Genomics and Proteomics

137. The Neurospora crassa community genome annotation project. Heather M. Hood¹, James E. Galagan², Bruce W. Birren², Jay C. Dunlap³, & Matthew S. Sachs¹. ¹Oregon Health & Science University, Beaverton, OR 97006; ²Broad Institute, Cambridge, MA 02141; ³Dartmouth Medical School, Hanover, NH 03755.

The Neurospora crassa genome sequence was released in 2001. Automated annotation predicted ~10,000 genes. While automated annotation, including prediction of protein-coding regions and intron-exon boundaries, is crucial, errors in automated processes occur and they have limited ability to assign functional properties to genes. To produce richer and more accurate annotation, manual annotation and curation is necessary. Our goals are to improve annotation by manual curation and to integrate phenotypic annotation with genomic sequence. Substantial phenotypic information already exists for approximately 1000 loci and ongoing targeted gene-knockout experiments are expanding phenotypic knowledge of genes identified directly from sequence data. To produce the most valuable annotation data are establishing a controlled vocabulary to describe phenotypes associated with mutations in specific genes. At this conference, we are releasing a prototype web-based community annotation resource. This will allow community experts to submit data for associating mutant allele phenotypic information with specific genes and for improving gene-structure models. Curators will assess submissions and integrate these data into the genome annotation. This project will vastly improve the utility of the Neurospora genome by providing manual and curated community annotation, and by integrating phenotypic data with sequence data.

138. The Podospora Genome Project. Philippe Silar¹, Patrick Wincker², Fabienne Malagnac¹, Eric Espagne¹, Antoine Boivin³, Olivier Lespinet¹, Amid Khalili¹, Robert Debuchy¹, Sylvie Arnaise¹, Veronique Berteaux-Lecellier¹, Corinne Clave⁴, Veronique Contamine¹, Evelyne Coppin¹, Arnaud Couloux², Corinne Dasilva², Fons Debets⁵, Michelle Dequard-Chablat¹, Rolf Hoekstra⁵, Marguerite Picard¹, Berangere Pinan-Lucarre⁴, Annie Sainsart-Chanet³, Sven Saupe⁴, Carole H. Sellem³ and Jean Weissenbach². ¹ Institut de Genetique et Microbiologie, UMR8621, Orsay, France. ² Genoscope, Evry, France. ³ Centre de Genetique Moleculaire, UPR2167, Gif sur Yvette, France. ⁴ Institut de Biochimie et Genetique Cellulaires, UMR5095, Bordeaux, France. ⁵ Laboratory of Genetics, Wageningen University, Wageningen, Netherland.

Using a whole genome shotgun approach, we sequenced the Podospora anserina genome to a 10X coverage. Arachne assembly led to 2911 contigs totaling 35 Mbp, close to pulsed-field gel estimate of genome size. The N50 consists of 232 contigs greater than 42 kb and 18 supercontigs greater than 550 kbp. Previously identified genetic markers and 150 newly discovered microsatellites allowed 53 supercontigs to be anchored to the genetic map. ESTs obtained from various stages of Podospora life cycle are currently sequenced. Data are posted at http://podospora.igmors.u-psud.fr .

139. Microarray analysis of vegetative incompatibility in Neurospora crassa. Sarah C. Brown, Takao Kasuga, Isao Kaneko and N. Louise Glass. University of California, Berkeley

In Neurospora crassa, hyphal fusion between genetically distinct individuals leads to the formation of heterokaryons. If hyphal fusion occurs between individuals with different alleles at any het locus, growth arrest, hyphal compartmentation and cell death occur. We are investigating this vegetative incompatibility using a temperature sensitive mutant capable of forming stable heterokaryons at 34̊C with an otherwise incompatible partner. After transfer to 20̊C the incompatible phenotype is evident. We are using oligonucleotide arrays to compare gene expression patterns in compatible and incompatible heterokaryons made with the temperature sensitive mutant at 34̊C and 20̊C. Preliminary results indicate the involvement of a number of known cell rescue genes and a high proportion of predicted genes of unknown function.

A microarray analysis of heterokaryons undergoing vegetative incompatibility will provide us with a global picture of changes in gene expression, giving a clearer understanding of the regulation of transcription during vegetative incompatibility in Neurospora crassa. We aim to identify genes that play an active role in the control of cell death during vegetative incompatibility.

140. A proteomic approach to identify extracellular and cell wall proteins involved in the Phytophthora infestans – plant interaction. Catherine R Bruce, Shuang Li, Neil AR Gow and Pieter van West. The Aberdeen Oomycete Group, College of Life Sciences and Medicine, University of Aberdeen, Foresterhill, Aberdeen, AB25 2ZD, Scotland, UK.

A thorough understanding of the molecular events taking place during interactions between Phytophthora infestans and host and non-host plants is crucial for developing new control strategies. At the plant-pathogen interface, an exchange of molecular signals is thought to determine the outcome of the interaction. We anticipate that secreted and cell wall proteins will be rich in important signalling molecules involved in disease resistance or establishing successful infection. A proteomic approach is employed to identify novel extracellular and cell wall proteins from mycelia cultured in vitro and from plant intercellular fluid during P. infestans – tomato interaction. At present we have identified over 40 protein spots. Several of these may represent effector molecules and these are characterised further. Here we present our latest results.

141. Molecular Dissection of Neotyphodium lolii / Perennial Ryegrass Symbiosis. Richard Johnson, Christine Voisey, Shalome Bassett, Susanne Rasmussen, Linda Johnson, Charlotte Gaborit and Gregory Bryan. AgResearch Ltd., Tennent Drive, Private Bag 11008, Palmerston North, New Zealand.

Neotyphodium lolii is a fungal endophyte that lives entirely within the intercellular spaces of its grass host, perennial ryegrass. Infection is symptomless and the endophyte relies on the host plant for dissemination via the seed. The association is mutually beneficial since the endophyte confers a number of biotic and abiotic advantages to the host, including enhanced plant growth, protection from certain mammalian and insect herbivores, enhanced resistance to nematodes and some fungal diseases.

We are using a multidisciplinary approach to dissect the molecular basis of this symbiosis and intend to link the knowledge gained from basic biology and cytology with various functional genomics approaches. We have used suppressive subtractive hybridisation, fungal EST and targeted genome sequencing to develop a sequence database containing over 5000 fungal genes. The ESTs and other genes have been used to develop a microarray (see L. Johnson et al.). We are using proteomics to correlate fungal proteins expressed in culture with those expressed in planta including secreted proteins which may play a role in signalling between the host and fungus.

By linking these approaches we hope to identify genes which are important in both the establishment and maintenance of symbiosis. We are interested in signalling between the fungus and its host and have isolated several signalling genes including adenylate cyclase and a MAP kinase for targeted gene disruption studies. We have also isolated several secondary metabolite biosynthesis gene clusters and are currently investigating their role in symbiosis (see R. Johnson et al. and D. Fleetwood et al.).

142. The role of SPO11 in meiosis and meiotic recombination in Neurospora crassa. F.J. Bowring, P.J. Yeadon, R.G. Stainer and D.E.A. Catcheside. School of Biological Sciences, Flinders University, PO Box 2100, Adelaide SA 5001, Australia.

A homologue of the yeast SPO11 gene is carried by numerous organisms, including humans, and is thought to code for the enzyme that catalyses the initiation of all meiotic recombination. We have generated three RIP mutant alleles and several deletion strains of the Neurospora crassa SPO11 homologue.

When homozygous, all three RIP alleles reduce fertility and spore viability, and an analysis of the remaining viable spores suggests extensive chromosome non-disjunction during meiosis. Microscopic analysis of mutant perithecial tissue reveals a spectrum of meiotic defects including a failure of homologous chromosomes to synapse during pachynema. While we expected that, as in other organisms, mutation of the SPO11 homologue would reduce or abolish recombination in Neurospora, our data indicate this is not the case. The frequency of allelic and non-allelic recombination near the recombination hotspot cog^L is at least preserved and possibly elevated in mutant crosses although recombination in another region of the genome may be suppressed. Preliminary analysis of meiosis in strains having the SPO11 homologue deleted suggests that the three RIP mutants harbour null alleles.

143. Use of yeast recombinational cloning and a Neurospora crassa strain defective in non-homologous end joining for high-throughput production of gene replacement mutants. Hildur V. Colot, Gyungsoon Park*, Carol Ringelberg, Susan Curilla, Christopher Crew*, Jennifer J. Loros, Katherine A. Borkovich* and Jay C. Dunlap. Genetics Department, Dartmouth Medical School, Hanover, NH; and *Department of Plant Pathology, University of California, Riverside, CA

We will be creating knockout (KO) mutants for all annotated Neurospora genes as part of an NIH-funded Program Project (PO1). We previously reported development of a strategy suitable for high-throughput gene deletions, involving the creation of hph-marked KO cassettes by recombination-mediated plasmid construction in S. cerevisiae (Asilomar 2004), followed by transformation of split-marker fragments into Neurospora. We have adapted our yeast recombinational cloning techniques for use on a Beckman Biomek NX robot and have now created KO cassettes for thousands of Neurospora genes. We have also improved our protocol for creating the final Neurospora KO mutants by utilizing a Neurospora mutant defective in nonhomologous end-joining (NHEJ) as the recipient for electroporation. Recent work from the Hirokazu Inoue laboratory (Ninomiya et. al., 2004) demonstrated that mutation of genes involved in NHEJ repair (mus-51/ku70 and mus-52/ku80; both genes replaced with hph) leads to a high rate of homologous recombination in Neurospora. We have subsequently created our own mus-51 and mus-52 KO strains using the selectable marker bar. We have confirmed that electroporation of either of these mutants with several KO constructs containing 1 kb of flanking DNA on either side of the hph marker results in 100% of the transformants possessing the correct gene replacement and no ectopic insertions of any portion of the cassette. Generation of homokaryotic KO mutants for an initial group of 100 transcription factor genes is in progress, and data on their characterization will be presented.

144. Saccharomyces Genome Database: a resource for fungal comparative genomics. Maria C. Costanzo¹, Rama Balakrishnan¹, Karen R. Christie¹, Kara Dolinski², Selina S. Dwight¹, Stacia R. Engel¹, Dianna G. Fisk¹, Jodi E. Hirschman¹, Eurie L. Hong¹, Robert Nash¹, Rose Oughtred², Anand Sethuraman¹, Marek S. Skrzypek¹, Chandra L. Theesfeld¹, Gail Binkley¹, Qing Dong¹, Stuart Miyasato¹, Mayank Thanawala¹, Shuai Weng¹, David Botstein², J. Michael Cherry¹. (1) Dept. of Genetics, Stanford University, 300 Pasteur Drive, Stanford, CA, 94305-5120, USA; (2) Department of Molecular Biology, Princeton University, Princeton, NJ 08544, USA.

The Saccharomyces Genome Database (SGD; www.yeastgenome.org ) is a comprehensive compilation of genome, gene, and protein information for the model fungus S. cerevisiae. SGD offers tools for comparison and analysis of information on a genome-wide scale, both within S. cerevisiae and between S. cerevisiae and other species. The PSI-BLAST pages present a precomputed, periodically updated PSI-BLAST analysis of each S. cerevisiae protein vs. the non-redundant protein set in GenBank, with results organized by taxonomic grouping. The Model Organism BLASTP Best Hits pages display links to curated database pages of proteins in other model organisms that are similar to S. cerevisiae proteins. The Fungal BLAST tool allows comparison of any input sequence against multiple fungal genomes. This open-source tool is based on the BLAST module developed by the GMOD project ( www.gmod.org ) and is readily adaptable to other databases. All information at SGD is freely available, both via the web interface and in downloadable files. SGD is supported by a grant from the National Human Genome Research Institute, NIH.

145. The Genus Alternaria as a Model for Necrotrophic Fungal Pathogens of Plants. K. D. Craven (1), R. Cramer (2), D. Knudson (3), C. B. Lawrence (4), T. K. Mitchell (1). (1) North Carolina State University, Fungal Genomics Laboratory, Raleigh, NC; (2) Duke University, Durham, NC; (3) Colorado State University, Fort Collins, CO; (4) Virginia Bioinformatics Institute, Blacksburg, VA.

We have created numerous resources to investigate the genome structure and mechanisms of pathogenicity and gene flow among plant pathogenic species in the fungal genus Alternaria. To study gene expression during compatible (disease-causing) interactions, we have generated and analyzed cDNAs from EST libraries created from Alternaria brassicicola-infected tissues of two related Brassicaceous hosts, and from fungal mycelium grown under nitrogen-starvation conditions. From a set of approximately 2000 unique fungal EST sequences, we have selected candidates for gene knockout to evaluate their effects on pathogenicity. A BAC library of the Alternaria brassicicola genome was created and is being used to elucidate genome structure, identify clusters of functionally related genes, and to contribute to the recently funded genome sequencing effort. Many distinct pathotypes of Alternaria alternata have been shown to produce host-selective toxins required for disease, and further, that the toxin biosynthetic genes are typically located on small, conditionally dispensable (CD) chromosomes. Despite their crucial role in disease development, such chromosomes are not characteristic of entire fungal species, and thus are likely to be under-represented in genome sequences of organisms containing them. We chose the tomato pathotype, A. alternata fsp. lycopersici, from which to isolate a known 1.0 Mb CD chromosome for library construction and sequencing. It has been postulated that the clustering of pathogenicity genes on CD chromosomes may allow for efficient horizontal transfer between fungal strains or species through hyphal anastomosis. We are investigating the potential for such gene flow through the construction of nitrate-utilization mutants of A. brassicicola and related Alternaria species.

146. Fungal comparative genomics – an update on the Fungal Genome Initiative. Li-Jun Ma, Christina Cuomo, Sarah Calvo, Dave DeCaprio, Jonathan Butler, Manfred Grabherr, David B. Jaffe, Tim Elkins, Shawn Samuel, Miriam Averbuch, Chinappa Kodira, Eric Lander, Chad Nusbaum, James Galagan, and Bruce Birren. The Broad/MIT, Cambridge, USA

The Fungal Genome Initiative (FGI) is generating genomic resources for organisms across the fungal kingdom by producing genome sequence, annotation and analysis. The FGI species represent the major branches of the fungal tree and will help elucidate the molecular basis for the tremendous diversity that arose in the 1 billion years of fungal evolution. In addition, we are sequencing clusters of phylogenetically closely related species to examine recent evolutionary events. These clusters include fungi that are human and plant pathogens, as well as important model systems. So far, a total of 26 fungal genome sequence projects have been completed or are in the queue. The data can be accessed at the Broad website (http://www.broad.mit.edu/annotation/fungi/fgi/ ). We will present an update on our sequencing progress and describe recent results.

147. Sequencing and analysis of the Fusarium graminearum genome. Christina Cuomo, Li-Jun Ma, Jonathan Butler, Sarah Calvo, Dave DeCaprio, Tim Elkins, James Galagan, Jin Rong Xu‡, Frances Trail†, Corby Kistler*, and Bruce Birren. Broad Institute of MIT and Harvard, Cambridge MA. *Univ. of Minnesota, St Paul MN. †Michigan State Univ., East Lansing, MI. ‡Purdue Univ., West Lafayette IN.

Fusarium graminearum is a major plant pathogen that causes head blight of wheat and barley. The Broad Institute and members of the Fusarium community have produced and analyzed a high-quality draft sequence of F. graminearum. The assembly contains 36 Mb of sequence at 10-fold depth. Nearly all (99.5%) of the sequence was anchored to the 4 chromosomes using genetic mapping. A total of 11,640 protein coding genes were predicted. 17% of proteins are unique to F. graminearum, showing no significant homology to any protein in the nonredundant (nr) protein database. Compared to other sequenced eukaryotes, F. graminearum contains very few high identity paralogous genes. Additionally, the genome is repeat poor, containing between 15 and 30-fold fewer repeats than other sequenced ascomycetes. One possible explanation for the lack of high identity sequences is that the process of repeat-induced point mutation (RIP) is active in F. graminearum. The longest repeat family, a group of Fot1-like inactive transposons, shows a mutation bias consistent with a low level of RIP. Further, a search of the protein set revealed a putative ortholog of the rid (rip-defective) gene of N. crassa. However, the genome contains very few remnants of repetitive elements as compared to N. crassa. This suggests that while there is some evidence for RIP activity in F. graminearum, RIP alone does not explain the absence of repeats in the genome. (This work was supported by the NSF/USDA, Microbial Genome Sequencing Project, award number 2002-35600-12782)

148. Genomic-directed development of new therapies to treat Pneumocystis infections. Melanie T. Cushion, Margaret Collins, Sandeep Bansil, Sandy Rebholz, Alan Ashbaugh, Peter D. Walzer. University of Cincinnati College of Medicine and the VAMC, 231 Albert Sabin Way, Cincinnati, OH 45267-0560

Species of the fungal genus Pneumocystis cause a lethal pneumonia (PcP) in mammalian hosts with compromised immune status. Pneumocystis lack detectable ergosterol rendering them refractory to standard anti-fungals. Increasing evidence indicates that P. jirovecii (the species infecting humans) is developing resistance to the most efficacious treatment, TMP-SMZ (targeting folate synthesis), and to second-line drugs (e.g. atovaquone). We re-constructed a sterol biosynthetic pathway for P. carinii with genomic data from the Pneumocystis Genome Project (http://pgp.cchmc.org). Use of specific inhibitors showed that enzymes in this pathway were functional. The effects of tolnaftate, tebuconazole, and simvastatin on gene regulation were evaluated using macroarrays. Simvastatin had little effect on sterol genes, although organism viability was decreased by 80%. Tebuconazole down-regulated its target, 14-alpha-demethylase, and other genes in the pathway. Tolnaftate down-regulated almost all sterol genes. Isobolograms showed synergy between simvastatin/tebuconazole. Simvastatin was effective in reducing lung burden in the mouse model of PcP (p=0.0258) vs untreated controls. Combination therapies targeting pathways other than folate synthesis hold promise for new treatment modalities for PcP while analyses of gene expression can suggest alternative drug targets (NIH-AI-50450, NO1 AI75319).

149. Phylogenomic Analyses of Heterokaryon Incompatibility Proteins in Aspergilli. Natalie D. Fedorova¹, Jennifer R Wortman¹, William C. Nierman¹. ¹The Institute for Genomic Research, 9712 Medical Center Drive, Rockville, MD 20850, USA.

Heterokaryon incompatibility is a form of programmed cell death in filamentous fungi, triggered by the fusion of filaments from two strains with different sets of het loci. The analyses of the completely sequenced genomes of Aspergillus fumigatus, Aspergillus fischerianus, Aspergillus nidulans and Aspergillus oryzae reveal remarkable diversity of heterokaryon incompatibility systems in aspergilli and other filamentous fungi. The Aspergillus genomes encode numerous homologs of heterokaryon incompatibility proteins from Neurospora crassa and Podospora anserina. Yet the number of homologous proteins, their domain architecture and, perhaps, the underlying mechanisms are quite different.

Notable examples include an aspergilli-specific expansion of the putative NACHT NTPases that function as signaling hubs during heterokaryon incompatibility in filamentous fungi, pathogen resistance in plants, or apoptosis in mammals. The Aspergillus NACHT proteins are most similar to GTP-binding protein Het-E and Het-D from P. anserina, but they do not have the HET domain, found at the N-terminus of the Podospora proteins. Instead many of Aspergillus NACHT proteins have a highly divergent nucleoside phosphorylase domain. Still unknown is whether all NATCH NTPases are involved in self/non-self recognition during heterokaryon incompatibility or some are part of another signaling pathway in aspergilli.

150. Demonstration of LOH by SNP microarray analysis and alterations in strain morphology in strains of Candida albicans during infection. Anja Forche, Paul T. Magee, and Georgiana May. University of Minnesota, Saint Paul MN

Candida albicans is a diploid yeast with a predominantly clonal mode of reproduction, and no complete sexual cycle is known. As a commensal organism, it inhabits a variety of niches in humans. It becomes an opportunistic pathogen in immunocompromised patients, and can cause both superficial and disseminated infections. It has been demonstrated that genome rearrangment and genetic variation in isolates of C. albicans is quite common. One possible mechanism for generating genome-level variation among individuals of this primarily clonal fungus is mutation and mitotic recombination leading to loss of heterozygosity (LOH). Taking advantage of a recently published genome-wide SNP map, a SNP microarray was developed for 23 SNP loci residing on chromosomes 5, 6, and 7. It was used to examine 21 strains previously shown to have undergone mitotic recombination at the GAL1 locus on chromosome 1 during infection in mice. In addition, karyotypes and morphological properties of these strains were evaluated. Our results show that during in vivo passaging, LOH events occur at observable frequencies and that such mitotic recombination events occur independently in different loci across the genome and that changes in karyotypes and alterations of phenotypic characteristics can be observed either alone, in combination, or together with LOH.

151. A genetic map of Gibberella zeae using sequence-tagged sites and AFLPs. Liane R. Gale¹, Je'Nise D. Bryant², Henriette Giese³, Talma Katan⁴, Kerry O'Donnell⁵, Haruisha Suga⁶, Thomas R. Usgaard⁵, Todd J. Ward⁵, and H. Corby Kistler¹. ¹CDL, USDA St. Paul, MN. ²University of Minnesota, St. Paul, MN. ³Royal Veterinary and Agricultural University, Copenhagen, Denmark. ⁴Volcani Center, Bet Dagan, Israel. ⁵NCAUR, USDA, Peoria, IL. ⁶Gifu University, Gifu, Japan.

A genetic map of Gibberella zeae (anamorph Fusarium graminearum) was constructed using a cross between nitrate-nonutilizing (nit) mutants of strain PH-1 and a Minnesota field strain, 00-676. A total of 111 ascospore progeny were analyzed for segregation at 237 loci. Genetic markers consisted of SNPs (detected as dCAPs, n=86 or CAPs, n=47), AFLPs (n=71), VNTRs (n=27), and six others. While 213 markers exhibited Mendelian inheritance, segregation distortion was noted for 17 markers at four genomic locations. A linkage map was generated using JoinMap 3.0 and a LOD threshold value of 4.0. Eleven linkage groups were obtained, covering 1154 cM and anchoring 99.8% of the sequence assembly. All linkage groups and anchored supercontigs were assembled into four chromosomes, leaving only 11 smaller supercontigs (76,055 bp total) of the nuclear DNA not anchored. Comparison between physical and genetic distance along chromosomes revealed genomic regions with reduced recombination and recombinational hotspots. More information can be found at http://www.broad.mit.edu/annotation/fungi/fusarium.

152. DelsGate a robust and rapid deletion plasmid construction method. María D. Garcia-Pedrajas and Scott E. Gold. Department of Plant Pathology, University of Georgia, Athens

A combination of PCR and gateway technology together with use of the I-SceI homing endonuclease provides a rapid robust universal method for construction of plasmids suitable for precise genomic deletions in two days. We employed this method for the deletion of several genes in the basidiomycete fungus Ustilago maydis which causes corn smut disease. This approach, however, is universal for fungi. The method involves 2 PCR steps followed by gateway cloning to generate the final product. The first PCR reaction employs specific primers (1 and 2) 1kbp 5' and 3' of the start and stop of the ORF. The 5' end of primer 1 includes the 18bp recognition sequence of I-SceI generally absent in fungal genomes. The 2nd reaction employs inverse PCR of the self-ligated 1st PCR product. Primers 3 and 4 are directed out from the ORF at the start and stop codons. These primers have at their 5' ends the attB1 and attB2 recognition sites. The 2nd PCR product, having 1kb gene flanks but lacking the entire ORF, is then inserted with B/P clonase into a donor vector with an appropriate selectable marker. After digestion with I-SceI the plasmid is transformed into the fungus of interest by standard methods.

153. Comparative Genomic Hybridization within the Genus Neurospora. Luz B. Gilbert, Takao Kasuga, Jeff Townsend, Louise Glass, and John W. Taylor. U.C. Berkeley, Plant and Microbial Biology.

Comparative Genomic Hybridizations (CGH) are becoming a popular way to determine similarity among strains and even species. A growing trend is to use CGH data to assess evolutionary history by developing phylogenetic trees from differences in hybridization between isolates. As yet few have questioned the reliability of CGH data to correctly assess sequence differences in hybridization and therefore the ability of this type of data to determine evolutionary relationships. The study of a simple eukaryote, the filamentous fungus Neurospora, offers a unique opportunity to rigorously address these questions using both experimental and simulated data.

The genus Neurospora consists of eight closely related conidiating species indistinguishable by morphology, as well as several non-conidiating species. An accurate phylogeny was published by Dettman et al. 2003. We have constructed a 70mer oligomer array for Neurospora crassa representing 10,000 genes. I have analyzed comparative genomic hybridizations for all eight conidiating species of Neurospora as well as a few non-conidiating isolates. These results were then compared to simulated data generated to mimic the design of the CGH experiment. The goal of the simulated data is to determine under what scenarios CGH data might accurately determine evolutionary relatedness. We have used both the simulated and empirical data to generate distance based dendograms for the different species that we can compare to the known phylogeny and assess the utility of CGH data for testing evolutionary relationships.

154. Identification from an EST database and genetic mapping of microsatellites in Mycosphaerella graminicola. Stephen B. Goodwin¹, Jessica R. Cavaletto¹, Theo van der Lee², Bas te Lintel Hekkert², Gert H. J. Kema². ¹ USDA-ARS, Crop Protection and Pest Control Research Unit, Department of Botany and Plant Pathology, Purdue University, West Lafayette, IN, USA; ² Plant Research International B.V., Wageningen, The Netherlands

Mycosphaerella graminicola, the cause of septoria tritici blotch of wheat, is developing rapidly as a genetic model for fungi in the order Dothideales. Further development of the genetics of this organism would be facilitated by the availability of microsatellite or simple-sequence repeat (SSR) markers. However, only nine microsatellite loci have been identified so far in M. graminicola, and none has been mapped. To identify additional polymorphic SSR loci, an EST database from M. graminicola was scanned for di- and trinucleotide units repeated tandemly six or more times. Among more than 30,000 EST sequences screened, 109 possible SSR loci were identified using an automated software pipeline. Primers flanking 99 of these SSRs were developed and tested for amplification and polymorphism on the two Dutch parents of the standard M. graminicola mapping population, one isolate from North Africa, and two from North America. Seventy-seven of the 99 primer pairs generated an easily scored banding pattern and 51 were polymorphic among the five field isolates of M. graminicola tested. Among these 51 loci, 23 were polymorphic between the parents of the mapping population; 21 of these plus two previously published microsatellites were integrated into the existing genetic map. These 23 microsatellite loci mapped to 12 of the 22 linkage groups of the M. graminicola genetic linkage map. Most (66%) of the primer pairs also amplified bands in the closely related barley pathogen Septoria passerinii, but only six were polymorphic among four isolates of that species tested. The EST database provided an excellent source of new microsatellites, some of which also may be useful in closely related species. These markers are highly polymorphic (up to 4 alleles per locus among five isolates tested) and can be multiplexed to facilitate integration of the different types of genetic analysis performed on this important plant pathogen.

155. Unfolded Protein Response in Aspergillus niger chemostat fermentations. T. Goosen^1,2, T. R. Jørgensen³, J.J.L. Iversen³, C.A.M.J.J. van den Hondel^1,2. ¹Leiden University, Institute of Biology Leiden, Fungal Genetics Research Group, Wassenaarseweg 64, 2333 AL Leiden, The Netherlands, ²Department of Microbiology, TNO-Nutrition and Food Research, P.O. Box 360, 3700 AJ Zeist, The Netherlands, ³Department of Biochemistry and Molecular Biology, University of Southern Denmark, Campusvej 55, 5230 Odense, Denmark

Unfolded Protein Response (UPR) is a universal reaction of eukaryotic cells to protein folding stress in the endoplasmic reticulum (ER). The expression of heterologous secreted proteins often results in such stress and thus UPR, also in filamentous fungi like Aspergillus niger which are renowned for their high secretion capacity. As a consequence, the production yield of secreted heterologous proteins usually is very low (milligrams/L at best).

UPR results in induction of expression of genes that allow the cell to cope with the surplus of protein folding intermediates. Identification of these genes and understanding the response pathway(s) should give leads to improve the folding and secretion capacity of A. niger. Towards this, two approaches are taken: transcriptomics with Affymetrix genome arrays to identify up- or down-regulated genes and genetic screening to select regulatory mutants.

For transcriptomics analysis, A. niger strains are constructed in which poorly folded (scFv) or unfoldable (yeast CPY*) proteins can be expressed under control of the regulated glaA promoter. RNA analysis is first performed with shake flask cultures, both under induced and non-induced conditions for the expression of the poorly folded proteins

For a sound transcriptomics analysis, tight control of all cultivation conditions is an absolute prerequisite. We therefore are developing protocols for steady-state fermentation of A. niger under repressing conditions for the glaA promoter and for transition to a glaA induced steady-state. The samples collected throughout these fermentations are used for transcriptomic and protein analysis.

Tools are developed to improve the genetic screen for the isolation of regulatory mutants with altered UPR characteristics.

We will report on the recent progress made in these project goals.

This research is carried out within the Kluyver Centre for Genomics of Industrial Fermentation.

156. Comparative fungal genomics using the MIPS comprehensive resources. U. Gueldener^1,*, M. Muensterkoetter¹, G. Mannhaupt², H.W. Mewes^1,3. ¹MIPS-Munich Information Center on Protein Sequences, GSF - National Research Center for Environment and Health, Ingolstädter Landstrasse 1, D-85764 Neuherberg, Germany. ²Max Planck Institute for Terrestrial Microbiology, Department of Organismic Interactions, Karl-von-Frisch-Strasse, D-35043 Marburg, Germany. ³Technische Universität München, Chair of Genome Oriented Bioinformatics, Center of Life and Food Science, D-85350 Freising-Weihenstephan, Germany

The number and quality of currently sequenced fungal genomes is far beyond what has been expected some years ago. Having 30 analyzed fungal genomes at hand, the transfer of knowledge from the well studied model species Saccharomyces cerevisiae and Neurospora crassa to the fungal key target organisms like Fusarium graminearum, Ustilago maydis or Phanerochaete chrysosporium will be the main challenge in the near future. Extensive comparative analysis methods not only provide more information, but in turn allow the annotation of distinct functions to up to now unknown proteins and thus significantly improve the quality of the genome annotation. Moreover comparative analysis on protein domain level and synteny between fungal as well as fungal and other species groups reveal evolutionary coherences and putative horizontal gene transfer. Mapping and comparison of protein-protein interactions (interologs) help to imagine interactomes. Recently intra- and inter-genomic promoter comparisons were started which, in combination with expression data, will give insight into regulatory networks. First comparative analysis results will be presented.

157. Genomic analysis of the secretion stress response in Aspergillus niger. Thomas Guillemette¹, Noel van Peij², Karin Lanthaler³, Geoff Robson³, Hein Stam², David Archer¹. ¹School of Biology, University Park, University of Nottingham, Nottingham NG7 2RD, UK. ²DSM (Food Specialities), P.O. Box 1, 2600 MA Delft 624-0295, The Netherlands. ³The Victoria University of Manchester, School of Biological Sciences, 2205 Stopford Building, Oxford Rd, M13 9PT Manchester, UK

Filamentous fungi such as Aspergillus niger have a well developed and high capacity secretory system and are therefore widely exploited for the industrial production of native and heterologous proteins. However, in most cases the yields of non-fungal proteins are significantly lower than those obtained for fungal proteins. One well-studied bottleneck appears to be the result of mis-folding of heterologous proteins in the ER during early stages of secretion, with related stress responses in the host, including the unfolded protein response (UPR), ER-associated (protein) degradation (ERAD) and a more recently discovered transcriptional repression of genes during secretion stress (RESS). The aim of our project is to use genomic approaches to gain insights into the bottlenecks that limit the application of A. niger as a host for recombinant protein production. A global analysis of secretion-related stress responses was performed using Affymetrix chips based on the sequenced genome available from DSM. In our experimental conditions, secretion stress in A. niger was induced by three means: (i) treatment of wild-type mycelium by chemical inducers of the UPR (DTT and tunicamycin), (ii) over-expression of a heterologous protein (t-PA), (iii) reduction of the level of ER-resident proteins involved in folding (PdiA). Overlapping (common to all conditions) and specific responses will be described in relation to secretion stress.

158. Characterization of the genes encoding tropomyosin and ARP4 in Neurospora crassa. Nahideh Haghighi¹, P. John Vierula¹. ¹Biology Department, Carleton Uinversity, Ottawa, Canada.

This study has focused on the genes encoding tropomyosin and ARP4 in the filamentous fungus Neurospora crassa. Tropomyosin is an extended coiled-coil protein that binds to actin filaments and influences many aspects of F-actin. Repeat Induced Point (RIP) mutagenesis was used to produce a tropomyosin mutant strain. Mating with strains containing multiple copies of the tpm gene produced 20% inviable ascospores and approximately 7% of all spores exhibited an abortive germination phenotype. Only one out of 120 viable progeny displayed a mutant phenotype suggesting that the N. crassa tpm gene is essential for germination of ascospores. Moreover, scanning electron microscopy (SEM) and light microscopy of the tpm-RIPed strain showed that it branches more frequently than wild type. Staining with FITC labelled mouse a-actin did not detect an apical actin distribution, instead actin positioned predominantly in subapical zones.

The actin related proteins (ARPs) have primary sequence homology to actin. In the present study, an ARP in N. crassa was identified that has 30-35% amino acid identity to ARP4 from other species. Like other ARP4s, NcARP4 has a nuclear localisation signal (NLS) suggesting that it is taken up by the nucleus. One morphological mutant strain was detected among 90 RIP mutagenesis progeny and its arp4 gene had very few G to A substitutions, suggesting that arp4 is also an essential gene. Approximately, 20% of the arp4-RIPed germlings have nuclei localized at the extreme tips of hyphae, in comparison to the 8-10 mm distance of the apical nucleus in wild type.

159. The utility of the incomplete genome: the Amanita bisporigera genome project. Heather E. Hallen¹ & Jonathan D. Walton². ¹Department of Plant Biology, and ²DOE Plant Research Laboratory, Michigan State University, East Lansing, MI 48824

As part of ongoing investigations into the ecology, biochemistry and evolutionary biology of Amanita section Phalloideae, we have initiated the Amanita genome project. This project originated as part of our investigations into amatoxin biosynthesis. Amatoxins are presumed to be synthesized via a non-ribosomal peptide synthetase (NRPS), predicted to be encoded by a 30-kb gene. Assuming random sampling across the genome and an average read size of 600 bp, the odds of hitting a 30 kb target in a 40 Mb genome are high: 95% chance in 4,000 random, independent sequences. Consequently, we are generating several thousand genomic sequence reads from Amanita bisporigera (10,000+ as of November 30, 2004). Sequencing an EST library was impractical, as amatoxin biosynthesis appears to take place in a narrow window at or near the time of button initiation, and transcription of amatoxin biosynthetic genes is therefore not observable in the macroscopic organism. We have a generated a public Amanita sequence database. Each sequence has been compared to GenBank's non-redundant database (using BLASTX) and the Coprinus and Phanerochaete genomes (using TBLASTX). Additionally, we maintain all sequences in BLAST-searchable format. This resource is available at . We envision it being of particular interest to the ectomycorrhizal research community.

160. Fungal and plant gene expression in endophyte-infected plant tissues. U. Hesse and C. L. Schardl. Plant Pathology Department , University of Kentucky

Analysing gene expression of symbiotic microorganisms in-planta is difficult due to relatively low biomass. We investigated the fungal endophyte Epichloë festucae, which lives in symbiosis with cool-season grasses. It represents an interesting model organism for studying plant-microbe interactions because it develops mutualistic, parasitic, and pleiotropic relationships with its host plants. For most of the life cycle, the endophyte grows benignly through the intercellular spaces of the above-ground plant organs (pseudostems, leaf blades, inflorescences, seeds) without damaging or entering host cells. Mutualistic effects are associated with production of bioprotective alkaloids by the fungus, and positive effects on the drought tolerance of the host. Reproductive development of the host induces production of external hyphae and development of a stroma (a compact mycelial layer which serves as a cradle for ascogenous perithecia) enveloping the inflorescence. In parasitic Epichloë sp.-grass interactions host reproduction is virtually terminated and the rare seeds from escaped inflorescences are endophyte-free. Epichloë festucae is an intermediate: it exhibits both transmission modes (benign colonization of seeds and stromata production) on one plant. Sequencing of normalized vs. non-normalized cDNA libraries from endophyte-infected plant tissues, and genomic DNA of the fungus allowed extensive identification of expressed fungal and plant unigenes, which can be used in future hybridization studies for discovery of differentially expressed symbiosis genes.

161. Live cell imaging of fungal hyphae using fluorescent dyes and recombinant probes. Patrick Hickey¹ and Nick Read². ¹LUX Biotechnology Ltd. Edinburgh, Scotland, EH9 3JL. ²University of Edinburgh, Scotland, EH9 3JH.

Recent developments in imaging technology have allowed us to explore new dimensions in cell biology. Using confocal microscopy we can simultaneously image living fungal hyphae capturing up to three fluorescent channels and a bright field image. In addition to recombinant probes like GFP, a number of different fluorescent dyes have been utilised to target organelles and components of fungal hyphae. The importance of live cell analysis is discussed and an overview of the techniques required for image processing and time lapse animation is provided. Methods for optimising image quality, and reducing photobleaching and phototoxicity are described. Finally, the use of fungi in high throughput imaging applications is introduced.

162. Transcription analysis of carbon repression of Aspergillus nidulans using High Density Microarrays. Jesper Mogensen¹, Henrik Bjørn Nielsen², Gerald Hofmann¹, and Jens Nielsen¹ . ¹Center for Microbial Biotechnology, BioCentrum, DTU, Denmark. ²Center for Biological Sequence Analysis, BioCentrum, DTU, Denmark

Glucose repression has been the subject of fungal research for many years now, and still there is a lot to be resolved, for example the many pleiotropic effects of mutations of the creA gene. Therefore this study aims at a better understanding of glucose and/or carbon repression in A. nidulans, utilising high density DNA-microarrays for high throughput transcriptional analysis.

The effect of CreA was investigated by comparison of the gene-expression of the creA mutant against a reference strain under growth conditions considered to be repressing (glucose) and de-repressing (ethanol) . Statistical analysis based on biological triplicates showed 200 significantly regulated genes (p-value less than 1.0 x 10^-2). Cluster analysis, based on the expression profiles of all four conditions, reveals a number of complex patterns, whose interpretation is hampered by the relative poor annotation of the majority of open reading frames (ORFs) the A. nidulans genome sequence. On the other hand, this enhances the value of this study, since it increases the knowledge about putative roles/functions of ORFs.

For further analysis and interpretation of the data, promotor analysis has been performed and the response of metabolic genes has been compared with data from C¹³ metabolic flux analysis of the same creA mutant strain.

163. Evolutionary differences in the fungal carbamoyl-phosphate synthetase small subunit gene and associated upstream open reading frame. Heather M. Hood and Matthew S. Sachs. Oregon Health & Science University, Beaverton, OR 97006

Neurospora crassa arg-2and Saccharomyces cerevisiae CPA1 encode the small subunit of arginine-specific carbamoyl-phosphate synthetase (CPS-A), which is required for arginine biosynthesis. Excess arginine negatively regulates the translation of each mRNA. This is accomplished via the action of a 5' upstream open reading frame (uORF) that encodes a cis-regulatory element, the arginine attenuator peptide (AAP). The AAP stalls ribosomes in response to arginine and by doing so reduces synthesis of the enzyme, which is produced from a downstream open reading frame. Here we show a comparative analysis of the genes and predicted regulatory elements of arg-2 homologs obtained from genomic sequences of Ascomycetes and Basidiomycetes. The AAP peptide and the structures of the genes specifying CPS-A has diverged between these phyla. Major differences detected included variation in intron number and position and varying degrees of regulatory motif conservation. Interestingly in Kluyveromyces lactis, the reading frames encoding the AAP and the CPS-A enzyme overlapped. Despite many differences across species, there are three perfectly conserved amino acid residues in the predicted AAP among all fungi, including an aspartic acid known to be crucial for arginine-dependent regulation of arg-2 and CPA1. We are following up on these in silico results by testing the efficiency with which the different isoforms of the AAP exert translational control.

164. Transcriptional Responses to Secretion Stress in the Fungi Trichoderma Reesei and S. cerevisiae Reveal Interesting Differences and Common Features. Mikko Arvas, Tiina Pakula, Karin Lanthaler*, Geoff Robson*, Markku Saloheimo and Merja Penttilä. VTT Biotechnology, Tietotie 2, Espoo, PL 1500, 02044 VTT, Finland. *Univ. of Manchester, Fac. of Life Sciences, 1800 Stopford Building, Oxford Road, Manchester M13 9 PT, UK.

Trichoderma reesei is an industrial protein production host known for its exceptional protein secretion capability. This study aims at uncovering the transcriptional responses occurring in T. reesei cells exposed to secretion stress and comparing these responses to similar experiments carried out in S. cerevisiae. Secretion stress is caused by compromised protein folding or transport in the secretory pathway. It induces a number of genes involved in different aspects of secretion through the unfolded protein response (UPR) pathway. In T. reesei it has also been shown that secretion stress down-regulates genes encoding secreted proteins.

 We constructed cDNA subtraction libraries and made cDNA-AFLP (amplified fragment length polymorphism) experiments from cells under secretion stress. A transformant expressing human tissue plasminogen activator (tPA), treatment with the chemical DTT (dithiothreitol) that prevents correct protein folding and a transformant over-expressing IREI protein (sensor protein of the UPR pathway) were analysed. Around two hundred unique ESTs were retrieved by these methods and the expression pattern of about 50 was confirmed by Northern experiments. A rank sum test for the Northern data was used to define those genes that show upregulation in all the three conditions. Data from DTT and tunicamycin treatment, foreign protein production and IRE1 and HAC1 (UPR transcription factor) deletion experiments in S. cerevisiae were combined from litterature. The transcriptional responses of T. reesei and S. cerevisiae show clear overlap, especially with respect to genes involved in protein translocation, folding and glycosylation in the ER. However, there seems to be differences in regulation of amino acid biosynthesis and nucleosome genes. The GCN4/CPC1 transcription factor and a limited set of its putative target genes are induced in T. reesei. This response points to the upregulation of glutathione synthesis to relieve oxidative stress caused by compromised protein folding. Interestingly also a set of nucleosome genes is upregulated in T. reesei without a clear connection to cell cycle.

165. A proteomic approach to identify proteins secreted by Fusarium oxysporum in xylem sap of tomato.. Petra M. Houterman¹, Dave Speijer³, Henk L. Dekker², Ben J.C. Cornelissen¹ and Martijn Rep¹. ¹Plant Pathology, ²Mass Spectrometry, Swammerdam Institute for Life Sciences, ³Medical Biochemistry, Academic Medical Center (AMC), University of Amsterdam, The Netherlands.

Fusarium oxysporum f. sp. lycopersici is a soil-born fungus that causes vascular wilt disease in tomato. Fusarium invades the plant via the roots and subsequently colonizes the plant via the xylem vessels. Proteins that play an important role in the interaction between plant and pathogen are therefore likely to be secreted into the xylem sap.

We analyzed the proteins that accumulate in the xylem sap of tomato upon infection with F. oxysporum. These proteins were identified using a combination of 2-dimensional gel electrophoresis (2-DE), peptide mass fingerprinting (MALDI MS) and mass spectrometric sequencing (LC MS/MS).

We had shown earlier that a number of pathogenesis related (PR) proteins from tomato and SIX1 (secreted in xylem 1) from Fusarium accumulate in xylem sap after infection (1,2). Here we report the identification of additional plant and possible fungal proteins. From tomato we identified a polygalacturonase (endoPG), a number of peroxidases, a xylogucan-specific endoglucanase inhibitor protein (XEGIP) and a xyloglucan endotranglycosylase (XET). A number of sequence tags were obtained from unidentified, possibly fungal, proteins. These will enable us to use degenerated PCR for identification of coding sequences.

(1) Rep et al. (2002) Plant Physiology 130: 904-917. (2) Rep et al. (2004) Molecular Microbiology 53(5): 1373-1383.

166. Gene disruption in Cryptococcus neoformans by in vitro transposition. G. G. Hu and J.W. Kronstad. Michael Smith Laboratories, The University of British Columbia, Vancouver, B.C. V6T 1Z4

Cryptococcus neoformans is a basidiomycetous fungus that infects immunocompromised and immunocompetent patients. Five serotypes (A, B, C, D and AD) have been documented based on antigenic differences in the polysaccharide capsule. In the present study, an in vitro transposon-based insertional mutagenesis strategy was developed and applied to C. neoformans. The goal was to establish a method to rapidly disrupt genes with interesting expression patterns, as determined by SAGE, or genes identified as part of our whole genome sequencing project for serotype B strains. To test the method, we disrupted the URA5 gene in the serotype A strain H99 and the CAP10 gene in the serotype B strains WM276 (VGI) and R265 (VGII). For the target DNA, we selected either plasmid DNA containing the cloned URA5 gene, or plasmid DNA containing the CAP10 gene from our 2 kb genomic library from the WM276 shotgun sequencing project. In the latter case, we have the end sequences of the 2 kb clones matched with the genomic sequence so that the clones for any gene can be selected for disruption. Modified transposons, containing the nourseothricin (NAT), neomycin (Neo), or hygromycin (Hyg) resistance cassettes, were randomly inserted into the target DNA (or gene) in an in vitro transposition reaction. Clones in which the gene of interest had been disrupted were identified by colony PCR and/or DNA sequencing and used to biolistically transform the strains of C. neoformans. Using this approach, we obtained H99 ura5:: NAT and R265 cap10:: Neo mutants. Confirmation and characterization of the mutants was performed by PCR, Southern blot and phenotype analysis. This technique was rapid and efficient for gene disruption compared to current strategies, and worked in a variety of strain backgrounds. The use of plasmid DNA from the shotgun-sequencing library combined with random in vitro insertions should allow high-throughout genetic analysis, particularly in serotype B strains of C. neoformans.

167. Haplotype comparison of an ultra gene dense island containing a G-protein Coupled Receptor gene in Phytophthora infestans. Iziah Sama, Keith O'Neill*, Francine Govers and Rays H.Y. Jiang. Plant Sciences Group, Laboratory of Phytopathology, Wageningen University, and Graduate School Experimental Plant Sciences, The Netherlands and *Broad Institute, Cambridge, MA 02141, USA.

Phytophthora infestans is a notorious oomycete pathogen causing potato late blight world wide. An ultra gene dense region of 8.9 kb containing the G-protein Coupled Receptor gene Pigcr1 is identified as a gene island in a BAC sequence comprising 107 kb. Haplotype comparison of the surrounding 40 kb region, using the sequence of a pair of allelic BACs, revealed very little polymorphism in this part of the genome. In addition to the high gene density (on average one gene per 1,8 kb) other remarkable features are found such as alternative splicing, shared core-promoter regions and overlapping 3' UTR's. To estimate the frequency of overlapping 3' UTRs in the P. infestans genome a bioinformatics tool was developed and a large P. infestans EST contig set was analysed. Synteny studies with the Phytophthora sojae and Phytophthora ramorum genomes demonstrated that the gene island is present in all three species and that the genes in this ultra gene dense region show a highly conserved order and orientation.

168. The external proteome of Histoplasma capsulatum. C Johnson¹, S Douangkesone, C Lichti, V Kruft, and J McEwen. J.L. McClellan VA Hospital, Little Rock, AR.

The human fungal pathogen, Histoplasma capsulatum, possesses a complex cadre of mechanisms for overcoming the immune system. Some of these are the ability to overcome the macrophage oxidative burst, to inhibit phagosome/lysosomal fusion and to regulate the internal pH of the phagolysosome. These abilities imply that H. capsulatum can regulate the conditions of its immediate external environment, possibly by secreting regulating molecules. In an effort to determine the composition of such an externally located regulating environment, we are characterizing the external proteome of H. capsulatum. To accomplish this, yeast proteins isolated from in vitro culture growth medium were resolved by 2-dimensional gel electrophoresis and the protein spots analyzed by mass spectrometry. Initial results indicate processed forms of the yps3 protein and H. capsulatum immunoreactive protein are part of the external proteome. Chitinase, in multiple modified forms, and the product of the catalase B gene, with a mass and amino terminus significantly different from that of the previously characterized mycelial M antigen (Zancope-Oliveira et al., 1999), also make up part of the yeast external proteome. As well, oxidative stress alters the composition of the external proteome by increasing the abundance of several of the proteome components, including the catalase B protein. In our earlier work, investigating gene expression at the level of transcription, we showed no significant increase in abundance of the catalase B transcript in yeast during oxidative stress (Johnson et al., 2002).

Our initial results show the external proteome of H. capsulatum has a number of secreted proteins in post-translationally modified forms. Chitinase is present in isoforms that differ by charge and mass and catalase B has a significant increase in mass. Furthermore, our findings show that regulation of catalase B gene expression is occurring at a post-transcriptional level under conditions of oxidative stress. Ongoing direct mass spectrometry analysis has identify additional components of the external proteome. Furthermore, we are attempting to define the specific post-translational modifications being made to the components of the external proteome.

169. Transcriptomics and metabolomics: An integrated approach to dissect endophyte-grass symbioses. Linda J. Johnson, Shalome Bassett, Gregory Bryan, Michael Christensen, Karl Fraser, Geoff Lane, Richard Johnson, Brian Tapper, Christine Voisey, Hong Xue, Susanne Rasmussen. AgResearch Grasslands, Tennent Drive, Private Bag 11008, Palmerston North, New Zealand.

Grass associations with fungal endophytes (genera Neotyphodium and Epichloë), display enhanced fitness as well as prolonged field persistence over their endophyte free equivalents. In order to understand how the endophyte globally affects its host plant's metabolism, and to gain a holistic understanding of the complex biological interactions that occur between a plant host and fungal symbiont, a combined metabolomics and transcriptomics approach using microarrays is currently being employed. The major metabolites from symbiotic interactions of perennial ryegrass (Lolium perenne) with endophytes N. lolii and E. festucae are being compared and analysed against endophyte-free perennial ryegrass. Transcriptomic and metabolite data generated from the perennial ryegrass-N. lolii interaction clearly show that the endophyte does affect its host plant's metabolism and these data will be presented.

An additional goal is to elucidate the biosynthetic function of fungal secondary metabolite gene clusters since there are numerous examples from other plant-fungal interactions where host functions can be manipulated via the production of fungal secondary metabolites. For this study, we have selected non-ribosomal peptide synthetase genes (NRPSs) of unknown function, as these genes are commonly associated with the production of important bioactive peptides. The metabolome and transcriptome of E. festucae endophyte strains containing targeted gene replacements in two of these pathways will be compared with the corresponding isogenic wild type E. festucae strains under culture conditions as well as in planta. From the integration of transcriptomic and metabolomic data, we hope to identify key regulatory and biochemical networks responsible for the maintenance of these plant-fungal associations.

170. Proteomic analysis of regulation and signalling. Meriel G Jones, Igor Morozov, Alisha Millard, Huw H Rees, Deborah Wood and Mark X Caddick. School of Biological Sciences, The University of Liverpool, UK

Our work investigates regulatory responses to extracellular pH and the availability of nitrogen and phosphate in the filamentous fungus A. nidulans. All are essential biological processes and have been implicated in fungal pathogenicity. The related species A. fumigatus is an important opportunistic pathogen in immunocompromised patients but the biology of its multifactoral pathogenicity determinants is poorly understood. The combination of proteomics with the recently released genome sequences of both A. nidulans and A. fumigatus, combined with the genetic resources offered by A. nidulans, offers an opportunity to advance knowledge of these processes. Although the pacC/pal system is known to regulate response to pH in A. nidulans, the combination of proteomics and mutant strains can provide leads to the extent, and identity, of additional factors. We have examined the response of the intracellular soluble and external secreted proteomes of A. nidulans to external pH during growth. We are also interested in applying proteomics to investigate signalling mechanisms, by looking at the primary response to an environmental shift. We are examining the response of the proteome to altered nitrogen status within two minutes to identify early consequences of nitrogen metabolite signalling. We will present a brief description of the biology of the systems under study and our recent work.

171. Serial Analysis of Gene Expression in Coccidioides posadasii. Ellen M. Kellner^1,2,3,4, M. Alejandra Mandel^1,2,3,4, John Galgiani^1,3 and Marc Orbach^1,2. ¹Valley Fever Center for Excellence and ²Department of Plant Sciences, University of Arizona, and ³Southern Arizona Veterans Administration Health Care System, Tucson, AZ. ⁴These authors contributed equally to this work.

Coccidioides posadasii is a primary fungal pathogen endemic to the desert southwest US, parts of Mexico and Central and South America. Coccidioidomycosis or Valley Fever is initiated upon the inhalation of arthroconidia from the soil. Coccidioides spp. are dimorphic and produce unique structures that are characteristic of its saprobic and parasitic growth phases. Mycelia and conidia are formed in soil whereas the host-specific phase involves the growth of spherules and production of endospores. We are taking a genomic approach to identify expressed genes in mycelia and spherules of C. posadasii in order to identify cell-type specific genes. Parasitic phase-specific genes may include regulatory factors controlling spherule morphogenesis, novel targets for pharmaceutical intervention or potential vaccine candidates. Serial analysis of gene expression (SAGE) was used to create libraries of expressed tags from 48-hour liquid-grown mycelium at 37 ^oC and 48-hour liquid-grown spherules at 39 ^oC. The SAGE libraries each contain several million tags of which approximately 100,000 were identified through sequencing. The genome of C. posadasii strain C735 was used to examine the flanking regions of expressed sequence tags to identify potential genes. Tags that were found to be enriched in 48hr spherule library relative to the mycelial library included a laccase, an alkaline phosphatase and an L-ornithine N5-monooxygenase. Tags present in highest numbers in both mycelia and spherules included many that mapped to genes encoding histones and ribosomal proteins. Real time RT-PCR was used to examine the expression patterns of selected gene classes during growth and spherule development both to validate the SAGE library and to extend our analyses.

172. Analysis of gene expression in the entomopathogenic fungus Beauveria bassiana. Eun-Min Cho and Nemat O. Keyhani. University of Florida, Microbiology and Cell Science, Bldg 981, Museum Rd. Gainesville, FL 32611

Beauveria bassiana is an important entomopathogenic fungus that displays a broad host range able to target a diverse number of arthropod species. Strains of B. bassiana have been selected for control of insects and other arthropods that act as disease vectors, crop and household pests, and have been used as biocontrol agents for limiting the spread of ecologically hazardous, invasive insects. B. bassiana produces at least three distinct single cell propagules including aerial conidia, vegetative cells termed blastospores, and microcycle conidia that can be isolated from agar plates, rich broth liquid cultures, and under conditions of nutrient limitation in submerged cultures, respectively. cDNA libraries were constructed from each B .bassiana cell type and a robust expressed sequence tagged (EST) dataset was generated. Approximately 2,000 clones from each library were sequenced and a unique sequence set was constructed. Comparative analysis of the expressed transcripts in each library indicated significant differences in gene expression pattern between the cell types in several broad categories including cell wall biosynthesis, secondary metabolism, and the production of proteases.

This work was supported in part by NSF grant # EF-0412137 (NOK).

173. Temperature-dependent Gene Expression in Aspergillus fumigatus Examined by DNA Microarray. H. Stanley Kim¹, Dan Chen¹, Christine E. Shamblin¹, Gregory S. May², Michael Anderson³, and William C. Nierman¹. ¹The Institute for Genomic Research, 9712 Medical Center Drive, Rockville, MD 20854, USA. ²The University of Texas M. D. Anderson Cancer Center, Houston, TX USA. ³University of Manchester, Oxford Road, Manchester M13 9PT, UK

Recent completion of the genome sequencing of the opportunistic human pathogen Aspergillus fumigatus provided a great opportunity to study its biology in greater detail. One of the characteristic phenotypes of A. fumigatus is its high adaptability to elevated environmental temperatures not tolerated by other members of the species. These environments, including soil, animal hosts, and compost, requires the fungus to adjust its biochemical constitution and physiology to survive and possibly resume growth after exposure to a temperature as high as 70o C. We examined genome-wide expression upon temperature shift from low to high and from high to low, within the range of 30¢ªC to 52¢ªC. We found that different genes have specific windows of expression over the temperature range and the duration at a given temperature. For example, many heat shock and stress-responsive genes were immediately up-regulated when the culture was shifted to high temperatures (above 48¢ªC), but some of these were also up at 37¢ªC after about an hour. On the other hand, many putative virulence genes were readily expressed at 37¢ªC, as if the fungus were in the host. This suggests that temperature change is an environmental signal by which the fungus adjusts its physiology in an environment appropriate manner. Another finding we made from this study is that many transposases, especially those of the Mariner-4 type, are highly expressed at high temperatures. This suggests that transposition might be activated at high temperatures perhaps as a means to allow improved adaptability to the high temperature stress through genome alterations. We are extending this study relating the temperature-responsive genes to the Comparative Genomic Hybridization (CGH) data with closely related stains.

174. The Alternaria brassicicola Genome Sequencing Project. La Rota C. Mauricio ¹, Brett Tyler ¹, Thomas Mitchell ², Susan Brown³, Dennis Knudson ³, Sandra Clifton ⁴ and Christopher Lawrence ¹. ¹Virginia Bioinformatics Institute, VPI & State University, Blacksburg, VA. ²NC State University, Raleigh, NC. 3. Colorado State University, Fort Collins, CO. 4. Washington University Genome Sequencing Center, St. Louis, MO.

This project aims to produce a draft genome sequence resource for the necrotrophic fungus, Alternaria brassicicola. A. brassicicola, with a predicted genome size of 30 Mb, is the causal organism of black spot disease and is an economically important pathogen of Brassicas. Moreover, A. brassicicola is representative of a genus of pathogens that inflicts substantial damage worldwide. The genus Alternaria as a whole has a dramatic impact on humans and human affairs, as it is also associated with mycotoxin contamination of food and food products, allergy, asthma, and opportunistic infection of immuno-suppressed patients. Importantly, A. brassicicola, which naturally infects cruciferous plants, has been used consistently as a model necrotrophic pathogen of Arabidopsis. No publicly available genome sequence exists for any true necrotrophic fungus that has a corresponding host genome publicly available. Excellent structural and functional genomic resources for A. brassicicola are currently being developed. In this project, the Washington University Genome Sequencing Center (WUGSC) will produce a whole genome shotgun sequence (WGS) of Altenaria brassicicola to 6-fold coverage. A BAC library will be used to generate a BAC-based fingerprint physical map of the genome. Assembly of the BAC library into contigs will be completed using restriction enzyme fingerprinting, and a minimum tile of BACs will be end-sequenced to provide a framework to facilitate assembly. A 0.3-fold coverage fosmid library will also be constructed and end sequenced at the WUGSC, as an additional resource to benefit assembly. Massively parallel signature sequencing (MPSS) will be used as an additional annotation feature for gene prediction. A bioinformatics platform for community finishing and annotation of the draft sequence will also be established. This project is funded by USDA-CSREES (Proj. # VAR-2004-05551).

175. Transcriptional Analysis of the Pathogenic Fungus Magnaporthe grisea During Rice Infection. Arnaud Lagorce*, Aurélie Darchis*, Fabien Munier*, Jean-Benoit Morel**, Rick De Rose*, Roland Beffa* et Marc-Henri Lebrun*. *Bayer CropScience / CNRS, 69263 Lyon Cedex 09, France. ** CIRAD, TA73/09, 34398 Montpellier Cedex 05, France

Magnapothe grisea – Oryza sativa is a relevant model to investigate the molecular mechanisms underlying fungal infection of plants. M. grisea infections are responsible of the main foliar disease of rice and cause severe crop losses world-wide. This fungus is well suitable to molecular biology since most genomic tools are available. Using microarrays representative respectively of 14 000 ORF from M. grisea, we performed a genome - wide transcriptional analysis to highlight fungal genes expressed during the infection process.

Leaves from the resistant rice cultivar IR64 carrying the resistance gene Pi33 were infected by the virulent M. grisea strain PH14. Infected rice leaves were harvested 5 days after infection when the first lesions appear. This time point corresponds to the active invasion and destruction of plant tissues by M. grisea. Out of the M.grisea 14 000 genes, we could detect hybridization signals for 3000 genes that are expressed during infection. The expression of a subset of genes that are either (1) highly expressed during infection or (2) are expressed during infection but not in the mycelium was monitored using real time RT-PCR. This strategy confirmed the expression during infection of 75% of these genes. Additionally, the expression of these genes was monitored in different fungal tissues (mycelium, spore, appressorium) revealing 8 genes that are only expressed during infection. These data were used for clustering analysis revealing 4 major expression groups (constrictively expressed, infection specific, spore/infection specific, spore/appressorium/infection specific).

In order to assess the role of these genes in the infection process, M. grisea mutants will be constructed by gene replacement or RNA interference. RNA interference constructs expressing hairpin structures under the control of the promoter from the highly expressed genes MPG1 were generated to inactivate known pathogenicity genes such as PLS1 and BIP1. Whereas, deletion mutants obtained by gene replacement of PLS1 or BIP1 are completely non-pathogenic, their inactivation by RNA interference only reduces the pathogenicity of the transformants by 50 to 90%. These observations illustrate the difficulty in turning completely off gene expression during infection using RNA interference.

176. Genome sequencing of the arbuscular mycorrhizal fungus Glomus intraradices. Peter J. Lammers^1,2, Swarnamala Ratnayaka², John B. Spalding², Jeongwon Jun², Tarik Ceylon¹, Ian R. Sanders³, J. Chris Detter⁴. ¹Dept. Chemistry and Biochemistry, New Mexico State University, Las Cruces, NM 88003, USA. ²Molecular Biology Program, New Mexico State University, Las Cruces, NM 88003 USA. ³Dept. Ecol & Evol, University of Lausanne, 1015 Lausanne, Switzerland. ⁴DOE Joint Genome Institute, Walnut Creek, CA 94598, USA

Arbuscular mycorrhizal (AM) fungi are coenocytic, multinucleate, asexual, obligate symbionts that colonize the majority of plant species. Approximately 150 species of AM fungi colonize over 200,000 species of plants resulting in a 5%-20% increase in photosynthesis and a large, if poorly understood contribution to the global carbon cycling budget. A genome sequencing project was funded by the DOE for a model AM fungus. Glomus intraradices was chosen because it a) has a worldwide distribution b) colonizes both annual and perennial plants of economic importance including maize, wheat, alfalfa, rice and the poplar tree c) can be grown aseptically in dual culture with Ri-transformed carrot roots and d) spores of this species are available commercially in large quantities (Premier Tech, Quebec Canada). Approximately 25 million base pairs of single-pass, unassembled sequence data was available by early December 2004 plus 3,500 pre-existing EST sequences. Key genes identified in the preliminary data include nutrient transporters, cell signaling and cell cycle regulation proteins, transcription factors, RNA polymerase subunits and splicing factors. An unexpected yet significant match was also found to a meiosis-specific recombination protein from yeast. Comparison of the genomic sequences with a mixed EST data set derived from the model legume Medicago truncatula colonized with Glomus versiforme was used to identify fungal genes expressed in symbiotic root tissue. Several dozen putative AM fungal sequences expressed in symbiotic root tissue were identified. Particularly well represented in the mixed EST database were Rab-like sequences highly similar to fungal proteins involved in protein trafficking and secretion.

177. Exploring new genomes, unveiling new enzymes: description of a new type of multicopper oxidase in Phanerochaete chrsysosporium and other fungal genomes. Luis. F. Larrondo¹, Rafael Vicuna¹ and Dan Cullen². ¹Departamento de Genética Molecular y Microbiología, Facultad de Ciencias Biológicas, Pontificia Universidad Católica de Chile, Santiago, Chile and Millenium Institute for Fundamental and Applied Biology.²USDA Forest Products Laboratory, Madison, Wisconsin 53705, USA

Laccases belong to the multicopper oxidase family, which also contains ascorbate oxidases, transmembrane Fet3 ferroxidases and ceruloplasmin. Until not long ago it was accepted that the only members of this family capable of oxidizing iron were fungal Fet3s and ceruloplasmin. It was recently shown that Phanerochaete chrysosporium, unlike other lignin-degrading fungi, does not have any sequence encoding conventional laccases (Larrondo et al., AEM 69: 6257-6263 2003). Instead it produces a novel multicopper oxidase possessing strong ferroxidase activity with catalytic parameters similar to those of yeast Fet3p. The physiological function of this protein (MCO1) is still uncertain.

The gene (mco1) is part of a cluster of 4 structurally related sequences located within a 25 kb region. All 4 genes are transcribed, but only mco1 has a clear secretion signal.

Multiple alignments of a large collection of fungal multicopper oxidase sequences, as well as structural comparisons of MCO1, show that these novel MCOs are closer to Fet3 proteins, than to conventional laccases. Together with iron permease Ftr1, Fet3 ferroxidase plays a key role in iron homeostasis. Clustal analysis of multicopper oxidases sequences from P. chrysosporium, Magnaporthe grisea, Fusarium graminearum and Ustilago maydis databases supports the existence of a new branch of the multicopper oxidase family. These fungal sequences group together, distinguishing themselves from the classical laccases or Fet3-ferroxidases by important features including: 1. the presence of the residue equivalent to Glu-185 from S. cerevisiae Fet3, which has been shown to be necessary for ferroxidase activity and; 2. the absence of a C-Terminal transmembrane domain, which is characteristic of Fet3-ferroxidases.

Based on the strong ferroxidase activity presented by P. chrysosporium MCO1, and by the presence of these sequences in plant-associated fungi genomes (not present in S. cerevisiae, C. albicans, N. crassa, P. anserina etc), we have hypothesized a possible role for this new type of enzyme in modulating redox reactions. Future studies will clarify the role and physiological significance of this new enzyme family.

178. An experimental screen for non-neutral evolving genes in symbiotic fungi. Le Quéré, A., Astrup-Eriksen, K., Rajashekar, B., Schützendübel, A., Canbäck, B., Johansson, T. and A. Tunlid. Lund University, Microbial Ecology, Lund, Sweden

Background: Ectomycorrhizae are formed by mutualistic interactions between fungi and the roots of woody plants. During the symbiosis, the two organisms exchange carbon and nutrients. Many fungal species are known to form ectomycorrhizae and there is a large variation in their host preferences and growth characteristics. The genomic basis for symbiotic adaptations and related phenotypic variations is poorly understood.

Results: The genomes of strains of the ectomycorrhizal fungus Paxillus involutus were compared using cDNA microarrays. A simple relative rate test was developed to identify genes that have evolved at a different rate within Paxillus as compared with the overall evolutionary rate in homobasidiomycetes. Approximately 17% of the genes were detected as non-neutral evolving in Paxillus. The cohort identified showed an overrepresentation of orphans, genes whose products are located at membranes, and genes encoding for components of stress/defense reactions.

Conclusions: We propose that data obtained from microarray-based comparative genomic hybridisation in combination with data on sequence divergence of the printed genes can be used for screening genomic variations associated with adaptive evolution. Using this approach, a subset of genes was identified in P. involutus that have an enhanced rate of evolution. Accelerated evolution may reveal functional differences in the corresponding proteins which are associated with symbiotic adaptations.

179. Proteomic analysis of Monascus strains reveals proteins associated with pigment production. Jo-Chi Wang, Huan-Yu Lin, Li-Ming Sung, Li-An Lai, Lina Huang, and Wen-Shen Chu. Bioresource Collection and Research Center, Food Industry Research and Development Institute, Hsinchu, Taiwan, ROC

Monascus fungi have been used for preparation of oriental fermented foods for centuries. They are rich sources of various secondary metabolites, such as pigments, antibiotics, mycotoxin and enzyme inhibitor. Organisms in the genus Monascus produce a mixture of six major pigments of polyketide origin. In addition to the well-known utilization as natural colorants, the pigments have been reported to exhibit some biological activities including antibacterial, antifungal and immunosuppressive effects. However, the mechanism of pigmentation has yet to be elucidated. A proteomic analysis of a non-pigment producing mutant, Monascus purpureus BCRC 38112, along with its parental strain Monascus purpureus BCRC 31499 having good red pigment production ability has been performed to investigate proteins related to pigment production. Proteomes of the two strains were separated by one- and two-dimensional electrophoresis followed by image analysis. A total of 80 protein spots with significantly different expression level were obtained. These protein spots were identified by tandem mass spectrometry and Mascot search system. Important qualitative differences between the strains were found in proteins involved in fatty acid metabolism, carbohydrate metabolism, ATP biosynthesis, protein folding and protein processing. This study shall provide useful information for further analysis of pigment production in Monascus spp..

180. A High Density SNP map for Neurospora crassa. Mi Shi¹, William Belden¹, Arun Mehra¹, Allan Froehlich¹, James E. Galagan², Bruce W. Birren², Jay C. Dunlap¹, and Jennifer J. Loros³. ¹Dept. of Genetics, Dartmouth Medical School, Hanover, NH 03755; ² Broad Institute, Cambridge, MA 02141; ³Dept. of Biochemistry, Dartmouth Medical School, Hanover, NH 03755.

The Neurospora crassa genome contains about 2000 map units. As a rule of thumb there are about 20 kb per 1% recombination in the midst of chromosome arms, suggesting that the population of a thousand SNPs on the Neurospora genome would allow rapid reduction of genetic map location to physical sequence of novel genes. In pursuit of this goal, we have compared sequences from two isolates of N. crassa – the Oak Ridge laboratory “wild-type” used as the basis for the genomic sequence completed in 2001 (Galagan et al Nature 422, 859 – 868, 2003) and the N. crassa isolate from Mauriceville Texas used previously as the basis for a widely used RFLP map (Metzenberg et al. Proc. Natl. Acad. Sci. U.S.A. 82:2067-2071) – and at this writing have populated a map with several hundred SNPs, a number we expect will soon exceed a thousand.

 Sequencing of cDNA libraries made from FGSC2225 (N. crassa Mauriceville, A) has revealed SNPs at a raw frequency of approximately 5 per 1000 nucleotides. These comprise the bulk of the newly identified SNPs. Additionally, we have focused attention on the centromeric region of LG 1 and on LG IV, deriving additional SNPs from directed local sequencing of FGSC 2225 DNA based on the available genomic sequence. In all cases, SNPs are detected amongst cross progeny using a PCR-based strategy (http://ausubellab.mgh.harvard.edu). Three oligos are needed to differentiate one SNP with certainty. Use of SNPs on LG IV has allowed the improvement of the current genome assembly, reversing the orientation of one supercontig as well as placing previously unordered sequences. Additionally, the SNPs have allowed the physical cloning of several genes of interest.

181. Development of a Genetic Linkage Map of Colletotrichum Lindemuthianum. F. Luna-Martinez^a, R. Rodriguez-Guerra^b, M. Victoria Campos^a and J. Simpson^a. ^aDepartment of Genetic Engineering, CINVESTAV, Unidad Irapuato, Apdo. Postal 629, Irapuato, Guanajuato, Mexico; ^bInstituto Nacional de Investigaciones forestales, Agricolas y Pecuarias (INIFAP), Campo Experimental del Bajio, Celaya, Guanajuato, Mexico.

A genetic map was being developed for C. lindemuthianum (Sacc. & Magnus) Lambs.-Scrib, the causal agent of anthracnose in Bean (Phaseolus vulgaris). Although classified as Deuteromycete, we have obtained a segregating F1 population of C. lindemuthianum from a sexual cross produced under laboratory conditions. We are currently developing a genetic map of the fungus using AFLP molecular markers. Currently the map is based on 49 F1 progeny derived from a cross between two Mexican isolates. A total of 131 markers have been mapped, covering 937 cM and spread over 14 major linkage groups. Significant deviation from the expected 1:1 segregation ratios was observed for only 16 markers (prob. of X²). C. lindemuthianum and will open the possibility to study the genetics of mating type specificity and avirulence factors in this pathogen.

182. Characterization of Aspergillus oryzae genome structure. Masayuki Machida¹, Motoaki Sano¹, Kiyoshi Asai¹, Toshitaka Kumagai¹, Taishin Kin¹, Hideki Nagasaki¹, Goro Terai², Takashi Komori², Toshihiro Tanaka³, Rie Igarashi³, Toshihiko Sawano³, Hisashi Kikuchi³, Toshihide Arima⁴, Osamu Akita⁴, Kenichi Kusumoto⁵, Yutaka Kashiwagi⁵, Keietsu Abe⁶, Katsuya Gomi⁶, Michio Takeuchi⁷, Tetsuo Kobayashi⁸, Hiroyuki Horiuchi⁹, Katsuhiko Kitamoto⁹, James E. Galagan¹⁰, David W. Denning¹¹, William C. Nierman¹², Jennifer Wortman¹², Jiujiang Yu¹³, Deepak Bhatnagar¹³, Thomas E. Cleveland¹³, A. oryzae genome sequencing consortium of Japan¹⁴. ¹Natl. Inst. Advanced Ind. Sci. Tech. ²INTEC Web and Genome Informatics, ³Natl. Inst. Tech. Eval., ⁴Natl. Res. Inst. Brewing, ⁵Natl. Food Res. Inst., ⁶Tohoku U., ⁷Tokyo U. Agric. Tech., ⁸Nagoya U., ⁹U. Tokyo, ¹⁰Broad Inst. MIT & Harvard, ¹¹U. Nottingham, ¹²The Inst. Genomic Res., ¹³South Regional Res. Center, ¹⁴Brewing Soc. Japan.

The genome sequencing of Aspergillus oryzae (AO) was completed. The total AO genome size was 36.8 Mb, from which more than 11,000 genes were predicted. Synteny analysis among AO, A. fumigatus (AF) and A. nidulans (AN) showed that AO has significantly more synteny breaks than between AF and AN and that AO genome had a mosaic structure consisting of AO-specific and common loci among the three species. Interestingly, metabolic genes were highly condensed on the AO-specific loci although they were repressed by mapping of ESTs. These results together with the existence of a pair of highly homologous and adjacently located genes in AO and Agrobacterium tumefacience suggest that AO might acquired foreign genes after it branched from AF and AN.

183. Development of an EST resource for the phytopathogenic oomycete Aphanomyces euteiches. Arnaud BOTTIN, Christophe JACQUET, Hélène SAN-CLEMENTE, Bernard DUMAS, and Elodie GAULIN. Université Paul-Sabatier, UMR5546 UPS-CNRS, Pôle de Biotechnologie Végétale, BP17, 31326 Castanet-Tolosan, FRANCE

The Aphanomyces genus is composed of saprophytic as well as parasitic species affecting either plant or animal hosts. Comparative genomics between Aphanomyces species, or between Aphanomyces and other oomycetes (e.g. Phytophthora), should help to identify the genes required for pathogenicity on plants or animals. Aphanomyces euteiches is a major root pathogen of Pea and Alfalfa in Northern America, and of Pea in Europe. Neither effective chemicals nor resistant cultivars are available to control the disease in Pea. In order to study the molecular basis of pathogenicity in A. euteiches, we are developping an EST sequencing project supported by the Genoscope d'Evry (France). Starting from mycelium grown in a standard medium or from starved mycelium placed in contact with host roots, two cDNA libraries are being constructed and 10.000 ESTs will be generated from each library. The sequences will be annotated and released on a publicly accessible web site hosted at our laboratory. Analysis of the sequences and of gene expression arrays will lead to a collection of putative pathogenicity genes. The functional characterization of these genes will involve their silencing after Agrobacterium-mediated transformation, which is under development at our laboratory. Progress of the project and preliminary results will be presented.

184. Gene modelling and Annotation of the complete Ustilago maydis Genome. Gertrud Mannhaupt, Ulrich Güldener and Werner Mewes. Max Planck Institute for Terrestrial Microbiology, Marburg, Germany

Our project was initiated to improve the predicted gene set of the Ustilago maydis genome, recently released by the Broad Institute. Manual inspection and correction are applied to set up a comprehensive Ustilago maydis Genome database. This will be done by comparing the predicted gene set with a set of gene prediction algorithms trained with various fungal coding sequences, available cDNAs, and a nonredundant protein database.

These combined data are graphically presented, enabling the annotator to retrieve information dynamically. Gene models will be changed according to cDNA matches and to obtain best homology to experimentally known proteins.

The protein set of the Broad Institutés release as well as a protein set independently generated at MIPS are already analysed and presented using the PEDANT system. PEDANT is a fully automated system using a wide spectrum of sequence analysis and structure prediction tools.

In the course of the project MUMDB (Mips Ustilago Maydis Data Base) is already established. Both protein data sets mentioned above are loaded into MUMDB omitting identical copies as well as all proteins derived from manually changed gene models. Beside annotating different features like INTERPRO, transmembrane domains or TargetP the proteins will be classified and assigned to functional categories. MUMDB will enable in depth intergenomic comparative analysis, especially using our recently developed SIMAP database containing all available protein sequences.

185. Cross species gene discovery using microarray analysis allows for the identification of D-galacturonic acid utilization pathway in Aspergillus niger and Aspergillus nidulans. Elena Martens-Uzunova, Peter Schaap, Johan van den Berg and Jacques Benen. Section Fungal Genomics, Laboratory of Microbiology, Wageningen University, Dreijenlaan 2, 6703 HA, The Netherlands

Contrary to the phosphorolytic degradation of D-galacturonic acid in bacteria and plants, non-phosphorolytic metabolism of D-galacturonic acid has been shown to occur in Aspergillus species (Uitzetter et al., 1985). This short pathway begins with the conversion of D-galacturonic acid to galactonate by an aldoketo reductase. Subsequently a dehydratase or racemase modifies galactonate to 2-keto-3-deoxygalactonate and an aldolase splits 2-keto-3-deoxygalactonate into pyruvate and glyceraldehyde. Although, A. nidulans mutants defective in the racemase and aldolase activities, known as gaaB and gaaA, respectively, were previously reported (Uitzetter et al., 1985), the genes encoding these activities have not been described yet. Transcriptome analysis of Aspergillus niger cultures grown on D-galacturonic acid, enabled the identification of a cluster of co-expressed genes that, among others, encode the necessary putative aldoketo reductase, racemase and aldolase. Sequence analysis of the genes orthologous to the A. niger gaaA and gaaB from the two A. nidulans D-galacturonate non utilizing mutants demonstrated the presence of point mutations in the coding regions of both A. nidulans gaaA and gaaB leading to the production of non-functional proteins and thus to the mutant phenotype. Comparative analysis of the A. niger gaaA and gaaB genes in publicly available fungal genomes revealed the presence of conserved orthologs to both genes in various distantly related fungi which suggests the presence of a common utilization pathway of D-galacturonic acid in fungi.

Reference: Uitzetter JH, Bos CJ, Visser J., J Gen Microbiol. 1986 132: 1167-72.

186. Location and Analysis of Syntenic Regions common to 3 Ascomycete Genomes. Diego Martinez, Michael R. Altherr. Los Alamos National Laboratory, P.O. Box 1663, Los Alamos, NM 87545.

The wealth of whole genome data now available within the ascomycete branch of fungi is able to speed the process and improve the quality of annotation. Here we utilize the close relationship of Neurospora crassa and Aspergilus nidulans to annotate and analyze the newly sequenced Trichoderma reesei genome. We aligned genetic markers from N. crassa and A.nidulans to each other and to T. reesei to search for syntenic regions that are conserved between two of the species or are conserved in all three of the species. Once identified these regions allow for the verification of genes based on position in the syntenic regions and other non-coding elements in the identified syntenic blocks to be verified by position as well as the standard amino-acid similarity. The creation of a database that include such large conserved regions will enrich our understanding of how genomes evolve or change over time, and will increase our knowledge of regions and motifs involved in regulation of gene transcription.

187. Withdrawn

188. A Genomic Assessment Of Novel Multi-functional Protein Sequences In The Soybean Pathogen Phytophthora sojae. Austin, Ryan ¹, Nicholas Provart¹, and Paul F. Morris ^{2 1}Department of Botany, University of Toronto, Toronto, Ont. CANADA; ²Biological Sciences, Bowling Green State University, Bowling Green OH, USA.

The close arrangement of genes in clusters in the Phytophthora sojae genome http://genome.jgi-psf.org/index.html may have facilitated a trend towards greater numbers of multifunctional proteins than in other eukaryotic organisms. For example, in the shikimate pathway leading to the synthesis of aromatic amino acids, we have identified four novel multifunctional proteins in addition to the pentafunctional enzyme (ARO1) that is also present in fungi. Multifunctional proteins in the oomycetes are not limited to metabolic pathways. They also include proteins involved in nuclear trafficking, signal transduction and membrane proteins. Novel multifunctional proteins are of interest for several reasons. A higher proportion of these genes may yield identifiable phenotypes by gene silencing. Multifuctional enzymes in key biosynthetic pathways are also a genetic resource that could be tapped to increase the nutritional quality of our foods. To identify the complete set of multifunctional proteins in the P. sojae genome, the predicted transcripts were downloaded from the DOE-JGI website. The sequences were subjected to BLASTx analysis against the NR database and protein motifs were identified by InterProScan software. The output was parsed into an SQL database on the Botany Beowolf Cluster. Candidate genes with multiple motifs, where the best BLASTx hit aligns to a fraction of the total length of the proteins are then selected for visual inspection on the DOE-JGI browser. Visual inspection is being used to evaluate models for inclusion in this data set. A partial survey has so far identified more than 50 such proteins.

189. Microtubule dynamics during hyphal growth and branching in Neurospora crassa. Rosa R. Mouriño-Pérez¹, Robert W. Roberson² and Salomon Bartnicki-García¹. ¹Departamento de Microbiología. Centro de Investigación Científica y Educación Superior de Ensenada. Ensenada, B. C. Mexico. ² School of Life Sciences. Arizona State University. Tempe, AZ. USA.

Neurospora crassa (N2524) with GFP-tagged microtubules (MTs) is an excellent strain to study dynamics of the microtubular cytoskeleton. By confocal microscopy, we analyzed MT behavior during hyphal growth and branching. Images were assembled in space and time for a precise picture of the 3-D organization of the microtubular cytoskeleton and a clear view of its dynamics. Cytoplasmic MTs are mainly arranged longitudinally along the hyphal tube. Straight segments are rare; most MTs show a distinct helical curvature with a long pitch and a tendency to intertwine with one another to form a loosely braided network throughout the cytoplasm. This study revealed that the microtubular cytoskeleton of a hypha advances as a unit: as the cell elongates, i.e., it moves forward by bulk flow. Nuclei appeared trapped in the microtubular network and were carried forward in unison as the hypha elongated. During branching, one or more MTs became associated with the incipient branch. Branch MTs are continuous with a set of adjacent MTs from the parent hypha. Originally transverse, MT orientation turned longitudinal as the branch elongated and the length and number of MTs increased. Although the exact origin of branch MTs remains an open question, the recorded evidence indicates both bulk insertion of parent-hypha MTs as well as de novo extension by anterograde polymerization. The latter conclusion was supported by FRAP studies showing evidence of MT assembly in the growing tip of the developing branch. Nuclei entered the branch entrapped in the advancing network of MTs.

190. Secondary metabolite biosynthetic gene clusters in filamentous fungi. William C. Nierman¹, Natalie Fedorova¹, Catherine M. Ronning¹, Resham Kulkarni¹, David Denning², Michael Anderson², Masayuki Machida³, Katsuhiko Kitamoto⁴, Kiyoshi Asai⁵, Joan Bennett⁶, Gary Payne⁷, Jiujiang Yu⁸, Deepak Bhatnagar⁸, Thomas E. Cleveland⁸, Jennifer Wortman¹, and Jacques Ravel¹. ¹The Institute for Genomic Research, Rockville, MD 20850, USA ²School of Medicine and Faculty of Life Sciences, The University of Manchester, Stopford Building, Manchester M13 9PT, UK, ³Research Center for Glycoscience, National Institute of Advanced Industrial Science and Technology , Tsukuba, Ibaraki, 305-8566, Japan. ⁴Department of Biotechnology, University of Tokyo, Bunkyo-ku, Tokyo 113-8657, Japan ⁵National Institute of Advanced Industrial Science and Technology, Tokyo, Japan. ⁶Tulane University, New Orleans, LA, USA ⁷Department of Plant Pathology, North Carolina State University, Raleigh, NC 27695, USA ⁸USDA/ARS, Southern Regional Research Center, 1100 Robert E. Lee Boulevard, New Orleans, Louisiana 70124 USA

Aspergillus fumigatus pathogenicity and competition for resources may be augmented by its numerous secondary metabolites. Secondary metabolite biosynthetic genes are typically organized in clusters containing most if not all of the structural enzyme genes required for product biosynthesis. The A. fumigatus genome contains 24 clusters with polyketide synthase, non-ribosomal peptide synthase or dimethylallyl tryptophan synthase genes which range in size from 20 kb to 60 kb and contain from 6 to 22 genes. These clusters are dispersed throughout the genome with only 8 of the 24 located in subtelomeric regions. Many A. fumigatus clusters contain regulatory genes, genes associated with resistance to the metabolite, and apparently unrelated genes with no obvious role in production of the metabolite in question. Twelve clusters contain genes encoding transcription factors involved in the regulation of polyketide production and other secondary metabolite biosynthetic pathways.

With the availability of several other sequenced Aspergilli and related filamentous fungi, we have undertaken a comparative analysis of the secondary metabolite biosynthetic genes and clusters. Some A. fumigatus clusters have orthologs in Aspergillus nidulans or Aspergillus oryzae while most are A. fumigatus specific. We will report on these and other comparative aspects of these clusters, including chromosomal localizations, cross-species cluster integrity, inter- and intra- species relatedness of paralogous genes and the likelihood of interspecific and perhaps interkingdom horizontal gene transfer.

191.Withdrawn

192.Withdrawn

193. Comparative genomics and cross-species microarray hybridization with Sordaria macrospora and Neurospora crassa. Minou Nowrousian¹, Carol Ringelberg², Jennifer J. Loros², Jay C. Dunlap², Christian Würtz¹, Stefanie Pöggeler¹, Ulrich Kück¹. ¹Lehrstuhl f. Allgemeine u. Molekulare Botanik, Ruhr-Universität Bochum, 44780 Bochum, Germany; ²Dartmouth Medical School, Departments of Genetics and Biochemistry, Hanover, NH 03755, USA

Comparative genomics is a powerful tool to identify functional regions within genomes. It requires sequence information from closely related organisms. Here, we present a comparison of 85 genes from Sordaria macrospora to their Neurospora crassa orthologues as well as a comparison of several regions of genomic DNA containing up to five genes (Nowrousian et al. Fungal Genet Biol 41: 285-292). These data show that S. macrospora and N. crassa share a high degree of sequence identity and that the genomes of the two species are highly syntenic. Thus, S. macrospora sequence information can be used to simplify the annotation of the N. crassa genome. Furthermore, N. crassa cDNA microarrays can be hybridized with S. macrospora targets. With this cross-species array hybridization, we have identified genes that are more than twofold up- or downregulated in three developmental mutants of S. macrospora. Among the genes that are up- or downregulated in the sterile mutant strains are the pheromone precursor genes, genes involved in cell wall biosynthesis, and putative regulatory genes. Gene expression analysis in double mutants using Northern blots and quantitative real time PCR indicates that several regulatory pathways form a complex network controlling gene expression during fruiting body development.

194. Proteome analysis of secreted proteins from Aspergillus oryzae in submerged and solid-state culture conditions. Ken Oda, Dararat Kakizono, Osamu Yamada, Haruyuki Iefuji, Osamu Akita, Kazuhiro Iwashita. National Research Institute of Brewing, 3-7-1, Kagamiyama, Higashihiroshima, Hiroshima, 739-0046, Japan

Filamentous fungi are widely used for the production of homogeneous and heterogeneous proteins. Aspergillus oryzae is focused on because of its ability of production of heterogeneous proteins in solid-state culture. To clarify whole overview of protein secretion by A.oryzae in solid-state culture, we carried out comparative proteome analysis of secreted proteins in solid-state culture and submerged culture. Secreted proteins, which were prepared from both culture for 0, 12, 24, 32, and 40hr, were subjected to 2-D electrophoresis, and protein spots of 40hr were identified by peptide mass finger printing using MALDI-TOF MS. Cell-wall bound proteins of 40hr submerged culture were also identified. One hundred twelve from solid-state and 110 spots from submerged culture were analyzed. From them, 82 spots were positively identified derived from 29 gene products in total. The protein secretion profile was revealed and the identified proteins were classified into 4 groups by secretion pattern. Northern analyses of 7 genes (TAA, bgl1, glaA, glaB, xynG1, xynG2, dppV) from these 4 groups were carried out to examine the mechanism that control solid-state specific protein production. These data suggested that A. oryzae altered the manner of protein secretion at posttranscriptional and transcriptional level with respond to culture condition.

195. Gene expression and functional analysis during spore germination and appressorium formation of Magnaporthe grisea. Y. Oh, N. Donofrio, H. Pan, T. Mitchell, R. A. Dean. Fungal Genomics Lab, North Carolina State University, Raleigh NC.

Magnaporthe grisea, the casual agent of rice blast disease, develops a specialized infection structure called appressorium to gain access into its host. To identify the genes uniquely expressed during spore germination and appressorium formation, we performed microarray experiments using the M. grisea 22K element oligonucleotide chip created in collaboration with Agilent Technology. The microarray contains 13,666 M. grisea and 7,124 rice elements. RNA was extracted from harvested spores as well as from spores incubated for 7 and 12 hours on either appressorium-inductive (hydrophobic) or non-inductive (hydrophilic) surfaces. At 7 hours of incubation on the inductive surface, the tip of the germ tube had started to swell and form an initial appressorium. Fully melanized appressoria were evident at 12 hours. Spores incubated on the non-inductive surface elaborated long germ tubes but no appressoria. Expression data was analyzed using the commercial software package Genespring and verified by reverse transcription PCR. Genes significantly up regulated during appressorium formation were subjected to functional characterization by targeted gene deletion. The results of gene expression studies and gene knock-out experiments will be presented.

196. Aspergillus flavus genome sequence: initial analysis. G. A. Payne, B.L. Pritchard, J. Yu, W.C. Nierman, R.A. Dean, D Bhatnagar and T.E. Cleveland.

Aspergillus flavus is a widely distributed filamentous fungus that normally occurs as a saprophyte in the soil or on decaying organic matter. It is pathogenic to plants, insects and animals and produces aflatoxin, one of the most toxic and carcinogenic naturally occurring compounds. A. flavus is also the second leading cause of aspergillosis in humans. A whole genome sequencing project funded by the USDA/NRI and USDA/ARS and conducted at TIGR is nearing completion. Preliminary studies of the 33.1 Mb draft sequence, which consists of 17 scaffolds representing 2995 contigs, indicate that the genome contains approximately 13,800 genes. Thus the A. flavus genome appears to be larger than either A. nidulans or A. fumigatus. Aspergillus species are of interest in part because of the large range of clinically and industrially important secondary metabolites they synthesize. An initial scan of the A. flavus genome for proteins involved in the production of these secondary metabolites revealed the presence of 24 putative polyketide synthases and 25 putative non-ribosomal peptide synthases. Manual annotation and analysis of the genome is being coordinated through North Carolina State University and will be made available at www.Aspergillusflavus.org.

197. Functional Genomics of early interactions ectomycorrhizal symbiosis: Study of Ras mediated signaling pathways. Podila, G.K. Dept. of Biological Sciences, University of Alabama in Huntsville, Huntsville, AL 35899, USA

The formation and maintenance of functional ectomycorrhiza requires dynamic signaling interactions between the plant host and the fungus, leading to recognition between the partners and establishment of a symbiotic organ. However, the signaling pathways involved in mycorrhizal symbiosis are not well understood. We have cloned a symbiosis-regulated gene, Lbras, from the ectomycorrhizal fungus Laccaria bicolor, which has been shown to be associated with signaling pathways controlling cell differentiation, proliferation and apoptosis in multicellular organisms. Functional similarity of Lbras to other known eukaryotic ras proteins was established by complementation in yeast and transformation of mammalian cells. Further analysis Ras-mediated signaling pathways in L. bicolor during symbiosis was studied through protein-protein interaction using yeast two-hybrid system. We have identified several interaction partners for Lbras including L. bicolor PF6.2, previously isolated and characterized in our laboratory. We have also identified Lbras interacting small hsp class proteins and vesicular transport proteins. These results will enable us to understand the ras-interacting partner(s) of the symbiosis-related pathway(s) and contribute towards understanding the signaling pathways involved in various stages of development of ectomycorrhizal symbiosis. The recent progress made in L. bicolor genome sequence analysis and its implications for deciphering signaling pathways will also be discussed.

198. Structure and dynamics of Magnaporthe grisea telomeres. Cathryn Rehmeyer¹, Motoaki Kusaba², Weixi Li¹, Chuck Staben¹, Bruce Birren³, and Mark Farman¹. ¹Department of Plant Pathology, University of Kentucky, Lexington, KY USA. ²Faculty of Agriculture, Saga University, Saga, Japan. ³The Broad Institute of Massachusetts Institute of Technology and Harvard University, Cambridge, MA USA.

In many eukaryotic microbes, telomere regions harbor genes with important roles in niche exploitation. Frequent recombination in these locations is believed to produce the variation that fuels adaptation. Some Magnaporthe grisea telomeres are known to harbor avirulence genes, suggesting that localization of genes to chromosome ends may contribute to pathogenic adaptation in this fungus. To test this hypothesis, we have sequenced approximately 40 kilobases from each telomere in M. grisea. We present a summary of genes identified at chromosome ends, focusing on those which may have roles in pathogenicity. Extensive duplications and exchanges are apparent among multiple chromosome ends, and may be a possible mechanism of variation for genes that reside in these regions. The characterization of sequence exchanges among ends, as well as the sequencing of a de novo telomere formed during culture, are providing further insights into the mechanisms of telomere hypervariability. In addition, we have identified tandem and dispersed repeats unique to subtelomeric regions and variable in copy number among ends. Of the 14 telomeres, 11 contain ordered arrays of short conserved repeats in subtelomeric regions, the composition and organization of which suggests a functional significance. Telomere-linked RecQ helicase genes are found exclusively between subtelomeric repeat arrays and the terminal repeats; we further characterize the members of this previously described gene family. Aside from yeast, this represents the first full characterization of a complete set of chromosome ends from an organism.

199. Absence of free iron induces the expression of non-ribosomal peptide synthetases in Aspergillus fumigatus. Reiber, Kathrin, Reeves, Emer P., Neville, Claire M., Winkler, Robert*, Gebhardt, Peter*, Kavanagh, Kevin and Doyle, Sean. National University of Ireland Maynooth

The filamentous fungus Aspergillus fumigatus is a human pathogen, causing devastating, invasive diseases mainly in the respiratory system of the immunocompromised hosts. Apart from an array of mycotoxins, contributing to its aggressive appearance, the fungus produces siderophores which are responsible for iron accumulation. Siderophore production appears to contribute to organismal virulence and it is becoming clear that non-ribosomal peptide synthetases (NRPS) are involved in hydroxamate siderophore biosynthesis. We report here the characterisation and analysis of two NRPS encoding genes, termed pesB and pesF. RT-PCR analysis confirms that expression of both pesB and F is up-regulated under conditions of iron limitation concomitant with increased siderophore production. In addition, a protein corresponding to the predicted molecular mass of pesB (~200 kDa) has been identified by MALDI TOF peptide mass fingerprinting and MALDI LIFT TOF/TOF MS peptide sequencing following SDS-PAGE and 2D-PAGE separation of protein extracts obtained from mycelia grown under iron-limiting conditions. Proteomic characterisation of other high molecular mass proteins is ongoing to enable identification of other siderophore synthetases in Aspergillus fumigatus. Because siderophore synthetases are unique in bacteria and fungi they display an attractive target for anti-fungal chemotherapy to reduce the virulence of Aspergillus fumigatus.

200. Functional Genomics and Proteomics of Trichoderma Antagonist Strains for Industry and Agriculture. Rey Barrera, M^*., Llobell, A. and Monte, E.. ^*Newbiotechnic, S.A. (NBT) Parque Industrial de Bollullos A-49 (PIBO). 41110 Bollullos de la Mitación. Sevilla. Spain.

The genome of the Trichoderma antagonists is poorly surveyed compared with other model microorganisms, due to the great diversity of its species, the absence of optimised systems for its exploration, and the great variety of genes expressed under a wide range of ambient conditions. The EU-project (QLK3-2002-02032) "TrichoEST" started under the 5th Framework Programme "Quality of Life" December 1. 2002. The aims are to identify genes and gene products from Trichoderma spp. with biotechnology value, to assess their industrial potential and, to exploit and commercialize them in concert with EU biotechnology strategy. The project is not a sequencing programme. Instead, it develops integrated functional genomic and proteomic approaches and innovative use of bioinformatics, leading to rapid exploitation of genes and gene products and their transformation into industrial processes. Technologies achieved will have applications in a range of agro-industrial, environmental and medical activities, including innovations in agricultural pest and pathogen management, novel antibiotics, and enzymes with industrial uses involving processing of plant-derived matter, including animal feed, composting and bioremediation. The project is coordinated by Dr. Manuel Rey Barrera, Newbiotechnic S.A., Spain, and the partners, besides us, are Dr. Fabrice Lefevre, Proteus S.A., France, Dr. Antonio Llobell González, University of Seville, Spain, Dr. Enrique Monte Vázquez, University of Salamanca, Spain, Dr. Matteo Lorito, University of Napoli, Italy, Dr. Christian Kubicek, Technical University of Vienna, Austria, Dr. Gabi Krczal, Centrum Grüne Gentechnic, and Dr. Paul Cannon, CAB International, UK.

201. Identification of genes in the Gibberella zeae sexual reproductive pathway by cross genome comparison. Barbara Robbertse and Gillian Turgeon Department of Plant Pathology, Cornell University Ithaca, NY 14853

The sexual development pathway of Saccharomyces cerevisiae (Sc) is almost certainly the best-understood fungal developmental pathway. How conserved is this pathway in filamentous fungi, specifically in Gibberella zeae (Gz)? Large-scale functional analysis of Sc, large-scale expression analysis of S. pombe (Sp), and perithecium-specific EST sequences of Neurospora crassa (Nc), Magnaporthe grisea (Mg), and Gz are available to study this question. We downloaded the predicted protein sequences of Sc, Sp, Gz, Nc, and Mg and performed blastp searches combined with Tribe-MCl and INPARANOID analyses to find all possible orthologs among the protein sequences of these fungi. In total, 3,034 putative Sc orthologs and 3,311 putative Sp orthologs were identified in Gz. Four large-scale functional studies and the SGD database have identified a non-redundant set of 610 genes as essential for Sc sexual development. Thus, the candidate Gz ortholog data set was filtered for all Sc genes with a mating/meiosis/sporulation-specific phenotype when deleted and for all Sp genes induced during mating, meiosis or sporulation. Approximately 44% of the 610 essential Sc genes have orthologs in Gz. Of these, 47 Gz genes were identified that had matches to perithecial ESTs. Fifty two percent (528) of induced Sp genes were orthologous to Gz and 82 of these Gz orthologs had matches to perithecium specific ESTs. The 47 Sc-Gz orthologs and 82 Sp-Gz orthologs had only 7 Gz orthologs in common. Deletion studies are underway to confirm that the Gz orthologs, identified by cross genome comparison, are essential for sexual development.

202. Defining the Secretome of Fusarium graminearum. John S. Scott-Craig, Kohhei Otani, Frances Trail$, Heather E. Hallen$, Brett Phinney*, Janet M. Paper, Neil Adhikari, and Jonathan D. Walton. MSU-DOE Plant Research Laboratory, *Genome Technology Support Facility, $Department of Plant Biology, Michigan State University, East Lansing, Mi 48824, USA

 We are using mass spectrometric-based proteomics to identify the secreted proteins of the plant pathogenic fungus Fusarium graminearum. Proteins and peptides are being identified from culture filtrates and mycelial wall extracts of F. graminearum grown on eight different media. The range of carbon sources used includes purified maize and carrot cell walls, pectin, xylan, casein, collagen, dried milk and sucrose. Intercellular wash fluids from F. graminearum-infected wheat heads are also being extracted and analyzed. By combining the publicly available, high-quality sequence of the F. graminearum genome with state-of-the-art instrumentation for proteomics, an analysis of the "secretome" (that is, the totality of secreted proteins) of a plant pathogenic fungus is now possible. These results complement and extend past biochemical, genetic, and genomic studies on secreted fungal proteins and their regulation. The results will contribute novel and useful information of interest to fungal biologists and especially to plant pathologists. In particular, the work should lead to the identification of proteins or peptides that contribute to the interaction between F. graminearum and its host.

203. Conservation of synteny among filamentous fungi. Michael R. Thon¹ and Ralph A. Dean². ¹Program for the Biology of Filamentous Fungi, Department of Plant Pathology and Microbiology,Texas A&M University, College Station, TX 77843, USA. ²Center for Integrated Fungal Research, Department of Plant Pathology, North Carolina State University, Raleigh, NC 27695, USA

We performed pairwise comparisons between Magnaporthe grisea chromosome VII and the genome sequences of Fusarium graminearum, Aspergillus nidulans, and Neurospora crassa to identify regions of conserved synteny among the fungi. Segmental homologs (i.e. syntenic blocks) containing 5 or more genes (P < 0.001) were identified using the algorithm and statistical test implemented in the FISH software package. 21 blocks were found in N. crassa, 17 in F. graminearum, and 2 in A. nidulans. In general, the blocks were roughly co-linear and interspersed with intervening, non-syntenic genes. The syntenic blocks were restricted to N. crassa chromosome I, F. graminearum chromosome II and IV, and A. nidulans chromosome III, suggesting that these chromosomes share common ancestry and/or a preference for intrachromosomal rearrangements. A segment of conserved synteny was found that spans all four species that is 50-75 genes in length, of which eight genes are conserved in all four species and 14 genes are conserved in three out of four species. We will present a summary of conservation of synteny between M. grisea chromosome VII and the genomes of related fungi, and discuss the implications for the evolution of fungal genomes.

204. Nutrient sensors in Candida albicans: role of Gpr1 in response to glucose limitation and amino acid deprivation. Hélène Tournu, Mykola Maydan, Larissa de Rop, Joke Serneels, Patrick Van Dijck Flanders Interuniversity Institute for Biotechnology, Department of Molecular Microbiology, VIB10, Laboratory of Molecular Cell Biology, K.U. Leuven, Kasteelpark Arenberg 31, B-3001 Leuven, Belgium

We and others have recently shown that the G-protein-coupled receptor Gpr1 is involved in the morphogenic transition between yeast and hyphae in the human pathogen Candida albicans. Phenotypic analyses on hyphal-inducing media such as the amino acids based medium Lee showed a clear defect in filamentous growth in a strain deleted for GPR1. Because of the nature of that particular medium, we investigated further the putative ligand of CaGpr1. Although glucose has been shown to be the ligand of Saccharomyces cerevisiae Gpr1, a strain lacking a functional Gpr1 in C. albicans is still capable of inducing a transient cAMP increase in response to the addition of glucose. Amongst the seven amino acids included in Lee's medium, methionine is essential to the wild type hyphal growth, but addition of methionine does not restore filamentation in a gpr1 null. Could CaGpr1 have an amino acid as a ligand? Preliminary results show that addition of methionine to cells growing in presence of limited amount of glucose induces an increase of cAMP in wild type but not in gpr1 null. Addition of a methionine analogue does not induce filamentation, and picomole amounts of L-methionine are sufficient to trigger the morphogenic switch in wild type cells suggesting a possible sensing mechanism.

205. Divergence in gene expression related to variation in host specificity of an ectomycorrhizal fungus. Antoine Le Quéré¹, Andres Schützendübel, Balaji Rajashekar¹, Björn Canbäck¹, Jenny Hedh¹, Susanne Erland¹, Tomas Johansson ¹and Anders Tunlid¹. ¹ Department of Microbial Ecology, Lund University, Lund, Sweden; ² Present address: Department of Forest Botany and Tree Physiology, Georg-August-University, Göttingen, Germany.

There is a considerable variation in the degree of host specificity among species and strains of ectomycorrhizal fungi. In this study, we have for the first time shown that this variation is associated with quantitative differences in gene expression, and with divergence in nucleotide sequences of symbiosis regulated genes. Gene expression and sequence evolution were compared in different strains of the ectomycorhizal fungus Paxillus involutus including Nau, that is not compatible with birch and poplar, and the two compatible strains Maj and ATCC200175. On genomic level, Nau and Maj were very similar. In contrast, 66 out of the 1,075 genes were differentially expressed in Maj compared to Nau after contact with birch roots. Thirty-seven of these symbiosis-regulated genes were also differentially expressed in the ATCC strain. Comparative analysis of DNA sequences of the symbiosis-regulated genes in different strains showed that two of them have evolved at an enhanced rate in Nau. The sequence divergence can be explained by a decreased selection pressure, which in turn is determined by lower functional constraints on these proteins in Nau as compared to the compatible strains.

206. Genome sequences of Phytophthora sojae and P. ramorum shed new light on their evolution and pathogenicity. Brett M. Tyler¹, Sucheta Tripathy¹, Igor Grigoriev², Daniel Rokhsar^2,3, Jeffrey Boore^2,3 and many others. ¹Virginia Bioinformatics Institute, Virginia Tech, Blacksburg VA 24061; ²DOE Joint Genome Institute, Walnut Creek, CA; ³ University of California, Berkeley.

The approximately 60 species of Phytophthora are all destructive pathogens, causing rots of roots, stems, leaves and fruits of a huge range of plants important to agriculture, horticulture and natural ecosystems. These pathogens are not fungi but belong to the oomycete group of the kingdom Stramenopiles. We have determined draft DNA sequences of the 95 Mb genome of the soybean pathogen Phytophthora sojae and of the 65 Mb genome of the sudden oak death pathogen Phytophthora ramorum, to a depth of 9x and 7x respectively. The genome sequences show clearly for the first time that oomycetes, which are heterotrophic, have evolved from a photosynthetic ancestor, which they share with photosynthetic Stramenopiles such as diatoms. The Phytophthora genomes also contain several bacterial genes not present in any other eukaryotes. The genomes show rapid diversification of proteins associated with plant infection such as hydrolases, ABC transporters, protein toxins, proteinase inhibitors and avirulence gene products, but relatively little diversification of biosynthetic genes for metabolite toxins. The P. sojae and P. ramorum genomes show substantial synteny (colinearity) except in regions encoding putative pathogenicity genes. The genome sequences have also enabled new tools such as an Affymetrix^TM gene expression microarray for P. sojae together with its host soybean, and an Affymetrix^TM SNP chip for tracking P. ramorum spread.

Supported by USDA grant 2002-35600-12747 and NSF grant MCB-0242131 to B. Tyler and DOE Genomes to Life funding to the DOE JGI.

207. The interplay of the pectinase spectrum of Aspergillus niger as revealed by DNA microarray studies. Elena Martens, Peter Schaap, Johan van den Berg and Jacques Benen. Fungal Genomics group, laboratory of Microbiology, dept. AgroTechnology and Food Sciences, Wageningen University, Dreijenlaan 2, 6703 HA Wageningen.

The saprophytic fungus Aspergillus niger is an efficient producer of extracellular enzymes. Previous research has revealed that many of these enzymes show carbohydrate modifying activities. These enzymes are Generally Regarded As Safe (GRAS status) and are therefore widely used in the food industry.

Recently DSM Food Specialties solved the genomic sequence of this fungus. Based on the new data, it was estimated that only a fraction of the potential of enzymes secreted by A. niger is currently characterized. Database mining using the proprietary genome sequence has resulted in the identification of at least total thirty-eight genes encoding enzymes involved in the depolymerisation of the backbone of the complex polysaccharide pectin. Additional enzymatic activities are required to degrade the arabinogalactan side chains attached to the main chain and to remove methyl and acetyl esters that are present in this main chain.

By applying the powerful technology of DNA microarrays we have sought to gain insight into the complex regulation of the expression of all the genes involved in pectin degradation. For this we have cultivated A. niger on sugar beet pectin and on the monomeric sugar constituents of pectin, viz galacturonic acid, rhamnose and xylose, and subsequently analysed the corresponding transcriptomes using microarrays. We will report on our findings concerning the regulation of the expression of the genes involved in the degradation of pectin and the consequences for the interplay of the encoded (novel) enzymes.

208. Differential gene expression by Metarhizium anisopliae growing in root exudate and host (Manduca sexta) cuticle or hemolymph reveals mechanisms of physiological adaptation. Chengshu Wang, Gang Hu and Raymond J. St. Leger. Department of Entomology, University of Maryland, 4112 Plant Sciences Building, College Park, MD 20742, USA.

Like many other fungal pathogens Metarhizium anisopliae is a facultative saprophyte with both soil-dwelling and insect pathogenic life-stages. In addition, as M. anisopliae traverses the cuticle and enters the hemolymph it must adapt to several different host environments. This implies considerable physiological plasticity. In this study we used expressed sequence tags (ESTs) and cDNA microarray analyses to demonstrate that physiological adaptation by M. anisopliae to insect cuticle, insect hemolymph, bean root exudate (a model for life in the soil) and nutrient rich Sabouraud dextrose broth (SDB) involves different subsets of genes. Overall, expression patterns in cuticle and hemolymph clustered separately from expression patterns in root exudates and SDB, indicative of critical differences in transcriptional control during pathogenic and saprophytic growth. However, there were differences in gene expression between hemolymph and cuticle and these mostly involved perception mechanisms, carbon metabolism, proteolysis, cell surface properties and synthesis of toxic metabolites. Some of these differences were of clear adaptive significance. Thus, cuticle-degrading proteases were switched off during growth in hemolymph, avoiding activation of the host prophenoloxidase system. In other cases differences suggested previously unsuspected stratagems of fungal pathogenicity that can be tested experimentally. Examples include large-scale changes in gene expression of cell wall proteins in hemolymph, as these may alter interactions with host defense responses. The results of this study will initiate a detailed molecular analysis of the ecological traits that adapt an ascomycete to live both in the soil and as a pathogen of insects.

209. The Evolutionary Genomics of 5S rRNA Genes in Filamentous Fungi: Patterns of Multigene Family Evolution and Gene Amplification. Todd J. Ward and Alejandro P. Rooney. Microbial Genomics & Bioprocessing Research Unit, National Center for Agricultural Utilization Research, USDA Agricultural Research Service, Peoria, IL 61604

The 5S gene is the smallest of the four nuclear ribosomal RNA (rRNA) genes in eukaryotes. In some species the gene is part of the regular rRNA tandem array, while in others the gene is dispersed. The 5S rRNA genes of filamentous fungi form a large multigene family of 50 to 100 copies. During the course of our studies on the genomic evolution of filamentous fungi, we found that these genes display varying degrees of divergence from one another, suggesting that they do not evolve in a concerted manner. This pattern is quite unusual because nuclear rRNA genes represent a paradigm of concerted evolution. The purpose of this study was to characterize the patterns of multigene family evolution and evolutionary genomics of 5S rRNA genes in the genomes of representative filamentous fungi. Our studies show that the genes undergo birth-and-death evolution under strong purifying selection. Furthermore, the rate of gene turnover appears to be quite high in these genomes. Because the 5S genes in these species are dispersed across the genome, our findings indicate that the mechanisms controlling the multiplication and movement of 5S genes across the genome are highly dynamic. As such it appears that a process resembling retroposition controls 5S rRNA gene amplification, dispersal and integration in the genomes of filamentous fungi.

210. Proteomic Identification And Functional Analysis of Extra-Cellular Proteins from Magnaporthe grisea. SHENG-CHENG WU, Kumar Kolli, Erica Berelc, Evan Conroy, Peter Albersheim, Alan Darvill, Ron Orlando. Complex Carbohydrate Center, University of Georgia, Athens, GA 30602

Magnaporthe grisea, the rice blast fungus, secretes copious amount of extracellular proteins (ECPs) in response to various growth conditions. These ECPs are believed to play indispensable roles in diverse biological activities such as nutrition uptake, growth and cell-cell communication and molecular interactions between the pathogen and its host. We have used high-throughput proteomics technologies such as multidimensional liquid chromatography-mass spectrometry (MDLC/MS) to identify a number of M. grisea ECPs. Although about half of the ECPs are putatively associated with a known function such as glycoside hydrolysis, the others are novel, with a few being "infection growth-specific". Gene knockout and necrosis/alkalinization analyses are being carried out to evaluate some ECP's role in pathogenicity and fungus-plant interactions, and tandem affinity purification (TAP) will assist with identifying host proteins possibly interacting with the ECPs. [Supported by U.S. Department of Energy grants DE-FG05-93ER20221 and DE-FG02-93ER20097, the National Science Foundation grant NSF-9626835 and the National Institutes of Health grant P41RRR05351]

211. Draft sequence of Monascus sp. BCRC 38072 genome. Gwo-Fang Yuan, Chun-Lin Wang, Chien-Chi Chen, Li-Ling Liaw, Wen-Shen. Chu, Shih-Hau Chiu, Jyh-Wei Chen, Tzu-Pei Chiu, Li-Ming Sung, Ing-Er Hwang, Hsiao-Chi Peng, Hui-Wen Chang, Mei-Chih Chien, Yi-Ju Liu and Chii-Cherng Liao,. Bioresource Collection and Research Center, Food Industry Research and Development Institute, Hsinchu, Taiwan, ROC

The filamentous fungus Monascus sp. can produce various secondary metabolites useful as food additives and/or pharmaceuticals. A Monascus strain, BCRC 38072, isolated from red rice can produce monacolin K, a cholesterol serum synthesis inhibitor. The genome sequence of Monascus sp. BCRC 38072 was obtained by whole-genome shotgun of a variety of clone types (plasmid, fosmid and BAC) at 11-fold sequence coverage. The Arachne package was used to assemble the genome sequence. A total of 673,853 highly qualified reads were input into the assembly program. The resulting draft consists of 709 contigs, larger than 2 kb in length, with the total length of 26.8 Mb. Seventeen major supercontigs were assembled covering 94.8% of the whole assembly length and 422 ungrouped contigs were assembled in the rest 5.2%. The N50 lengths of supercontig and contig are 2.5 Mb and 224 Kb, respectively. A total of 6,326 genes were predicted with 6,226 encoding proteins longer than 100 amino acids. Among the predicted proteins, 1,739 (27.5%) and 2,312 (36.6%) have significant matches to the known Saccharomyces cerevisiae proteins and proteins from SwissProt, respectively.