Tom Hsiang

Comparative analysis of the genomes of two field isolates of the rice blast fungus Magnaporthe oryzae.

Rice blast caused by Magnaporthe oryzae is one of the most destructive diseases of rice worldwide. The fungal pathogen is notorious for its ability to overcome host resistance. To better understand its genetic variation in nature, we sequenced the genomes of two field isolates, Y34 and P131. In comparison with the previously sequenced laboratory strain 70-15, both field isolates had a similar genome size but slightly more genes. Sequences from the field isolates were used to improve genome assembly and gene prediction of 70-15. Although the overall genome structure is similar, a number of gene families that are likely involved in plant-fungal interactions are expanded in the field isolates. Genome-wide analysis on asynonymous to synonymous nucleotide substitution rates revealed that many infection-related genes underwent diversifying selection. The field isolates also have hundreds of isolate-specific genes and a number of isolate-specific gene duplication events. Functional characterization of randomly selected isolate-specific genes revealed that they play diverse roles, some of which affect virulence. Furthermore, each genome contains thousands of loci of transposon-like elements, but less than 30% of them are conserved among different isolates, suggesting active transposition events in M. oryzae. A total of approximately 200 genes were disrupted in these three strains by transposable elements. Interestingly, transposon-like elements tend to be associated with isolate-specific or duplicated sequences. Overall, our results indicate that gain or loss of unique genes, DNA duplication, gene family expansion, and frequent translocation of transposon-like elements are important factors in genome variation of the rice blast fungus.

Baseline sensitivity and cross-resistance to demethylation-inhibiting fungicides in Ontario isolates of Sclerotinia homoeocarpa

Four hundred and thirty-five isolates of Sclerotinia homoeocarpa from eight populations in southern Ontario were tested for sensitivity to the demethylation-inhibiting (DMI) fungicides, propiconazole, myclobutanil, fenarimol and tebuconazole. The isolates were collected in summer 1994 just prior to legal DMI fungicide use on turfgrass in Ontario. There were wide variations in sensitivities, and seven of the eight populations were very sensitive to the fungicides. Based on mean EC50 and the distribution of DMI sensitivity, one population near the U.S. border was suspected of having been previously exposed to DMI fungicide. Pairwise comparisons of EC50 values for the different fungicides showed low to moderate correlations between fungicides. EC50 values of myclobutanil and propiconazole had the best correlation, followed by the pair of tebuconazole and fenarimol. Other pairwise comparisons were not statistically significant except for a barely significant relationship between EC50 values of myclobutanil and tebuconazole. For field populations of plant pathogens, cross-resistance to different DMI fungicides may not be as strong as conventionally thought. The data collected here will allow comparison to subsequent years to look for detectable shifts in S. homoeocarpa sensitivity to DMI fungicides as they become more frequently used in Ontario.

Specific adaptation of Ustilaginoidea virens in occupying host florets revealed by comparative and functional genomics

Ustilaginoidea virens (Cooke) Takah is an ascomycetous fungus that causes rice false smut, a devastating emerging disease worldwide. Here we report a 39.4 Mb draft genome sequence of U. virens that encodes 8,426 predicted genes. The genome has ~25% repetitive sequences that have been affected by repeat-induced point mutations. Evolutionarily, U. virens is close to the entomopathogenic Metarhizium spp., suggesting potential host jumping across kingdoms. U. virens possesses reduced gene inventories for polysaccharide degradation, nutrient uptake and secondary metabolism, which may result from adaptations to the specific floret infection and biotrophic lifestyles. Consistent with their potential roles in pathogenicity, genes for secreted proteins and secondary metabolism and the pathogen-host interaction database genes are highly enriched in the transcriptome during early infection. We further show that 18 candidate effectors can suppress plant hypersensitive responses. Together, our analyses offer new insights into molecular mechanisms of evolution, biotrophy and pathogenesis of U. virens.

Use of green fluorescent protein to quantify the growth of Colletotrichum during infection of tobacco

To develop a quantitative assay of fungal growth inside plant tissues, strains of Colletotrichum destructivum and Colletotrichum orbiculare were transformed with a modified green fluorescent protein (GFP) gene fused with a glyceraldehyde-3-phosphate dehydrogenase promoter from Aspergillus nidulans. Transformants expressed GFP in culture and had the same growth rate and general appearance as the wild type. GFP was observed in all fungal structures during infection of leaves of Nicotiana benthamiana, except for the melanized appressoria and setae. The timing and appearance of the fungal structures in the host appeared to be identical to that of the wild type. GFP accumulation in inoculated leaves of N. benthamiana was quantified in leaf extracts using a fluorescence microplate reader, and the quantity of fluorescence was strongly correlated with the growth of the fungus as measured by the amount of fungal actin gene expression using Northern blot hybridizations. These results demonstrated that assaying green fluorescence levels from a GFP-transformed fungus is an accurate, fast and easy means of quantifying fungal growth inside host plant cells.

Microbial populations and suppression of dollar spot disease in creeping bentgrass with inorganic and organic amendments

Alginate, ammonium nitrate, Bovamura, Milorganite, Ringer lawn Restore, Ringer Greens Super, Ringer Turf Restore, Sandaid, sewage sludge and sulphur-coated urea were evaluated from 1991 to 1993 on a creeping bentgrass (Agrostis palustris) green and a Kentucky bluegrass (Poa pratensis) lawn for their effects on soil bacterial and fungal populations and dollar spot disease incidence. Over the 3 yr, fertilizers were applied every 4 wk at recommended rates from early June to September, and once again in November. Application of Ringer fertilizers, ammonium nitrate, and sulfur-coated urea gave rise to significantly higher microbial populations on turfgrass leaves and in thatch and soil than did most other fertilizers. In most experiments, Ringer fertilizers also improved water retention in thatch compared with other treatments. Ringer Greens Super, Ringer Turf Restore, or ammonium nitrate on the creeping bentgrass green significantly suppressed dollar spot disease compared with the other amendments or the untreated control, but for most of the season they did not control disease as well as the fungicide chlorothalonil did

Hemibiotrophic infection and identity of the fungus, Colletotrichum destructivum , causing anthracnose of tobacco

The causal agent of tobacco anthracnose was identified as Colletotrichum destructivum based on the morphology of the fungus and a comparison of the sequence of the rDNA ITS with those of other Colletotrichum species. The infection process on tobacco (Nicotiana tabacum and N. benthamiana) was examined by light microscopy, which revealed that the pathogen acted as an intracellular hemibiotroph. Penetration occurred preferentially at the anticlinal walls of epidermal cells by an appressorium and penetration peg. An infection vesicle formed in the penetrated host cell by 48 h after inoculation, and out of this, a multi-lobed infection vesicle grew which remained limited to the initially infected cell. The interaction at this point was biotrophic, which was confirmed by plasmolysis and accumulation of a vital stain by the infected host cells. Thin secondary hyphae arose from multi-lobed infection vesicles at 60 h after inoculation, which then penetrated the host cell wall and began the necrotrophic phase of the infection. Acervuli formed on the plant surface by 96 h after inoculation, typically with a single melanized seta. In addition to tobacco, the fungus could infect alfalfa, cowpea, and Medicago truncatula, but not soybean. The process of infection of C. destructivum in tobacco was very similar to that previously reported in alfalfa and cowpea.

Infection of Nicotiana species by the anthracnose fungus, Colletotrichum orbiculare

Colletotrichum gloeosporioides f. sp. malvae, isolate Biomal®, ATCC 20767, was originally isolated from round-leaved mallow (Malva pusilla) and developed as a weed biocontrol agent. Ribosomal DNA sequence analysis was recently used to re-classify this fungus as C. orbiculare, which is an aggregate species with a number of formae speciales. Several morphological features of ATCC 20767 were examined that were consistent with those described for C. orbiculare, and inoculation of a number of Nicotiana species and several cultivars of N. tabacum showed that this fungus was pathogenic to many of these previously undescribed hosts. Spore germination and appressorium formation were higher on tobacco than previously observed on round-leaved mallow. The pathogen produced melanized appressoria on N. tabacum leaves that formed preferentially at the anticlinal epidermal cell wall. A symptomless phase of infection persisted for 72–96 h postinoculation, during which time the fungus first produced a spherical infection vesicle from an infection peg, and then large primary hyphae which grew through the epidermal cells. The large primary hyphae were highly constricted at the points of penetration of the host cell walls. Thin secondary hyphae appeared at 96–120 h postinoculation coinciding with the appearance of light green, water-soaked spots and the formation of acervuli. The infection of tobacco by C. orbiculare ATCC 20767 is not a non-specific interaction but appears to follow an intracellular hemibiotrophic infection process that is very similar to that established for the C. orbiculare infection of round-leaved mallow, cucurbits and beans.

Quantifying fungal infection of plant leaves by digital image analysis using Scion Image software

A digital image analysis method previously used to evaluate leaf color changes due to nutritional changes was modified to measure the severity of several foliar fungal diseases. Images captured with a flatbed scanner or digital camera were analyzed with a freely available software package, Scion Image, to measure changes in leaf color caused by fungal sporulation or tissue damage. High correlations were observed between the percent diseased leaf area estimated by Scion Image analysis and the percent diseased leaf area from leaf drawings. These drawings of various foliar diseases came from a disease key previously developed to aid in visual estimation of disease severity. For leaves of Nicotiana benthamiana inoculated with different spore concentrations of the anthracnose fungus Colletotrichum destructivum, a high correlation was found between the percent diseased tissue measured by Scion Image analysis and the number of leaf spots. The method was adapted to quantify percent diseased leaf area ranging from 0 to 90% for anthracnose of lily-of-the-valley, apple scab, powdery mildew of phlox and rust of golden rod. In some cases, the brightness and contrast of the images were adjusted and other modifications were made, but these were standardized for each disease. Detached leaves were used with the flatbed scanner, but a method using attached leaves with a digital camera was also developed to make serial measurements of individual leaves to quantify symptom progression. This was successfully applied to monitor anthracnose on N. benthamiana leaves. Digital image analysis using Scion Image software is a useful tool for quantifying a wide variety of fungal interactions with plant leaves.

Biological control of fusarium head blight of wheat with Clonostachys rosea strain ACM941

Fusarium head blight (FHB), caused by Gibberella zeae, is a devastating disease of wheat. A strain of Clonostachys rosea, ACM941 (American Type Culture Collection ATCC 74447), was evaluated for antibiosis against G. zeae in vitro and for control of FHB under greenhouse and field conditions in comparison to the registered fungicide Folicur (tebuconazole). Strain ACM941 reduced mycelial growth of the pathogen by 52.6% in dual-culture after 6 days and completely suppressed spore germination for 6 h when cocultured with a macroconidial suspension of G. zeae. Strain ACM941 reduced G. zeae perithecial production by more than 99% in a leaf disk assay, 60%–77% on infected corn kernels, and 32%–57% on spikelet debris in the field. These effects were significant (P < 0.05) and not statistically different from those produced by tebuconazole. When strain ACM941 was sprayed onto wheat heads 2 days prior to inoculation with G. zeae, it significantly reduced infected spikelets (IS) by 64% and Fusarium -damaged kernels (FDK) by 65% in greenhouse experiments. Under simulated disease epidemic conditions during 2005–2007, strain ACM941 reduced the FHB index by 58%, IS by 46%, FDK by 49%, and deoxynivalenol (DON) in kernels by 21%. These effects were significant but lesser in magnitude than those achieved by tebuconazole, which reduced FHB index by 97%, IS by 82%, FDK by 73%, and DON by 62%. Results from this research suggest that strain ACM941 of C. rosea is a promising biocontrol agent against G. zeae and may be used as a control measure in an integrated FHB management program.

Comparison of the Yeast Proteome to Other Fungal Genomes to Find Core Fungal Genes

The purpose of this research was to search for evolutionarily conserved fungal sequences to test the hypothesis that fungi have a set of core genes that are not found in other organisms, as these genes may indicate what makes fungi different from other organisms. By comparing 6355 predicted or known yeast (Saccharomyces cerevisiae) genes to the genomes of 13 other fungi using Standalone TBLASTN at an e-value <1E-5, a list of 3340 yeast genes was obtained with homologs present in at least 12 of 14 fungal genomes. By comparing these common fungal genes to complete genomes of animals (Fugu rubripes, Caenorhabditis elegans), plants (Arabidopsis thaliana, Oryza sativa), and bacteria (Agrobacterium tumefaciens, Xylella fastidiosa), a list of common fungal genes with homologs in these plants, animals, and bacteria was produced (938 genes), as well as a list of exclusively fungal genes without homologs in these other genomes (60 genes). To ensure that the 60 genes were exclusively fungal, these were compared using TBLASTN to the major sequence databases at GenBank: NR (nonredundant), EST (expressed sequence tags), GSS (genome survey sequences), and HTGS (unfinished high-throughput genome sequences). This resulted in 17 yeast genes with homologs in other fungal genomes, but without known homologs in other organisms. These 17 core, fungal genes were not found to differ from other yeast genes in GC content or codon usage patterns. More intensive study is required of these 17 genes and other common fungal genes to discover unique features of fungi compared to other organisms.