Population and Evolutionary Biology

75. Some Like it Hot: differences in Thermotolerance of Coccidioides species

Bridget M. Barker1,2, Sarah N. Statt1, John N. Galgiani2, Marc J. Orbach1. 1Department of Plant Sciences, Division of Plant Pathology and Microbiology, University of Arizona, Tucson, AZ. 2 Southern Arizona Veterans Administration Health Care System, Tucson, AZ.

Recent division of Coccidioides into two species, C. immitis and C. posadasii, was based on molecular phylogenetic evidence defining two distinct clades, but very little phenotypic difference has been seen between the two species (Fisher et al. 2002). A subtle salt tolerance difference was observed where initial growth was slower for C. posadasii, but by two weeks,was not significant. While surveying strains to develop selective conditions for recovery of C. posadasii from the desert soil, we have identified a significant difference in growth rate. A subset of 22 C. posadasii and 34 C. immitis isolates, representing its full geographical range were screened at 37ºC and 28ºC. Radial growth rate was measured over 16 days. We found that C. posadasii grew significantly faster at 37ºC, when compared to C. immitis. There was no significant difference in growth rate at 28ºC. The division of Coccidioides into two species has caused concern among many people in the clinical setting. Diagnosis of coccidioidomycosis does not require that clinicians know the species that caused the disease, but there are questions about disease severity and treatment. Our research indicates there are differences related to temperature, which is the first significant phenotypic difference between the species and is easily assayable in a clinical setting.

76. The relative contributions of seedborne and windborne inoculum to foliar epidemics of Phaeosphaeria nodorum. Rebecca S. Bennett1, Michael G. Milgroom1, Barry M. Cunfer2, and Gary C. Bergstrom1. 1Department of Plant Pathology, Cornell Univ., Ithaca, NY. 2Department of Plant Pathology, Univ. of Georgia, Griffin, GA.

Population genetic and epidemiological data, when examined separately, have resulted in different hypotheses about the predominant source of inoculum in the Phaeosphaeria nodorum-wheat pathosystem, i.e., sexually-derived, windborne ascospores versus asexual, seedborne inoculum, respectively. We are addressing these competing hypotheses by using a mark-recapture experiment in which seedborne isolates of P. nodorum can be identified by rare alleles. We planted infected wheat seed in experimental plots and sampled P. nodorum throughout the growing season. If inoculum comes primarily from seed, we expect to find that isolates collected have the same rare alleles as in our released isolates. If the inoculum is coming primarily from immigrant, windborne ascospores, we expect to find that isolates collected from different experimental plots are genetically similar to each other but distinct from artificially constructed seed populations. Analyses of early foliar samples indicate that infected seed was likely to be more important that ascospore inoculum. Data from late foliar samples and harvested seed will also be presented.

77. Cross fertility of lineages of Gibberella zeae. R.L. Bowden1, J.F. Leslie2, Jungkwan Lee2, and Yin-Won Lee3. 1USDA-ARS Plant Science and Entomology Research Unit, Manhattan, KS; 2Department of Plant Pathology, Kansas State University, Manhattan, KS; 3School of Agricultural Biotechnology, Seoul National University, Seoul, Korea

O'Donnell et al. (2000) divided Gibberella zeae into seven phylogenetic lineages and these were later extended to nine and given species rank (O'Donnell et al. (2004)). The objective of this study was to estimate the potential for genetic exchange between these lineages by quantifying cross fertility in the laboratory. Crosses were conducted on carrot agar as described by Bowden and Leslie (1999). Three strains of G. zeae lineage 7 with an insertion in the MAT1-2 locus that renders them heterothallic were used as females (Lee et al., 2003). Standardized suspensions of macroconidia from strains of each of the nine lineages were used as males to fertilize the females. Ten days after fertilization, carrot agar plates were inverted over water agar plates and fertility was measured by counting ascospores deposited overnight. Homothallic cultures and unfertilized heterothallic strains served as controls. Cross fertility was highly variable and differed for the three female strains. All males from all lineages produced viable progeny with at least two lineage 7 female strains. Certain pairings of lineage 7 females with males of other lineages showed fertility levels comparable to lineage 7 x 7 crosses. No evidence for consistent fertility barriers between lineage 7 and the others was found.

78. Conservation of non-ribosomal peptide synthetases in closely related Cochliobolus species. Kathryn E. Bushley and B. Gillian Turgeon. Department of Plant Pathology, Cornell University, Ithaca, NY, 14853

Non-ribosomal peptide synthetases (NRPSs) are multimodular enzymes, found in ascomycete fungi and bacteria that make peptides (NRPs) through a thiotemplate mechanism independent of ribosomes. NRPs are structurally diverse and have a broad spectrum of biological activities, both beneficial and harmful. A minimal NRPS module, encoded by a non-ribosomal peptide synthetase gene (NPS), is composed of an adenylation (AMP) and a thiolation domain. A condensation domain functions in peptide bond formation and elongation of the nascent peptide. Previous studies suggest that genes encoding NRPSs have a discontinuous distribution in the genomes of distantly related filamentous ascomycetes (Kroken et al, unpublished). Various evolutionary processes could explain this pattern: 1) gene duplication and differential loss, 2) recombination 3) gene conversion, and 4) horizontal gene transfer. We are investigating diversity and evolutionary history of NPSs among closely related taxa, using three Cochliobolus species (C. heterostrophus (Ch), C. carbonum (Cc), and C. victoriae (Cv) to address which mechanism(s) may explain the diversity across divergent taxa. For this, sequence data from conserved motifs of 25 AMP domains, extracted from 11 NPSs in identified in Ch (Lee et al, in press), were used to design degenerate and specific primers to identify counterparts of the known ChNPSs, as well as previously unidentified NPSs, in Cv and Cc. Results suggest that the set of ChAMP domains is largely conserved in Cv and Cc, however differences exist. Six previously unidentified Cv- and one Cc-unique AMP domain have been identified, while attempts to amplify ChNPS8 and ChNPS11 AMP domains and one ChNPS9 AMP domain have failed. Current work focuses on whether AMP domains within each Chgene are syntenous across species. Initial results suggest that the order of AMP domains in at least two genes, NPS1 and NPS2, is conserved across species. These data will allow us to evaluate whether recombination plays a role in generating new NPSs.

79. The Evolutionary Genetics of Dikaryosis Versus Diploidy. Travis A. Clark and James B. Anderson. Department of Botany, University of Toronto, Ontario, Canada.

While many eukaryotic organisms exist as diploids, with the two gametic genomes residing in the same nucleus, most basidiomycete fungi exist as dikaryons, with the two gametic genomes residing in separate nuclei that are physically paired. To determine if haploid monokaryotic and dikaryotic mycelia of Schizophyllum commune adapt to novel environments under natural selection, we serially transferred replicate populations of each ploidy state on minimal medium for 24 months (17,500 generations). Dikaryotic mycelia responded to selection with increases in growth rate, while haploid monokaryotic mycelia did not. To determine if the haploid components of the dikaryon adapt reciprocally to one another's presence over time, we recovered the intact haploid components of dikaryotic mycelia at different time points (without meiosis) and mated them with nuclei of different evolutionary histories. We found evidence for coadaptation between nuclei in one dikaryotic line, in which a dominant deleterious mutation in one nucleus was followed by a compensatory mutation in the other nucleus; the mutant nuclei that evolved together had the best overall fitness. In other lines, nuclei had equal or higher fitness when paired with nuclei of other histories, indicating a heterozygote advantage. To determine if genetic exchange occurs between the two nuclei of a dikaryon, we developed a 24-locus genotyping system based on single nucleotide polymorphisms to monitor somatic exchange. We observed genetic exchange and recombination between the nuclei of several different dikaryons, resulting in genotypic variation in these mitotic cell lineages. We are currently conducting experiments to compare the adaptive potential of the dikaryotic state with that of the corresponding, isogenic diploid state. The hypothesis is that dikaryons and diploids differ with respect to their patterns of genetic change over time, with accompanying differences in the patterns of gene regulation and phenotypic plasticity.

80. Diversity in natural populations of Colletotrichum from the tallgrass prairie. Jo Anne Crouch, Bruce B. Clarke, and Bradley I. Hillman. Department of Plant Biology and Pathology, Rutgers University.

The fungal genus Colletotrichum contains several species infecting monocot hosts in both cultivated and natural grass communities. In monocultured agroecosystems, these fungi are often found as destructive pathogens, capable of inducing significant disease in the host plant. In contrast, our recent sampling of Colletotrichum in a natural tallgrass prairie ecosystem suggests that populations of this fungus living in diverse grass communities maintain a non-pathogenic lifestyle, with their presence never correlated with substantial disease. Preliminary multi-locus phylogenetic analysis supports the presence of a single lineage of Colletotrichum in the grassland environment, but transposon distribution data and RFLP patterns suggest that these populations may actually represent a hybrid zone between distinct Colletotrichum phylogenetic species. We are currently employing microsatellite markers to evaluate whether natural grasslands represent regions of hybridization for Colletotrichum species that inhabit Pooideae grasses.

81. Evolutionary relationships of Colletotrichum species pathogenic to grasses of the Poaceae. Jo Anne Crouch, Bruce B. Clarke, and Bradley I. Hillman. Rutgers University, New Brunswick, NJ

Colletotrichum graminicola is a filamentous, primarily clonally-reproducing fungus inhabiting a wide range of grasses and cereals. During the past several years, this fungus has emerged as one of the most important pathogens in turfgrass systems. Control of the disease is often unpredictable, possibly due to an imperfect understanding of population structure and pathogen variability. Using Bayesian likelihood analyses of nucleotide sequences from three unlinked nuclear loci, we provide compelling evidence of sympatric, species-level phylogenetic divergence between C. graminicola pathogenic to Zea mays and Colletotrichum specimens causing disease in grasses of the subfamily Pooideae. These data also revealed two lineages emerging along distinct evolutionary trajectories among isolates sampled from Pooideae grasses, including several Pooid species cultivated as turfgrasses. Nucleotide substitution rates consistent with positive adaptive selection at the fungal mating-type gene suggests that reinforcement may have played a role in reproductively isolating these unique sibling species as they emerged in host-range restricted ecological niches. Transposon distribution patterns and RFLP fingerprint analysis extend the phylogenetic estimations, independently supporting the conclusion that several closely related cryptic species of Colleotrichum inhabit the grasses of North America.

82. Het-S spore killing in Podospora anserina: Non-Mendelian nuclear inheritance caused by the non-Mendelian cytoplasmic HET-s prion-element. Henk Dalstra, Klaas Swart, Sven Saupe and Fons Debets. Laboratory of Genetics, Wageningen, The Netherlands

Two alleles of the het-s/S locus occur naturally in Podospora anserina, het-s and het-S. The het-s encoded protein can form a prion that propagates a self-perpetuating amyloid aggregate, resulting in two phenotypes for the het-s strains. The prion infected [Het-s] shows an antagonistic interaction to het-S whereas the prion-free [Het-s*] is neutral in interaction to het-S. The antagonism between [Het-s] and het-S is seen as heterokaryon incompatibility at the somatic level and as het-S spore killing in the sexual cycle. The consequences of the unique transition from a coenocytic to a cellular state in the sexual phase and the timing and localization of paternal and maternal HET-s and HET-S expression that are pertinent to prion transmission and het-S spore killing are elaborated. A model for het-S spore killing is proposed.

83. Experimental Evolution of Reproductive Isolation in Neurospora. Jeremy Dettman, James Anderson, and Linda Kohn. Department of Botany, University of Toronto, Ontario, Canada

Reproductive isolation may develop between populations by genetic drift alone, or as a by-product of adaptation to divergent environmental conditions (ecological speciation). Interspecific hybridization may also produce novel combinations of alleles which may facilitate speciation by creating new opportunities for adaptive evolution (hybrid speciation). The evolution of reproductive isolation typically occurs over long time scales and is difficult to observe directly in nature. To study these modes of speciation, we are experimentally evolving populations of Neurospora in controlled environments and are monitoring changes in reproductive behavior as they occur. Populations of N. crassa and N. crassa-N. intermedia hybrids are being serially propagated asexually, with periodic sexual cycles to maintain the capability for sexual reproduction within lineages. To promote divergent adaptation, replicate lineages are being evolved under two different suboptimal growth conditions (high salinity and low temperature). At the end of the evolution regime, individuals from the evolved lineages will be mated and reproductive success will be measured. Several questions will be addressed: Do the populations adapt to the novel environments? Are lineages evolved in divergent environments more likely to develop reproduction isolation than lineages evolved in the same environment? Does the increased genetic diversity of hybrid lineages allow for more rapid development of reproductive isolation among hybrid lineages? Do similar phenotypes of reproductive isolation develop between hybrid lineage pairs and non-hybrid lineage pairs? Preliminary tests after two months of evolution indicate that some forms of reproductive defects have arisen. Results of these and additional tests will be presented and discussed.

84. Landscape genetics of Penicillium marneffei in southeast Asia. Matthew C. Fisher Dpt. Infectious Disease, Imperial College London, UK

High-throughput multilocus sequence typing (MLST) and multilocus microsatellite typing (MLMT) techniques are uncovering a wealth of genetic diversity hidden within morphologically-defined fungal species. New statistical tools (such as Bayesian approaches) are enabling us to analyse the structure of this genetic diversity within species; these analyses are then allowing us to understand the microevolutionary processes that are generating genetic structure. MLMT of Penicillium marneffei has shown the existence of high levels of genetic diversity; visualizing this diversity within a spatial context shows that landform impacts strongly on genetic structure. Despite the high dispersive capacity of fungal spores, gene-flow in P. marneffei appear to be low, suggesting that microevolution of isolates is causing local adaptation to their immediate environment. In order to test this hypothesis, highly sensitive environmental PCRs are being used to detect P. marneffei in the environment. Remotely-sensed GIS databases are then being used to correlate the presence/absence of the fungus with specific environmental variables in order to determine the ecotypes within which P. marneffei is associated. It is then possible to use MLMT data to determine the extent to which P. marneffei genetic structure is determined by landform and ecotype.

85. Convergent evolution of chromosomal sex-determining regions in the animal, plant and fungal kingdoms. James A. Fraser, Stephanie Diezmann, Ryan L. Subaran, Andria Allen, Klaus B. Lengeler, Fred S. Dietrich, and Joseph Heitman. Duke University Medical Center

Sexual identity is governed by sex chromosomes in plants and animals, and mating type loci (MAT) in fungi. Comparative analysis of the MAT locus from a species cluster of the human fungal pathogen Cryptococcus revealed sequential evolutionary events that fashioned this large, highly unusual region. We hypothesize MAT evolved via four main steps beginning with gene acquisition into two unlinked sex-determining regions, forming independent gene clusters involved in pheromone production/sensing and meiosis/karyogamy. These cluster then fused via chromosomal translocation. The transitional tripolar intermediate state created was then converted to a bipolar system via gene conversion or recombination between the linked and unlinked sex determining regions. Experimental tests of this model are currently in progress. MAT was subsequently subjected to intra- and interallelic gene conversion and inversions that suppress recombination. Large-scale analysis of MAT structure of >100 isolates revealed a major force driving this evolution was the acquisition of transposable elements, with the locus transposon content over 4 times greater than elsewhere in the genome. These events resemble those that shaped mammalian and plant sex chromosomes, illustrating convergent evolution in sex-determining structures in the animal, plant and fungal kingdoms.

86. Fusarium graminearum in the U.S.: heterogeneous and in flux. Liane Rosewich Gale1, Je'Nise D. Bryant2, Gerald E. Ochocki1, Todd J. Ward3, and H. Corby Kistler1. 1USDA, CDL, St. Paul, MN. 2University of Minnesota, St. Paul, MN. 3USDA, NCAUR, Peoria, IL.

From a previous survey of diversity in Fusarium graminearum collected from 86 fields in 53 counties in 9 midwestern states in 1999 and 2000, we identified a small population in ND and MN (7%) that produced 3-acetyl, deoxynivalenol (3ADON) that were genetically distinct (Nm = 0.5) from the predominant 15-acetyl, deoxynivalenol (15ADON) producing F. graminearum. Collections in 2003 from 40 wheat fields in 24 counties in ND, and 9 fields in 7 counties in MN resulted in 2,133 isolates. The 3ADON type was widespread and at high frequency (21% in ND, 24% in MN). Further analysis with primers targeting three VNTR loci demonstrated that recombination in F. graminearum, although occurring, may be an infrequent event, as only 70 potential recombinants between the two populations were identified. Chemotyping in collections from 2001-2003 indicates that 15ADON is still the only type in other midwestern states, though the nivalenol type was the most frequent in isolates from LA. The predominance of the nivalenol type in LA and the build-up of the 3ADON type in MN and ND suggest that selection is a principal evolutionary force acting on populations of F. graminearum. At the same time this study demonstrates the need for continual monitoring of the population composition, as F. graminearum in the U.S. is certainly not as homogeneous as previously believed.

87. Association of clinical, environmental and plant pathogenic isolates within the Fusarium solani species complex. Ning Zhang1, Kerry O'Donnell2, and David M. Geiser1. 1Department of Plant Pathology, Penn State University, University Park, PA USA.


Members of the Fusarium solani species complex are increasingly implicated as the causative agents of human mycoses, particularly in the immunocompromised patient population. Previously members of this complex were shown to comprise at least 26 phylogenetically distinct species, including ubiquitous saprophytes and economically important plant pathogens. To identify phylogenetic species associated with human infections, we generated three partial gene sequences (1726 bp) from 471 isolates from clinical (human N= 278, hospital environment N =21) and non-clinical (N = 172) sources chosen to represent the known phylogenetic diversity in the complex. We identified 218 four locus haplotypes among these isolates, 99 of which included clinical isolates. Nineteen haplotypes included both clinical and non-clinical isolates, and monophyly of the clinical isolates was rejected. Members of the most common environmental haplotypes, including those found in hospitals, were associated with the majority of F. solani infections, suggesting that F. solani infections are truly opportunistic across a wide range of phylogenetic species and haplotype diversity. The majority (77%) of clinical isolates fell into one of four clades of haplotypes, each of which included isolates from non-clinical sources. One of these clades corresponded to Mating Population V of the F. solani species complex.

88. Improved bioinformatic tools for identification of fungi from environmental samples based on ITS and LSU rDNA sequences. József Geml, James Long, Shawn Houston, Thomas G. Marr, D. Lee Taylor. Institute of Arctic Biology, University of Alaska Fairbanks, 311 Irving I, Fairbanks, AK, 99775, U.S.A.

To study the fungal diversity in boreal forests of Alaska we are sequencing 30,000 clones from PCR libraries of fungal ribosomal gene-regions amplified from soil DNA extracts. In addition, we will compare sporocarp versus soil views of fungal diversity by sequencing 4000 sporocarps from the UAF Mycological Herbarium. Database similarity searches play crucial role in matching unknown environmental sequences to sequences from identified fruitbodies. In this paper, we assess the performance of different similarity search algorithms (BLASTN, FASTA) - with and without masking highly conserved regions - against newly assembled databases of fungal ITS and LSU rDNA sequences from public (GenBank, TreeBASE) and private databases. We report new bioinformatic tools for identification of fungi publicly available on our website (http://iab-devel.arsc.edu/metagenomics/): 1) Similarity search with automated masking against a database of more than 28,000 fungal ITS rDNA sequences; 2) Similarity search against a database of more than 8,000 fungal LSU rDNA sequences; 3) Similarity score and sequence output of the best 50 hits in FASTA format; 4) Multiple alignment of the search result sequences. Database searching is done with an MPI (Message Passing Interface) version of FASTA across 16 cpus in a computer cluster consisting of 2 GHz Apple G5's. An MPI version of ClutstalW is used to align the top 50 hits. Automatic masking is accomplished with RepeatMasker using a custom masking library. Users enter their query sequences in a web interfaced derived from the Pise system. These tools help mycologists, fungal systematists, microbial ecologists and others to identify unknown fungi, such as species that may be recovered from the soil PCR libraries.

89. High degree of structure is detected between populations separated by short and long distances in two sympatric Rhizopogon species. .Lisa C. Grubisha and Thomas D. Bruns. Department of Plant and Microbial Biology, 111 Koshland, University of California, Berkeley, CA 94720-3102 USA.

We are investigating the relationship between genetic and geographic structure in two sympatric species of Rhizopogon. R. vulgaris and R. occidentalis are hypogeous, ectomycorrhizal fungi associated with pines. Spore dispersal is by animals that consume fruiting bodies, thus dispersal should be restricted by geographic barriers and distance between populations. R. vulgaris and R. occidentalis were sampled from native pine populations on two islands and coastal mainland sites in California. Results from analysis of microsatellite data reveal that in both species populations separated by very short distances can exhibit a high degree of genetic differentiation if intervening areas do not contain suitable habitat. Within Santa Cruz Island R. vulgaris populations showed a high degree of structure even though they were only separated by 11-18 km. Physical barriers, such as the lack of pines and a large dry valley, separated these populations, that were located on two mountain ranges. Very high levels of genetic differentiation between the northern California and the island populations suggest the possibility of cryptic speciation for both species. These results are consistent with the idea that in isolated pine populations gene flow between populations of these fungi is restricted by the movement of the mammals that disperse their spores.

90. Simple sequence repeat abundance and distribution in two oomycete genomes: Phytophthora sojae and P. ramorum. N. Grunwald1, S. Tripathy2, K. Ivors3, K. Lamour4 and B. Tyler2. 1USDA ARS, Corvallis, OR; 2VPI, Blacksburg, VA; 3NC State, Fletcher, NC; 4Univ. Tennessee, Knoxville, TN.

The first whole genome assembly of the two oomycete plant pathogens Phytophthora sojae (95 Mb) and P. ramorum (65 Mb) were examined to determine types, abundance and distribution of different simple sequence repeats (SSRs) ranging between 2 to 6 bp in motif length. We found 2,128 and 1,000 SSRs in Ps and Pr, respectively. In general, the density of SSRs (bp per Mb) in P. sojae is about 1.5 times that of Pr. Whereas AC dinucleotide repeats appear at a higher density in Pr, AG and AT repeats appear at a higher density in Ps. Interestingly, density of most trinucleotide repeats was higher in Ps than in Pr. Although Ps has a larger genome, the percentage of SSR loci located in coding regions is higher at 17.4% compared to Pr at 14.8%. Compared to other species, including Saccharomyces cerevisiae, repeats of length 4, 5, and 6 bp are considerably underrepresented in both Phytophthora genomes. Whereas in most genomes studied to date dinucleotide repeat stretches tended to be longer than other repeats, in the case of Phytophthora only tetranucleotide repeats were occasionally considerably longer (ACAG, ACAT, and AGAT). As expected frequency of trinucleotide repeats in exons was considerably higher when compared to di, tetra- or pentanucleotide repeats.

91. Host-associated population structure in Septobasidium ramorum, a fungal symbiont of scale insects. Daniel A. Henk and Rytas Vilgalys. Duke University, Durham, NC

Septobasidium ramorum is a fungal symbiont of diaspidid scale insects. The fungi form mycelial mats over entire colonies of scale insects including infected and uninfected insects. The fungus-insect relationship may be mutualistic from the perspective of a scale insect colony, but infected insects are sterile. The accepted fungal lifecycle requires frequent anastomosis of putatively recombinant genotypes within colonies and codispersal with local host scale insects. We use DNA sequence data from single spore isolates, tissue isolates, and isolates from within infected scale insects to confirm the recombinant nature of fungi infecting individual insects and detect genetic structure within fungal colonies. We also use DNA sequence data to quantify the relative host specificity and selectivity of S. ramorum and to assess copopulation genetic structure of S. ramorum and its primary host scale insect, Chionaspis nyssae. Results suggest that S. ramorum has a wide host range but its population structure largely mirrors that of its primary host insect. Comparisons to species with overlapping geographic range but different host specificity will be discussed.

92. The Coprinopsis cinerea laccase multi-gene family and evolution of multi-copper oxidase genes in basidiomycetes. P. Hoegger1, S. Kilaru1, J. Thacker2, T. James2, R. Vilgalys2 and U. Kües1. 1Georg-August-University Göttingen, Institute of Forest Botany, Göttingen, Germany; 2Duke University, Department of Biology, Durham, NC.

Laccases belong to the group of multi-copper oxidases and catalyze the oxidation of phenolic compounds. We have identified a laccase multi-gene family in the white-rot Coprinopsis cinerea comprising 17 non-allelic laccase genes. Based on intron distributions and deduced amino acid sequences, they divide into two distinct groups. Several ancient and more recent duplication and recombination events within the two groups seem to be responsible for the observed number and genomic organisation of the genes. In phylogenetic analysis of almost 300 multi-copper proteins from insects, plants, fungi and bacteria all formed distinct clades containing all the sequences from a given kingdom and mostly according to their function (where known). In other completely sequenced genomes of basidiomycetes (Phanerochaete, Ustilago, Cryptococcus), we found only multi-copper oxidase genes with rather low similarities to the C. cinerea laccase proteins. These sequences cluster with ferroxidases, ascorbate oxidases, or enzymes of melanin synthesis. Within the fungi, the basidiomycetes appear to form a clade of typical laccases distinct from the ascomycetes. The basidiomycete sequences seem to diverge according to different life styles of the organisms, partially independent from organismal relationships.

Work in Göttingen is supported by the Deutsche Bundesstiftung Umwelt (DBU).

93. Horizontal transfer of selfing in the ascomycete genus Stemphylium (Pleosporales). Patrik Inderbitzin, Jennifer Harkness and Mary L. Berbee. University of British Columbia, Department of Botany, Vancouver, Canada, V6T 1Z4

The genus Stemphylium contains selfing species that evolved from outcrossing ancestors. To find out how selfing originated, we analyzed the Stemphylium MAT loci that regulate sexual reproduction in ascomycetes, and compared MAT structures and phylogeny with a multigene Stemphylium species phylogeny. We found that some Stemphylium species' MAT loci contained a single gene, either MAT1-1 or MAT1-2, while others contained a unique fusion of the MAT1-1 and MAT1-2 regions. In all fused MAT regions, MAT1-1 was inverted and joined to a forward oriented MAT1-2 region. As in the closely related Cochliobolus, Stemphylium species with fused MAT regions were able to self. Structural and phylogenetic analyses of the MAT loci showed that the selfing-conferring fused MAT regions were monophyletic with strong support. However, in an organismal phylogeny of Stemphylium species based on 114 isolates and the four loci ITS, GPD, EF-1 alpha and vmaA-vpsA, selfers with identical fused MAT regions arose in two clades, each time with strong support. We showed that a one-time origin of the fused MAT loci followed by a horizontal transfer across lineages was compatible with the data. Another group of selfers in Stemphylium only had forward oriented MAT1-1 at their MAT loci, constituting an additional and third origin of selfing in Stemphylium. We would like to acknowledge financial support from NSERC (grants to M. Berbee and P. Inderbitzin), NSF subcontract to M. Berbee (P.I. B. G. Turgeon), University of British Columbia Graduate Fellowship and a Swiss National Science Foundation Postdoctoral Fellowship to P.Inderbitzin.

94. Early diverging lineages on the fungal tree of life: phylogenetic analysis using six gene regions. Tim James1, Joyce Longcore2, Rytas Vilgalys1, and the Assembling the Fungal Tree of Life (AFTOL) Working Group. 1Dept. of Biology, Duke University, Durham, NC USA, 2Dept. of Biol. Sciences, U. Maine, Orono, ME USA

Much of the phylogenetic and ecological diversity of the Fungi can be found in those lineages that diverged early in the history of the kingdom. These lineages include the Chytridiomycota, Zygomycota, Glomeromycota, and possibly the Microsporidia. Understanding the history of the Fungi includes unraveling the branching at the base of the kingdom, but resolution of the deepest nodes is difficult because these divergence times are ancient. The goal of the Assembling the Fungal Tree of Life project is to provide a global phylogenetic hypothesis for the Fungi using DNA sequences from multiple loci. We present preliminary results from the phylogenetic analysis of over 50 taxa of basal fungi using DNA sequences for nuclear 18S and 28S rRNA genes, ATP6, EF1-alpha, RPB1, and RPB2. There is a general lack of congruence among the phylogenies estimated from each of the individual genes. Whether the Chytridiomycetes or Zygomycetes are monophyletic remains uncertain. The rDNA data suggest the Glomeromycetes form a clade with the Ascomycota+Basidiomycota, while the data from protein encoding genes place the Glomeromycetes with the Zygomycetes. No relationship was observed between the Microsporidia and any other basal fungal lineage sampled.

95. Avirulence genes in European populations of Leptosphaeria maculans, the cause of stem canker of crucifers. Anna Stachowiak 1, Julia Olechnowicz 1, Malgorzata Jedryczka 1, Thierry Rouxel 2, Marie-Hélene Balesdent 2, Ingrid Happstadius 3, Peter Gladders 4, Neal Evans5. 1Institute of Plant Genetics PAS, Poznan, Poland. 2INRA, PMDV, Versailles, France. 3Svalöf Weibull AB, Svalöv, Sweden. 4ADAS Boxworth, Cambridge, UK. 5Rothamsted Research, Harpenden, UK

The fungus Leptosphaeria maculans is the cause of stem canker (blackleg) of crucifers, the damaging disease of oilseed rape in Europe, Canada and Australia. At present, nine avirulence genes of the pathogen (AvrLm1-AvrLm9) are known, and these correspond to RLm1-RLm9 resistance genes in the host. A survey was done to investigate the distribution of avirulence genes in the L. maculans population across Europe. The EU-funded project SECURE (StEm Canker of oilseed rape: molecular tools and mathematical modelling to deploy dUrable REsistance: QLK5-CT-2002-01813) followed a sampling protocol used during a previous large scale survey done in France. Isolates were then screened using a cotyledon test on a differential set of nine cultivars developed by INRA-Versailles. The race structure in Poland, UK, Germany and Sweden resembled the structure observed in France, with all isolates having AvrLm6 and AvrLm7, but no AvrLm2, AvrLm3 and AvrLm9. The distribution of AvrLm1, AvrLm4 and AvrLm5 ranged between countries and experiment sites, with a prevalence of a virulent allele.

96. Genetic diversity of Ophiostoma clavigerum associated with the mountain pine beetle, Dendroctonus ponderosae Hopkins in North America. S. Lee and C. Breuil. Department of Wood Science, University of British Columbia, 2424 Main Mall, Vancouver, BC, V6T 1Z4, Canada

In this work, the genetic diversity of the Ophiostoma clavigerum (Robinson-Jeffrey & Davidson) Harrington was assessed. This pathogenic, sapstaining fungus is associated with the mountain pine beetle (Dendroctonus ponderosae Hopkins), which is a native insect in North America. As of 2004, in British Columbia only, 108 million m3 of lodgepole pine (Pinus contorta) has been infested by the mountain pine beetle. In 2003, O. clavigerum was isolated from the infested lodgepole pines in Canada (Banff in Alberta, Ft.St.James, Houston, Williams Lake, and Kamloops in British Columbia) and USA (Hidden Valley in Montana, and Hell Roaring in Idaho). A total of one hundred eighty six single-spore isolates were analyzed for their genetic variability by the amplified fragment length polymorphism (AFLP) markers generated with six primer sets. In this work, we described the genetic variation and differentiation within/among the O. clavigerum populations. We also examined the possible correlation of genetic distances with geographic distances among sampling locations. Overall, it appeared that the genetic polymorphism in O. clavigerum is low, which might be related to relatively rare sexual reproduction.

97. Progesterone decreases cAMP content in filamentous fungus Rhizopus nigricans. Helena Lenasi, Natasa Jeraj and Katja Breskvar. Institute of Biochemistry, Medical Faculty, University of Ljubljana, Ljubljana, Slovenia

Mammalian steroid hormones are not present in the natural environment of a saprophytic fungus Rhizopus nigricans, but when cultivated in vitro fungal growth is selectively inhibited by progesterone, deoxycorticosterone, testosterone and estradiol. The most efficient inhibitor is progesterone. Previous studies revealed progesterone receptors (Kd = 70nM) coupled to G proteins in the plasma membrane fraction of the fungus. The biological role of progesterone binding to membrane receptors as well as the resulting signalling pathway is not known yet. In this report we examined possible signalling pathways (MAPK pathway, routes including inositol-3-phosphate and cAMP) which might follow membrane receptor activation. A remarkable decrease of cAMP content (24+14.3 to 17.1+11.0 pmol/mg protein) was observed in the presence of progesterone (>15 M). Progesterone decrease of cAMP content was not influenced by pertussis toxin or cholera toxin indicating the possibility that in R. nigricans Galpha subunits other than mammalian Galphas or Galphai were involved in this signalling. Progesterone also drastically changed fungal morphology. However, cAMP-mediated progesterone signalling contributes only a small part to the total growth inhibition by progesterone, therefore an additional non-specific action of progesterone should be taken into consideration.

98. Molecular evolution in the endopolygalacturonase I of Botrytis cinerea. C. Lévis, C. Giraud-Delville and E. Fournier. INRA-PMDV, Route de Saint-Cyr, F-78026 Versailles, France.

Identifying genes responsible for adaptive differences between populations and species may help understanding the mechanisms of evolutionary diversification. Adaptive differences should leave a signature of positive selection at the molecular level. In fungal plant pathogens, genes involved in fungus-plant interactions are candidate genes on which rapid positive selection may operate.

Endopolygalacturonases are fungal enzymes produced early during plant-pathogen interactions, e.g. the endopolygalacturonase I (BCPG1) of the Ascomycete Botrytis cinerea Pers. (ten Have et al. 2001, Fungal Genet Biol. 33). This species complex comprises two cryptic groups : Group II, attacking about 200 plants, and Group I with a more restricted host range (Fournier et al. 2003, Mycologia 95). In this multi-host pathogen, selection by the host may have operate. It has been shown that BCPG I less aggresive B. cinerea isolates BCPG I elicites a defense reaction in the plant (Poinsot et al. 2003, MPMI 16). BCPG I is therefore a good candidate gene that may be shaped by positive selection.

We tested this hypothesis by sequencing BCPG I and surrounding regions (1000 bp upstream, 2500 bp downstream) in 12 B. cinerea isolates (2 Group I, 10 Group II). An ORF was found in the 3' region, coding for a putative N-acetyl-transeferase. Tree topologies and evolution speeds were compared in each region. Ratio of nonsynonymous to synonymous mutation were estimated in the two coding regions. Results will be discussed in the poster.

99. Natural Variation in the Amino Terminal Polyglutamine Domain of Neurospora crassa WHITE COLLAR-1 Confers Circadian Acuity. Todd P. Michael1, Sohyun Park2, Tae-Sung Kim2, Joanne Chory1, and Kwangwon Lee2. 1Plant Biology Laboratory, The Salk Institute, 10010 N. Torrey Pines Rd., La Jolla, CA 92037. 2Department of Plant Pathology, 201 Bradfield Hall, Cornell University, Ithaca, NY 14853

Natural variation in the circadian clock provides acuity for specific environments. The circadian clock synchronizes the internal biology of an organism to the exact 24-hour period of the local external environment so biological processes occur at precise times over the day despite seasonal changes. Utilizing an inverted race tube assay, which eliminates the need for the band mutation, we found that 144 Neurospora crassa accessions display variation in circadian parameters such as period, phase and temperature compensation. In order to understand the impact of natural variation on the underlying molecular mechanisms of the circadian clock we evaluated known clock components. We found that WHITE COLLAR-1 (WC-1), which plays dual roles as both a blue light photoreceptor and central clock component, displays variation across the 144 accessions in its amino-terminal and carboxy-terminal glutamine repeats. Utilizing F1 hybrids between accessions with different N-terminal glutamine repeats we confirm a relationship between the N-terminal glutamine repeat size and circadian period. We propose that the plastic nature of the activation domains in WC-1 may increase the fitness of an organism through optimizing the acuity of the circadian clock.

100. The impact of commercial applications of the entomopathogenic fungus Beauveria bassiana on the genetic diversity of indigenous conspecific populations. Prashant K. Mishra1, Louela A. Castrillo2, Eleanor Groden1, John D. Vandenberg3, and Seanna L. Annis1. 1Department of Biological Sciences, University of Maine, Orono, ME04469; 2Department of Entomology, Cornell University, Ithaca, NY14853; 3Plant, Soil, and Nutrition Lab, USDA-ARS, Ithaca, NY14853.

The entomopathogenic fungus Beauveria bassiana has been widely used for the biological control of various agricultural insect pests. In this study, we have analyzed the genetic diversity in the indigenous populations of this fungus and the changes that have occurred after the application of a commercial formulation of B. bassiana strain GHA (Emerald BioAgriculture Corp., Lansing, MI). Amplified fragment length polymorphisms and random amplified polymorphic DNA markers were generated from B. bassiana isolates collected from three sprayed and three unsprayed fields. There was displacement of indigenous genotypes 1 to 2 years after application, however the frequency of indigenous genotypes increased 4 years after treatment. Isolates were assigned to four ancestral genetic populations, 1 GHA type and 3 indigenous types. Isolates from nonsprayed fields were only assigned to indigenous genetic populations. In sprayed fields, some isolates had mixed ancestry from GHA and indigenous genetic populations indicating the occurrence of recombination. This study can serve as a model for assessing the risk associated with the release of genetically modified fungi on their conspecifics.

101. Origin and migration of Ustilago maydis. Andrew Munkacsi1 and Georgiana May 2. 1Plant Biological Sciences Graduate Program, University of Minnesota. 2Department of Ecology, Evolution and Behavior, University of Minnesota.

Ustilago maydis is a fungus that only infects maize and the wild relatives of maize, the teosintes. The domestication and spread of maize is well understood. Maize was domesticated by humans in southern Mexico from a teosinte 6,000-10,000 years ago, and exported by humans to the United States 1,000-2,000 years ago and to South America 2,000-4,000 years ago. Unlike the evolutionary history of maize, the origin and migration of U. maydis has yet to be elucidated. A popular hypothesis is that the ancestral U. maydis population was in Mexico and U. maydis followed maize as it was traded to new locations. To test this hypothesis, we genotyped 900 collections of U. maydis teliospores from Mexico, the United States, and South America using 10 microsatellite markers. Based on this dataset, we infer the geographic location of the ancestral population, predict whether the host of that population was maize or a teosinte, and estimate the divergence times between genetically distinct populations. Together, we compare the origin and migration of U. maydis to the origin and migration of maize.

102. Phylogenetic Analysis of Septin Gene Family. Fangfang Pan, Russell Malmberg, Michelle Momany. Plant Biology Department, University of Georgia, 30602

Members of the septin gene family have been found in animals, fungi, and some primitive eukaryotes. They play important roles in cytokinesis and cell surface growth. To understand the evolution of the septin gene family, criteria for deciding whether a sequence may be classified as a septin have been developed. GTP-binding domains are highly conserved in septins. These motifs can be used to distinguish septins from other members of the GTPase superfamily. Some septins also have a distinctive coiled-coil domain. Approximately 170 septin and septin-like genes were identified by PSI-Blast database search for this analysis.

Phylogenetic trees were built for this family using MrBayes. There was a large diversification of septin sequences prior to the fungal/animal split. Most members of the septin gene family can be grouped with one of the S. cerevisiae septins Cdc3, Cdc10, Cdc11 or Cdc12. Seven possible new members of the septin gene family were found, including septin-like genes in more ancient organisms. Future experimental studies are needed to verify the identity of the newly found septin genes.

103. Evolution of Ascochyta spp. on wild and cultivated legumes. Tobin L. Peever, Marco Hernandez-Bello, Maneesha Barve & Walter J. Kaiser. Department of Plant Pathology, Washington State University, Pullman, WA 99164-6430, USA.

Ascochyta spp. were sampled from chickpea (Cicer arietinum), faba bean (Vicia faba), lentil (Lens culinaris), pea (Pisum sativum), hairy vetch (Vicia villosa) and several of their wild relatives in various geographic locations worldwide. Previous results from laboratory crosses indicated that Ascochyta spp. infecting faba bean, lentil and chickpea should each be considered biological species. We tested the hypothesis that these biological species were also phylogenetic species and extended the phylogeny to include isolates sampled from other cultivated legumes and their related wild relatives. We sequenced the nuclear ribosomal internal transcribed spacer (ITS) and 3 protein-coding loci and estimated a phylogeny for each dataset independently and for the combined dataset. Phylogenies estimated from all genomic regions were congruent and the protein-coding loci revealed significantly more variation than ITS. The combined phylogeny revealed 6 clades, each with high bootstrap support and associated with a single cultivated host. Isolates from wild chickpea (C. ervoides and C. monbretti), wild pea (P. elatius) and several wild vetch species clustered with isolates from their respective cultivated hosts. Isolates from bigflower vetch (V. grandiflora) were polyphyletic. Ascochyta spp. isolated from pea and faba bean were host specific and easily crossed in the laboratory. Progeny were phenotypically normal but most were unable to infect either pea or faba bean. Phylogenetic analyses, host inoculations and in vitro genetic crosses with Ascochyta spp. will be useful in studying fungal speciation and the genetics of host specificity at the species level in plant-pathogenic fungi.

104. Analysis of the complete mitochondrial genomes from Phakopsora pachyrhizi and Phakopsora meibomiae. Martha L. Posada-Buitrago1, Jeffrey L. Boore1 and Reid D. Frederick2. 1DOE Joint Genome Institute. 2800 Mitchell Drive, Walnut Creek, CA 94598. USDA-ARS Foreign Disease-Weed Science Research Unit, 1301 Ditto Avenue, Fort Detrick, MD 21702.

The complete nucleotide sequence of the mitochondrial (mt) genome was determined for two obligate fungal pathogens, Phakopsora pachyrhizi and P. meibomiae, the causing agents of the soybean rust. These 32 kb genomes contain the genes encoding ATP synthase subunits 6, 8, and 9 (atp6, atp8, and atp9), cytochrome oxidase subunits I, II, and III (cox1, cox2, and cox3), apocytochrome b (cob), reduced nicotinamide adenine dinucleotide ubiquinone oxireductase subunits (nad1, nad2, nad3, nad4, nad4L, nad5, and nad6), the large and small mitochondrial ribosomal RNAs and tRNAs for all amino acids. Gene order and gene sequences were compared with all available complete fungal mtDNAs-representatives of all four Phyla.

105. Fungal Inteins and Horizontal Gene Transfer. Russell Poulter, Margi Butler, Jeremy Gray and Tim Goodwin. Department of Biochemistry, University of Otago, Dunedin New Zealand

There are only two described nuclear encoded inteins. One is the VMA intein (which includes a homing endonuclease domain) found in some hemiascomycetous yeasts. We have described a second intein in the basidiomycete yeast Cryptococcus neoformans, a fungal pathogen of humans. This intein is encoded as an in-frame insertion within the nuclear PRP8 gene. The intein includes the domains associated with intein splicing, but lacks the homing endonuclease domain associated with intein mobility. This intein is present in all C. neoformans strains examined and in all strains of the closely related pathogenic species, Cryptococcus bacillisporus. The intein was not found in most related Cryptococcus species or in other basidiomycetes. PRP8 inteins were found in several ascomycetes including Histoplasma capsulatum, Aspergillus nidulans, Aspergillus fumigatus, and its close relative Neosartorya fischeri. These inteins are at the identical site in the PRP8 sequence. PRP8 inteins were not found in other ascomycetes. Unlike the Cryptococcus neoformans and C. bacillisporus inteins, some of the ascomycete PRP8 inteins contain homing endonuclease domains and are thus potentially mobile. Comparisons of the shared splicing domains reveal that the ascomycete and cryptococcal inteins are surprisingly similar in sequence, given the time since these fungi diverged from a common ancestor (~550 Mya). We propose that the discontinuous phylogenetic distribution of the intein is the result of a horizontal gene transfer.

106. Transposition, Recombination and Gene Genesis in Magnaporthe grisea. Powell AJ, Pan H, Diener SE and Dean RA. Fungal Genomics Laboratory North Carolina State University

Magnaporthe grisea (f. sp. oryzae) is the premier model organism for study of fungal pathogenic interactions with cereal crops. Recent advances in this model include acquisition of the genome sequence, large-scale mutational analyses and global transcriptional studies using oligo-based microarrays. The latter global transcriptional studies revealed the apparent upregulated expression of LTR retrotransposon maggy in asexual spores (conidia), relative to mycelia and appressoria. RT PCR analyses of gag and pol, constituent ORFs of maggy, corroborated our microarray findings. Two formal possibilities could account for the observed pattern of expression: either gag and pol ORFs were autonomously transposing or these ORFs had inserted into upregulated resident genes, and were thus "hitchhikers," and not truly actively transposing. Rigorous bioinformatic analyses revealed the presence of known and predicted genes containing different partial copies of maggy. These partial copies represent the full spectrum of maggy structural features, from gag and pol ORFs, to LTR regions. The maggy-derived genes are therefore chimeric ORFs, where recipient genes have acquired modules of non-homologous sequence, from which evolution can shape novel forms and functions. Preliminary RT PCR and EST data indicate that maggy-derived chimeric ORFs are expressed. Our results are significant because they represent the discovery of a novel class of chimeric genes that show developmentally distinct patterns of expression. These chimeras are likely the product of transpositional recombination, an expected source of genomic novelty. Detailed structural, functional and evolutionary characterization of these maggy-derived chimeras, as well as those arising from other transposable elements, promises to broaden our understanding of genomic, organismal and population levels of biological complexity.

107. Withdrawn

108. Assessing reproductive mode in the mitotic entomopathogen, Beauveria bassiana using gene genealogies, allelic associations and distribution of mating-type idiomorphs. Stephen A. Rehner and Ellen P. Buckley. USDA-ARS, Insect Biocontrol Laboratory, Beltsville, MD, USA.

The entomopathogenic fungus, Beauveria bassiana , is renowned for its copious production of mitotic, haploid conidia and the conspicuous absence of a sexual stage throughout most of its global range. To determine reproductive mode in this species, an essential first step is the circumscription of appropriate population samples in which the contrasting reproductive alternatives of clonality and recombination can be critically investigated. We outline our implementation of the phylogenetic species recognition (PSR) criterion to identify regionally sympatric groups of related individuals for population-level analyses. We focus on two closely related terminal clades: one from eastern Asia, which includes a mix of sexual and asexual isolates, and a second from western Europe, in which only asexual reproduction has been observed. Statistical tests of allele associations among polymorphic microsatellite loci indicate that each population is recombining. The structural organization of the mating-type locus is described, demonstrating that B. bassiana possesses a heterothallic mating-type system with canonical MAT1 and MAT2 specificities. Evidence is presented that the MAT1 and MAT2 mating-types are randomly distributed in the European population, supporting the conclusion that genetic recombination arises through sexual reproduction.

109. Populations, clones and individual mycelia of arbuscular mycorrhizal fungi. Søren Rosendahl and Eva H Stukenbrock, Department of Microbiology, Institute of Biology, University of Copenhagen, Denmark

Population genetic analyses of arbuscular mycorrhizal fungi (AMF) are often hampered by the obligate biotrophy of the fungi. Multiple markers from single spores can be obtained by nested multiplex PCR method using a combination of primers for simultaneous amplification of five loci in one PCR reaction. This approach was used for multilocus genotyping of arbuscular mycorrhizal fungal populations sampled hierarchically from an organically and a conventionally cultured field. All spore genotypes were unique suggesting that no recombination was taking place in the populations. There were no overall differences in the distribution of genotypes in the two fields and identical genotypes could be sampled from both fields. Analysis of gene diversity showed that Glomus populations are subdivided between plots within each field, while there was no subdivision between fields. This suggests that individuals related to the dominant mycelial network are favoured while the establishment of genetically different individuals is restricted. This hypothesis is supported by studies of AMF in natural vegetation where the community is dominated by even larger mycelial networks that may interconnect different plant species.

110. Mating system transitions in the euascomycete genus Aspergillus subgenus Fumigati section Fumigati. Carla Rydholm1, Mathieu Paoletti2, Paul S. Dyer2, Francois Lutzoni1. 1 Duke University, North Carolina. 2 University of Nottingham, Nottingham, UK.

The evolutionary origin of Aspergillus fumigatus is of applied and theoretical interest. This study has uncovered data which suggests the divergence of this species was coincident with a mating system transition from homothallism to heterothallism. To test this hypothesis, a combined approach is used where the arrangements of mating loci and their flanking genes are characterized for multiple heterothallic and homothallic species from Aspergillus subgenus Fumigati section Fumigati. The likelihood of heterothallism as the derived state for A. fumigatus and other heterothallics of the group is estimated within a phylogenetic framework using ancestral character state reconstruction techniques. At this point, preliminary results for both types of evidence, from the organization of the mating loci and reconstruction of ancestral state, support a transition from homothallism to heterothallism having taken place within the genus Aspergillus subgenus Fumigati section Fumigati.

111. Population genetics of Aspergillus fumigatus and its close relative Neosartorya fischeri. Carla Rydholm1, George Szakacs2, Francois Lutzoni1. 1 Duke University, North Carolina, USA. 2 Technical University of Budapest, Budapest, Hungary.

Aspergillus fumigatus is a saprophytic euascomycete mold with a ubiquitous presence worldwide. It is also a pervasive pathogen to immunosuppressed patients. Despite intensive work to understand its success as a pathogen, little information is known regarding the population dynamics and recent evolutionary history of this species. We examined patterns of variation at three intergenic loci in a sample of 60 natural isolates from various parts of the world. These loci were also used to analyze site specific variation for 30 strains isolated from five localities. For both data sets no evidence of population structure was detected and there was no association between the genetic and geographic distances among different natural isolates. Thus, nothwithstanding its nearly worldwide occurrence, little evidence was found for local adaptation in A. fumigatus strains derived from diverse locations. Since conflicting observations have been made as to the amount of genetic variation within A. fumigatus, we contrast its effective population size with that of its closest relative, Neosartorya fischeri, and conclude A. fumigatus has comparatively low intraspecific genetic variation. Based on linkage disequilibrium measures A. fumigatus is inferred to have a recombining population structure.

112. Naturally Occurring Hybrids of Fusarium graminearum. Amgad A. Saleh, John F. Leslie, and Robert L. Bowden. Department of Plant Pathology, Kansas State University, Manhattan, KS

Fusarium graminearum (Gibberella zeae) is a major cause of cereal scab, in particular Fusarium scab (head blight) of wheat. DNA sequence analyses have been used to define nine phylogenetic lineages that have been accorded species rank. In this study, we used DNA sequence from four nuclear genes (MAT-1-13, alpha-tubulin, reductase, and Tri101) and AFLP analyses to determine if hybrids between these lineages are present under field conditions. We analyzed 200 isolates of F. graminearum collected from wheat and sorghum at eleven locations in Brazil and Uruguay. Three fields contained isolates of more than one lineage. In these fields, lineage 7 strains represented more than 90% of the sample. We found hybrids between lineages 1 and 7; 2 and 7; and 7 and 8. Our data suggest that there are no substantial barriers to gene flow between different lineages in the same location, and that the different phylogenetic lineages belong to a single biological species.

113. A Phylogenomic Approach to Reconstructing the Diversification of Proteases in Fungi. G. Hu and R.J. St. Leger. University of Maryland, Department of Entomology, College Park, MD 20742

Gene duplication and loss are important mechanistic antecedents in generating functional diversification. Here we adopted a phylogenomic approach with fungi of very different virulence and habitat to survey and characterize their serine proteinases (subtilases and trypsins) with the goal of providing a framework of information on these important enzymes, as well as improving understanding of general processes in fungal gene family evolution. The survey was based on 9 fungal genomes and expressed sequence tags from the insect pathogen Metarhizium anisopliae. Comparing subtilases between species revealed that basidiomycetes (Cryptococcus neoformans, Coprinus cinereus, Ustilago maydis) and saprophytic ascomycetes (Saccharomyces cerevisiae, Schizosaccharomyces pombe, Aspergillus nidulans, Neurospora crassa) lack the large gene families encoding secreted enzymes found in the pathogenic ascomycetes (M. anisopliae, Magnaporthe grisea, Fusarium graminearum). Patterns of intron loss and the degree of divergence between paralogs indicate that the proliferation of subtilisins classes I and II in pathogens occurred after the basidiomycete/ascomycete split but predated radiation of ascomycete lineages. This suggests that the early ascomycetes had a lifestyle that selected for multiple proteases, while the current disparity in gene numbers between ascomycete lineages results from retention of genes in pathogens that have been lost in saprophytes. However, the pathogens retained and occasionally expanded different gene families. Thus, M. grisea has 15 class I subtilisins and 6 class II subtilisins, while M. anisopliae and F. graminearum each possess 11 class II subtilisins but three or fewer class I subtilisins. This reveals evolutionary selection of different gene families among Ascomycete fungi. A prevailing trend towards lineage specific gene loss was shown by the distribution of trypsins across 35 representative fungi. Trypsin genes are lacking in most saprophytes, but are present in a basidiomycete insect symbiont (Septobasidium canescens), most zygomycetes and many ascomycete plant and insect pathogens. The patchy distribution of trypsins suggest that their phylogenetic breadth will have been much wider in early fungi than currently and confirms the dynamic nature of the fungal genome.

114. Molecular Adaptations for Host Range Diversification in a Fungal Pathogen. G. Hu, F.M. Freimoser and R.J. St. Leger. University of Maryland, Department of Entomology, College Park, MD 20742

Host pathogen interactions are an important force shaping organismal diversity, yet little is known about the evolution of genes responsible for virulence in pathogens. The tremendous amount of genetic variation, distinct disease phenotypes and host ranges of strains of the insect pathogen Metarhizium anisopliae have made it an excellent model to study the role of gene duplication/divergence in generating the functional diversification of enzymes and toxins necessary for adaptation to different hosts. To illustrate this, we present examples where strains with broad host ranges and strains with very narrow host ranges have diverged through changes in gene regulation, gene duplication/loss, and gene divergence.

115. Fungal Intron Evolution. Jason E Stajich1, Scott W Roy2, Fred S Dietrich1. 1Department of Molecular Genetics and Microbiology Duke University, Durham, NC. 2 Department of Molecular and Cellular Biology, Harvard University, Cambridge, MA

Understanding the evolutionary changes in exon and intron structure is an important aspect of gene evolution. The recent genome sequencing projects have given us a rich sampling of fungal genomes to study the changes in gene structure. The availability of the sequence for over 25 fungal genomes including intron rich genomes of Basidiomycete and Euascomycete fungi and the intron poor Hemiascomycetes permit the investigation of intron evolution in the fungal kingdom.

We have evaluated alignments of orthologous sequences to identify putatively orthologous introns, those which fell with in or between the same codons in the alignment of the genes. This analysis allowed us to observe the number of shared intron positions among the different fungal genes and identify many introns which are putatively ancient among the Euascomycete and Basidiomycete clades.

We used a maximum likelihood framework to reconstruct the ancestral density of introns in 800 orthologous genes across the Basidiomycota, Hemiascomycota, and Euascomycota and found evidence of massive and recurrent episodes of intron loss. This analysis suggests while intron gain may play some role, intron loss is the dominant force in fungal gene structure evolution.

116. Delineation of species boundaries within Fusarium graminearum, the causative agent of Fusarium Head Blight. D. E. Starkey, USDA, NCAUR, MGB, Peoria, IL, 61604; T. J. Ward, USDA, NCAUR, MGB, Peoria, IL, 61604; K. L. O'Donnell, USDA, NCAUR, MGB, Peoria, IL, 61604; D. M. Geiser, Dept. of Plant Pathology, Pennsylvania State Univ., University Park, PA 16802; G. Kuldau, Dept. of Plant Pathology, Pennsylvania State Univ., University Park, PA 16802; R. M. Clear, Canadian Grain Commission, Winnipeg, MB R3C 3G8; L. R. Gale, USDA Cereal Disease Lab, St. Paul, MN 55108; H. C. Kistler, USDA Cereal Disease Lab, St. Paul, MN 55108, and T. Aoki, MAFF, Tsukuba, Japan.

The primary etiological agent of Fusarium Head Blight (FHB), Fusarium graminearum, has been regarded as a single panmictic species. However, phylogenetic analyses of multilocus DNA sequence data has indicated that this morphospecies comprises at least 9 phylogenetically distinct, and biogeographically structured species (hereafter referred to as the Fg clade) using genealogical concordance phylogenetic species recognition (O'Donnell et al., 2000; Taylor et al., 2000; Ward et al., 2002). In a preliminary screening of FHB strains, O'Donnell et al. (2004) formally proposed species rank for the eight unnamed cryptic species within the Fg clade. In this study, a panel of 2520 FHB strains from around the world was screened to determine the species' geographic and host distribution, using portions of two phylogenetically informative genes (EF1-alpha and reductase, 1601 bp/strain). The trichothecene toxin chemotype of each strain was also determined using a multiplex PCR assay targeting TRI3 and TRI12 (Ward et al. 2002). The present study resulted in the discovery of 4 cryptic species, including a novel clade in Australia and the presence of 6 Fg clade species within the U.S. The development and application of robust molecular tools, such as those described herein, for Fg clade species identification and chemotype determination will significantly improve disease surveillance and global monitoring efforts, and will make available for the first time detailed information on the geographic and host distributions of FHB pathogens and enhance the current knowledge of the ecology, epidemiology and population dynamics of these mycotoxigenic cereal pathogens.

117. Development and amplification of multiple co-dominant genetic markers from single spores of arbuscular mycorrhizal fungi by nested multiplex PCR. Eva H. Stukenbrock and Søren Rosendahl. Department of Microbiology, Institute of Biology, University of Copenhagen, Oester Farimagsgade 2D, DK-1353 Copenhagen K, Denmark

Multilocus genotyping of arbuscular mycorrhizal fungi has so far only been carried out using dominant markers as AFLP and ISSR. The interpretation of these dominant markers may however be difficult as homology of the polymorphic loci cannot be ensured. Here we report the development of a nested multiplex PCR method which allows amplification of multiple co-dominant markers from single spores of Glomus mosseae, G. caledonium and G. geosporum in one PCR reaction.

The polymorphic co-dominant markers were identified and characterized in intron regions in the three protein coding genes GmFOX2, GmTOR2 and GmGIN1 by sequencing and SSCP. Additionally, we included Glomus specific primers for the well characterized LSU rDNA D2 region in the nested multiplex PCR. Seven different isolates of G. mosseae were characterized as seven unique multilocus genotypes. We found only one sequence type of each protein coding gene in each spore suggesting a homokaryotic structure of the multikaryotic Glomus spores.

Multilocus genotyping of Glomus single spores using co-dominant markers provide an important tool for future studies of evolution and population genetics of these obligate biotrophic fungi.

118. Development of a PCR-RFLP-based identification system for Fusarium asiaticum and genetic characterization of western Japanese isolates. H. Suga1, G.W. Karugia2, T. Ward3, L. R. Gale4, K. Tomimura5, T. Nakajima5, K. Kageyama6, M. Hyakumachi2

1Life Science Research Center, Gifu University, Japan; 2Fac. of Applied Biological Sciences, Gifu University, Japan; 3National Center for Agricultural Utilization Research, USDA-ARS, Peoria, IL, USA; 4USDA-ARS Cereal Disease Laboratory, St. Paul, MN, USA; 5KONARC, National Agriculture and Bio-oriented Research Organization, Japan; 6River Basin Research Center, Gifu University, Japan

The Fusarium graminearum clade consists of at least nine phylogenetically and sometimes biogeographically distinct species. Two Japanese isolates have previously been shown to belong to Fusarium asiaticum O'Donnell, T. Aoki, Kistler et Geiser (=Fusarium graminearum lineage 6). To further investigate the Japanese population, 175 isolates were obtained from wheat and barley with head blight symptoms collected from western Japan in 2002. All 18 randomly selected isolates had F. asiaticum-unique nucleotides in their reductase gene although two main sequence types were found among them. One of the unique nucleotides was present in a MseI recognition site of both main types and this restriction site was used for the development of a PCR-RFLP-based identification system for Fusarium asiaticum and characterization of all isolates. Additionally, isolates were subjected to PCR-based trichothecene chemotyping (15 acetyl deoxynivalenol: 15ADON, 3 acetyl deoxynivalenol : 3ADON or nivalenol: NIV). Among 170 isolates that were identified as F. asiaticum, 107 (63%), 62 (36%), and 1(1%) were of the NIV, 3ADON and 15ADON type, respectively. Reductase gene sequencing identified three isolates as F. graminearum O'Donnell, T. Aoki, Kistler et Geiser (=Fusarium graminearum lineage 7). Among these, two isolates were of the 3ADON and one of the 15ADON type. Using 10 VNTR markers, all 26 randomly selected F. asiaticum isolates displayed different haplotypes; therefore a high level of genetic diversity was observed.

119. Development of Simple Sequence Repeat Markers for Puccinia graminis and P. triticina. Les J. Szabo and Kim Phuong Nguyen, USDA ARS Cereal Disease Lab, University of Minnesota, St. Paul, Minnesota, USA

Rusts are obligate plant parasitic fungi with a complex life cycle that often include five different spore stages. As a result, molecular studies are often performed with the asexual, dikaryotic uredinial stage (urediniospores) using dominant markers (RAPDs, AFLPs). Currently, molecular genetic and population genetic studies are underway with the wheat rust fungi Puccinia graminis f.sp. tritici and P. triticina the casual agents of wheat stem rust and wheat leaf rust, respectively. The lack of co-dominant molecular markers for genetic analysis has been a significant impediment. Simple sequence repeat (SSR) enriched libraries were constructed from P. graminis and P. triticina genomic DNA. Approximately, 200 clones from each library were screened and sequenced. Sixty SSR markers have been developed from the P. graminis clones and the majority consists of di-nucleotide (AG and GT) repeats ranging from 10 to 28 repeats in length. Eleven of the repeats were complex and often contained degenerate motifs. Thirty of the markers have been screened against ten P. graminis North American isolates, 27 of which were polymorphic and detected from 2 to 10 alleles per marker. Fifty-six markers have been developed from the P. triticina clones, the majority of which consists of di-nucleotide (AG and GT) repeats and 13 consist of tri-nucleotide repeats (CAA). Screening of these markers against a select set of P. triticina isolates have begun.

120. Unusual insertions of phylogenetic relevance in the nuclear ITS region of some Peronosporaceae (Oomycetes). Marco Thines1, Hedwig Komjati2, Mark Bachofer1, Otmar Spring1. 1Institute of Botany of the University of Hohenheim, Stuttgart, Germany, 2Department of Plant Protection of the Szent Istvan University, Goedoelloe, Hungary

The ITS region is a sequence routinely screened for the elucidation of the phylogeny of related species. For the Peronosporales, some of which cause severe economic loss in agriculture, no such sequence data are available for a broad array of taxa. Amplification of the nuclear ITS region in Plasmopara halstedii resulted in a 2700bp fragment, which could not be sequenced by primer walking, due to repeated elements in the ITS-2 part. A restriction-ligation technique uncovered four tandemly arranged elements of 322bp in length each. These are part of an insertion found neither in Phytophthora nor in Peronospora species, in which total ITS-lengths of about 900bp were observed. The elements revealed variable parts in comparison to each other and were divided by short spacers of 34bp to 38bp in length. The tandemly arranged elements showed sequence variation in samples of different geographic origin, in contrast to the uniformity of the other parts of the ITS. Other species of Plasmopara and Bremia revealed lengths of ITS up to 3kB. These lengths are also derived from similar insertions, which may become a tool to trace the phylogeny of the Peronosporaceae. A comparative analysis of the repetitions in Plasmopara halstedii and Bremia lactucae will be presented.

121. Population genetics of the aflatoxigenic species, Aspergillus flavus and Aspergillus parasiticus. Nai Tran-Dinh1, John Pitt1 and Dee Carter2. 1Food Science Australia, CSIRO, North Ryde, NSW, Australia. 2School of Molecular and Microbial Biosciences, The University of Sydney, Sydney, NSW, Australia.

Aspergillus flavus and Aspergillus parasiticus are closely related, morphologically similar species belonging to the Aspergillus section Flavi. Both species can produce aflatoxins, but not all strains of either species do so. Aflatoxin contamination of crops such as peanuts and corn is a worldwide problem. A possible solution is the use of nontoxigenic strains of Aspergillus as biocontrol agents to competitively exclude their toxigenic counterparts in the field. Knowledge of genetic diversity, strain identification and the potential for genetic exchange are essential for predicting the likely success of such a strategy. RAPD and microsatellite analysis of 35 Australia strains distinguished A. flavus from A. parasiticus strains. A. flavus strains were further divided into two distinct groups; Groups 1 and 2. Group 1 and 2 A. flavus strains differed in their toxin production profiles, and mode of reproduction analysis revealed evidence for recombination in Group 1 A. flavus but not in Group 2. These differences suggested that Group 2 A. flavus are a new species. A global survey of 296 strains, including strains from Africa, Asia, Oceania, Europe, North and South America, using microsatellites, revealed 271 different genotypes and that both A. flavus and A. parasiticus strains showed no correlation between genotype and geographic origin or ability to produce aflatoxin. The majority of A. flavus strains studied clustered within Group 1 A. flavus. Group 2 A. flavus were mainly restricted to the Southern Hemisphere. Implications for a biocontrol strategy are discussed.

122. Molecular markers used for speciation in the genus Alternaria. Nai Tran-Dinh, Ailsa Hocking and John Pitt. Food Science Australia, CSIRO, North Ryde, NSW, Australia.

The genus Alternaria is a major problem in Australian grains (wheat, sorghum, barley) and is also involved in post-harvest loss of citrus fruit and tomatoes. Some Alternaria species are toxigenic, producing a range of toxins including tenuazonic acid, which has been shown to be toxic to plants, animals and insects. Despite its importance, speciation of Alternaria infecting crops is poorly understood and delineated. Alternaria species are principally identified by microscopic examination of conidial morphology and occasionally by some biochemical properties. Identification of Alternaria species by such methods is often difficult and requires specialised training. Conidial morphologies of various species have overlapping features, and some morphological characters are not consistently expressed in culture. Environmental conditions also affect the variability of morphological characteristics. With over 1000 Alternaria species names having been published since the genus was named in 1817, some confusion exists in its taxonomy. The advent of molecular techniques and their application to fungal taxonomy has provided an additional tool for speciation. This study uses various molecular markers for the identification of Alternaria species including amplified fragment length polymorphisms and microsatellite markers. A collection of type strains, including the Alternaria species A. alternata, A. citri, A. infectoria, A. solani, and A. tomato, were used to access the suitability of these molecular markers for the identification of Alternaria species.

123. Application of three molecular marker techniques to characterise strains of Aspergillus niger. Alexandre Esteban1, 2, Su-lin L. Leong1, 3, 4 and Nai Tran-Dinh1. 1Food Science Australia, PO Box 52, North Ryde NSW 1670, Australia. 2Departament de Sanitat i d'Anatomia Animals, Facultat de Veterinària, Universitat Autònoma de Barcelona, E-08193 Bellaterra, Barcelona, Spain. 3School of Agriculture and Wine, University of Adelaide, PMB 1, Glen Osmond, SA 5064, Australia. 4Cooperative Research Centre for Viticulture, PO Box 154, Glen Osmond, SA 5064, Australia

Aspergillus niger is a member of the black aspergilli (Aspergillus sect. Nigri) and is a common food spoilage fungus. A. niger holds GRAS (Generally Regarded As Safe) status and is widely used in the food industry as a source of hydrolytic enzymes and organic acids. However, some strains of A. niger are able to produce the nephrotoxin, ochratoxin A (OTA), and have been associated with OTA in coffee and grape products. Despite its importance, the taxonomy of A. niger, and other members of the black aspergilli, remains unclear. Species identification is primarily based on morphological criteria, but a growing number of molecular techniques are being applied to the A. niger aggregate. RFLP analyses have divided the A. niger aggregate into types N and T, and RAPD analyses have shown a high level of intraspecific variability. The molecular techniques, Enterobacterial Repetitive Intergenic Consensus (ERIC) PCR, Amplified Fragment Length Polymorphisms (AFLP) and microsatellite markers were evaluated for their suitability in typing strains of A. niger. Strains assessed included isolates from culture collections and those isolated from natural substrates including grapes, coffee, animal feed and soil. The collection included both toxigenic and nontoxigenic isolates. ERIC-PCR differentiated A. niger from other closely related black aspergilli. AFLP analysis separated A. niger strains into types N and T, and this was confirmed by analysis using six novel microsatellite markers, developed for A. niger. Furthermore, both the AFLP and microsatellite analyses separated type N strains into two distinct groups. No correlation was seen between toxin production and genotype.

124. Multi-gene phylogeny and evolution of thraustochytrids. Clement K.M. Tsui, W. Marshall, D. Honda*, R. Yokoyama*, M.L. Berbee. Department of Botany, #3529-6270 University Blvd., The University of British Columbia, Vancouver, B.C., Canada V6T 1Z4 *Department of Biology, Faculty of Science and Engineering, Konan University, Kobe

The labyrinthulomycetes are fungal-like protists composed of thraustochytrids, aplanochytrids and labyrinthulids. The thraustochytrids, including three major genera – Thraustochytrium, Schizochytrium, and Ulkenia, are abundant heterotrophs in marine and mangroves habitats. Previous phylogenetic investigation based on SSU rDNA revealed that none of the genera Thraustochytrium, Schizochytrium and Ulkenia is monophyletic, indicating that the morphological characters are unreliable as taxonomic criteria. So we are applying multi protein-coding gene phylogenies (Elongation factor 1-alpha, beta-tubulin, actin) to re-evaluate the relationships among these genera and species. Preliminary results from analysis of individual gene and combined data appear to be congruent to rDNA that these genera do not form monophyletic groups. We will also attempt to evaluate the phylogenetic relationships among different groups of stramenopiles that include non-photosynthetic and photosynthetic protists whose zoospore flagella bear tripartite tubular hairs. Additional strains are also being isolated from marine animals, as part of the investigation of eukaryotic diversity. Their SSU rDNA sequences are determined and included in the phylogenetic analysis. Results indicated that there are at least eight distinct taxa within labyrinthulomycetes, and their protein genes will be further investigated and analysed.

125. Partitioning reproductive isolation in Neurospora: measuring the strength of component reproductive isolation barriers and their relative contributions to isolation between species. Elizabeth Turner, David J. Jacobson, John W. Taylor. University of California, Berkeley, USA

Sexual isolation between closely related species results from the cumulative effects of several isolating barriers acting at various points throughout the reproductive pathway. To understand how reproductive isolation evolves and how genetic isolation is maintained, it is necessary to understand the nature of the barriers, the strength of the barriers, and the contribution of each barrier to total reproductive isolation. Experiments that dissect the effects of individual barriers allow us to estimate what proportion of heterospecific gene flow a given barrier prevents and to assess natural variation in barrier strength and the effects that different mating conditions have on barrier performance. The contribution of a given barrier depends both on the strength of that barrier and the strengths of all barriers that precede it in the sexual pathway, and barriers acting earlier in the pathway will have a disproportionately large effect. We demonstrate the utility of this method by analyzing experiments that partition reproductive isolation between Neurospora crassa and N. intermedia into the following components: trichogyne/conidium interactions, perithecia production, spore melanization, spore viability, and viable spore production. We find that barrier strength varies between pairs of strains and that reproductive isolation profile is altered when females are simultaneously fertilized by conspecific and heterospecific males.

126. Reinforcement of reproductive isolation barriers in Neurospora: selective advantage and genetic basis of the early abortion of hybrid perithecia. Elizabeth Turner, David J. Jacobson, John W. Taylor. University of California, Berkeley, USA.

Reproductive isolation barriers between lineages can evolve by several different mechanisms. Both neutral and selected changes will accumulate in diverging genomes and these can render hybrids between lineages inviable or unfit due to negative epistasis. In a process termed "reinforcement," isolation barriers can themselves be positively selected if they prevent the wasteful investment of reproductive effort in the formation of unfit hybrids. The signature of reinforcement is reproductive isolation barriers that are stronger in sympatry versus allopatry. In particular, barriers will be positively selected that permit females to discriminate between hybrid and conspecific matings. Previous work from this lab has shown that hybrid matings between Neurospora individuals from sympatric populations do experience stronger barriers than matings between allopatric individuals. We now present results of two sets of experiments exploring a potential fitness advantage and the genetic basis of one putative reinforcement barrier, the very early abortion of N. crassa peritheicia fertilized by sympatric males of N. intermedia. First we show that this barrier can dramatically increase the overall fecundity of these females if they have additional opportunities to mate with conspecific males. Second we identify a major quantitative trait locus associated with the reinforcement phenotype.

127. Improved gene replacement in Aspergillus awamori by Agrobacterium-mediated transformation. C.B. Michielse1, M. Arentshorst1, A.F.J. Ram1,2 and C.A.M.J.J. van den Hondel1,2. 1Leiden University, Institute of Biology Leiden, Fungal Genetics Research Group, Wassenaarseweg 64, 2333 AL Leiden, The Netherlands, 2Department of Applied Microbiology and Gene Technology, TNO-Nutrition, 3700 AJ Zeist, The Netherlands.

The efficiency of gene replacement in A. awamori between Agrobacterium-mediated transformation and CaCl2/PEG mediated transformation has been compared. For two genes targeted, it was found that the homologous recombination frequencies obtained by Agrobacterium-mediated transformation were 3 to 6-fold higher than the frequencies obtained with the CaCl2/PEG mediated transformation method. Furthermore, it was found that efficient homologous recombination with shorter DNA flanks could be obtained in Agrobacterium-mediated transformation. Finally, the addition of the dominant amdS marker as a second selection marker to the gene replacement cassette led to a further 2-fold enrichment of putative gene replacement transformants, resulting in a homologous recombination frequency of 55%. Based on the data it can be concluded that Agrobacterium-mediated transformation is an efficient tool for gene replacement and that the amdS gene can be successfully used as a second selection marker to enrich the pool of putative gene replacement transformants

128. Generation of a large set of microsatellite-markers for Phytophthora infestans by mining sequence data. Theo van der Lee1, Odette Mendes1, Hannneke van der Schoot1, Caroline Ruyter-Spira1, Bas te Lintel Hekkert1, Francine Govers2 and Gert H. J. Kema1. Plant Sciences Group, 1Plant Research International and 2Laboratory of Phytopathology, Wageningen University, Wageningen, The Netherlands

Microsatellite or simple-sequence repeat (SSR) markers are extremely useful in population studies particularly in diploid species. So far, however, only a limited number of informative microsatellite loci have been identified in the potato late blight pathogen Phytophthora infestans and none have been mapped. To identify additional polymorphic SSR loci, genomic and EST sequences were scanned for the presence of di- and trinucleotide units that are repeated six or more times. We identified 333 unique SSR loci using an automated software pipeline. Primers flanking these SSRs were developed and tested on a set of ten previously characterized P. infestans field isolates. Of the 300 primers pairs tested, 203 pairs generated a clear fragment of the expected length. More than half of these amplified microsatellites (110) showed length differences among the different isolates. Some microsatellites seemed to be very variable in the P. infestans population with up to 9 different alleles detected in the ten genotypes. The most informative microsatellites are currently being positioned on the genetic linkage map of cross 71 (80029 x 88133). The genomic and EST sequences (xgi.ncgr.org/spc/; Randall et al., 2005 MPMI in press) proved to be an excellent source for new microsatellites. We anticipate that these SSR markers will be instrumental for efficient analyses of P. infestans populations and will facilitate integration of the various genetic datasets available for this important plant pathogen.

129. Phylogeography of the Ustilago maydis Virus in the USA and Mexico. Voth, Peter1, Lockhart, Ben E.2 and May, Georgiana1. Depts. of Plant Biological Sciences1 and Plant Pathology2, University of Minnesota, St. Paul, MN USA

Symbiotic interactions strongly affect the evolutionary trajectory of the participating symbionts and are important in structuring communities. These interactions are such that the population genetic structure of one symbiont can affect that of the other. One such symbiosis is that of Ustilago maydis, Ustilago maydis Virus (UMV), and Zea mays. Ustilago maydis, commonly known as corn smut, causes infections on vegetative and reproductive tissues of corn. UMV is vertically transmitted through cytoplasmic fusion during mating of U. maydis individuals and, thus, the population dynamics of U. maydis can affect the population structure of UMV. In this work, I investigate the evolutionary interaction of a dsRNA virus (UMV) with U. maydis to determine the phylogeography of UMV throughout the USA and Mexico. I have collected sequence data from regions of two genes on the UMV genome and assessed infection frequency, phylogenetics, genetic diversity, pairwise Fst, and recombination frequencies. These analyses have revealed that Mexican populations of UMV demonstrate higher genetic diversity and infection rates than do populations from the USA, local populations in the USA are composed of multiple UMV lineages, and populations of UMV are not in equilibrium with mutation, migration, and drift.

130. A Multilocus SNP Microsphere Array for Identification of Fusarium Head Blight Species and Chemotypes. Todd J. Ward, David Starkey, Brent Page, and Kerry O'Donnell. Microbial Genomics & Bioprocessing Research Unit, National Center for Agricultural Utilization Research, USDA Agricultural Research Service, Peoria, IL 61604

Fusarium head blight (FHB) poses a serious constraint to the production of wheat and barley worldwide. Infested grains are often unsuitable for food or feed due to contamination with trichothecene toxins and estrogenic compounds. In addition, FHB generally causes a significant reduction in seed quality and yields. A detailed understanding of pathogen diversity is crucial for effective disease control programs that minimize the threat of FHB. It appears that only a fraction of FHB species/chemotype diversity is currently represented within North America. Therefore, the introduction of novel FHB pathogens or chemotypes via global trade in agricultural products has the potential to exacerbate the FHB problem in the U.S. Using a unique multi-locus DNA sequence database (11 nuclear genes, 13.6 kb of DNA sequence) we have developed a high-throughput single tube assay for the simultaneous identification of all known B-trichothecene FHB species and chemotypes in order to improve disease surveillance efforts and to facilitate a greater understanding of the ecology, epidemiology, and population dynamics of these FHB pathogens.

131. Vegetative compatibility groups in Botryotinia fuckeliana. Pauline L Weeds, and Ross E Beever. Landcare Research, Auckland, New Zealand.

A vegetative compatibility system is known to occur in the plant pathogenic fungus Botryotinia fuckeliana (Botrytis cinerea). We have used complementing nitrate-non utilising (Nit) mutants to study the distribution and inheritance of vegetative compatibility groups (VCGs). Field populations were found to comprise many VCGs, with 60 identified amongst 71 field strains. While most VCGs contained a single strain, 5 contained 2 or more members. A sample of 28 single ascospore progeny from a sexual cross between 2 field strains in different VCGs generated 9 VCGs, including both parental types. Considering both field and single ascospore strains we now recognise 67 VCGs for B. fuckeliana. Our findings are consistent with the presence of multiple vegetative compatibility (vic) genes; if 2 alleles are assumed for each gene, our data suggest the existence of at least 7 such genes. Genetic diversity of the field populations was assessed using 8 microsatellite markers. Some field strains (25%) showed 2 alleles for one or more markers, consistent with the hypothesis they are heterokaryons. Excluding these strains, most field strains had different haplotypes, except for those collected in a hierachical manner in one vineyard block, where there was evidence of clonality and inbreeding. The parents of the sexual cross differed for 5 of the microsatellite markers. These segregated approximately 1:1 and none were closely linked. A total of 14 haplotypes were detected amongst the 28 progeny, although neither of the parental haplotypes was recovered.

132. Examination of the phylogeny and possible function of genes specific to A. fumigatus and A. fischerianus. Jennifer R. Wortman1, Natalie D. Fedorova1, Charles Lu1, Jonathan Badger1, Michael J. Anderson2, Neil Hall1, Jonathan A. Eisen1, William C. Nierman1. 1The Institute for Genomic Research, Rockville, MD, USA. 2The University of Manchester, Manchester, UK.

Comparative analysis of the genomes of Aspergillus fumigatus, a human pathogen, with the model organism Aspergillus nidulans and the industrial agent Aspergillus oryzae has led to the identification of more than 500 genes that appear specific to A. fumigatus. Provocatively, many of these A. fumigatus genes have few or no homologs in the genomes of other eukaryotes available in public databases. Instead, a subset shows significant sequence similarity to bacterial or archaeal proteins. Interesting examples include heat shock protein GroES, death-on-curing family protein Doc, and arsenate reductase ArsC.

The preliminary genomic sequence and automated annotation of a fourth Aspergillus genome, Aspergillus fischerianus (Neosartorya fischeri), has allowed us to investigate further the A. fumigatus-specific gene set. As A. fischerianus is the most closely related species to A. fumigatus that has been identified (apart from the sister species A fumigatus var. ellipticus), we can now determine that approximately half of the genes previously identified as A. fumigatus-specific genes are actually shared between A. fumigatus and A. fischerianus. We are also investigating genes that appear unique to A. fischerianus.

Some of these A. fumigatus and A. fischerianus-specific genes are likely to confer significant selective advantage to these fungi in competing for resources and in adapting to diverse environments. We are currently exploring the origin of these genes using phylogenetic methods in order to posit whether they were selectively retained by these species or acquired through lateral gene transfer.

133. Biodiversity in Ten members of Stachybotrys from Ukraine through molecular analysis of the small subunit ribosomal DNA. Veronica Amaku, Oliver Rojas, Gary Andersen, Tamas Torok, Duane Pierson, Mark Ott, and Olufisayo Jejelowo

Members of the genus Stachybotrys are of interest due to the implication of Stachybotrys chartarum in sick building syndrome and infant pulmonary hemosiderosis. Current identification techniques which involve cultural and microscopical identification of colonies, conidia and hyphae, are time consuming and subjective. They depend on human evaluation, which varies according to the level of the analyst's expertise and experience. Recent advances in molecular biology provide tools that can be exploited to improve accuracy and time required for fungal identification. Although the use of molecular techniques for fungal identification is very promising, it is in infant stage and still being explored. In this study, we analyzed the small subunit ribosomal DNA of ten members of the genus Stachybotrys using PCR and Sanger sequencing. Four of the isolates matched similar sequences in existing nucleotide databases, three isolates could not be amplified, and sequences from the remaining three isolates matched Penicillium species.

134. Sequencing and annotation of the genome of the human pathogenic basidiomycete Cryptococcus neoformans serotype D, Strain JEC21. Paolo Amedeo, Paola Roncaglia, Iain Anderson, Rama Maiti, Bernard B. Suh, Brian J. Haas, Jonathan E. Allen, Mihaela Pertea, Jessica Vamathevan, Viktoriya Grinberg, Florenta R. Riggs, Terry R. Utterback, Steven L. Salzberg, Jennifer R. Wortman, Brendan J. Loftus, Claire M. Fraser

Cryptococcus neoformans is a soil borne basidiomycete growing as encapsulated yeast and responsible of serious respiratory diseases and meningitis in humans. Endemic of Australia, Vancouver Island (since recently) and many other places having mild climate, it is frequently isolated ubiquitously from soil and, more copiously, from birds' excrements.

It's clinical importance has risen during the last decades due to AIDS epidemic and the usage of immunosuppressive drugs to treat autoimmune diseases and after organ transplantations.

Five main serotypes (A, B, C, D, and A-D) of C. neoformans sensu lato are known, being serotype A the most important from a clinical perspective, but serotype D the first of which strains of opposite mating types were isolated allowing genetic studies.

Strain JEC21 has been completely sequenced and fully assembled in chromosomes virtually without any sequence gap. Gene structure have been manually curated and alternatively spliced isoforms have been identified thanks the alignment of the end-sequences of about 23000 full-length cDNAs.

In this poster we describe the sequencing and annotation of this strain and we compare it with the sequence draft of a partially co-genic strain (B3501A) sequenced by Stanford Genome Technology Center.

135. Viewing Annotated Aspergillus Genomes on the Central Aspergillus Data Repository (Cadre) Website. JEM Gilsenan MJ Anderson TK Attwood SG Oliver NW Paton GD Robson DW Denning. The University of Manchester, Manchester, UK.

CADRE has been funded to house publicly available genomic data for all Aspergillus species. The Ensembl database schema and software have been adapted and used to set up a Website (http://www.cadre.man.ac.uk/). Using a Web browser, the user can view the position of mapped features, such as protein-coding genes, on sequence contigs. For each feature, links are provided to allow the user to retrieve further data. For protein-coding genes, such data include: chromosomal co-ordinates; a description of the encoded protein's function; similarity matches; and displays of transcript structure and protein features. Two complete Aspergillus genomes are currently available on the Website. The A. nidulans genomic sequence and annotation has been provided by the Broad Institute and consists of 28.6Mb of DNA with 9520 genes. The sequence consists of 248 contigs assembled into 27 supercontigs. The next release of the A. nidulans sequence will incorporate manually corrected gene reports brought to our attention during the writing of the genome sequence paper. The A. fumigatus sequence was determined by The Institute for Genomic Research (TIGR) and the Sanger Institute. Annotation was generated automatically and manual changes have been incorporated by TIGR during the process of writing the genome paper. The genome consists of 28.8Mb of DNA with 10,034 genes. The sequence has been assembled into 16 chromosomal arms stretching from the telomere to the centromere. Two arms contain gaps: one of these is only 5kb and the other one represents the ribosomal DNA repeat.

136. The Candida Genome Database (CGD), a curated community resource for Candida albicans gene and protein information. Martha B. Arnaud, Maria C. Costanzo, Marek S. Skrzypek, Gail Binkley, Christopher Lane, Stuart R. Miyasato, and Gavin Sherlock. Stanford University School of Medicine, Stanford, CA

The Candida Genome Database (CGD) contains genomic information and provides community resources for researchers who are interested in the opportunistic fungal pathogen Candida albicans. CGD is freely available at http://www.candidagenome.org/. CGD displays C. albicans information collected from the scientific literature, including C. albicans gene names and aliases; Gene Ontology (GO) terms describing the molecular function, biological process, and subcellular localization of each gene product; mutant phenotypes; and free-text description lines to summarize the function and biological context of each gene product. CGD also provides community resources. At the request of the community, CGD has implemented a gene name reservation system to facilitate gene name assignment before publication and to avoid future publication of conflicting gene names. CGD also hosts a colleague registry through which Candida researchers can share contact information and research interests. CGD is currently under development, and we welcome comments and suggestions. To contact CGD curators, send email to candida-curator@genome.stanford.edu. CGD is supported by NIH grant R01 DE15873-01 from the NIDCR at the NIH.