Trajectory and genomic determinants of fungal-pathogen speciation and host adaptation
Edited by Jay C. Dunlap, Geisel School of Medicine at Dartmouth, Hanover, NH, and approved October 3, 2014 (received for review July 4, 2014)
Significance
Fossil records have provided compelling evidence for evolution, but lack of existing transitional species has hindered our understanding of speciation at the molecular level. Genomic analyses of seven Metarhizium species revealed a directional speciation continuum from specialists with narrow host ranges to transitional species and then to generalists that paralleled insect evolution. This diversification was coupled with a complex interplay between an array of genomic features that worked together to drive fungal speciation at an accelerating rate and provided a roadmap for identifying variation underlying adaptation and speciation. In particular, specialization was associated with retention of sexuality and rapid evolution of existing protein sequences whereas generalization was associated with loss of sexuality and protein-family expansion.
Abstract
Much remains unknown regarding speciation. Host–pathogen interactions are a major driving force for diversification, but the genomic basis for speciation and host shifting remains unclear. The fungal genus Metarhizium contains species ranging from specialists with very narrow host ranges to generalists that attack a wide range of insects. By genomic analyses of seven species, we demonstrated that generalists evolved from specialists via transitional species with intermediate host ranges and that this shift paralleled insect evolution. We found that specialization was associated with retention of sexuality and rapid evolution of existing protein sequences whereas generalization was associated with protein-family expansion, loss of genome-defense mechanisms, genome restructuring, horizontal gene transfer, and positive selection that accelerated after reinforcement of reproductive isolation. These results advance understanding of speciation and genomic signatures that underlie pathogen adaptation to hosts.
Sign up for PNAS alerts.
Get alerts for new articles, or get an alert when an article is cited.
Speciation is a central component of biological diversification and is increasingly viewed as a continuum or process rather than an event. However, failure to identify transitional species has hindered progress in understanding genomic patterns of divergence along the speciation continuum (1). Plant or animal pathogenic fungi are genetically tractable models for the study of speciation due to their diverse lifestyles and the occurrence of sibling species that differ from each other principally in host specificity (2, 3). However, fundamental questions remain, including whether generalization or specialization to particular hosts is the ancestral condition, whether we can identify the existence of transitional forms, and what are the underlying molecular mechanisms driving speciation (4).
We exploited the ascomycete genus Metarhizium, a radiating lineage of insect pathogens that are frequently used as biological insecticides (5, 6) and for genomic studies into the nature of adaptive differences by which novel pathogens emerge and form new species. Besides the previously sequenced Metarhizium robertsii (abbreviated as MAA) and Metarhizium acridum (MAC) (7), five new species were sequenced: Metarhizium album (MAM), Metarhizium majus (MAJ), Metarhizium guizhouense (MGU), Metarhizium brunneum (MBR), and Metarhizium anisopliae (MAN) (Dataset S1, Table S1). MAM is specific for hemipteran insects (8) whereas MAJ and MGU have intermediate host ranges as they are predominately associated with coleopteran insects but can also infect lepidopterans (9). Like MAA, MBR and MAN are generalists parasitizing a broad range of insects representing more than seven orders (10, 11). Generalist species such as MAA and MBR can also colonize the roots of plants (12), consistent with increased phenotypic flexibility.
Our analyses revealed that the evolutionary trajectory of Metarhizium spp. was from specialists via intermediate host range species to generalists that coincided with host insect diversification and availability. This host adaptation was coupled with a complex interplay between an array of genomic features that worked together to drive fungal speciation and provide a roadmap for identifying variation underlying adaptation and speciation.
Results
Genome Features.
Additional sequencing with a Roche 454 sequencer plus optical mapping-assisted scaffolding was used to improve the previously sequenced MAA genome. The improved genome contains 15 large scaffolds, 4 of them representing full chromosomes and 6 of them containing a single chromosome end with characteristic telomere repeats (TTAGGG/CCCTAA)n. The total genome size (41.7 Mb vs. 39.0 Mb) and number of predicted genes (11,689 vs. 10,582) were increased over previous estimations (Dataset S1, Table S2). We used the improved MAA genome as a reference to assist assembly of the other genomes. The completeness of each genome reached >98% (Dataset S1, Table S1). The hemipteran specialist MAM has a genome size (30.5 Mb) that is about 12 Mb smaller than the genomes of intermediate species MGU and MAJ (Dataset S1, Table S1). MGU and MAJ have more genes (average 11,318) than the generalists (11,089) and specialists (9,160). The generalists, however, have more protein families (average 2,992) than do the intermediate (2,918) and specialist species (2,841). An analysis of phylogenetically independent contrasts (PICs) (13) (SI Materials and Methods) revealed a significant relationship across the seven Metarhizium species between their genome size and gene-coding capacity (F = 26.75, P = 0.0035) and between their protein family size and ability to form appressoria on different substrates indicative of host range (F = 21.02, P = 0.0059). Thus, gene content is related to Metarhizium genome size and linked to fungal-host ranges.
A pan-genome analysis for Metarhizium spp. indicated that the core genome (genes present in all species) reaches a constant value of 6,466 genes (Fig. 1 A and B). Consistent with previous studies (14, 15), >50% of the species-specific genes in each Metarhizium sp. lack conserved domains (Dataset S1, Table S3). Most of the remaining lineage-specific genes have domains found in genes listed in the Pathogen–Host Interaction (PHI) database (16) or are effector-like small secreted cysteine-rich protein (SSCP) genes. SSCPs are frequently associated with host adaptation or specialization (17). The plot for estimating the number of new genes added by each genomic sequence fitted a decaying exponential, and mathematical extrapolation predicts that an average of 923 new genes will be identified for every new genome sequenced (Fig. 1C). The estimated exponent γ (= 0.261 ± 0.015) > 0 indicates that Metarhizium species have an open pan-genome (Fig. 1D), which is normally a feature of species that colonize multiple environments and have multiple ways of exchanging genetic materials (18).
Fig. 1.
Phenotyping and Phylogeny Reconstruction.
Metarhizium species differ from each other in spore size and shape (Fig. 2A and Dataset S1, Table S1). Formation of appresorial infection structures is the hallmark of host recognition and specificity (10, 11). Unlike the specialist and transitional species, the generalists formed appressoria on diverse insect substrates (SI Materials and Methods and Fig. S1), consistent with their wide host ranges (9). To infer their phylogenetic relationships, we used concatenated nucleotide sequences of 457 single-copy genes present in all species for a maximum-likelihood analysis. Consistent with a previous analysis (19), the results confirmed that the hemipteran-specific MAM is a basal species and that generalist Metarhizium spp. evolved from specialists (Fig. 2A). A consensus tree based on the frequency of individually inferred single-copy gene trees also supported this pattern of Metarhizium speciation (Fig. S2A).
Fig. 2.
Using a tree based on 316 single-copy universal genes, we placed Metarhizium spp. in the context of other fungal taxa and provided a time line for reconstructing their evolutionary history. We found that the monophyletic Metarhizium lineage diverged from clavicipitacean plant pathogens and endophytes about 231 million years ago (MYA), and MAM diverged from other Metarhizium species about 117 MYA (Fig. S2B). We speculate that the close physical proximity of the plant-associated ancestor of MAM to plant sap-sucking hemipteran bugs likely facilitated this particular host switch to entomopathogenicity. The acridid-specific MAC diverged 48 MYA within the mid-Eocene when newly evolved grasses were growing along riverbanks and grass-feeding acridids first appeared (20), supporting the principle of coevolution/cospeciation between hosts and their pathogens/symbionts (Fig. 2B) (17, 21). The intermediate species split 15 MYA followed by the generalists MAA and MAN that diverged from each other 7 MYA (Fig. 2A). These divergence times are consistent with the average nucleotide sequence identities estimated from orthologous genes: i.e., the basal MAM has the lowest nucleotide sequence identities (∼80%) with other species whereas the generalists share about 98% identity (Dataset S1, Table S4) and show that recent speciation processes are associated with increased phenotypic plasticity (Fig. 2B and Fig. S1). This plasticity coincided with climate change that was critical for a massive diversification of flowering plants, trees, and associated insects (22).
Protein-Family Expansion Associated with Metarhizium Speciation.
Analysis of protein families putatively involved in fungal virulence shows a dynamic loss and gain of genes (Fig. S3), but overall the generalists have larger gene families (Table 1 and Dataset S1, Tables S5–S13). In particular, we found a marked expansion of PHI genes in MAJ and MGU (average 1,958), and the generalists (average 1,899), relative to the specialists (average 1,468). For example, the MAM genome highlights the early expansion of genes involved in cuticle degradation as it has more than threefold more trypsin genes than the plant endophyte and phytopathogens (Dataset S1, Table S5). However, compared with MAM (87 proteases) and MAC (116 proteases), there has been additional expansion of proteolytic capacity in other Metarhizium species (average 165 proteases) (Table 1 and Dataset S1, Table S7).
Table 1.
| M. album (MAM) | M. acridum (MAC) | M. majus (MAJ) | M. guizhouense (MGU) | M. brunneum (MBR) | M. anisopliae (MAN) | M. robertsii (MAA) | |
|---|---|---|---|---|---|---|---|
| Total proteins | 8,472 | 9,849 | 11,535 | 11,787 | 10,689 | 10,891 | 11,689 |
| Protein families | 2,790 | 2,892 | 2,911 | 2,925 | 2,947 | 2,976 | 3,054 |
| Secreted proteins | 754 | 886 | 1,127 | 1,208 | 1,149 | 1,133 | 1,278 |
| PHI proteins | 1,308 | 1,628 | 1,903 | 2,013 | 1,868 | 1,874 | 1,956 |
| SSCPs | 209 | 206 | 298 | 326 | 288 | 285 | 333 |
| Proteases | 320 | 398 | 470 | 538 | 476 | 518 | 491 |
| Secreted proteases | 87 | 116 | 150 | 167 | 180 | 152 | 178 |
| Subtilisins | 28 | 43 | 56 | 58 | 60 | 56 | 63 |
| Trypsins | 12 | 15 | 25 | 29 | 32 | 26 | 36 |
| CAZy | 232 | 303 | 387 | 375 | 348 | 349 | 367 |
| GH18/GH | 12/108 | 14/126 | 20/147 | 28/162 | 16/140 | 20/148 | 24/154 |
| GPCR | 32 | 37 | 62 | 67 | 63 | 61 | 63 |
| Cytochrome P450 | 74 | 108 | 138 | 136 | 134 | 127 | 139 |
| Dehydrogenases | 191 | 279 | 294 | 320 | 309 | 310 | 313 |
| Transcription factors | 117 | 140 | 162 | 169 | 154 | 163 | 176 |
| SM clusters | 34 | 40 | 57 | 66 | 62 | 62 | 60 |
| HET proteins | 10 | 28 | 42 | 44 | 40 | 43 | 47 |
PHI proteins, pathogen–host interaction proteins; SSCPs, small secreted cysteine-rich proteins; CAZy, carbohydrate active enzyme; GH, glycoside hydrolase (GH18 for chitinase); GPCR, G protein-coupled receptor; SM clusters, secondary metabolite biosynthetic gene clusters; HET proteins, heterokaryon incompatibility control proteins.
Fungal G protein-coupled receptors (GPCRs) mediate host recognition and activation of downstream pathways to control fungal differentiation and development (23). Compared with MAM and MAC, there was a major expansion of GPCR-related proteins in MAJ, MGU, and the generalists (Table 1). In particular, the generalists showed a more than twofold expansion of Pth11-like receptors (average 51 vs. 23 in specialists) (Dataset S1, Table S8), with established roles in fungal virulence (23). There was a particular expansion in the SF1 and SF5 subfamilies (Dataset S1, Table S9), which are developmentally up-regulated during Metarhizium infection processes (7). A PIC test showed that the number of GPCRs was significantly correlated with the ability to produce appressoria on different substrates (F = 20.96, P = 0.006), consistent with receptors being important to fungal-host recognition.
Secondary metabolites produced by Metarhizium contribute to virulence and host specificity (9). The specialists MAM and MAC have fewer gene clusters encoding bioactive metabolites than other Metarhizium species or plant-associated fungi (Table 1 and Dataset S1, Table S10). Consistent with this observation, a PIC test revealed a significant relationship between these gene clusters and appressorial formation (F = 25.32, P = 0.004). Additional evidence about the lifestyles of nonspecialists could be found in their comparatively large number of effector-like proteins potentially involved in blocking host immune responses (Table 1), as well as an expansion of families that contribute to detoxification, including dehydrogenases (Dataset S1, Table S11), cytochrome P450s (Dataset S1, Table S12), and transporters (Dataset S1, Table S13). These proteins contribute to both primary and secondary metabolism and could contribute to host range through production of mycotoxins or detoxification of host metabolites (7, 24).
The genes involved in sex (25), asexual cytonuclear incompatibility (CI) (26), and heterokaryon incompatibility (HI) play pivotal roles in reproductive isolation (RI) (27, 28). Sex- and CI-related orthologs are similarly distributed in all Metarhizium species (Dataset S1, Table S14). Intriguingly, however, the number of HI proteins (HETs) is much expanded in the transitional and generalist species (Table 1). Based on HET genes that have been functionally studied (29), Metarhizium proteins were classified (Dataset S1, Table S15). The specialists have fewer HI-responsive HET-6 (29) genes (average 1 vs. 4 in nonspecialists) and HET genes without functional counterparts (average 11 vs. 28 in nonspecialists). This observation suggests that the nonspecialists have evolved tighter controls for achieving RI. In addition, PIC tests confirmed correlations between numbers of HET proteins and GPCRs (F = 21.22, P = 0.0058) and between the HETs and appressorial formation (F = 9.32, P = 0.0283), consistent with HET proteins being linked with fungal host specificity. We therefore investigated the evolutionary force(s) driving protein-family expansion and Metarhizium speciation.
Gene Duplication and Repeat-Induced Point Mutation.
Chromosomal or genome duplication, gene retroposition, or unequal crossing-over can result in gene duplication that leads to protein-family expansion, one of the recognized mechanisms of adaptive innovation (26). Our analysis of gene paralogy showed that, compared with specialists, the nonspecialists have higher percentages of paralogs with more than 50% nucleotide identity (Fig. S4A), consistent with their resulting from recent gene duplication events (26). For example, MAJ has 110 paralogous pairs of identical coding sequences whereas MGU has only 42 identical pairs. For duplicated genes to function, repeat-induced point mutation (RIP) must be disabled (30). Consistent with their different gene-family sizes, calculations of RIP indices indicate that RIP occurs in MAM and MAC, but not in the nonspecialists (Fig. S4B). RIP functions only during meiosis (30), which also suggests retention of sexuality in specialists although their sexual stages have not been verified (7).
Transposable Elements and Genome-Structure Diversification.
Consistent with RIP operating exclusively in specialist species, MAM and MAC have fewer transposable elements (TEs) encoding both DNA-type transposases (averages 39 vs. 57) and RNA-type retrotransposases (averages 9 vs. 57) than other Metarhizium species (Fig. S5A). TEs cause gene mutation, gene duplication, or genome-structure reorganization (31) so, not surprisingly, specialists have fewer pseudogenes than other species (average 219 vs. 351) (Dataset S1, Table S1 and Fig. S5B), and the nonspecialists exhibit numerous genome-structure rearrangements (Fig. S6). In particular, k-mer and contig length distribution analyses identified two peaks in the MAJ genome (Fig. 3 A and B). One possibility is that MAJ is a heterozygote because fungal asexual sporulation is a process of haploidization (32). The mating-type (MAT) locus represents the sex minichromosome in fungi (33). Analysis of the MAT loci showed that MAJ is unique among Metarhizium species in having both MAT1-1 and MAT1-2 loci (Fig. 4A), which was confirmed by PCR (Fig. 4B). We also verified that the spores of each Metarhizium species are uninucleate (Fig. 4C), precluding MAJ being a heterokaryon (i.e., two nuclei in one spore). Synteny analysis of MAJ and the closely related MGU confirmed the presence of duplicated segments in the MAJ genome (Fig. 3C).
Fig. 3.
Fig. 4.
Horizontal Gene Transfer.
Genome-wide analysis of horizontal gene transfer (HGT) events revealed that, as in other fungi (34), Metarhizium species acquired diverse genes from bacteria and archaea and even arthropods, plants, and vertebrates. The nonspecialist species have obtained more bacterial genes than the specialists (average 63 vs. 27) (Fig. S7A). Orthology analysis revealed that only eight HGT genes are present in all Metarhizium species whereas the remaining genes were probably acquired clade- or species-specifically because they are absent from multiple species (Fig. S7B). For example, experimentally verified bacterial-like proteins in MAA encoding a chymotrypsin (MAA_00986) (35), a pentose metabolizing phosphoketolase (MAA_04563) (36), a cold shock protein (MAA_05998) (37), and a host-to-fungus sterol carrier protein (MAA_03817) (38), as well as a putative chitinase (MAA_08315), are present in all seven Metarhizium species, indicating that these genes were acquired before Metarhizium lineages radiated. More recent acquisitions include a putative insect-like trypsin (MAA_03017) that is absent only in MAM and a putative TcdB-like insecticidal toxin protein (MAA_08634) that is unique to MAA.
HGT events were also implicated in orthology analysis of the secondary metabolism gene clusters. We found that only 15 clusters are present in all seven Metarhizium species (Fig. S8A and Dataset S1, Table S16). Around half of the clusters in MAM (11/19) and MAC (11/25) are species-specific (vs. an average of 7 in the other species) although most are also present in other fungal genera (Fig. S8B), implying that they were laterally acquired during Metarhizium speciation. For example, the gene cluster for producing insecticidal cyclopeptide destruxins (dtx) is absent in MAM and MAC (Fig. 2A and Fig. S8C) but is present in plant pathogenic Alternaria species (39). Thus, the dtx cluster may have spread laterally between fungal genera and been acquired by Metarhizium about 15 MYA. However, differential gene loss in MAM and MAC could have yielded similar gene-distribution patterns. In either event, the high diversity of gene clusters for secondary metabolites is evidence for a long-term arms race between host and pathogens.
Positive Selection.
Rapidly evolving genes are assumed to be under positive selection (PS) for evolutionary innovation (40). Based on the established phylogeny of Metarhizium species (Fig. 2A), we performed a series of likelihood ratio tests (LRTs) to detect positively selected genes (PSGs) [P < 0.01; false discovery rate (FDR) < 0.05] in the seven Metarhizium lineages (40). This analysis indicated that PSGs were unevenly distributed across Metarhizium lineages (Fig. 2A and Dataset S1, Tables S17–S23). There were 184 genes under PS in all seven Metarhizium genomes whereas 289 PSGs were under selection in MAC, significantly more than in other species or clades (Fig. 2A). Of the identified PSGs, 16–40% lack conserved domains (Fig. 2A), consistent with PS being a powerful force for creating rapidly changing sequences (41). For example, the Metarhizium-specific MCL1 gene (MAA_01665) is positively selected (Dataset S1, Table S14). MCL1 is crucial for evading host immunity (42) so its positive selection may be key to how Metarhizium spp. adapt to host defenses in new hosts. We also found that 47 of the 184 PSGs evolving in all Metarhizium genomes were organized into 23 clusters containing 2 to 11 genes; however, these proximal PSGs lacked similarity with each other and so were not the result of gene duplication (Dataset S1, Table S17). This pattern is different from plant pathogens where duplicated genes arranged in tandem undergo positive selection (43).
Gene categories enriched in PSGs included orthologs of putative PHI genes in plant pathogens (44) and genes previously identified as virulence determinants in Metarhizium spp., including serine proteases, chitinases, and CFEM (cysteine-rich fungal extracellular membrane)-containing proteins (Dataset S1, Table S17). Different types of transcription factors (TFs) were also rapidly evolving in Metarhizium spp. (Fig. 2A), particularly C2H2- and Zn2Cys6-type TFs (Dataset S1, Tables S17–S23). The former are largely involved in regulating fungal adaptation to environmental conditions whereas Zn2Cys6-type TFs control secondary metabolism and drug resistance (45). Positively selected C2H2-type and Zn2Cys6-type TFs were among those previously found to be highly expressed during early infection of insects (7). As well as changes in gene coding regions, pathogen host adaptation has also therefore involved integrating molecularly changed TFs into existing gene regulatory networks to produce lineage-unique repertoires of gene expression.
Discussion
It is still frequently argued that, if evolution were true, there should be more transitional species existing today (1). In this paper, we established that the evolutionary direction of Metarhizium speciation is from specialists to transitional species with an intermediate host range and then to generalists able to attack diverse hosts. Our analyses suggest that MAJ and MGU represent transitional species during the shift from narrow to broad host range Metarhizium species. In addition to their intermediate evolutionary positioning (Fig. 2A), MAJ and MGU have a host range limited to two insect orders and resemble specialists in not producing infection structures in glycerol (Fig. S1). On the other hand, MAJ and MGU genome sizes and encoding capacities are similar to those of generalist species. In addition, the presence of both MAT1-1 and MAT1-2 in MAJ suggests a homothallic pattern that differs from the heterothallic nature of other clavicipitacean species (46). Analysis of the MAT loci in mammalian pathogenic Cryptococcus spp. also revealed a transitional mating system, which has been converted to a heterothallic system in most species (33). MAJ may represent a similar transitional state. It is easier for fungi to lose sex than to regain it (47); therefore, the absence of an opposite MAT locus in the closely related MGU suggests that MGU might have recently diverged from MAJ and sustained gene loss. In general, Metarhizium speciation could provide a model for the evolution of host preference, specificity, and virulence in other pathogens. The analysis of three Fusarium species also suggested a transitional pattern from specialist to generalist, coupled with genome and protein family expansions (14). However, a few oomycete plant-pathogenic Phytophthora spp. with clear host records seem to have transitioned from generalist to specialist (48) so the general applicability of our data to other pathogens remains to be determined.
Clearly, there are factors missing from specialists that limit their ability to cause disease in multiple insects as demonstrated by an increase in host range after transfer of genes from generalist strains to the specialist MAC (10). Nevertheless, the large number of rapidly evolving genes in MAC shows that it has not remained functionally static, but specialization of MAC to acridids has involved rapid evolution of existing protein sequences rather than the extensive gene duplication shown in generalists. Classical theory predicts that specialists will be more likely than generalists to lose sex to prevent reassortment of well-adapted gene combinations (47), but lack of RIP in generalist Metarhizium species suggests that they are less likely than specialists to have cryptic sex or a recent sexual history. We speculate that an increase in protein-coding composition after loss of RIP in generalists facilitated their opportunistic lifestyles for different insect hosts. Similarly, the broad host range plant pathogen Fusarium oxysporum lacks RIP whereas the cereal specialists Fusarium graminearum and Fusarium verticillioides have strong RIP effects (14), suggesting that this pattern may be common in related fungal species with a wide range of host preferences. The vast majority (∼95%) of cataloged strains of the seven sequenced Metarhizium species in the Agricultural Research Service Collection of Entomopathogenic Fungal Cultures (ARSEF) belong to generalist species (www.ars.usda.gov/News/docs.htm?docid=12125), suggesting that this relationship is where most biodiversity within the Metarhizium genus resides and that broad host range is linked to ecological fitness.
Consistent with observations in other organisms (1, 4), speciation in Metarhizium was coupled with TE-mediated gene mutations and genome restructuring, HGT events, gene duplications, and positive selection. Reproductive isolation (RI) to halt genomic homogenization is the hallmark of new species emergence (1). In animals and plants, RI is reinforced by selection against maladapted hybrids (49). We found that sex-related genes are similarly distributed in Metarhizium species, but the nonspecialist species have many more HET proteins than do the specialists. In fungi, RI-like vegetative incompatibility mediated by HET proteins determines nonself recognition and cell death (30) and contributes to speciation (28). Thus, fungi may resemble plants and animals (49) in exhibiting RI reinforcement but mediated through the function of vegetative incompatibility.
In conclusion, our results provide clear connections between phenotype, genotype, and pathogen fitness that underlie the evolutionary trajectory of speciation and host adaptation. Except for examples from the fossil records, failure to identify existing transitional species has long hindered our understanding of speciation continuums (1). Having multiple genomes is providing a dynamic picture of the evolving species boundary and how genomic differentiation unfolds through time during speciation. In particular, the identification of existing transitional species not only solidifies the evolutionary theory of speciation continuum but also provides a roadmap for identifying variation crucial in the origins of biodiversity and host shifting.
Materials and Methods
Fungi and Culture Conditions.
Besides the previously sequenced species MAA (strain ARSEF 23, previously classified as M. anisopliae) and MAC (strain CQMa 102) (7), five additional Metarhizium species were genome-sequenced in this study, including MAM strain ARSEF 1941, MAJ strain ARSEF 297, MGU strain ARSEF 977, MBR strain ARSEF 3297, and MAN strain ARSEF 549 (Dataset S1, Table S1). Single-spore isolates were obtained from each strain and used for DNA isolation and genome sequencing. Fungal cultures were maintained on a potato dextrose agar (Difco) at 25 °C for 2 wk for sporulation.
Genome Sequencing.
To create a better-quality reference genome, M. robertsii was sequenced again with the Roche GS-GLX platform using the manufacturer’s standard protocols. To improve assembly with the previously acquired Illumina sequencing data (7), an optical-mapping analysis of MAA was performed using the Argus Optical Mapping System (OpGen). The restriction maps were assembled using the OM Assembler. De novo sequencing of MAM, MAJ, MBR, MGU, and MAN was performed with the Illumina Hiseq2000 system at the BGI-Shenzhen Company. The details of genome assembly, gene annotation, and analyses of TEs, RIP, and positively selected genes are provided in SI Materials and Methods.
Pan-Genome Analysis.
The number of genes shared by all Metarhizium species and the number of species-specific genes were extrapolated by fitting the exponential decaying functions
and
, respectively, where n is the number of sequenced genomes and κc, τc, κs, τs, Ω, and tg(θ) are free parameters (18). Pan-genome size was estimated using the power law fit
, where n is the number of sequenced genomes. The exponent γ > 0 indicates that the pan-genome is open whereas γ < 0 means that the pan-genome is closed (18).
Phylogenomic Analysis.
Based on the results from OrthoMCL analysis, the nucleotide sequences of 457 single-copy orthologous genes from each Metarhizium species were obtained and aligned individually with the software MUSCLE (www.ebi.ac.uk/Tools/msa/muscle/). A maximum-likelihood (ML) tree was generated with the concatenated sequences using the program RAxML, version 7.0.4, with 1,000 bootstrap searches for best models (50). The divergence time among different Metarhizium species was estimated as described (7). In addition, individual ML trees were generated using PhyML 3.1 (51), with 100 bootstrap replicates for each single-copy gene. A frequency-consensus tree was then generated using the program Phylip, version 3.695 (evolution.genetics.washington.edu/phylip.html). To infer insect-host phylogeny, the concatenated amino acid sequences of single-copy universal genes (1,491 genes) were used for maximum-likelihood analysis of selected insect-host species.
Mating-Type Gene Analysis and Nucleus Staining.
The presence of the MAT1-1-1 or MAT1-2-1 gene in different species was verified by PCR with the primer pairs mat111F/mat111R and mat121F/mat121R (15), respectively. Conidia were stained with the dye 4,6-diamidino-2-phenylindole to determine the number of nuclei.
Data Availability
Data deposition: The genome sequences reported in this paper have been deposited in the National Center for Biotechnology Information (NCBI) database [NCBI accession nos. AZHE00000000 (Metarhizium album), ADNI00000000 (Metarhizium acridum), AZNE00000000 (Metarhizium majus), AZNH00000000 (Metarhizium guizhouense), AZNG00000000 (Metarhizium brunneum), AZNF00000000 (Metarhizium anisopliae), and ADNJ02000000 (Metarhizium robertsii)].
Acknowledgments
We thank BGI-Shenzhen for genome-sequencing services. This work was supported by Strategic Priority Research Program of the Chinese Academy of Sciences Grant XDB11030100, National Nature Science Foundation of China Grant 31225023, and United States National Science Foundation Grant 1257685.
Supporting Information
Supporting Information (PDF)
Supporting Information
- Download
- 1.75 MB
pnas.1412662111.sd01.xlsx
- Download
- 341.40 KB
References
1
O Seehausen, et al., Genomics and the origin of species. Nat Rev Genet 15, 176–192 (2014).
2
LM Kohn, Mechanisms of fungal speciation. Annu Rev Phytopathol 43, 279–308 (2005).
3
T Giraud, G Refrégier, M Le Gac, DM de Vienne, ME Hood, Speciation in fungi. Fungal Genet Biol 45, 791–802 (2008).
4
EH Stukenbrock, Evolution, selection and isolation: A genomic view of speciation in fungal plant pathogens. New Phytol 199, 895–907 (2013).
5
RJ St Leger, C Wang, Genetic engineering of fungal biocontrol agents to achieve greater efficacy against insect pests. Appl Microbiol Biotechnol 85, 901–907 (2010).
6
CS Wang, M-G Feng, Advances in fundamental and applied studies in China of fungal biocontrol agents for use against arthropod pests. Biol Control 68, 129–135 (2014).
7
Q Gao, et al., Genome sequencing and comparative transcriptomics of the model entomopathogenic fungi Metarhizium anisopliae and M. acridum. PLoS Genet 7, e1001264 (2011).
8
MC Rombach, RA Humber, HC Evans, Metarhizium album, a fungal pathogen of leaf- and planthoppers of rice. Trans Br Mycol Soc 88, 451–459 (1987).
9
B Wang, Q Kang, Y Lu, L Bai, C Wang, Unveiling the biosynthetic puzzle of destruxins in Metarhizium species. Proc Natl Acad Sci USA 109, 1287–1292 (2012).
10
S Wang, W Fang, C Wang, RJ St Leger, Insertion of an esterase gene into a specific locust pathogen (Metarhizium acridum) enables it to infect caterpillars. PLoS Pathog 7, e1002097 (2011).
11
C Wang, RJ St Leger, Developmental and transcriptional responses to host and nonhost cuticles by the specific locust pathogen Metarhizium anisopliae var. acridum. Eukaryot Cell 4, 937–947 (2005).
12
SW Behie, PM Zelisko, MJ Bidochka, Endophytic insect-parasitic fungi translocate nitrogen directly from insects to plants. Science 336, 1576–1577 (2012).
13
J Felsenstein, Phylogenies and the comparative method. Am Nat 125, 1–15 (1985).
14
LJ Ma, et al., Comparative genomics reveals mobile pathogenicity chromosomes in Fusarium. Nature 464, 367–373 (2010).
15
P Zheng, et al., Genome sequence of the insect pathogenic fungus Cordyceps militaris, a valued traditional Chinese medicine. Genome Biol 12, R116 (2011).
16
R Winnenburg, et al., PHI-base update: additions to the pathogen host interaction database. Nucleic Acids Res 36, D572–D576 (2008).
17
P Schulze-Lefert, R Panstruga, A molecular evolutionary concept connecting nonhost resistance, pathogen host range, and pathogen speciation. Trends Plant Sci 16, 117–125 (2011).
18
H Tettelin, D Riley, C Cattuto, D Medini, Comparative genomics: The bacterial pan-genome. Curr Opin Microbiol 11, 472–477 (2008).
19
JW Spatafora, GH Sung, JM Sung, NL Hywel-Jones, Jr JF White, Phylogenetic evidence for an animal pathogen origin of ergot and the grass endophytes. Mol Ecol 16, 1701–1711 (2007).
20
TA Stidham, JA Stidham, A new Miocene band-winged grasshopper (Orthoptera: Acrididae) from Nevada. Ann Entomol Soc Am 93, 405–407 (2000).
21
AP Jackson, A reconciliation analysis of host switching in plant-fungal symbioses. Evolution 58, 1909–1923 (2004).
22
WW Hay, E Soeding, RM DeConto, CN Wold, The late Cenozoic uplift-climate change paradox. Int J Earth Sci 91, 746–774 (2002).
23
C Xue, YP Hsueh, J Heitman, Magnificent seven: Roles of G protein-coupled receptors in extracellular sensing in fungi. FEMS Microbiol Rev 32, 1010–1032 (2008).
24
NA Milani, DP Lawrence, AE Arnold, HD VanEtten, Origin of pisatin demethylase (PDA) in the genus Fusarium. Fungal Genet Biol 49, 933–942 (2012).
25
J Heitman, S Sun, TY James, Evolution of fungal sexual reproduction. Mycologia 105, 1–27 (2013).
26
P Gladieux, et al., Fungal evolutionary genomics provides insight into the mechanisms of adaptive divergence in eukaryotes. Mol Ecol 23, 753–773 (2014).
27
NL Glass, K Dementhon, Non-self recognition and programmed cell death in filamentous fungi. Curr Opin Microbiol 9, 553–558 (2006).
28
SJ Saupe, The [Het-s] prion of Podospora anserina and its role in heterokaryon incompatibility. Semin Cell Dev Biol 22, 460–468 (2011).
29
MA Van der Nest, et al., Distribution and evolution of het gene homologs in the basidiomycota. Fungal Genet Biol 64, 45–57 (2014).
30
JE Galagan, EU Selker, RIP: The evolutionary cost of genome defense. Trends Genet 20, 417–423 (2004).
31
MJ Daboussi, P Capy, Transposable elements in filamentous fungi. Annu Rev Microbiol 57, 275–299 (2003).
32
AM Neiman, Sporulation in the budding yeast Saccharomyces cerevisiae. Genetics 189, 737–765 (2011).
33
JA Fraser, et al., Convergent evolution of chromosomal sex-determining regions in the animal and fungal kingdoms. PLoS Biol 2, e384 (2004).
34
R Mehrabi, et al., Horizontal gene and chromosome transfer in plant pathogenic fungi affecting host range. FEMS Microbiol Rev 35, 542–554 (2011).
35
SE Screen, RJ St Leger, Cloning, expression, and substrate specificity of a fungal chymotrypsin: Evidence for lateral gene transfer from an actinomycete bacterium. J Biol Chem 275, 6689–6694 (2000).
36
Z Duan, Y Shang, Q Gao, P Zheng, C Wang, A phosphoketolase Mpk1 of bacterial origin is adaptively required for full virulence in the insect-pathogenic fungus Metarhizium anisopliae. Environ Microbiol 11, 2351–2360 (2009).
37
W Fang, RJ St Leger, RNA binding proteins mediate the ability of a fungus to adapt to the cold. Environ Microbiol 12, 810–820 (2010).
38
H Zhao, et al., Host-to-pathogen gene transfer facilitated infection of insects by a pathogenic fungus. PLoS Pathog 10, e1004009 (2014).
39
L Buchwaldt, H Green, Phytotoxicity of destruxin B and its possible role in the pathogenesis of Alternaria brassicae. Plant Pathol 41, 55–63 (1992).
40
C Kosiol, et al., Patterns of positive selection in six Mammalian genomes. PLoS Genet 4, e1000144 (2008).
41
M Long, E Betrán, K Thornton, W Wang, The origin of new genes: Glimpses from the young and old. Nat Rev Genet 4, 865–875 (2003).
42
C Wang, RJ St Leger, A collagenous protective coat enables Metarhizium anisopliae to evade insect immune responses. Proc Natl Acad Sci USA 103, 6647–6652 (2006).
43
RH Jiang, S Tripathy, F Govers, BM Tyler, RXLR effector reservoir in two Phytophthora species is dominated by a single rapidly evolving superfamily with more than 700 members. Proc Natl Acad Sci USA 105, 4874–4879 (2008).
44
T Wicker, et al., The wheat powdery mildew genome shows the unique evolution of an obligate biotroph. Nat Genet 45, 1092–1096 (2013).
45
AM Calvo, RA Wilson, JW Bok, NP Keller, Relationship between secondary metabolism and fungal development. Microbiol Mol Biol Rev 66, 447–459 (2002).
46
P Zheng, Y Xia, S Zhang, C Wang, Genetics of Cordyceps and related fungi. Appl Microbiol Biotechnol 97, 2797–2804 (2013).
47
S Sun, J Heitman, Is sex necessary? BMC Biol 9, 56–59 (2011).
48
JE Blair, MD Coffey, SY Park, DM Geiser, S Kang, A multi-locus phylogeny for Phytophthora utilizing markers derived from complete genome sequences. Fungal Genet Biol 45, 266–277 (2008).
49
R Hopkins, Reinforcement in plants. New Phytol 197, 1095–1103 (2013).
50
A Stamatakis, P Hoover, J Rougemont, A rapid bootstrap algorithm for the RAxML Web servers. Syst Biol 57, 758–771 (2008).
51
S Guindon, et al., New algorithms and methods to estimate maximum-likelihood phylogenies: Assessing the performance of PhyML 3.0. Syst Biol 59, 307–321 (2010).
Information & Authors
Information
Published in
Classifications
Copyright
Freely available online through the PNAS open access option.
Data Availability
Data deposition: The genome sequences reported in this paper have been deposited in the National Center for Biotechnology Information (NCBI) database [NCBI accession nos. AZHE00000000 (Metarhizium album), ADNI00000000 (Metarhizium acridum), AZNE00000000 (Metarhizium majus), AZNH00000000 (Metarhizium guizhouense), AZNG00000000 (Metarhizium brunneum), AZNF00000000 (Metarhizium anisopliae), and ADNJ02000000 (Metarhizium robertsii)].
Submission history
Published online: November 3, 2014
Published in issue: November 25, 2014
Keywords
Acknowledgments
We thank BGI-Shenzhen for genome-sequencing services. This work was supported by Strategic Priority Research Program of the Chinese Academy of Sciences Grant XDB11030100, National Nature Science Foundation of China Grant 31225023, and United States National Science Foundation Grant 1257685.
Notes
This article is a PNAS Direct Submission.
Authors
Competing Interests
The authors declare no conflict of interest.
Metrics & Citations
Metrics
Citation statements
Altmetrics
Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
Cited by
Loading...
View Options
View options
PDF format
Download this article as a PDF file
DOWNLOAD PDFGet Access
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Personal login Institutional LoginRecommend to a librarian
Recommend PNAS to a LibrarianPurchase options
Purchase this article to get full access to it.