The pathogenic mechanisms of Tilletia horrida as revealed by comparative and functional genomics

Abstract

Tilletia horrida is a soil-borne, mononucleate basidiomycete fungus with a biotrophic lifestyle that causes rice kernel smut, a disease that is distributed throughout hybrid rice growing areas worldwide. Here we report on the high-quality genome sequence of T. horrida; it is composed of 23.2 Mb that encode 7,729 predicted genes and 6,973 genes supported by RNA-seq. The genome contains few repetitive elements that account for 8.45% of the total. Evolutionarily, T. horrida lies close to the Ustilago fungi, suggesting grass species as potential hosts, but co-linearity was not observed between T. horrida and the barley smut Ustilago hordei. Genes and functions relevant to pathogenicity were presumed. T. horrida possesses a smaller set of carbohydrate-active enzymes and secondary metabolites, which probably reflect the specific characteristics of its infection and biotrophic lifestyle. Genes that encode secreted proteins and enzymes of secondary metabolism, and genes that are represented in the pathogen-host interaction gene database genes, are highly expressed during early infection; this is consistent with their potential roles in pathogenicity. Furthermore, among the 131 candidate pathogen effectors identified according to their expression patterns and functionality, we validated two that trigger leaf cell death in Nicotiana benthamiana. In summary, we have revealed new molecular mechanisms involved in the evolution, biotrophy, and pathogenesis of T. horrida.

Figure 1

Characteristics of the Tilletia horrida strain. (a) A single nuclear mycelium stained with 4′, 6-diamidino-2-phenylindole (DAPI) after culture on PDA medium for 5 days, observed using a fluorescence microscope. Scale bars, 20 µm; (b) Teliospores under scanning microscopy; (c) and (d) Teliospores germination. (e) Morphology of secondary microspores. Scale bars, b-c: 10 µm; (f) Colony morphology of T. horrida after 15 d on PSA; (g) Hypha infection in rice kernels and growing points; (h) Kernel smut balls formed in rice spikelets.

Figure 2

Phylogenetic relationship of Tilletia horrida with other Basidiomycetes fungi. (a) The proportion (%) of different types of repetitive sequences in the T. horrida JY-521 genome. LINE: long interspersed elements, LTR: long terminal repeat retrotransposons. (b) Venn diagram showing orthologs between the five sequenced smut fungi. The values explain the counts of ortholog groups and the counts of genes in parentheses. (c) The phylogeny of 12 Basidiomycota fungi and 2 Ascomycota fungi. The phylogeny was constructed using Mega 6 with 806 single-copy gene. Protein alignments were analyzed using MUSCLE3.8.31.

Figure 3

Up-regulated and down-regulated genes according to Gene Ontology (GO) annotation.

Figure 4

Summary of Tilletia horrida genes assigned with CAZyme functional annotations.

Figure 5

The expression patterns of genes coding carbohydrate degradative enzymes of Tilletia horrida.

Figure 6

Genes and gene clusters encoding putative effectors in the Tilletia horrida genome. 120 putative effector genes are aggregated into 50 clusters. Shaded arrows represent putative effector genes while white arrows denote other genes.

Figure 7

Physical locations of predicted secreted protein genes, all protein-coding genes and PHI-base genes in relation to regions of repetitive sequences and GC content distribution in the assembled genome of T. horrida. From outside to in: Location of genes encoding predicted secreted proteins (purple) in the assembled genome; The distribution of transposable elements (blue) in the T. horrida genome; Single-copy DNA regions (red) of the T. horrida genome; Locations of the PHI-base gene homologues (green) involved in pathogen–host interactions; Graphs of GC (black) contents. Areas of low GC correspond well to regions of repetitive DNA. The maps were drawn with OmniMapFree.

Figure 8

Candidate effectors cause cell-death in Nicotiana benthamiana. (a) Phenotypes observed on N. benthamiana. The effectors are encoded by the smut_2965 and smut_5844 genes, respectively, and the cell death phenotypes were visible 4 d after inoculation with purified proteins. BAX gene was used as a positive control, and the GFP as the negative control. (b) Genes with signal peptides (red) and domain structures. Ribonuclease domains (blue) are identified from the Pfam database. (c) Clustering of predicted effectors.