Large-scale genomic analyses with machine learning uncover predictive patterns associated with fungal phytopathogenic lifestyles and traits.

Invasive plant pathogenic fungi have a global impact, with devastating economic and environmental effects on crops and forests. Biosurveillance, a critical component of threat mitigation, requires risk prediction based on fungal lifestyles and traits. Recent studies have revealed distinct genomic patterns associated with specific groups of plant pathogenic fungi. We sought to establish whether these phytopathogenic genomic patterns hold across diverse taxonomic and ecological groups from the Ascomycota and Basidiomycota, and furthermore, if those patterns can be used in a predictive capacity for biosurveillance. Using a supervised machine learning approach that integrates phylogenetic and genomic data, we analyzed 387 fungal genomes to test a proof-of-concept for the use of genomic signatures in predicting fungal phytopathogenic lifestyles and traits during biosurveillance activities. Our machine learning feature sets were derived from genome annotation data of carbohydrate-active enzymes (CAZymes), peptidases, secondary metabolite clusters (SMCs), transporters, and transcription factors. We found that machine learning could successfully predict fungal lifestyles and traits across taxonomic groups, with the best predictive performance coming from feature sets comprising CAZyme, peptidase, and SMC data. While phylogeny was an important component in most predictions, the inclusion of genomic data improved prediction performance for every lifestyle and trait tested. Plant pathogenicity was one of the best-predicted traits, showing the promise of predictive genomics for biosurveillance applications. Furthermore, our machine learning approach revealed expansions in the number of genes from specific CAZyme and peptidase families in the genomes of plant pathogens compared to non-phytopathogenic genomes (saprotrophs, endo- and ectomycorrhizal fungi). Such genomic feature profiles give insight into the evolution of fungal phytopathogenicity and could be useful to predict the risks of unknown fungi in future biosurveillance activities.

Population Genomic Analyses Reveal Connectivity via Human-Mediated Transport across Populus Plantations in North America and an Undescribed Subpopulation of Sphaerulina musiva.

Domestication of plant species has affected the evolutionary dynamics of plant pathogens in agriculture and forestry. A model system for studying the consequences of plant domestication on the evolution of an emergent plant disease is the fungal pathogen Sphaerulina musiva. This ascomycete causes leaf spot and stem canker disease of Populus spp. and their hybrids. A population genomics approach was used to determine the degree of population structure and evidence for selection on the North American population of S. musiva. In total, 122 samples of the fungus were genotyped identifying 120,016 single-nucleotide polymorphisms after quality filtering. In North America, S. musiva has low to moderate degrees of differentiation among locations. Three main genetic clusters were detected: southeastern United States, midwestern United States and Canada, and a new British Columbia cluster (BC2). Population genomics suggest that BC2 is a novel genetic cluster from central British Columbia, clearly differentiated from previously reported S. musiva from coastal British Columbia, and the product of a single migration event. Phenotypic measurements from greenhouse experiments indicate lower aggressiveness of BC2 on Populus trichocarpa. In summary, S. musiva has geographic structure across broad regions indicative of gene flow among clusters. The interconnectedness of the North American S. musiva populations across large geographic distances further supports the hypothesis of anthropogenic-facilitated transport of the pathogen.

Cryptic Speciation in Western North America and Eastern Eurasia of the Pathogens Responsible for Laminated Root Rot.

Coniferiporia sulphurascens is a facultative fungal pathogen that causes laminated root rot (LRR) in commercially important coniferous species worldwide. This fungus spreads primarily by way of vegetative mycelium transferring at points of contact between infected and healthy roots. Successful intervention to control LRR requires a better understanding of the population structure and genetic variability of C. sulphurascens. In this study, we investigated the population genetic structure and origin of C. sulphurascens populations in western North America and eastern Eurasia collected from multiple coniferous hosts. By analyzing the small and large mitochondrial ribosomal RNA subunit genes combined with six nuclear loci (internal transcribed spacer region, actin, RNA polymerase II largest subunit, RNA polymerase II second-largest subunit, laccase-like multicopper oxidase, and translation elongation factor 1-α), we observed that none of the alleles among the loci were shared between North American (NA) and Eurasian C. sulphurascens populations. In total, 55 multilocus genotypes (MLGs) were retrieved in C. sulphurascens isolates occurring in these two continental regions. Of these, 41 MLGs were observed among 58 isolates collected from widespread locations in British Columbia (Canada) and the northwestern United States, while 14 MLGs were observed among 16 isolates sampled in Siberia and Japan. Our data showed that the levels of genetic differentiation between the NA and Eurasian populations are much greater than the populations from within each continental region; the two continental populations formed clearly divergent phylogenetic clades or lineages since they were separated approximately 7.5 million years ago. Moreover, the Eurasian population could be the source of the NA population. Our study indicates the existence of cryptic diversity in this pathogen species, and strongly suggests that the NA and Eurasian populations represent two lineages, which have progressively diverged from each other in allopatry.

Characterization of microsatellite loci in the fungus, Grosmannia clavigera, a pine pathogen associated with the mountain pine beetle.

The largest forest pest epidemic in Canadian history caused by the mountain pine beetle (MPB) and its fungal associates has killed over 15 million hectares of forest. Sixty simple sequence repeat regions were identified from Grosmannia clavigera, an MPB associated fungus. Eight loci genotyped in 53 isolates from two populations in British Columbia, Canada revealed three to 10 alleles per locus and gene diversities of 0 to 0.79. All but two of these loci showed length polymorphism in Leptographium longiclavatum, a related MPB fungal associate. These microsatellites will be useful in population genetic studies of these fungi.

First Report of Shining Willow as a Host Plant for Septoria musiva.

During the summer of 2001, leaf spots resembling those caused by Septoria musiva Peck. were observed on shining willow (Salix lucida Mühl. subsp. lucida) at Leclerville, Québec, Canada (46°34'19″N,71°59'35″W). Affected leaves had brown, necrotic leaf spots (>5mm in diameter) surrounded by a darker brown halo. Conidia were cylindrical, straight to curved with 1 to 4 septa, 28 to 54 × 3.5 to 4 µm, and were produced in pycnidia located on the abaxial surface in the center of the leaf spots. The causal agent of this disease was successfully isolated by germinating the conidia on corn meal agar that was supplemented with streptomycin (50mg/ml) and chloramphenicol (300mg/ml) and followed with the transfer of the germinated conidia to potato dextrose agar. Leaf symptoms and morphology matched those of S. musiva, the cause of leaf spot and stem canker of hybrid poplars in North America (2,4). The internal transcribed spacers and the 5.8S portion of the rDNA were amplified using PCR with the ITS1 (5'-TCC GTA GGT GAA CCT GCG G-3') and ITS2 (5'-GCT GCG TTC TTC ATC GAT GC-3') primer pair on total genomic DNA extracted from a pure culture of the pathogen. The rDNA sequence obtained (GenBank Accession No. AY555277) had 100% identity at 506 base positions with the ITS1, 5.8S, and ITS2 of three S. musiva isolates from Québec and one from Wisconsin (GenBank Accession Nos. AY549464 to AY549467). To test for pathogenicity, excised leaf disks from plants propagated by softwood cuttings of the source plant and from one hybrid poplar clone (Populus maximowiczii × P. xjackii) were inoculated with 3 µl of a suspension of ground mycelium or sterile water (control). Disks were placed in a 24-well tissue culture plate with 1 ml of distilled water per well and incubated in a growth room maintained at 22°C with a 16-h photoperiod. After 1 month, symptoms were similar to those previously observed. Isolates collected from shining willow or hybrid poplar were able to induce S. musiva leaf spot symptoms on leaf disks excised from shining willow or the hybrid poplar clone. From symptomatic leaf disks, S. musiva was consistently reisolated. To our knowledge, this is the first report of S. musiva on a member of the genus Salix. S. didyma, S. salicicola, and S. salicina have been reported from leaves of species of Salix (1,3). Only a vague morphological description of S. didyma was found (3). Moreover, conidia of S. salicicola (20 to 50 × 2.5 to 3.5 µm) and S. salicina (40 to 60 µm long, unspecified width) overlap dimensions of S. musiva conidia (1). There is a need to reexamine the relationships between these species of Septoria. Evidently, the complete host range of S. musiva is not yet known. References: (1) L. Lanier et al. Mycologie et Pathologie Forestières. Masson. Paris, 1978. (2) M. E. Ostry. Eur. J. For. Pathol. 17:158, 1987. (3) P. A. Saccardo. Sylloge fungurum omnium hucusque cognitorum. Patavii: Sumptibus Auctoris, 1882. (4) L. J. Spielman et al. Plant Dis. 70:968, 1986.