKAC gene from the hemi-biotroph C. trifolii caused small reductions in mycelial growth, altered conidiation patterns and germ ination, and delayed appressorium development. These mutants were only able to cause disease on mechanically wounded alfalfa plants (Yang and Dickman 1999). In U. maydis , the biotrophic maize smut pathogen, two genes encoding pka catalytic subunit genes have been cloned, and functionally characterized. Uka1 and Adr1 encode PKA catalytic subunits , but appear to function differently in cAMP-mediated signaling events. Durrenberger et al. (1998) have shown that adr1 mutants lead to constitutive filamentous growth, were non-pathogenic and accounted for the majority of PKA activity in U. maydis . Conversely, uka1 deletion mutants were no different from wild type in that cell morphology, and pathogenicity were unaffected. Unlike the double PKA mutants in M. grisea , adr1 / uka1 mutants were viable, suggesting the existence of a third pka catalytic subunit gene or functional redundancy in U. maydis. I am interested in determining if a postulated pka1 paralog ( pka2 ) from S. sclerotiorum has a conserved function in host penetrati on, despite the differences in penetration structures and pathogenic lif estyle from other fungi. Phylogenetic Evidence for Two pka Catalytic Subunit Genes in S. sclerotiorum The recently sequenced and annotated genomes of M. grisea and N. crassa were investigated to determine presence and relationships of pka genes among the serine/threonine kinase genes . The serine/threonine kinomes from both fungi possess only two genes bearing all of the conserved hallmarks of a pka catalytic subunit gene. Phylogenetic relationships established among PKAs from filamentous fungi presented here distinguished two clades, co mprising two distinct classes of pka genes. These clades

PAGE 55

45 are well supported by bootstrap values of 100 for each branch. The upper clade is composed of pka catalytic subunit genes that when mutated to nonfunctional alleles give rise to an obvious phenotype. The lowe r clade is composed of a cluster of pka genes that when mutated have no discernable phenotype, pka1 belongs to this clade. I hypothesize, based on the evidence presen ted in this work, that a paralog of pka1 is the major contributor of PKA activity in S. sclerotiorum . Cloning and characterization of a pka1 paralog are currently underway. I predic t, based on previous pharmacological evidence (Rollins and Dickman 1998) that loss of function mutants in this gene may result in defects in oxalate accumulati on/production, may not produce sclerotia or produce sclerotia that do not mature, and ma y be non-pathogenic due to aberrations in infection cushion formation and occurrence of altered oxalate levels. These investigations are predicted to shed new insights into the biological role of PKA-mediated signaling events in the context of a necrot rophic plant-parasite interaction.

PAGE 56

46 CHAPTER 3 ADENYLATE CYCLASE DELETION MUTANTS IN Sclerotinia sclerotiorum DISPLAYED ALTERED MYCELIAL BR ANCHING, PRODUCED ABERRANT SCLEROTIA AND WERE NON-PATHOGENIC Introduction The fungus Sclerotinia sclerotiorum (Lib.) de Bary is a necrotrophic filamentous ascomycete plant pathogen that has been documented to cause disease on at least 408 plant species from 278 genera in 75 families (Boland and Hall 1994). Most of these hosts are dicots, but a number of agriculturally important crop hosts are monocots. The biochemical basis for the broad host range of S. sclerotiorum is postulated to result from: the production of copious amounts of oxalic acid a nd the secretion of an array of cell wall degrading enzymes (CWDE) (i.e. pectinas es, proteases, cellulases, etc). Oxalate accumulates early during infection and is im portant for lowering the pH of infected tissues to allow optimal function of CWDEs. Previous physiological investigations have demonstrated and confirmed the synerg istic role of oxalate and CWDEs in S. sclerotiorum pathogenicity (de Bary 1887; Marciano et al. 1983; Maxwell and Lumsden 1970). S. sclerotiorum does not produce mitotic spores, but is vegetatively dispersed by multihyphal, melanized, long-term resting stru ctures termed sclerotia. Sclerotia are capable of resisting physical, chemical and microbial degradation and can remain viable for up to 8 years in the soil (Adams and Ayers 1979; Chet and Henis 1975; Willets and Wong 1980; Willets and Bullock 1992). Town send (1957) described three stages of

PAGE 57

47 sclerotial development: 1) in itiation, 2) development, and 3) maturation. The mature sclerotium is composed of three distinct la yers; the rind, cortex a nd medulla (Mordue and Holliday 1979). Melanin present in high concentr ation in the cell walls of outer rind cells is responsible for the dark coloration of sclerotia (Jones 1970). Sclerotia are capable of myceliogenic germination, and under appropria te physiological conditions (temperature, light, moisture, etc.), can undergo carpogenic germination to produce apothecia. An individual sclerotium can give rise to multip le apothecia that can each produce millions of ascospores. Forcibly discharged ascospores have been shown to be the primary source of inoculum in most Sclerotinia diseases a nd are critical for the maintenance and spread of inoculum in the field (Steadman, 1979). Since sclerotia play a crucial role in the spread and propagation of S. sclerotiorum , previous studies have focused on dete rmining the physiological factors that regulate sclerotial development (Chet and Henis 1975; Le Tourneau 1979; Willets and Bullock 1992; and Willets and Wong 1980). Ho wever, few studies have been aimed at identifying the underlying molecular and bioc hemical mechanisms th at control sclerotial development. In 1998 Rollins and Dickman im plemented a pharmacological screen to identify signaling components that affected sclerotial development in Sclerotinia sclerotiorum . They found that high cAMP levels inhibited sclerotial initiation and concomitantly raised oxalate levels. The au thors’ concluded that a cAMP-dependent signaling pathway may be involved in regulati ng the transition from mycelial growth to sclerotial initiation. The primary source of cellular cAMP is adenylate cyclase (AC), which uses ATP to form cAMP and pyrophosphate. Cyclic AMP is degraded into 5’ adenosine

PAGE 58

48 monophosphate (5’ AMP) by cAMP-specific phosphodiesterases (PDEs). AC’s and PDE’s work together to modulate the interc ellular levels of this important second messenger. Mammalian AC isoforms are membrane localized containing two transmembrane domains and two catalytic domains (CycC) separated by a domain of unknown function (DUF) (Tang a nd Gilman 1992). However, fungal adenylate cyclases only share the catalytic doma in in common with mammalia n AC isoforms and guanylate cyclases and are thought to be membrane associated via the amino-terminus. Fungal adenylate cyclases contain the following fi ve domains: an N-terminally located Rasassociation domain, followed by multiple le ucine rich repeats, a PP2C phosphatase domain, a catalytic domain and 2 cyclic AMPassociated protein bi nding sites (Klimpel et al . 2002). Recent molecular investigations invo lving adenylate cyclase mutants in phytopathogenic filamentous fungi have show n that cAMP and adenylate cyclase are important for proper vegetative growth, inf ection structure forma tion and pathogenicity. Choi and Dean (1997) showed that stra ins of the ascomycete plant pathogen, Magnaporthe grisea, deficient in adenylate cyclase ( MAC1 ) were unable to produce appressoria and did not penetrat e susceptible rice leaves. In the dimorphic basidiomycete plant pathogen Ustilago maydis , deletion of the adenylate cyclase gene, uac1, resulted in constitutive filamentous growth and were non-pathogenic (Gold et al . 1994). In the post harvest necrotroph, Botrytis cinerea , gene replacement of an adenylate cyclase gene ( BAC ) resulted in reduced vege tative growth, low levels of cAMP, slow lesion development, and lack of sporulation in planta .

PAGE 59

49 The major focus of this study was to dete rmine the biological affects of cAMP through analysis of adenylate cyclase deficient mutants in S. sclerotiorum . A targeted gene deletion approach was implemented and it was hypothesized based on previous pharmacological data (Rollins and Dickman 1998) that adenylate cyclase loss-of-function mutants may exhibit aberrations in sclerotial development and or morphogenesis, and that oxalic acid levels may be reduced compared to wild type. It was also hypothesized that adenylate cyclase–deficient mutants may be af fected in pathogenici ty and or virulence due to the loss of infection st ructure formation based on results from AC mutants in other filamentous fungal plant pathogens. Results from this study will provide new insights into the role that cAMP plays in the bi ology of a broad host ra nge necrotrophic plant pathogen like S. sclerotiorum . Materials and Methods Fungal Strains, Growth Conditions and Media Wild type (Wt) S. sclerotiorum isolate 1980 (Godoy et al . 1990) was used to derive all strains examined in this stu dy. Cultures were routinely grown on potato dextrose agar (PDA) (Difco, Detroit, MI, U.S.A.) and transferred in mass hyphal form. Transformants were cultured on PDA supplemented with either 100 g/ml hygromycin B (EMD Biosciences, USA) or 10 g/ml Bialaphos (PhytoT echnology Laboratories, Shawnee Mission, Kansas) . Permanent stocks were maintained as desiccated myceliacolonized filter paper and sclerotia at -20 C. Liquid shake cultures containing 50 ml of YPSuc medium (4g/L yeast extract [Difco] + 15g/L sucrose + 1g/L K2HPO4 + 0.5g/L MgSO4 pH 6.5) and were cultured as previously described (Rollins, 2001). Growth and analysis of cultures on cAMP-amended PDA was executed as reported by Rollins and

PAGE 60

50 Dickman (1998). Apothecial induction wa s carried out using PDA culture–derived sclerotia (Russo and Van Etten 1982). Basic Procedures for Nucleic Acid Manipulation Mycelia from liquid shake cultures we re flash frozen in liquid nitrogen, lyophilized and stored at -80C. Lyophilized mycelia were used to isolate genomic DNA as described by Yelton et al . (1984). Total RNA was extrac ted from lyophilized mycelia using Trizol reagent (Gibco BRL, Rockv ille, MD) according to the manufacturer’s instructions. Agarose gel fractionation of total RNA was conducted as described by Rollins and Dickman (2001). E. coli strain JM109 was used to propagate a ll plasmid DNA. Plasmid isolations, agarose gel electrophoresis, DNA restrict ion digests, ligation reactions, and E. coli transformations were conducte d using standard procedures (Sambrook and Russell 2001). For Southern blot analyses, restriction en zyme-digested genomic DNA were transferred to MagnaGraph nylon membrane (Micron Separations Inc., Westborough, MA.) by downward alkaline transfer (Chomczynski 1992) and fixed to the membrane using UV light. For Northern blot analyses, tota l RNA was transferred to MagnaGraph nylon membrane by standard procedures (Ausubel et al . 1997). Radioactive probes for all hybridizati ons were generated using the Random Primers DNA Labeling System (Invitrogen Life Technologies, Carlsbad, CA.) as per the manufacturers’ instru ctions. A full length 2.1kb hph (hygromycin phosphotransferase) gene cassette (Redman and Rodriguez 1994) and a 500bp amplicon corresponding to the catalytic region (CycC) of the sac1 gene was used to probe both genomic DNA and RNA blots.

PAGE 61

51 Cloning and Identification of the sac1 Adenylate Cyclase Gene A partial adenylate cyclase genomic clone was obtained from the Dickman Lab at the University of Nebraska, Lincoln. This pa rtial genomic subclone was used as a probe to obtain a full-length genomic clone from a S. sclerotiorum pMOcosTEL genomic cosmid library. Approximately 12,000 colonies with an average insert size of 11kb were screened (~4x genome coverage) and two posi tive clones were identified. These clones had similar restriction patterns and one was chosen for further characterization. Sequence information from the full-length sac1 (Sclerotinia adenylate cyclase 1) genomic clone was obtained by primer walking. Both strands were sequenced, and data was compiled into a contig using the Sequencher softwa re (Gene Codes Corp, Ann Arbor, MI). Construction of sac1 Gene Replacement and Complementation vectors A pair of gene specific primers FL -AC-5’-II (5’ TTA TCG ACG GCT TAT TAG AAC GTA CG 3’) and AC-5 ’+AscI-R (5’ AGG CGC GCC GCT AAA AAC CGT CCA TCC 3’) were designed to amplify ~2.0kb of the 5’ untranslated region of the sac1 gene using standard PCR conditions and S. sclerotiorum isolate 1980 genomic DNA as a template. To facilitate cloning an Asc I restriction enzyme recognition site was attached to AC-5’+AscI-R (underlined ). The 2.0 kb amplicon was gel purified and cloned using the TOPO TA cloning kit (Invitrogen, Carlsb ad, California). This clone was designated AC-5’ and to determine the identity and orie ntation of insert, the clone was submitted for sequence analysis at the University of Fl orida ICBR sequencing core. A pair of gene specific primers AC-3’Fla nk-AscI (5’ AGG CGC GCC CCT GAA ACA GCA ATG CTT GAG) and AC-3’Flank-No AscI (5’ GGG CTG GTA AAT GGC GTA ATC 3’) were designed to amplify ~2.2 kb of 3’ UTR a nd ~0.5 kb of coding sequence corresponding to the 3’ end of the sac1 gene. One of the gene spec ific primers contained an Asc I

PAGE 62

52 restriction enzyme site (underl ined) to facilitate directiona l cloning of the fragment. The 2.7 kb 3’ UTR amplicon was generated using standard PCR conditions and S. sclerotiorum isolate 1980 DNA as a template. This product was cloned using the TOPO TA cloning Kit and designated AC-3’ and orie ntation and identity of the insert was determined by sequence analysis at the ICBR se quencing core, University of Florida. The hygromycin phosphotransferase ( hph ) cassette containing the TrpC promoter and terminator with flanking Asc I restriction enzyme recognition sites was obtained from Mr. Andrew R. Hutchens III. Details concerning its origin and synthesi s can be found in his thesis online at the Universi ty of Florida Library enti tled “Ambient pHand CarbonRegulated Gene Expression in the Necrotrophic Phytopathogen Sclerotinia sclerotiorum .” The sac1 gene deletion replacement v ector was constructed in the following manner. Both AC-5’ and AC-3’ were double digested with Not I/ Asc I and separated on a 0.8% agarose TBE gel. The linearized 6.0kb AC-5’ and the 2.7kb AC-3’ insert were gel purified and ligated together to form a 8.7kb vector designated AC-5’+3’. The hph cassette and the AC-5’-3’ vectors were digested with Asc I and separated on a 0.8% agarose gel. The linearized 8.7kb vector was gel purified and ligated to the 2.2kb Asc I digested hph cassette which gave rise to a 10.9 kb sac1 gene deletion vector. This vector was linearized with Not I, gel purified and used to transform fungal protoplasts. The sac1 knock-out vector was used as a template to produce two sac1 hph hybrid PCR products referred to as split marker fragments (Catlett et al. 2004). One primer set consisting of Split Marker-A C-5’ (5’ ATC CAG GGA CCT CGA ACG GCA TTT G 3’) annealed to the 5’ UTR of sac1 and Split Marker-HY (5’ AAA TTG CCG TCA ACC AAG CTC TGA TAG 3’) a nnealed internal to the hph gene cassette. The

PAGE 63

53 cDNA polymerase and primer set (Split Mark er-AC-5’+Split Marker-HY) was used according to the manufacturers instructions (B D Biosciences, Palo Alto, California), to amplify a 3.2kb amplicon that was designated AC -5’-HY. A pair of gene specific primers was designed to amplify the 3’ portion of the AC UTR using Split Marker-AC-3’ (5’ TGA CCT ACT TGC CGT CTT TCA GTG C 3’) and an internal hph cassette primer Split Marker-YG (5’ TTT CAG CTT CGA TGT AGG AGG GCG 3’). Using cDNA polymerase and primer set (Split Marker-AC-3’+Split Marker-YG) a 4.3kb amplicon was generated and designated AC-5’-YG. These two split marker fragments, AC-5’-HY and AC-3’-YG were gel purified, quantified via spectrophotometric analysis and 5 g of each were used to transform wild type fungal protoplasts as mentioned in the previous Materials and Methods section. To complement the sac1 deletion mutant, the pBAR KS1 vector containing a bar gene cassette (Pall and Brunell i 1994) was digested with Not I restriction enzyme (New England Biolabs; USA) and tr eated with shrimp alkaline phosphatase according to the manufactures instructions (Promega, Madis on, Wisconsin). A pair of gene specific primers AC-FL-5’-1.5kb UTR (5’ TAC CC T GTG CTC TAA ATT TGG ATC ACC 3’) and AC-FL-3’-1.0kb UTR (5’ AAT CCA AGC CAT CCA ACC TAT CTA ACC 3’) were designed to both 5’ and 3’ flanking genomic regions of the sac1 gene and used to amplify 100 nanograms of wild type 1980 S. sclerotiorum genomic DNA. PCR was carried out using a cDNA polymerase accordin g to the manufactures instructions (BD Biosciences, Palo Alto, California) and yielded a 9.5 kb product which contained the sac1 coding sequence and ~1.5kb of 5’ and ~ 1kb of 3’ untranslated region. The 9.5 kb sac1 amplicon was cloned into the pGEM-T easy ve ctor (Promega, USA) and digested with

PAGE 64

54 Not I restriction enzyme. The Not I digested sac1 amplicon was ligated into the 4.5kb Not I-linearized pBARKS1 vector which ga ve rise to a 14kb plasmid designated sac1 pBARKS1. Transformation of Fungal Protoplasts a nd Evaluation of Transformant Strains Fungal protoplasts of the wild type S. sclerotiorum 1980 strain were prepared and transformed with the sac1 gene replacement vector as described by Rollins (2003). The sac1 deletion strain was complemented with sac1 -pBARKS1 plasmid and transformants were selected on RM medium containing 10 g/ml Bialophos (PhytoTechnology Laboratories, Shawnee Mission, Kansas). Southern blot hybridization was used to screen integration events in all transformant strains. This was acco mplished by digesting genomic DNA from hygromycin-resistant strains with Stu I and hybridizing it with a 500 bp probe which was made from the catalytic region of the sac1 gene. A 2.1 kb fragment corresponding to the hygromycin cassette was also used to probe genomic DNA blots. Two distinct PCR reactions were used to further evaluate tr ansformant strains. The first PCR reaction implemented Taq polymerase (New England Bi olabs, USA) and primer set 1 containing gene specific primers AC-CycC-5’-II (5’ TTT TCA CTG ATA AGA GC 3’) and ACCycC-3’-II (ATA GGA CCA AAG TAG TCC) that flank the catalytic domain of the adenylate cyclase sac1 gene resulting in a ~450 bp amplicon. The second PCR reaction utilized cDNA polymerase (BD Biosciences, USA) and primer set 2 consisting of two gene specific primers AC-5’XTRA-UT R-F (5’ AAG ACT CCA TCT GAT ACA TGC AGA CTC 3’) AC-3’XTRA-R (AAA TTC CA T AAC AGT GCC TTT TTG GG 3’). The PCR reaction yielded two different sized produc ts depending on the inte gration event that occurred in a given strain.

PAGE 65

55 Radial Growth Analysis, Determination of Oxalate, and Pathogenicity Assays Radial growth analysis, and oxalate accu mulation kinetics were carried out as described by Rollins (2001). Pathogenicity assays were conducted according to Rollins (2003) except that the percentage of leaf area colonized was quan tified 7 days after inoculation using Spot Adva nced Software program (Diagnostic Instruments, USA). Infection Cushion Assay Glass cover slips (FisherBrand Microsc ope Cover Glass, Pittsburg, PA) of 22x22 mm size were cut in half using a diamond knife a nd a straight edge as a guide. They were then immersed into 95% ethanol using forcep s and flamed with a Bunsen burner until the alcohol evaporated. After a cooling, four c over slips were placed around the periphery of each 9 cm potato dextrose agar plate (PDA). PDA plates were inoculated in the center with a 2.5 mm3 plugs cut using a cork borer w ith either wild type (Wt) or sac1 deletion (KO1) strains. Mycelial-coloni zed cover slips were removed from the PDA plates after ~7 days and directly placed onto glass s lides (FisherBrand Microscope Slides-Plain, Pittsburg, PA) for examination using a com pound microscope (Leica model DM R HC, Germany). Pictures were take n with a digital camera (Diagno stic Instruments Inc-Model 3.2.0, USA) attached to the microscope us ing the Spot basic software program (Diagnostic Instruments, USA). Extraction and Determinatio n of Total cAMP Levels Lyophilized mycelium (15 mg) was ground to a fine powder in a 1.5 ml eppendorf tube using a metal spatula. Twenty l of lysis reagent 1A and 180 ul of cAMP assay buffer were added and the samples, mixe d by inversion and allowed to incubate at room temperature for 10 minutes. All samp les were spun in a microcentrifuge for 5 minutes at 14 K rpm. The supernatant was re moved and diluted with cAMP assay buffer

PAGE 66

56 and used directly in the ELISA-based BIOTRAK cellular communication assay (GE Healthcare, USA.) as per the manufacturers’ instructions. Results Cloning and Characterization of sac1 Analysis of the nucleotide and amino aci d sequences using BLAST (Basic Local Alignment Search Tool) (Altschul et al. 1997) revealed that sac1 was most similar to the BAC adenylate cyclase-encoding gene from B. cinerea and to other adenylate cyclase genes from filamentous fungi. Mu ltiple sequence alignment of sac1 with other filamentous fungal adenylate cy clases was used to determin e the putative start and stop codons and locations of intron/exon junctions. The sac1 gene was determined to contain 4 exons interrupted by 3 introns. The predicte d joined open reading frame yielded a 6.4kb sequence that encodes 2157 am ino acids (Fig. 3-1A). The deduced polypeptide sequence contains five domains typical of filament ous fungi: an N-terminal Ras-association domain, multiple leucine-rich repeats, a PP2C phosphatase, a catal ytic domain, and two cAMP-associated protein binding sites (Fig. 31B). Southern blot analysis at low and high strigency using 500 bp of the AC catal ytic region as a probe suggested that sac1 is a single copy gene (Fig. 3-1C). Mi ning sequence data from the S. sclerotiorum Genome Project ( Sclerotinia sclerotiorum Sequencing Project. Broad Institute of Harvard and MIT. http://www.broad.mit.edu ) indicated that adenylate cyclase sac1 is singly represented. Deletion of the sac1 Locus The strategy to delete the sac1 locus with the sac1 hph gene replacement construct described in the materials a nd methods is shown in Fig. 3-2A. This Not Ilinearized construct was used to transform wild type Ss1980 protoplasts and ~50 strains

PAGE 67

57 were isolated and further analyzed. Southern blot analysis of one wild type (Wt), three deletion (AC-KO 1, 2, & 3), a complemented (C1), an ectopic (E1) and one noncomplemented control (Nc1) strain is s hown in Fig 3-2B. Hybridization using the catalytic region as a probe yi elded a single 4.8 kb band for Wt and E1 strains and did not hybridize to the KO1 or Nc1 strains as expe cted. DNA from all stra ins was probed with the full length hph gene as a hybridization control and as expected, all strains except wild type possessed a 6.4 kb band corresponding to the hph gene replacement construct. Primer set (1) flank the catalytic region of adenylate cyclase, was used for PCR to detect an intact sac1 gene by production of a 500bp amplicon. Fig. 3-3A shows that the 500bp amplicon was only present in wild type, ectopic and complemented strains, and was absent in all three KO strains and th e non-complemented control. A second primer set (2), were designed to analyze site-specific integration of the sac1 replacement locus and yielded a 4.5kb product as shown in figure 3-3A. Northern hybridization was used to de termine if AC deletion mutants possessed the wild type sac1 transcript. Fig. 3-3B shows an R NA blot after hybrid ization with the AC-CycC probe. Wt, E1 and C1 strains contained the 6.4kb sac1 transcript, while the KO1 and Nc1 strains lacked the sac1 transcript. Visualizatio n of ethidium bromidestained rRNA confirmed that RNA from each sample was equally loaded and of high quality. Evaluation of Total Cellular cAMP Levels Total cellular cAMP was extracted and assayed from three biological replicates corresponding to Wt and KO1 strain. The grap h in Fig. 3-4 shows that the KO1 strain produced ~ 25% of Wt cellular cAMP levels.

PAGE 68

58 Adenylate Cyclase Deletion Mutants Exhibit Many Morphological Defects Figs. 3-5A and 3-5B illustrate a variet y of morphological abnormalities observed in the KO1 strain. When grown on potato dext rose agar plates (PDA), KO1 produced an abundance of aerial hyphae. Sclerotial developm ent occurred in concentric rings with clearing zones between each ring of sclerotia. Sclerotia were aberrant in size and shape but contain a dark melanized rind like wild type. KO1 sclerotia were capable of myceliogenic germination, however did not pr oduce apothecia under standard conditions. Close scrutiny of the hyphal branching pattern in PDAgrown expanding colonies revealed that hyphal branches are short and arise from the main hypha at 90 angles instead of acute angle branching found in w ild type strains. Branching pattern and cultural growth was restored in the presen ce of 10 mM exogenous cAMP (Fig. 3-6A). However, the KO1 strain remained cAMP-re sponsive as sclerotial development was inhibited. Addition of exogenous 10mM cGMP was unable to complement the growth habit of the KO1 and the Nc1 strains (Fig.3-6B ). In addition to the aberrant concentric pattern of sclerotial development, KO1 exhibited a significantly reduced growth rate relative to Wt. The rate of ra dial expansion of the KO1 stra in was ~8X below that of Wt (Fig. 3-7) sac1 Deletion Mutants are Non-Pathogenic Inoculation of detached tomato leafle ts revealed that th e KO1 strain was nonpathogenic (Fig 3-8). All strains including wild type, ectopic, and complemented were fully pathogenic and had completely colonized the tomato leaflets within 5 days of inoculation. In very few instances, a brown spot of dead cells was obs erved directly under the KO1 inoculation plug, however, no internal colonization of the leaves was observed when the area was examined microscopically using the fungal-specific stain trypan blue

PAGE 69

59 (not shown). Wounding of the tomato leaves caused some lesion formation as shown in fig. 3-9. I sought to investigate the bioche mical source of the pathogenicity defect observed in the KO mutant, and therefore assayed oxalic acid accumulation kinetics. Interestingly, the KO1 produced as much oxa late as the Wt and control strains and accumulation kinetics for all strains were vi rtually indistinguishable (Fig. 3-10A). The production of infection cushions was investig ated in Wt and KO1 strains (Fig. 3-10B). Wt typically produces complex, slightly me lanized multihyphal inf ection cushions in excess on hard surfaces. When infection cushions development was examined on microscope cover slips, the Wt develope d slightly melanized, multihyphal infection cushions. However, the KO1 strain was comp letely devoid of either simple and or complex types of infection cushions.

PAGE 70

60 A. B. C Figure 3-1 . A .Genomic organization of the sac1 locus illustrating the location and relative size of exons and introns and putative stop and start codons. B . Cartoon of Sac1 polypeptide illustrating the five functional domains possessed by fungal adenylate cyclases. (CAP = cyclase associated protein) C. Southern Blot hybridization under low (55C) a nd high stringency (65C) conditions of wild type (Wt) Sclerotinia sclerotiorum DNA digested with three different restriction enzymes Bam H I (B), Hind III (H), and Spe I (S) and hybridized with probe sequence 1 corresponding to the adenylate cy clase catalytic domain. 3 1 2 4 N C RAS-assoc domain Leucine rich repeat domain Phosphatase domain Catalytic domain CAP-Binding Domains Low Stringency High Stringency B H S 12kb 9kb 7kb B H S ATG TAA 3’ UTR 5’ UTR

PAGE 71

61 A. B. Figure 3-2. Construction and analysis of sac1 gene replacement strains. A. sac1 locus and gene replacement construct contai ning the hygromycin phosphotransferase gene ( hph ) cassette in place of the sac1 gene sequence. The double cross-over homologous recombination event re sulting in the replacement of sac1 coding sequence with the hph sequence is shown. Location of sequences and primers used for hybridization in this study are shown. B. Southern hybridization analysis of sac1 gene replacement strains. Genomic DNAs from wild type (Wt), sac1 deletion (KO1), ectopic (E1), complemented (C1), and noncomplemented control (Nc1) st rains were digested with Stu I. The blot on the left was hybridized with probe #1 corresponding to a portion of the AC catalytic region, and the blot on the ri ght was hybridized with probe #2 which corresponds to the hph gene cassette. hph sac1 locus sac1replacement locus Probe 1 1 hph Probe 2 sac1 Stu I 1kb Wt 123E1 C 1 Nc 1 Wt 1 2 3 E1 C1 Nc1 6.4 kb 4.8 kb KO KO 2 2Stu I sac 1 hph

PAGE 72

62 A. B. Figure 3-3 . PCR and Northern Blot analysis of sac1 deletion and control strains. A. PCR analysis of wild type (Wt), sac1 deletion (KO1), ectopic (E1), complemented (C1), and non-complemented (Nc1) st rains using primer set (1) which amplifies a portion of the AC catalyt ic domain and primer set (2) which amplifies the 5’ portion of the AC gene. L = 1kb ladder B. Northern blot analysis of Wt, KO1, E1, C1, and Nc1 st rains. A 6.4kb transcript is detectable in all strains except KO1 and Nc1 as expected. Ethidium bromide-stained rRNA is shown to ensure quality a nd equal loading of RNA samples. Northern blot analysis was performe d using RNA from three biological replicates, and a representative blot is shown. Wt 1 2 3 E1 C1 Nc1 L 1 2 3 Nc1 500bp 4.5kb Wt KO 1 E 1 C 1 Nc 1 28S rRNA 6.4kb KO KO

PAGE 73

63 Figure 3-4 . Analysis of total cellular cAMP levels in wild type (Wt) and sac1 deletion (KO1). Three biological replicates of each strain were analyzed in duplicate and the mean values from all three replicates are represented. Total Cellular cAMP0 100 200 300 400 500 600 700 800 900 1000fMol cAMP wt KO

PAGE 74

64 A. B. Figure 3-5. A. Cultural morphology of wild type (Wt), sac1 deletion (KO11), ectopic (E1), complemented (C1), and non-complemented strains (Nc1) grown on potato dextrose agar for 14 days. B. Branching pattern of Wt, KO1, and C1 strains viewed at 600X total magnifi cation using a dissecting microscope. Wt KO1 C1 Wt KO1 E1 C1 Nc1

PAGE 75

65 A. B. Figure 3-6. A. Growth of wild type (Wt), sac1 deletion (KO1), ectopic (E1), complemented (C1), and non-complemented (Nc1) strains on 10mM cAMPamended potato dextrose agar for 7 days. B. Growth of Wt, KO1, E1, C1, and Nc1 strains on 10mM cGMP-amended potat o dextrose agar for 14 days. Both of these experiments were repeated 3 tim es with 3 replicatio ns per strain. The results from one representa tive experiment is shown. Wt KO1 E1 C1 Nc1 Wt KO1 E1 C1 Nc1 +10mM cAMP +10mM cGMP

PAGE 76

66 Figure 3-7 . Radial growth analysis of wild type (Wt), sac1 deletion, (KO1), ectopic (E1), complemented (C1), and non-complemented (Nc1) strains. Colony diameters were measured at 12 hour intervals. Each point represents the mean and standard deviation from three indepe ndent cultures. This experiment was repeated three times, each with three independent cultures. Results from all three experiments were similar and one representative experiment is shown. Radial Growth Analysis 0 1 2 3 4 5 6 7 8 9 10 0122436486072 Time (h)Radial Growth (cm) Wt KO E C Nc

PAGE 77

67 Figure 3-8. Pathogenicity assay with wild t ype and transformant strains on detached tomato leaflets. Tomato leaflets (cv. B onnie Best) were mock inoculated with an uncolonized potato dextrose agar (P DA) plug (M), inoculated with a PDA plug colonized with wild type (Wt), sac1 deletion (KO1), ectopic (E1), complemented (C1), and non-complemented control strain (Nc1). Photographs were taken 7 days pos t inoculation. One representative replication from four experiments is shown. M Wt KO1 E1 C1 Nc1

PAGE 78

68 Figure 3-9. Pathogenicity assay of wild type and transformant strains on wounded detached tomato leaflets. Tomato le aflets (c.v. Bonnie Best) were wounded by making a single hole with a dissecting needle and inoculation plugs were placed directly over the wound. Tomato leaves were inoculated with an uncolonized potato dextro se agar plug (M), wild type (Wt), and sac1 deletion strain (KO1). Percentage of area colonized was measured 7 days post inoculation and these values appear under each leaf. One representative replication from four experiments is shown. M Wt KO1 0 100 9.2

PAGE 79

69 A. Oxalate Accumulation Kinetics Exp #30 0.5 1 1.5 2 2.5 0102030 Time (hours)mg/ml oxalic acid wt ko e1 c1 nc1 B. Figure 3-10. A . Oxalic acid accumulation kinetics from 0.5M MOPS-buffered YPSU cultures pH 7.0 of wild type (Wt), sac1 deletion (KO1), ectopic (E1), complemented (C1), and non-complemented (Nc1) strains. Data points represent the mean from three independent cultures B . Infection cushion formation in the Wt and KO1 strains (400X magnification). Wt KO1

PAGE 80

70 Discussion Molecular Cloning and Characterization of sac1 in S. sclerotiorum The predicted Sac1 amino acid sequence c ontains five functional domains typical of fungal adenylate cyclases: an N-terminal Ra s association domain, multiple leucine-rich repeats, a PP2C class phosphatase, a catalyt ic domain and two C-terminally located cAMP-associated binding protein domains. The Ras-association domain has been shown to be involved in activati on of adenylate cyclase via the g-alpha protein Ras in mammalian systems (Bahn and Sundstrom 2001). However, no regulatory role for Ras on adenylate cyclase has been identified t hus far in filamentous fungi (Fillinger et al . 2002). Leucine rich repeats are known to be involve d in protein-protein interactions and may facilitate dimerization of ACs and or serve as interaction points for other proteins. The PP2C phosphatase domain is found in all fungal adenylate cyclases and may mediate the Ras-GTP activation of AC activity. Neverthe less, the PP2C domain has not been shown to possess phosphatase activity in yeast (Barford et al . 1998). The catalytic domain of adenylate cyclases is shared in common with mammalian ACs and guanylate cyclases having the consensus sequen ce (V/IKTXGDA) (Roelofs et al . 2001a). In yeast, the 70 kDa CAP protein (cyclic AMP-associat ed protein), is required for proper in vivo response of adenylate cyclase to Ras (Shima et al . 1997). To date, functi onal analysis of CAP proteins and or their associated sequences have not been carried out in filamentous fungi. Multiple Morphological Aberrations Exist in sac 1 Deletion Mutants A number of physiological and functionalgenetic studies have revealed that cAMP controls a variety of biological pro cesses in filamentous fungal plant pathogens (Choi and Dean 1997; Hall and Gurr 2000; Klimpel et al . 2002; Lee and Dean 1993; Robson et al . 1991). Based on data from the functiona l analysis of AC -deficient fungal

PAGE 81

71 phytopathogens and findings and a previ ous pharmacological investigation in S. sclerotiorum by Rollins and Dickman (1998), I hypothesized that sac1 deletion mutants may exhibit multiple morphological defects, produce aberrant sclerotia and would be non-pathogenic and or reduced in virulence. In this study I sought to determine the role(s) of cAMP in the biology of th e wide host range necrotroph S. sclerotiorum . This was accomplished by cloning the single copy adenylate cyclase-encoding gene, sac1 , and deleting it from the genome to create a lossof-function mutant. Nucleotide analysis via BLAST (Altschul 1997) revealed that sac1 encodes an adenylate cyclase which closely resembles the B. cinerea adenylate cyclase BAC gene. Three independent sac1 mutants were generated and all displayed identical morphological phenotypes as a representative deletion strain (KO1) grew ~eight times slower than wild type (Fig. 3-7), possesse d dense aerial hyphae, a nd produced aberrant sclerotia in concentric rings in culture. Dense, compact cultural phenotypes have been described for other adenylate cyclas e mutants like the crisp strain ( cr-1 ) in N. crassa (Lindegren, 1936), MAC1 in M. grisea (Choi and Dean 1997) and the BAC mutant in B. cinerea (Klimpel et al . 2002). Interestingl y, the abnormal appearance of sclerotia produced by sac1 knock out strains has not been docum ented in other sclerotial-forming fungi. Another vegetative defect observe d was the predominance of 90-branched hyphae. Wild type S. sclerotiorum produces acute (60 and 45) hyphal branching patterns in culture. The wild type branching pattern could not be complemented with 10 mM cGMP, but could be restored with the addition of 10 mM cAMP and by complementing the AC mutant with the sac1 gene (Fig. 3-5 and 3-6). This branching

PAGE 82

72 defect also appears to be limited to sac1 mutants as it has not been documented for other fungal AC mutants. cAMP is Detectable in sac1 KO mutants Deletion of the entire sac1 gene, except for 400 bp encoding one cAMPassociated protein binding s ite at the 3’ end, from S. sclerotiorum was unable to completely abolish cAMP levels in sac1 deletion strains. There was a ~4 fold reduction in cAMP levels in the KO1 strain compared to wild type (Fig 3-4). Examination of cAMP levels for AC mutants N. crassa cr-1 and the Aspergillus nidulans cya strains, reported “no detectable” amounts of cA MP were present (Fillinger et al. 2002 and Terenzi et al. 1976). However, detectable levels of cAMP in an AC-deficient strain were reported for the BAC mutant (Klimpel et al. 2002). BAC mutants showed ~5 fold reduction in cAMP compared to wild type strains. Since cAMP has been shown to be involved in numerous biological processes, it is l ogical to assume that this important second messenger might be required for viability. Since all of the AC-deficient mutant s characterized to date have not been lethal, a basal level of cAMP in these mutant strains might be expected. The only known source of cAMP in the cell is ad enylate cyclase. However, it is well known that the catalytic domain of adenylate cycl ase and guanylate cyclase are very similar (Roelofs et al . 2001a). Therefore, I believe that it is possible that guany late cyclase is capable of producing the low basal level of cAMP that was observed in the adenylate cyclase-deficient sac1 mutant. The Role of Adenylate Cyclase and cAMP in Pathogencity and Virulence Pathogenicity assays involving detached tomato leaves revealed that sac1 mutants were non-pathogenic (Figs. 3-8). However, they were able to partially colonize mechanically wounded tomato leaflets (F ig. 3-9). Therefore I hypothesized that sac1

PAGE 83

73 mutants may possess undetectable or reduced levels of oxalate th at would render the sac1 deletion mutants non-pathogenic (Godoy et al. 1990). Previous pharmacological data also supported this hypothesis as treatment with exogenous cAMP resulted in increased oxalate levels compared to wild type (R ollins and Dickman, 1998). However, production and accumulation of oxalate was no different in the KO1 compared to wild type and control strains (Fig. 3-10A). The lack of reduced oxalate levels may be due to the presence of low cAMP levels in the KO1 muta nt. It is possible that the KO is producing a threshold level of cAMP which facilitates wild type levels of oxalate to accumulate. In contrast, it is possible that the pharmacological observation of increased oxalate levels in conjunction with high cAMP levels may not be directly responsible for oxalate accumulation. Pathogenicity was restored, but vi rulence remained severely attenuated in wounded tomato leaflets. Reduction in pathogenicity and virulence in AC-deficient mutants has been shown in B. cinerea (Klimel et al . 2000), Ustilago maydis (Gold 1994) and M. grisea (Choi and Dean 1997). Defects in infection struct ure formation were documented in MAC1 mutants as they did not produce appressoria a nd were non-pathogenic on susceptible rice cultivars. However, appressorium formati on could be restored with application of exogenous cAMP and mutant strains were able to colonize wounded rice leaves. I sought to determine if the defect in pathogenicity was due to a lack of infection cushion (appressorium) formation in sac1 deletion strains. Interesti ngly, I was unable to visualize infection cushions on glass cover slips col onized by KO1 strain. Therefore, the absence of infection cushion formation in the KO1 stra in coupled with the greatly reduced growth rate is responsible for th e loss of pathogenicity.

PAGE 84

74 Reduction in pathogenicity and virulence in AC-deficient mutants has been shown in B. cinerea (Klimpel et al . 2000), U. maydis (Gold 1994) and M. grisea (Choi and Dean 1997). Defects in infection structur e formation were documented in MAC1 mutants as they did not produce appressoria and were non-pathogenic on susceptible rice cultivars. However, appressorium formation could be restored with application of exogenous cAMP and mutant strains were able to colonize wounded rice leaves. I sought to determine if the defect in pathogenicity was due to a lack of infection cushion (appressorium) formation in sac1 deletion strains. Interesti ngly, I was unable to visualize infection cushions on glass cover slips col onized by KO1 strain. Therefore, the absence of infection cushion formation in the KO1 st rain appears to play the primary role in eliminating pathogenicity in intact plants. Pathogenicity is restor ed by wounding the host but the reduced growth rate of the KO1 mu tants appears to be responsible for the attenuated virulence.

PAGE 85

75 APPENDIX IDENTIFICATION OF GENES INVOLVE D IN cAMP SIGNALING FROM THE DRAFT GENOME SEQUENCE OF Sclerotinia sclerotiorum (Lib.) de Bary Genome sequence data from S. sclerotiorum has facilitated rapid identification and annotation of genes involved in cAMP-m ediated signal transduction. Therefore, I have chosen to construct a table containing the name, Locus ID number, protein class, PFAM domain(s) and pBLAST hit informati on for each gene. I have also included a detailed annotation of the nucleotide and amino acid sequence that corresponds to the pka2 catalytic subunit gene. Gene Name Locus ID # Protein Class PFAM Domain(s) Highest pBLAST Hit e-value sac1 SS1G_07715 Adenylate Cyclase Catalytic, LRR, Ras-Assoc, Phosphatase, (2) CAP-Binding Adenylate Cyclase ( B. fuckeliana) 0 spde1 SS1G_12594 cAMP PDEase I PDEase Type I Hypothetical Protein (M. grisea) 0 spde2 SS1G_07412 cAMP PDEase II PDEase Type II Hypothetical Protein (G. zeae) 1.00E-107 scap1 SS1G_13327 CAP CAP Protein AC-Associated Protein (A. fumigatus) 1.00E-106 sgpa1 SS1G_07597 G-protein Alpha G-Alpha Subunit G-Protein Alpha Subunit (S. sclerotiorum) 2.00E-180 sgpa2 SS1G_10286 G-protein Alpha G-Alpha Subunit G-Protein Alpha Subunit (B. fuckeliana) 0 sgpa3 SS1G_12343 G-protein Alpha G-Alpha subunit G-Protein Alpha Subunit (B. fuckeliana) 0 sgpb1 SS1G_03482 G-protein Beta WD40 G-Protein Alpha Subunit (B. fuckeliana) 0 sgpg1 SS1G_12567 G-protein Gamma N/A G-Protein Gamma Subunit (N. crassa) 2.00E-24 scbp1 SS1G_13049 Hypothetical (2) CBS, Patatin-Like Phospholipase Hypothetical Protein (M. grisea) 0 scbp2 SS1G_05838 Hypothetical (2) CBS, (7) LRR Hypothetical Protein (A. nidulans) 0 scbp3 SS1G_01159 Hypothetical Sulfate Transporte r, CBS Hypothetical Protein (G. zeae) 0 pka1 SS1G_03171 PKA Catalytic S/T Kinase PKA catalytic subunit (S. sclerotiorum) 0 spka2 SS1G_13577 PKA Catalytic S/T Ki nase PKA Catalytic Subunit ( B. graminis) 0 spkar1 SS1G_10536 PKA Regulatory (2) CBS PKA Regulatory Subunit ( B. graminis) 1.00E-150 AC – adenylate cyclase cAMP – 3’, 5’ cyclic adenosine monophosphate CAP – cyclase-associated protein CBS – cAMP-binding site G-protein – guanine nucleotide-binding protein LRR – Leucine rich repeat N/A – not available PDEase – phosphodiesterase PKA – cAMP-dependent protein kinase A S/T – serine/threonine

PAGE 86

76 pka2 nucleotide sequence ATG CCGACACTCGGATTTTTGAAGAAGAAACGGAC GAAGGAGTCTCCGGGTATAACCTCTC CGACATCTAGTCAAGCCACAAG TCCGATAACCCCCACATCATC AAGAGGCTTCGATAACTCC ATAGCTTCCATATCTACCCAGTCTACTTTACC CCAATCCCACGAGCACGAGGTTCTTCAAACT ACAACCTCTCGCACAGCAGACCAGCCTGC TGCTGTGGCCAAAAAAGCCT CAATGACACAAG GGATAACTCAACAGCAGCAAT CTGCGCAAATTCAT TCCGACCAACATCAGCATAGGCAAAAC TCCCCAAGTATCAGCAACTTGAT TCATCCACCACAGCATGACGGCGCGAGCCAGTCATATCC AACCCCACCAACCTCCTCGCAAAC TTCGCTGCATCCTGTAGCTGGTAACACCAGCACACTCC AAATGGATGCCTCGCACCGACAT TCCCAGCAAGCACAGCAGCAGCCACAGATCAGACAAAC CAAGGGCAAATATTCATTGACCGACTTCGAAAT CCAGCGCACATTAGGTACTGGAAGTTTTG GGCGAGTTCATTTAGTTCGATCGAAGCACAAC CAACGATACTATGCTGTCAAGGTATTGAAG AAGGCTCAGGTGCACAAGATGAAGCAGGTCGAACATACCAATGATGAGAGGAGAATGTTGC AAGAAGTCAAACACCCATTCCTCATCACTCTATGGGGTACATTCCAAGATTCAAAGAATTTAT ACATGGTTATGGATTTTGTAGAGGGAGGTGAGCTATTCTCTTTACTGAGAAAATCACAAGTATG TTCAATGGTACCAATTACTCGATATATACTAATGGTTCGTTTCAGAGATTCCCTAACCCAGTAGCCAAATTT TATGCCGCTGAAGTTACATTGGCCCTTGAATAT CTCCATTCAAAACAAATCATCTATCGGGAT CTAAAGCCAGAGAATTTACTTCTTGACAGGCAT GGACATCTTAAGATTACTGATTTTGGATTC GCCAAGAAGGTCCCAGATATCACATGGACATT GTGCGGGACCCCGGATTACTTAGCACCAG AAGTAGTCTCAAGTAAAGGTTATAATAAGTCTGTAGATTGGTAAGTATTGGTACCTTCATTCCCCTGCC ATCACTGCCCCTCCAAAGCGAAGCG AAAATGCTTTCGCGATAGT GACCCAGCGGAGCTTCGTCAATTAAACCTTCCT CTCTTCATACTTTCTCTTATGGGGG CTTGCCCCCATACCCGTCCTCGCT ACGCTCGAATGGAAACTTAGAGGTTGTTT ACTAACATTTCGAATCAGGTGGTCGCTTGGGATA CTGATTTTCGAAATGCTT TGTGGTTTCACGCC ATTTTGGGATGGTGGCTCACCTATGAAGA TTTATGAGAATATCCTTAAAGGGAGGGTCAAAT ATCCTCCATATATTCACCCAGATGCACAAGA TTTATTACAACGATTGATCACAGCAGATCTTA CCAAACGTCTTGGCAACTTA CATGGAGGCTCTGAGGACG TCAAAAATCATCAATGGTTTGCC GAAGTAACTTGGGAAAGGTTATCCAAGAAAGATA TTGATGCGCCTTATG TCCCTCCTGTCAA GGCCGGAGTCGGTGATGCTAGTCAGTTTGAT AAATATCCCGAGGAAACGGAGATATATGGT CAAGGTGGGCCTGATGAGTAAGTAGATTTCTTGAACTTCAT CAAATCTAGGCTAATATTATTAGATTTGGGC ACCTATTTGTCGATTTT TAA Exon 1 = 801bp Intron 1 = 50bp Exon 2 = 254bp Intron 2 = 202bp Exon 3 = 375bp Intron 3 = 47bp Exon 4 = 28bp Introns appear in a small font and puta tive start and stop c odons are in bold.

PAGE 87

77 Pka2 amino acid sequence MPTLGFLKKKRTKESPGITSPTSSQATSPITPTSSRGFDNSIASISTQST LPQSHEHEVLQTTTSRTADQPAAVAKKASMTQGITQQQQSAQIHSDQHQH RQNSPSISNLIHPPQHDGASQSYPTPPTSSQTSLHPVAGNTSTLQMDASH RHSQQAQQQPQIRQTKGKYSLTDFEIQRTLGTGSFGRVHLVRSKHNQRYY AVKVLKKAQVHKMKQVEHTNDERRMLQEVKHPFLITLWGTFQDSKNLYMV MDFVEGGELFSLLRKSQRFPNPVAKFYAAEVTLALEYLHSKQIIYRDLKP ENLLLDRHGHLKITDFGFAKKVPDITWTLCGTPDYLAPEVVSSKGYNKSV DWWSLGILIFEMLCGFTPFWDGGSPMKIYENILKGRVKYPPYIHPDAQDL LQRLITADLTKRLGNLHGGSEDVKNHQWFAEVTWERLSKKDIDAPYVPPV KAGVGDASQFDKYPEETEIYGQGGPDEFGHLFVDF Roman numerals designate the 12 highly conser ved domains of serine/threonine protein kinases. Filled circles represent amino aci d residues involved in regulatory subunit binding and a filled square represents the threonine residue involved in autophosphorylation. The solid filled bar de notes the 5 amino acid residues known to bind the PKI inhibitor peptide. I II III IV V VIa V Ib VII VIII IX X XI • • •

PAGE 88

78 Phylogenetic analysis of Pka catalytic subunits was accomplished using PAUP and statistical analysis was ca rried out using Bootstrap. Clade 1 Clade 2

PAGE 89

79 LIST OF REFERENCES Abawi, G.S., and Grogan, R.G., 1979. Sclerotinia sclerotiorum : epidemiology of diseases caused by Sclerotinia species. Phytopathology. 69:899-904. Adams, P.B., and Ayers, W.A., 1979. Sclerotinia sclerotiorum : ecology of Sclerotinia species. Phytopathology. 69:896-899. Altschul, S.F., Madden, T.L., Schaffer, A. A., Zhang, J., Zhang, Z., Miller, W., and Lipman, D.J. 1997. Gapped BLAST and PSI-B LAST: a new generation of protein database search programs. Nucleic Acids Res. 25:3389-3402. Ausubel, F.M., Brent, R., Kingston, R.E., M oore, D.D., Seidman, J.G., Smith, J.A., and Struhl, K. 1991. Current Protocols in Mo lecular Biology. John Wiley and Sons, New York. Ayers, W.A. and Adams, P.B. 1981. Mycopara sitism and its applic ation to biological control of plant disease. In Biological Control. in Crop Production Beltsville Symposium in Agricultural Research. 5:91-103. Bahn, Y.S., and Sundstrom, P. 2001. CAP1, an adenylate cyclase-associated protein gene, regulates bud-hypha transi tions, filamentous growth, a nd cyclic AMP levels and is required for virulence in Candida albicans . J. Bacteriol. 183:3211-3223. Bahn, Y.S., Staab, J., and Sundstrom, P. 2003. Increased high-affinity phosphodiesterase PDE2 gene expression in germ tubes counteracts CAP1 -dependent synthesis of cyclic AMP, limits hypha production and promotes virulence in Candida albicans . Mol. Microbiol. 50:391-409. Bar-Sagi, D. and Hall, A. 2000. Ras and Rho GTPases: A family reunion. Cell. 103:227238. Bardin, S.D., and Huang, H.C. 2001. Research on biology and cont rol of Sclerotinia diseases in Canada. Can. J. Plant Pathol. 23:88-98. Barford, D., Das, A.K., and Engloff, M.P. 1998. The structure and mechanism of protein phosphatases: insights into catalysis and regulation. A nnu. Rev. Biophys. Biomol. Str. 27:133-164.

PAGE 90

80 Bateman, D.F., and Beer, S.V., 1964. Simultane ous production of synergistic action of oxalic acid and polygalacturona se during pathogenesis by Sclerotium rolfsii . Phytopathology. 55:204-211. Beavo, J.A., and Brunton, L.L. 2002. Cyclic nucleotide research – still expanding after half a century. Nature Re v. Mol. Cell Biol. 3:710-718. Beckman. 1951. Synthesis of oxalate from s odium formate and sodium hydroxide. U.S. Patent #2687433. Boland, M., and Hall, R. 1987. Evaluating s oybean cultivars fo r resistance to Sclerotinia sclerotiorum under field conditions. Plant Dis. 71:934-936. Boland, M., and Hall, R. 1994. Index of plant hosts of Sclerotinia sclerotiorum . Can. J. Plant Pathol. 16:93-108. Bradford, M.M. 1976. A rapid and sensitive me thod for the quantification of microgram quantities of proteins utilizing the principle of dye binding. Anal. Biochem . 72:248. Canaves, J.M. and Taylor, S.S. 2002. Classi fication and Phylogenetic Analysis of the cAMP-Dependent Regulatory Subunit Family. J. Mol. Evol. 54:17-29. Catlett, N.L., Lee, B.N., Yoder, O.C., and Turgeon, G. 2004. Split-Marker Recombination for Efficient Targeted Deleti on of Fungal Genes. Fungal Genet. Newsl. 50:9-11. Cessna, S.G., Sears, V.E., Dickman, M.B., and Low, P.S. 2000. Oxalic acid, a pathogenicity factor for Sclerotinia sclerotiorum , suppresses the oxidative burst of the host plant. Plant Cell. 12: 2191-2199. Chandrasekaran, K., and Jayaraman, J. 1978. Eff ect of cyclic AMP on the biogenesis of cytochrome oxidase in y east. FEBS Lett. 87:52-54. Chet, I., and Henis, Y. 1975. Sclerotial mo rphogenesis in fungi. Ann. Rev. Phytopathol. 13:169-192. Chin, K.V., Yang, W.L., Ravatan, R., Kita, T., Reitman, E., Vettori, D., Cvijic, M.E., Shin, M., and Iacono, L. 2002. Reinventing th e Wheel of cAMP: Novel Mechanisms of cAMP Signaling. Ann. N.Y. Acad. Sci 968:49-64. Choi, W. and Dean, R.A. 1997. The Adenylate Cyclase Gene MAC1 of Magnaporthe grisea Controls Appressorium Formation and Other Aspects of Growth and Development. Plant Cell. 9:1973-1983. Chomczynski, P. 1992. One-hour downward alka line capillary transf er for blotting of DNA and RNA. Anal. Biochem. 201:134-139.

PAGE 91

81 Cui, J., Melman, Y., Palma, E., Fishman, G.I., and McDonald, T.V. 2000. Cyclic AMP regulates the HERG K+ channel by du al pathways. Curr. Biol. 10:671-674. Cuomo, C., Dickman, M.B., Kohn, L.M., and Rollins, J.A. S. sclerotiorum Sequencing Project. 2005. Broad Institute of Harvard and MIT. 2/28/2006. Davis, M.R., Subbarao, K.V., Raid, R.N., a nd Kurtz, E.A. 1997. Compendium of Lettuce diseases. APS Press, Saint Paul, Minnesota. de Bary, A., 1887. Comparative Morphology an d Biology of the Fungi, Mycetozoa and Bacteria. Oxford, UK, Clarendon Press. Dickman, M.B., Mitra, A. 1992. Arabidopsis thaliana as a model for studying Sclerotinia sclerotiorum pathogenesis. Physiol. Mo l. Plant Pathol. 41:255-263. Dickman, M.B., and Yarden, O. 1999. Seri ne/Threonine Protein Kinases and Phosphatases in Filamentous Fungi . Fungal Genet. Biol. 26:99-117. Donaldson, P.A., Anderson, T., Lane, B.G., Davidson, A.L., and Simmons, D.H. 2001. Soybean plants expressing an active oligom eric oxalate oxidase from wheat gf-2.8 (germin) gene are resistant to the oxalate secreting pathogen Sclerotinia sclerotiorum . Physiol. Mol. Plan t Pathol. 59:297-307. D’Souza, C.A., Alspaugh, J.A., Yue, C., Hara shima, T., Cox, G.M., Perfect, J.R., and Heitman, J. 2001. Cyclic AMP-dependent protein kinase controls virulence of the fungal pathogen Cryptococcus neoformans . Mol. Cell Biol. 21:3179-3191. Durrenberger, F., Wong, K., and Kronstad, J.W. 1998. Identification of a cAMPdependent protein kinase catalytic subunit required for virulence and morphogenesis in Ustilago maydis . Proc. Natl. Acad. Sci. USA. 95:5684-5689. Dutton, M.V., and Evans, C.S. 1996. Oxal ate production by fungi: its role in pathogenicity and ecology in the soil e nvironment. Can. J. Microbiol. 42:881-895. Fillinger, S., Chaveroche, M.K., Shimizu, K., Keller, N., and d’ Enfert, C. 2002. cAMP and Ras signaling independently control s pore germination in the filamentous fungus Aspergillus nidulans . Mol. Microbiol. 44:1001-1016. Gilman, A. 1984. G-proteins and dual contro l of adenylate cyclase. Cell. 36:577-579. Godoy, G., Steadman, J.R., Dickman, M.B., a nd Dam, R. 1990. Use of mutants to demonstrate the role of oxalic acid in pathogenicity of Sclerotinia sclerotiorum on Phaseolus vulgaris . Physiol. Mol. Plant. Pathol. 37:179-191.

PAGE 92

82 Gold, M.H., Farrrand, R.J., Livioni, J.P., and Segel, I.H. 1974. Neurospora crassa glycogen phosphorylase interconve rsion and kinetic properties of the “active” form. Arch. Biochem. Biophys. 161:515-527. Gold, S.E., Brogdon, S.M., Mayorga, M.E., and Kronstad, J.W. 1997. The Ustilago maydis regulatory subunit of a cA MP-Dependent protein kinase is required for gall formation in maize. Plant Cell. 9:1585-1594. Grau, C.R., Radke, V.L. and Gillespie, F.L. 1982. Resistance of soybean cultivars to Sclerotinia sclerotiorum. Plant Dis. 66:506-508. Guimaraes,R.L. and Stotz, H.U. 2004. Oxalate production by Sclerotinia sclerotiorum deregulates guard cells during in fection. Plant Physiol. 136:3703-3711. Hall, A.A., and Gurr, S.J. 2000. Initiation and appressorial germ t ube differentiation and appressorial hooking: distinct morphologica l events regulated by cAMP signaling in Blumeria graminis f.sp. hordei . Physiol. Mol. Plant Pathol. 56:39-46. Hartman, G.L., Sinclair, J.B., and Rupe, J. C. 1999. Compendium of Soybean Diseases, Fourth Edition. APS Press, Saint Paul Minnesota. Hicks, J.K., Bahn, Y.S., and Heitman, J. 2005. Pde1 Phosphodiesterase modulates cAMP levels through a protein kinase A-me diated negative feedback loop in Cryptococcus neoformans . Eukar. Cell. 4:1971-1981. Hu, X., Bidney, D.L., Yalpani, N., Duvick, J. P., Crasta, O., Folkerts, O., and Lu, G. 2003. Overexpression of a gene encoding hydrogen peroxide-generating oxa late oxidase evokes defense responses in sunflower. Plant Physiol. 133:170-181. Huang, H.C. and Kozub, G.C. 1989. A simple method for production of apothecia from sclerotia of Sclerotinia sclerotiorum . Plant Prot. Bull. 31:333-345. Huang, H.C., and Kozub, G.C. 1993. Influen ce of inoculum production temperature on carpogenic germination of sclerotia of Sclerotinia sclerotiorum . Can. J. Microbiol. 39:548-550. Hutchens III, A.R. 2005. Ambient pH-and Carbon-Regulated Gene expression in the necrotrophic phytopathogen Sclerotinia sclerotiorum . Thesis. University of Florida. Inglis, G.D. and Boland, G.J. 1990. The microflo ra of bean and rapeseed petals and the influence of the microflora of bean petals on white mold. Can. J. Plant Pathol. 12:129134. Jones, D., 1970. Ultrastructure and composition of the cell walls of Sclerotinia sclerotiorum . Trans. Brit. Mycol. Soc. 54:351-360.

PAGE 93

83 Jung, W.H. Warn, P., Ragni, I., Popolo, C.D ., Nunn, M.P., and Stateva, L. 2005. Deletion of PDE2 , the gene encoding the high-affinity cAMP phosphodiesterase, results in changes in the cell wall and membrane in Candida albicans . Yeast. 22:285-294. Jurick II, W.M., Dickman, M.B., and Rollins, J.A. 2004. Characterization and Functional Analysis of a cAMP-dependent protei n kinase A catalytic subunit gene ( pka1 ) in Sclerotinia sclerotiorum . Physiol. Mol. Plant Pathol. 64:155-163. Kawasaki, H., Springett, G.M., Mochizuki, N., Toki, S., Nakaya, M., Matsuda, M., Housman, D.E., and Graybiel, A.N. 1998. A fa mily of cAMP-binding proteins that directly activate Rap1. Science. 282:2275-2279. Kesarwani, M., Azam, M., Natajaran, K ., Mehta, A., and Datta, A. 2000. Oxalate decarboxylase from Collybia velutipes. Molecular cloning and its overexpression to confer resistance to fungal infection in tr ansgenic tobacco and tomato. J. Biol. Chem. 275:7230-7238. Kim, H.S., and Diers, B.W. 2000. Inheritance of partial resistance to Sclerotinia stem rot in soybean. Crop Sci. 40:55-61. Klein, P.S., Sun, T.J., Saxe III, C.L., Kimmel , A.R., Johnson, R.L., and Devreotes, P.N. 1988. A chemoattractant receptor controls development in Dictyostelium discoideum . Science. 241:1467-1472. Klimpel, A., Gronover, C.S., Williamson, B., Stewart, J.A., and Tudzinski, B. 2002. The adenylate cyclase (BAC) in Botrytis cinerea is required for full pa thogenicity. Mol. Plant Pathol. 3:439-450. Knighton, D.R., Zheng, J., Ten Eyck, J.F., As hford, V.A., Xuong, N.H., Taylor, S.S., and Sowadski, T.M. 1991. Crystal structure of th e catalytic subunit of cyclic adenosine monophosphate dependent kina se. Science. 253:414-420. Kohn, L.M., 1979a. Sclerotinia sclerotiorum : delimitation of the economically important plant pathogenic Sclerotinia species. Phytopathology. 69:881-886. Kohn, L.M., 1979b. A monographic revision of the genus Sclerotinia. Mycotaxon. 10: 365-444. Kokalis, N. Porter, D.M., Rodriguez, R ., Smith, D.H., and Subrahmanyam, P. 1997. Compendium of Peanut Diseases, Second Ed ition. APS Press. Saint Paul, Minnesota. Kor-eda, S., Murayama, T., and Uno, I. 1991. Suppression of the cr-1 mutation in Neurospora crassa . Jpn. J. Genet. 66:77-83. Kosasih, B.D., Willetts, H.J., 1975. Ontogene tic and histochemical studies of the apothecium of Sclerotinia sclerotiorum . Ann. Bot. 39:185-191.

PAGE 94

84 Kucharek, T., 2003. Chemical control guide for diseases of vegetables, University of Florida, Institute of Food and Agricultural Sciences. Plant Protection Pointers. Report no. 6, rev. 16. Lee,Y.H. and Dean, R.A. 1993. cAMP regulated infection structure fo rmation in the plant pathogenic fungi Magnaporthe grisea . Plant Cell. 5:193-199. Le Tourneau, D. 1979. Morphology, cy tology, physiology, of Sclerotinia species in culture. Phytopathology. 69:887-890. Libert, M.A. 1837. PL. Crypt. Arduennae, Fasc. IV: 326. Lindegren, C.C. 1936. A si x-point map of the sex chromosome of Neurospora crassa . J.Genet. 32:243-256. Livingstone, D.M., Hampton, J.L., Phipps , P.M., and Grabau, E.A. 2005. Enhancing resistance to Sclerotinia minor in peanut by expressing a barley oxalate oxidase gene. Plant Physiol. 137:1354-1362. Lumsden, R.D., 1979. Sclerotinia sclerotiorum : histology and physiology of pathogenesis in plant diseases caused by Sclerotin ia species. Phytopathology. 69:890-896. Lumsden, R.D., and Dow, R. L. 1973. Histopathology of Sclerotinia sclerotiorum infection of bean. Phytopathology. 63:708-715. Maeda, T., Mochizuki, N., and Yamamoto, M. 1990. Adenylyl cyclase is dispensable for vegetative cell growth in the fission yeast Schizosaccharomyces pombe . Proc. Natl. Acad. Sci. 87:7814-7818. Maeda, T., Watanabe, Y., Kunitomo, H ., and Yamamoto, M. 1994. Cloning of the pka1 gene encoding the catalytic subunit of cAMP-Depende nt Protein Kinase in Schizosaccharomyces pombe . J. Biol. Chem. 269:9632-9637. Marciano, P., Di Lenna, P., and Magro, P ., 1983. Oxalic acid, cell wall-degrading enzymes and pH in pathogenesis and thei r significance in the virulence of two Sclerotinia sclerotiorum isolates on sunflower. Phys iol. Plant Pathol. 22:339-345. Matsumoto, K., Uno, I., Oshima, Y., and Ishikawa, T. 1982. Isolation and Characterization of Yeast Mutants Deficient in Adenylate Cyclase and cAMP-Dependent Protein Kinase. Proc. Na tl. Acad. Sci. 79:2355-2359. Maxwell, D.P., and Lumsden, R. D., 1970. Oxalic acid production by Sclerotinia sclerotiorum in infected bean and in culture. Phytopathology. 60:1395-1398. Maxwell, D.P., Williams, P.H., and Lu msden, R.D. 1972. Studies on the possible relationships of microbodies and mu tlivesicular bodies to oxalate,

PAGE 95

85 endopolygalacturonase, and cel lulase (Cx) production by Sclerotinia sclerotiorum . Can. J. Bot. 50:17443-1748. Mclean, D.M. 1959. Sclerotinia stalk rot of cabba ge and seed plants. Wash. State. Agric. Exp. Stn. Circ. 359:1-4. Mertz, L.M., and Rashtchian, A. 1994. Nucleotide imbalance and polymerase chain reaction: effects on DNA amplification and s ynthesis of high specific activity radio labeled DNA probes. Anal. Biochem. 221:160-165. Mitchell, T.K. and Dean, R.A. 1995. The cA MP-dependent protein kinase catalytic subunit is required for appressorium form ation and pathogenesis by the rice blast pathogen Magnaporthe grisea . Plant Cell. 7:1869-1879. Moore, W.D., 1949. Flooding as a means of destroying the sclerotia of Sclerotinia sclerotiorum . Phytopathology. 39:920-927. Mordue, J.E.M., and Holliday, P. 1976. Sclerotinia sclerotiorum (sclerotial state). CMI Descriptions of Pathogenic Fungi and B acteria. No. 513.Morgan, O.D. 1971. A study of four weed hosts of Sclerotinia species in alfalfa fields . Plant Dis. Rep. 55:1087-1089. Morral, R.A. 1977. A preliminary study of the influence of water potential on sclerotium germination in Sclerotinia sclerotiorum . Can. J. Bot. 55:8-11. Nelson, B.D., Helsm, T.C., Olson, M.A. 1991. Comparison of laboratory and field evaluations of resist ance in soybean to Sclerotinia sclerotiorum . Plant Dis. 75:662-665. Pall, M.L. 1981. Adenosine 3’, 5’-Phospha te in Fungi. Microbiol. Rev. 45:462-481. Pall, M.L., and Brunelli, J.P. 1993. A series of six compact fungal transformation vectors containing polylinkers with multiple unique restriction sites. Fungal Genet. Newsl. 40:5962. Pan, X., and Heitman, J. 1999. cAMP-Depe ndent Protein Kina se A Regulates Pseudohyphal Differentiation in Saccharomyces cereviseae . Mol. Cell Biol. 19:48744887. Pan, X., and Heitman, J. 2002. Protein Kinase A Operates a Molecular Switch that Governs Yeast Pseudohyphal grow th. Mol. Cell Biol. 22:3981-3993. Perkins, D.D. 1959. New markers and multiple point linkage data in Neurospora. Genetics. 44:1185-1202. Pohronezny, K., and Purdy, L.H. Sclerotinia diseases of vegetabl e and field crops in Florida. Plant Pathology Fact Sheet. PP-22:1-3.

PAGE 96

86 Rall, T.W., and Sutherland, E.W. 1958. Form ation of a Cyclic Ribonucleotide by Tissue Particles. J. Biol. Chem. 232:1065-1076. Redman, R.S., and Rodriguez, R.J. 1994. Factor s affecting the efficient transformation of Colletotrichum species. Exp. Mycology. 18:230-246. Robertson, L.S. and Fink, G.R. 1998. Three yeast A kinases have specific signaling functions in pseudohyphal growth. Proc. Natl. Acad. Sci. 95:13783-13787. Robson, G.D., Wiebe, M.G., and Trinci, A.P.J. 1991. Exogenous cAMP and cGMP modulate branching in Fusarium graminearum . J. Gen. Microbiol. 137:963-969. Roelofs, J., Meima, M., Schapp, P., and Van Haastert, P.J.M. 2001a. The Dictyostelium homologue of mammalian soluble adenylate cy clase encodes a guanylate cyclase. EMBO J. 20:4341-4348. Rollins, J.A., and Dickman, M.B. 1998. Increases in endogenous and exogenous cAMP levels inhibits sclerotial development in Sclerotinia sclerotiorum. Appl. Environ. Microb. 64:2539-2544. Rollins, J.A., and Dickman, M.B. 2001. pH signaling in Sclerotinia sclerotiorum : Identification of a pacC/RIM1 homologue. Appl. Environ. Microb. 67:75-81. Rollins, J.A. 2003. The Sclerotinia sclerotiorum pac1 gene is required for sclerotial development and virulence. Mol. Plant-Microb. Interact. 16:785-795. Russo, G.M., Dahlberg, K.R., and Van Etten, J.L. 1982. Identification of a developmentspecific protein in sclerotia of Sclerotinia sclerotiorum . Exper. Mycol. 6:259-267. Sambrook, J., and Russell, D.W. 2001. Molecu lar Cloning: A laborat ory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. Shima, F., Katoaka, Y., Yanagihara, C., Tama da, M., Okada, T., Kariya, K., and Kataoka, T. 1997. Effect of the association with ade nylyl cyclase-associat ed protein in the interaction of yeast adenyl yl cyclase with Ras protei n. Mol. Cell. Biol. 17:1057-1064. Sippel, C.J., Goewert, R.R., and Slachm an, F.N. 1979. The regulation of ubiquinone-6 biosynthesis by Saccharomyces cerevieae. Fed. Proc. 38:508. Soderling, S.H., and Beavo, J.A. 2000. Regula tion of cAMP and cG MP signaling: new phosphodiesterases and new functions. Curr. Opin. Cell Biol. 12:174-179. Sonneborn, A., Bockmuhl, D.P., Gerads, M., Ku rpanek, K., Sanglard, D., and Ernsy, J.F. 2000. Protein Kinase A encoded by TP K2 regulates dimorphism in Candida albicans . Mol. Microbiol. 35:386-396.

PAGE 97

87 Steadman, J. R. 1979. Control of plant di seases caused by Sclerotinia species . Phytopathology. 69:904-907. Rall, T.W., and Sutherland, E.W. 1958. Form ation of a Cyclic Adenine Ribonucleotide by Tissue Particles. J. Biol. Chem. 232:1077-1092. Tang, W.J. and Gilman, A.G. 1992. Ad enylyl Cyclases. Cell. 70:869-872. Taylor, S.S., Buechler, J.A., and Yonemot o, W. 1990. cAMP-depende nt protein kinase: Framework for a diverse family of regul atory enzymes. Annu. Rev. Biochem. 59:9711005. Tellez-Inon, M.T., and Torres, H.N. 1970. Interconvertable forms of glycogen phosphorylase in Neurospora crassa . Proc. Natl. Acad. Sci. 66:459-463. Terenzi, H.F., Flawia, M.M. and Torres, H.N. 1974. A Neurospora crassa Morphological Mutant Showing Reduced Adenylate Cyclas e Activity. Biochem. Biophys. Res. Comm. 58:990-996. Terenzi, H.F., Jorge, J.A., Roselino, J.E., and Migliorini, R.H. 1979. Adenylyl cyclase deficient cr-1 (Crisp) mutant of Neurospora crassa : cyclic AMP-dependent nutritional deficiencies. Arch. Micrbiol. 123:251-258. Thevelein, J.M., and de Winde, J.H. 1999. Nove l sensing mechanisms and targets for the cAMP-protein kinase A pathway in the yeast Saccharomyces cerevisiae . Mol. Microbiol. 33:904-919. Thompson, J. D., Gibson, T. J., Plewniak, F ., Jeanmougin, F. and D. G. Higgins. 1997. The ClustalX windows interface: flexible st rategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 24:4876-4882. Townsend, B.B. 1957. Nutritional factors in fluencing the producti on of sclerotia by certain fungi. Ann. Rev. Bot. 21:153-166. Turkington, T.K. and Morrall, R.A. 1993. Use of petal infestation to forecast Sclerotinia stem rot of canola: the influence of inoc ulum variation over the flowering period and canopy density. Phytopathology. 83:682-689. Uno, I., and Ishikawa, T. 1976. Effect of cyclic AMP on glycogen phosphorylation in Coprinus macrorhizus . Biochim. Biophys. Acta. 452:112-120. van der Platt, J.V., and van Solingen, P. 1974. Cyclic 3’,5’ adenosine monophosphate stimulates trehalose degradation in bakers yeast. Biochem. Biophys. Res. Commun.56:580-587.

PAGE 98

88 van Solingen, P., and van der Platt, J.B. 1975. Partial purification of the protein system controlling the breakdown of tr ehalose in bakers yeast. Biochem. Biophys. Res. Comm. 62:553-560. Vega, R.R., Corsini, D., and Le Tournea u, D. 1970. Nonvolatile organic acids produced by Sclerotinia sclerotiorum in synthetic liquid medi a. Mycologia. 62:332-338. Wallace. 1926. Synthesis of oxalic acid from carbon monoxide and sodium hydroxide. U.S. Patent #1602802. Walsh, D.A, Perkins, J.P., and Krebs, E.G. 1968. An adenosine 3’,5’ monophosphatedependent protein kinase from rabbit sk eletal muscle. J. Biol. Chem. 243:3763-3765. Whetzel, H.H., 1945. A synopsis of the genera and species of the Sclerotiniaceae, a family of stromatic inoperculate discomycetes. Mycologia. 37:648-714. Willets, H.J., and Bullock, S. 1992. Developmental Biology of Sclerotinia. Mycol. Res. 96:801-816. Willets, H.J. and Wong, A.-L., 1979. Cytology of Sclerotinia sclerotiorum and related species. J. Gen. Microbiol. 112:29-34. Willets, H.J., and Wong, J.A.-L. 1980. The biology of Sclerotinia sclerotiorum , S. trifoliorum, and S. minor with emphasis on specific nomenclature. Bot. Rev. 46:101-165. Wiseman, A. and Lim, K.A. 1975. Removal of glucose repression of glucosidase synthesis in yeast protoplasts by adenosine 3’,5’ cyclic monophosphate addition and its prevention by chloramphenicol. Biochem. Soc. Trans. 2:932. Wiseman, A. and Lim. T.K. 1977. Adenosine 3’,5’ monophosphate in lifting of glucose repression of glucosidase by fluoride in yeast. Biochem. Soc. Trans. 5:970. Wittinghofer, A. 1998. Signal transduction via Ras. J. Biol. Chem. 379:933-937. Yang, Zhonghui., and Dickman, M.B. 1999. Colletotrichum trifolii mutants disrupted in the catalytic subunit of cAMP-dependent prot ein kinase are non-pathogenic. Mol. PlantMicrob. Interact. 12:430-439. Yelton, M.A., Hamer, J.E. and Timberlake, W.E. 1984. Transformation of Aspergillus nidulans by using a trpC plasmid. Proc. Natl. Acad. Sci. USA. 81:1470-1474. Xu, Jin-Rong, Urban, M., Sweigard, J.A., and Hamer, J.E. 1997. The CPKA gene of Magnaporthe grisea is essential for appressorial pene tration. Mol. Plan t-Microb. Interact. 10:187-194. Zubay, G.L. 1998. Biochemistry. Fourth Edition. McGraw-Hill, Boston.

PAGE 99

89 BIOGRAPHICAL SKETCH Wayne M. Jurick II was born July 4, 1976 in Naples, Florida to Ronda C. Jurick and Wayne M. Jurick Sr. He grew up in Fort Myers, Florida and enjoyed outdoor activities such as boating, fishing, hunting, pl aying baseball, basketball, and working on cars. After graduating from Bishop Verot Catho lic High School in Fort Myers, Florida, Wayne attended Edison Community College where he obtained an A.A. degree. He transferred to the University of Florida in August of 1996 where he received a Bachelor of Science degree in Plant Science with emphasis in plant pathology in August 1998. In the Fall of 1998 he started graduate school in the department of Plant Pathology under the supervision of Dr. Prem S. Chourey. The ma ster’s research project focused on sugarmodulated gene expression of a maize chitinase gene and he received a Master of Science degree in August of 2001. Near the end of his master’s studie s, Wayne had the good fortune of meeting Dr. Jeff A. Rollins when he was applying for a faculty position in the plant pathology department and gave an outst anding seminar. Intrigued by Rollins’ work Wayne contemplated joining Rollins laborator y and in August 2001, began to work in the area of fungal molecular biology after being awarded an Alumni Fellowship from the University of Florida. He completed his doctoral studies involvi ng cAMP-signaling in S. sclerotiorum in May 2006 and was awarded a Doctor of Philosophy degree. After graduation Wayne would like to continue in the area of pl ant pathology as a research scientist.