- Research article
- Open Access
Sequencing the genome of Marssonina brunnea reveals fungus-poplar co-evolution
BMC Genomics volume 13, Article number: 382 (2012)
The fungus Marssonina brunnea is a causal pathogen of Marssonina leaf spot that devastates poplar plantations by defoliating susceptible trees before normal fall leaf drop.
We sequence the genome of M. brunnea with a size of 52 Mb assembled into 89 scaffolds, representing the first sequenced Dermateaceae genome. By inoculating this fungus onto a poplar hybrid clone, we investigate how M. brunnea interacts and co-evolves with its host to colonize poplar leaves. While a handful of virulence genes in M. brunnea, mostly from the LysM family, are detected to up-regulate during infection, the poplar down-regulates its resistance genes, such as nucleotide binding site domains and leucine rich repeats, in response to infection. From 10,027 predicted proteins of M. brunnea in a comparison with those from poplar, we identify four poplar transferases that stimulate the host to resist M. brunnea. These transferas-encoding genes may have driven the co-evolution of M. brunnea and Populus during the process of infection and anti-infection.
Our results from the draft sequence of the M. brunnea genome provide evidence for genome-genome interactions that play an important role in poplar-pathogen co-evolution. This knowledge could help to design effective strategies for controlling Marssonina leaf spot in poplar.
Marssonina, belonging to the family Dermateaceae, is an important fungus that causes Marssonina leaf spot, one of the most important and widespread foliage diseases, on all species of Populus[1–3]. Poplars infected with Marssonina are symptomized by small, scattered, circular to oval dead areas in the leaves, resulting in premature defoliation and, ultimately, weakening and dieback of the tree. Because of the continuing shrinkage of natural forests, fast-growing hybrid poplars have been increasingly planted worldwide in a short rotation intensive culture, aimed to maximize carbon sequestration and woody biomass production[4–6]. However, the infection of Marssonina leaf spot severely reduces the growth and productivity of hybrid poplars, leading to significant economic and ecological losses.
Marssonina leaf spot is caused mainly by three species, M. brunnea, M. castagnei, and M. populi. M. brunnea is a filamentous fungus with a relatively narrow host range. Figure1 describes the life history of this fungus, its morphologies and cytological karyotyping. Both macroconidia and microconidia of M. brunnea are hyaline, but the former are unequally 2-celled and ovate or pear-shaped, whereas the latter is 1-celled and elliptical. In China, M. brunnea can be classified into two specialized forms, M. brunnea f. sp. multigermtubi and M. brunnea f. sp. monogermtubi. Both forms infect leaves of poplar from Aigeiros, Tacamahaca, and Leuce, three of six sections of the Populus genus. At present, no fungicides are available for controlling Marssonina leaf spot and, thus, the most promising control is to plant poplar varieties resistant or tolerant to this disease. However, since the genetic mechanisms by which the fungus interacts with poplar to form Marssonina leaf spot are still elusive, our success to breed and select resistant poplar clones through marker-assisted and biotechnological approaches is very limited.
As a first step toward the selection of Marssonina-resistant poplars, we initiated a project for sequencing the genome of M. brunnea. In the past several years, more than 40 fungal genomes have been sequenced (http://www.broadinstitute.org). However, most sequenced phytopathogenic fungi are those that colonize herbaceous plants, such as the rice blast pathogen Magnaporthe grisea, the corn smut pathogen Ustilago maydis, and the wheat head blight pathogen Fusarium graminearum. Only a few studies have reported on the genome sequences of fungi parasitizing woody plants. Apart from its economic value, M. brunnea can be used as a model system for studying pathogen-woody plant interactions because of its easy experimental manipulation, small genome size on three chromosomes and high genetic diversity[13, 14].
In this study, we use a combination of Roche 454, ABI SOLiD, and Illumina/Solexa GA-II sequencing to sequence the genome of M. brunnea, in order to study the function of pathogenicity genes in this fungus. By comparing the M. brunnea genome with the genomes of two related fungi, Botrytis cinerea and Sclerotinia sclerotiorum, which have each evolved a different lifecycle, we further study the evolution and speciation of pathogenicity. In particular, by integrating it with the sequenced genome of the host poplar, the M. brunnea genome is used to identify protein-protein interactions between the pathogen and host. These findings could be translated into the development of effective and efficient strategies for controlling the pathogenesis of the disease and selecting resistant poplar clones.
Results and Discussion
The genome of M. brunnea
Using a combination of Roche 454, ABI SOLiD, and Illumina/Solexa GA-II sequencing, the genome of M. brunnea was sequenced to approximately 34-fold coverage, yielding 2,990 contigs and 155 scaffolds from a specialized form M. brunnea f. sp. multigermtubi (Additional file1). The N50 scaffold length is 33,873 bp from the 4,532,414 Roche 454 reads with Newbler (v2.3). After gap filling, fewer contigs (2420) were assembled into 90 scaffolds with a larger N50 size, generating 52 Mb of assembled genome sequence (Table1). We identified 28 s rRNA, 18 s rRNA and Internal Transcribed Spacer (ITS) using RNAmmer (v1.2) (Additional file2).
Of the 192 gaps within the scaffolds that were filled using the Solexa contigs, three were coincident with the 27 gaps closed by primer walking, PCR, and sequencing. A preliminary finishing effort closed approximate 10% of the remaining genome gaps, some of which contained important regions, such as ITS and complete mitochondrial DNA. As an evaluation of the genome assembly scaffolds, 80.27% Solexa reads were mapped to the original 90 scaffolds as paired-end alignments using Bowtie (v0.12.7). Reads from Illumina/Solexa GA-II were de novo assembled into 53,924 contigs with a total of 51 Mb using Velvet (v1.0.02), of which 53,519 (99.25%) were aligned to the scaffolds.
Table2 compares genome-wide proteins among the three closely related fungi, B. cinerea, S. sclerotiorum and M. brunnea. Of 14,522 proteins in B. cinerea, 10,699 (73.67%) were aligned to 9,928 proteins (68.37%) in S. sclerotiorum. Of 10,040 proteins in M. brunnea, 7,508 and 7528 were homologous to 8,154 of 9,928 proteins (82.13%) in B. cinerea and 8,907 of 10,699 proteins (83.25%) in S. sclerotiorum, respectively Table3.
Relatively little is known about the phylogenetic history of fungi because of a lack of their fossil records. The concatenated amino acid sequences were used to construct a phylogenetic tree for 23 fungi[18, 19] (Figure2A), where B. cinerea and S. sclerotiorum are most closely related to M. brunnea, followed by M. grisea, F. graminearum, and N. crassa (Additional file3), as supported by taxonomic positions among these fungi (http://www.ncbi.nlm.nih.gov/Taxonomy/). However, pairwise comparisons indicated that M. brunnea only have 1,370 kb and 1,354 kb sequences similar to B. cinere a and S. sclerotiorum, respectively, suggesting that the former is evolutionarily distant to the latter two.
Due to a relatively rapid pace of change within the ITS1 and ITS2 sequences (Additional file2), these regions can be suitably used to assess phylogenetic relationships among closely related taxa, including filamentous fungi at the species or genus level[21, 22]. For example, ITS sequences were used for the phylogenetic analysis of genus Lens Mill and species Fusarium oxysporum. A neighbor-joining (NJ) phylogenetic tree was obtained using B. cinerea and S. sclerotiorum as outgroups (Figure2B; Additontial file 4). The ITS sequence of M. brunnea, M. rosae, and M. coronariae were clustered as a group and were further subdivided into three sister subgroups. However, the ITS sequence of M. brunnea was also very similar to those of B. cinerea and S. sclerotiorum. By global alignment analysis with Needle (EMBOSS (v6.2.0) (http://emboss.sourceforge.net/)), the ITS sequences of Marssonina brunnea f. sp. multigermtubi had a level of similarity of 59% with B. cinerea strain “FSU6300”, 68% with B. cinerea strain BC12, 70% with S. sclerotiorum strain “ms82”, and 72% with S. sclerotiorum strain “ms83”.
A total of 10,027 protein-coding genes were identified in the genome of M. brunnea. To measure the reliability of gene prediction, these predicted genes were compared by BLAST (BLASTP, E-value ≤ 1e-10) against CEGs (core eukaryotic genes) for orthologues[24, 25]. The result from the comparative analysis showed 99% of orthologues (or 245 of 248 CEGs) found as full or partial genes and also indirectly suggested a relatively high reliability of gene prediction and completeness of the assembly. In addition, ~93% of the gene models (9340 predicted genes) were supported with unique RNA-seq reads.
There were 7,257 predicted proteins that were assigned potential functions by BLAST based on protein databases, including NR, UniProt, and KEGG. A total of 2,736 protein families containing 6,774 predicted proteins (Additional files4 &5 ) were identified in M. brunnea using HMMER (v3.0) search against Pfam (v24). In addition, 288 (398 predicted proteins) and 61 (83 predicted proteins) protein families were identified by HMMER searching against Superfamily (v1.0) and TIGRFAM (v9.0), respectively.
Phi-base (pathogen-host interaction database) is a database that collects pathogenicity, virulence, and effector genes from fungi, oomycetes, and bacterial pathogens. A total of 793 predicted genes shared homology to 622 of 924 genes in Phi-base (v3.2), when we used BLASTP with an E-value of <1E-10. Table S6 shows the number of proteins with more than 10 homologs from M. brunnea. By comparative functional analysis, the pathogenic genes were classified into six categories: (1) genes involved in recognizing the host and signal pathways, (2) genes affecting the biosynthesis of fungal cell wall and infection structure, (3) genes involved in degradation of the plant cuticle and cell wall, (4) genes involved in the pathogen protection mechanism during infection process (Additional files6 &7), (5) genes whose roles are in fungal toxin biosynthesis (Additional file8), and (6) fungal genes whose roles are in nutrient acquisition (Additional files9 &10). We used a BLAST approach to infer the function of some of these genes, e.g., those related to sexual reproduction.
In the samples of M. brunnea, obtained from the eastern region of China, we did observe the asexual state but not the sexual state. Likewise, no sexual reproduction was found for Marssonina species in New Zealand Farm Forestry (http://www.nzffa.org.nz/ farm-forestry-model/the-essentials/forest-health-pests-and-diseases/diseases/Marssonina/Poplar-anthracnose). To identity whether this fungus undergoes a sexual cycle, we used BLAST searching for orthologues of all genes related to sexual reproduction and meiosis (Additional file11). Most of these sex-related genes were not found in M. brunnea. Some genes required for meiosis were present in M. brunnea, but they were involved in regulation as transcription factors or as supplementaries in syngenesis. For example, DMC1 related to meiosis were observed in M. brunnea, whereas those genes required for the formation of DMC1-containing nucleoprotein filaments were absent (Additional file9). All these supported that M. brunnea might have no capacity to perform sexual reproduction.
To compare gene-gene interactions between the pathogen and host, we used parallel massive sequencing of cDNA (RNA-seq) to estimate the relative expression levels of genes from M. brunnea and the host, poplar clone NL895 (P. euramericana CL“NL895”). Three cDNA libraries were constructed, including sample M6 of M. brunnea spores collected from potato dextrose agor, sample 895-M6 of clone NL895 leaves after 96 hours of infection by M. brunnea, and sample 895 of clone NL895 leaves with no infection.
RNA-seq reads from three different samples were aligned against the genome sequences of M. brunnea and Populus (http://genome.jgi-psf.org/poplar/poplar.home.html) using TopHat (v1.1.4). Additional file12 gives the mapping results of RNA-seq reads. Of the 10,040 predicted genes in M .brunnea and the 45,554 predicted genes in Populus, 9,340 (93%) and 31,794 (70%) were identified through RNA-seq, respectively, suggesting a relatively high coverage of the transcriptome.
The relative level of expression was calculated by using the amount of uniquely mapped reads for the annotated genes (Figure3). There are 2,559 M. brunnea predicted genes that display different levels of expression from Sample M6 to Sample 895-M6 (p < 0.001), of which 1,898 are up-regulated. Most up-regulated genes come from the LysM family that functions effectors to suppress plant basal immunity during the colonization of plants[28–30], with 30 predicted genes from a total of 33 being significantly up-regulated (p < 0.001) (Additional files13 &14). Other families that contain up-regulated genes (p < 0.001) are SNF2 family proteins, involved in such biological processes as transcription, DNA repair, chromatin-remodeling and hyphae development, AMP-binding enzymes, playing a key role in degradation and synthesis of amino acids and lipids, and GDSL-like lipase/acylhydrolases family proteins (Additional file13). The numbers of up-regulated genes for these three families are 11, 11, and 10 from a total of 25, 26, and 15, respectively. No genes from the four families above were down-regulated, indicating that these gene families may play a pivotal role in the early stage of infection (96 hour post-inoculation).
We also compared differences of gene expression between Sample M6-895 and Sample 895. Of the 13,053 Populus genes that display such differences, 4,811 and 8,242 were up- and down-regulated, respectively. This comparison allows us to identify resistance proteins by which plants resist pathogenic attack. A majority of plant resistant (R) genes contain nucleotide binding site domains (NBS) and leucine rich repeats (LRR), which are involved in the recognition of, and resistance to, pathogens[36, 37]. Nine putative Populus R-genes were highly up-regulated at 96 hpi (p < 0.001) (Additional file15), of which seven were the NBS-LRR type and two were the NBS and LRR (lrr1) types. Two Populus proteins, 815301 and 723016, similar to aminotransferases were significantly down-regulated (p < 0.001) for infected leaves, compared to uninfected ones. As aminotransferases regulating resistance to P. cubensis for melon, proteins At1 and At2, were significantly down-regulated in poplar (p < 0.001). These two proteins have a similar function to NSP-interacting kinases (NIKs) that mediate defense pathways in plants. In Arabidopsis, NIK1 serves as a defense receptor that elicits the plant’s defense response.
Chitin widely exists in fungal cell walls and can be recognized by many LysM receptors in plants. The innate immunity of Arabidopsis was elicited when the LysM receptor CERK1 bounds to chitin[41, 42]. There are 32 proteins containing the LysM domain in poplar, of which two (171810 and 233480) were significant down-regulated (p < 0.001) and shared homology with plant LysM receptor kinases, such as CERK1 in Arabidopsis. Perhaps it is possible that the putative LysM receptors in poplar were inhibited by LysM proteins in M. brunnea through competitive combination with fungal chitin.
All in all, most predict genes of M. brunnea and Populus could be detected in RNA-seq, some of which may play a crucial role in pathogen-host interactions, such as LysM motif-containing genes. The molecular mechanisms of the interactions between fungi and poplar have been studied through a complete description of the transcriptome of fungus-plant interactions.
The co-evolution of M. brunnea and Populus
Like Melampsora larici-populina causing leaf rust of poplar, M. brunnea was an obligate plant pathogen to parasite poplar. There has been some evidence that obligate plant pathogens have co-evolved with their hosts expressed at the protein level. Using the BLAST (BLASTP, E-value ≤ 1E-5) analysis, we found 8,093 predicted proteins of M. brunnea that are homologous to other eight fungal genomes, including B. cinerea (strain B05.10), S. sclerotiorum (strain 1980), M. grisea (strain 70–15), F. graminearum (strain PH-1), U. maydis (strain 521), Schizosaccharomyces pombe (strain 972 h-), Saccharomyces cerevisiae (strain RM11-1a), and M. larici-populina (strain 98AG31), as well as the Populus genome P. trichocarpa (poplar, v1.1), respectively. Of these proteins, 2,901 are homologous to each of the nine species, 265 are homologous to only one of the species, i.e., 96 to B. cinerea, 70 to S. sclerotiorum, 41 to M. grisea, 42 to F. graminearum, four to U. maydis, four to S. pombe, only one to S. cerevisiae, three to M. larici-populina, and four to P. trichocarpa.
The detection of more homologues to B. cinerea and S. sclerotiorum indirectly supported that M. brunnea is relatively more closely related to these two species than to the other species. One of the four homologues between M. brunnea and P. trichocarpa is M6_05232 that contains RNIG finger domain. The other three are glcG (M6_00711), 4-hydroxythreonine-4-phosphate dehydrogenase (M6_06189), and phosphomannomutase (M6_04436). Phosphomannomutase (PMM,EC188.8.131.52) pervading eukaryotes, such as SEC53 in S. cerevisiae, PMM1 in Candida albicans, and At2g45790 in Arabidopsis, is a kind of phosphotransferases that participates in mannose metabolism. The genes encoding phosphomannomutase in M. brunnea has nonsignificant similarity to those in other species, which may be due to the high specificity of these genes in sequence and function generated in the co-evolution of M. brunnea and Populus as well their convergent evolution with a certain host genes. In addition, these genes decreased their expression remarkably 96 hours after the leaves of poplar were inoculated with M. brunnea (p < 0.001), but their Populus homologues did not produce any significant change in the level of expression. The change of expression of the PMM-encoding genes may arise from the alteration of how M. brunnea acquires energy after it invades poplar.
M. brunnea has three proteins that only have a significant sequence similarity to those in M. larici-populina, which are a secretory protein, a dynein heavy chain-like protein, and a glycosyltransferase 8 domain-containing (Pfam: F01501) protein. Glycosyltransferase 8 domain-containing gene was significantly up-regulated in the leaves of poplar infected after 96 hours (p < 0.001). In poplar fungal pathogens, such as M. brunnea and M. larici-populina, the glycosyltransferase 8 domain-containing protein may have played an important role in the assimilation of nutrients and the transportation of energy and carbohydrates from the poplar host. The four Populus-homologous genes, M6_05232, M6_00711, M6_06189, and M6_04436, derived from M. brunnea and M. larici-populina functions similarly in the infection of poplar leaves, suggesting that the environment where the two types of fungi live is an impetus for their genes to evolve into the same direction.
It is interesting to find that a putative galactokinase (EC 184.108.40.206) protein (M6_02750) of M. brunnea had a homologue only in M. larici-populina (jgi|Mellp1|115317) and Populus genomes (jgi|Poptr1_1|811685). Galactokinase is a phosphotransferase which has well been studied in many species, such as yeasts and plants. It is likely that this gene experiences co-evolution between the host and pathogen as well as between different obligate fungal pathogens that infect the same host, thus leading these three of species to produce a relatively high similarity in gene sequence. Also, the genes encoding galactokinase in M. brunnea and M. larici-populina produce convergent evolution with the homologues of their poplar host.
The genome of Marssonina brunnea, a woody plant pathogenic fungus that causes foliar disease in poplar, was sequenced and assembled with next-generation sequencing techniques, followed by a preliminary finishing effort that closed approximately 10% of the remaining gaps. After gap closure, the numbers of contigs and scaffolds decreased largely, accompanying an increasing size of N50 and the completion of a circular mtDNA. The genome sequence of M. brunnea reveals many important biological characteristics of the fungus, not only useful for studying the structure, organization and evolution of microbe genomes, but also shedding light on the molecular mechanisms of how pathogens and hosts interact and co-evolve.
A total of 28 LysM-containing proteins in M. brunnea were predicted as secreted proteins, which were significantly up-regulated during the process of infection. Interestingly, two LysM receptor-like kinases in Populus were significantly down-regulated after poplar is infected. These discoveries suggest that LysM proteins may play an important role in inhibiting the immunity system of poplar through competitive binding to chitin with the plant LysM receptor proteins.
We identified the genes that encode three types of transferases, i.e., phosphoglucomutase/ phosphomannomutase, glycosyltransferase 8 domain-containing protein, and galactokinase. These genes have driven M. brunnea and its Populus host to co-evolve, providing new insights into the genetic machineries of how obligate pathogenic fungi infect obligate hosts and how matter and energy flow and exchange between the pathogen and host. The completion of sequencing the M. brunnea genome opens a new resource for understanding the fundamental questions regarding pathogen-plant interactions, developing novel disease-control strategies and producing new disease-resistant varieties of tree crops.
Marssonina brunnea f. sp. multigermtubi was obtained from the eastern region of China, including Shandong, Jiangsu, Henan, Shanxi, Jilin Provinces, and Beijing. It has been studied in our laboratory for approximately 30 years. M. brunnea f. sp. multigermtubi (M6), which infects Populus species from Sections Aigeiros and Tacamahaca, was used as a sequenced reference strain.
The sequencing of the genome of M. brunnea f.sp multigermtubi was performed at CHGC (Chinese National Human Genome Center at Shanghai). This yielded 4.5×106 Roche 454 reads (4RUN) with an average length of 383 nt and a total size of 1.7 Gb. 4.7×107 pairs of mate-paired reads (35 nt) with insert sizes of 5 kb were obtained from the SOLiD System. 2.1×107 pairs of paired-end reads (120 nt) with insert sizes of 200 bp were obtained from the Illumina/Solexa GA-II. All PCR products for gap closure were sequenced using ABI 3730 xl DNA Analyzers.
Three RNA samples, i.e., M6 (M. brunnea f. sp. multigermtubi), 895 (leaves of poplar clone NL895 (P. euramericana CL“NL895”)) and 895-M6 (leaves of clone NL895 infected by M. brunnea f. sp. multigermtubi), were sequenced by the Illumina/Solexa GA-II. A dataset with 19.8 Gb or 73,228,774 reads with 113 nt reads length was produced.
Assembly and gap closure
First, 4.5×106 Roche 454 reads were assembled into 2,990 contigs by Newbler (v2.3). Then, 155 scaffolds were constructed using mate-paired information from SOLiD mate-paired reads and based on the algorithm of ConPath. Using velvet (v1.0.02), 2.1×107 pairs of paired-end reads (120 nt) from Illumina/Solexa GA-II were de novo assembled into 53,924 Solexa contigs, with a total of 51 Mb. Based the information of order and direction of contigs within scaffolds, 192 gaps within scaffolds were closed using the 53,924 Solexa contigs. A total of 50 pairs of primers were designed to fill gaps between both adjacent contigs within scaffolds. A total of 27 gap sequences (with an average length of ~130 bp) were successfully filled, of which three gaps were coincident with that of 192 gaps using the Solexa contigs. After gap closure, the amount of initial contigs was reduced to just 2,420. Finally, a total of 90 scaffolds were reconstructed, with a total length of 52 Mb.
Next generation sequencing (NGS) short reads were mapped against the genome using Bowtie (v0.12.7). Solexa contigs were located to the genome sequences of M. brunnea using MEGABLAST (http://www.ncbi.nlm.nih.gov/blast/megablast.shtml) with identity cut-off of 90%.
The gene prediction of M. brunnea was performed independently with a combination of three gene prediction program, including GeneMark (v2.3), Augustus (v2.3.1), and Exonhunter. The gene models were selected and manually curated by Argo Genome Browser (v 1.0.31,http://www.broadinstitute.org/annotation/argo/). The gene models were aligned using BLASTP against the protein sequence of B. cinerea and S. sclerotiorum (http://www.broadinstitute.org/). The predicted proteins were identified using BLASTP against NR, KEGG, and UniProt.
The classification of protein families was done using HMMER (v3.0) against Pfam (v24), SupperFamily (v1.0), and TIGRFAM (v9.0). tRNA genes were detected using tRNAScan-SE (v1.23). Repetitive elements were screened using RepeatModeler (v 1.0.4) and RepeatMasker (v 3.2.9) (Additional files16 &17). The analyses of putative transposon/retrotransposons were performed using Repbase (v 16.01). Secretory proteins were identified by a combination of SignalP (v 3.0) and TMHMM (v 2.0) (http://www.cbs.dtu.dk/services/). The predicted secreted proteins in M. brunnea were aligned to the secretory proteins of six fungi (U. maydis, M. grisea, B. cinerea, S. sclerotiorum, S. cerevisiae, and S. pombe) from the Fungal Secretome Knowledge base (http://proteomics.ysu.edu/secretomes/fungi.php), using BLASTP with a cutoff E-value <1e-5 (Additional files18 &19). Aligning genome-scale proteins against PHI-base (v3.2,http://www.phi-base.org/) was performed by BLAST with an E-value of less than 1E-10 and to find putative gene involved in pathogenicity or virulence.
Orthology and phyogenetic analysis
There are a total of 621 orthologous proteins which were obtained from M. brunnea, B. cinerea (B05.10), and 21 species which included 19 fungi, Caenorhabditis brenneri and Marssonina coronariae (Inparanoid (v7.0),http://inparanoid.sbc.su.se/cgi-bin/index.cgi). Multiple sequence alignments were done with ClustalW (v 2.1). A neighbor-joining (NJ) phylogenetic tree was constructed, based on concatenated protein sequences by MEGA (v4.0.2) with a bootstrap value of 1000.
To find potential synteny blocks between the M. brunnea genome and the genomes of B. cinerea and S. sclerotiorum, we used the BLAST analysis (BLASTN, with the alignment length of longer than 60 bp and over 75% identity) of the M. brunnea genome against the genomes of B. cinerea and S. sclerotiorum.
ITS (Internal transcribed spacer) sequences for B. cinerea and S. sclerotiorum were downloaded from the NCBI (Additional file20). ITS sequences from M. brunnea were identified by ITS1 (tccgtaggtgaaccttgcgg) and ITS5 (ggaagtaaaagtcgtaacaagg). A Neighbor-joining (NJ) phylogenetic tree was constructed based on ITS sequences by MEGA (v4.0.2) with a bootstrap value of 1000.
Digital transcriptome analysis
Poplar clone NL895, highly resistant to M. brunnea f. sp. multigermtubi, is one of the most important commercial planting clones in China. Cuttings of clone NL895 were cultured in the greenhouse at 22°C with a 12-hour photoperiod, until the cuttings were 0.5 ~ 1 m high and had 10 to 20 fully expanded leaves. Five or six fully expanded leaves were taken and placed on 2% water agar in sterile culture dishes with the abaxial surface uppermost. Conidia of M. brunnea f. sp. multigermtubi were suspended in sterile water. The spore suspensions were adjusted to 80,000 spores/ml and sprayed on the abaxial surface of the poplar leaves. Treated leaves were incubated in an illuminated incubator under 100% relative humidity (RH) at 22°C with a 12-hour photoperiod. Treated leaves were harvested at 4 days post-inoculation (dpi), then frozen quickly using liquid nitrogen, and stored at −70°C. RNA of the M. brunnea f. sp. multigermtubi conidia, uninfected leaves, and infected leaves were all extracted using Trizol reagent according to the manufacturer’s instructions (Invitrogen, Carlsbad, CA). Genomic DNA was removed by DNase I (TaKaRa, Japan).
RNA-seq reads were generated on an Illumina/Solexa GA-II. RNA-seq reads were mapped onto the genome of M. brunnea and Populus trichocarpa (v1.1,http://genome.jgi-psf.org/poplar/poplar.home.html), using a splice junction mapper named Tophat (v1.1.4,http://tophat.cbcb.umd.edu/). Differentially expressed genes were identified by determining the number of raw reads that uniquely mapped to genes, as a basis for determining significance by Fisher’s exact test and chi-square test.
The whole genome shotgun project has been submitted to GenBank/EMBL/DDBJ for Marssonina brunnea .f.sp multigermtubi (GeneBank accession No: AFXC00000000). ITS (GeneBank accession No: JN172909) and mitochondrial sequences (GenBank accession No: JN204424) of M. brunnea .f.sp multigermtubi are available.
Beare JA, Archer SA, Bell JNB: Marssonina leafspot disease of poplar under elevated ozone: pre-fumigated host and in vitro studies. Environ Poll. 1999, 105: 409-417.
Han ZM, Yin TM, Li CD, Huang MR, Wu RL: Host effect on genetic variation of Marssonina brunnea pathogenic to poplars. Theor Appl Genet. 2000, 100: 614-620.
Erickson JE, Stanosz GR, Kruger EL: Photosynthetic consequences of Marssonina leaf spot differ between two poplar hybrids. New Phytol. 2004, 161: 577-583.
Wu RL, Wang MX, Huang MR: Quantitative genetics of yield breeding for Populus short rotation culture. I. Dynamics of genetic control and selection model of yield traits. Can J For Res. 1992, 22: 175-
Pellis A, Laureysens I, Ceulemans R: Growth and production of a short rotation coppice culture of poplar I. Clonal differences in leaf characteristics in relation to biomass production. Biomass Bioenergy. 2004, 27: 9-19.
FAO (Food and Agriculture Organization of the United Nations): State of the World's Forests. 2011, Rome, 164-
Han ZM, Li CD, Huang MR: Comparative studies of ioslates of Marssonina brunnea in china. Sci Sil Sin. 1998, 34: 59-65.
Duplessis S, Major I, Martin F, Séguin A: Poplar and pathogen interactions: Insights from Populus genome-wide analyses of resistance and defense gene families and gene expression profiling. Crit Rev Plant Sci. 2009, 28: 309-334.
Dean RA, Talbot NJ, Ebbole DJ, Farman ML, Mitchell TK, Orbach MC, Thon M, Kulkarni R, Xu JR, Pan H, et al: The genome sequence of the rice blast fungus Magnaporthe grisea. Nature. 2005, 434: 980-986.
Kamper J, Kahmann R, Bolker M, Ma LJ, Brefort T, Saville BJ, Banuett F, Kronstad JW, Gold SE, Müller O: Insights from the genome of the biotrophic fungal plant pathogen Ustilago maydis. Nature. 2006, 444: 97-101.
Cuomo CA, Guldener U, Xu JR, Trail F, Turgeon BG, et al: The Fusarium graminearum genome reveals a link between localized polymorphism and pathogen specialization. Science. 2007, 317: 1400-1402.
Duplessis S, Cuomo CA, Lin YC, Aerts A, Tisserant E, et al: Obligate biotrophy features unraveled by the genomic analysis of rust fungi. Proc Natl Acad Sci U S A. 2001, 108: 9166-9171.
Zhang Y, Zhang X, Chen Y, Wang Q, Wang M, et al: Function and chromosomal localization of differentially expressed genes induced by Marssonina brunnea f. sp. multigermtubi in Populus deltoides. J Genet Genomics. 2007, 34: 641-648.
Cheng Q, Cao Y, Jiang C, Xu L, Wang M, et al: Identifying secreted proteins of Marssonina brunnea by degenerate PCR. Proteomics. 2010, 10: 2406-2417.
Amselem J, Cuomo CA, van Kan JA, Viaud M, Benito EP, et al: Genomic analysis of the necrotrophic fungal pathogens sclerotinia sclerotiorum and botrytis cinerea. PLoS Genet. 2011, 7: e1002230-
Tuskan GA, Difazio S, Jansson S, Bohlmann J, Grigoriev I, et al: The genome of black cottonwood, Populus trichocarpa (Torr. & Gray). Science. 2006, 313: 1596-1604.
Redecker D, Kodner R, Graham LE: Glomalean fungi from the Ordovician. Science. 2000, 289: 1920-1921.
Langmead B, Trapnell C, Pop M, Salzberg SL: Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol. 2009, 10: R25-
James TY, Kauff F, Schoch CL, Matheny PB, Hofstetter V, et al: Reconstructing the early evolution of Fungi using a six-gene phylogeny. Nature. 2006, 443: 818-822.
Sonnante G, Galasso I, Pignone D: ITS sequence analysis and phylogenetic inference in the genus Lens mill. Ann Bot. 2003, 91: 49-54.
Peay KG, Kennedy PG, Bruns TD: Fungal community ecology: a hybrid beast with a molecular master. Bioscience. 2008, 58: 799-810.
Martin KJ, Rygiewicz PT: Fungal-specific PCR primers developed for analysis of the ITS region of environmental DNA extracts. BMC Microbiol. 2005, 5: 28-
Gurjar G, Barve M, Giri A, Gupta V: Identification of Indian pathogenic races of Fusarium oxysporum f. sp. ciceris with gene specific, ITS and random markers. Mycologia. 2009, 101: 484-495.
Parra G, Bradnam K, Korf I: CEGMA: a pipeline to accurately annotate core genes in eukaryotic genomes. Bioinformatics. 2007, 23: 1061-1067.
Parra G, Bradnam K, Ning Z, Keane T, Korf I: Assessing the gene space in draft genomes. Nucleic Acids Res. 2009, 37: 289-297.
Coleman JJ, Mylonakis E: Efflux in fungi: la piece de resistance. PLoS Pathog. 2009, 5: e1000486-
Tsubouchi H, Roeder GS: The budding yeast mei5 and sae3 proteins act together with dmc1 during meiotic recombination. Genetics. 2004, 168: 1219-1230.
de Jonge R, Thomma BP: Fungal LysM effectors: extinguishers of host immunity?. Trend Microbiol. 2009, 17: 151-157.33.
de Jonge R, van Esse HP, Kombrink A, Shinya T, Desaki Y, et al: Conserved fungal LysM effector Ecp6 prevents chitin-triggered immunity in plants. Science. 2010, 329: 953-955.
de Jonge R, Bolton MD, Thomma BP: How filamentous pathogens co-opt plants: the ins and outs of fungal effectors. Curr Opin Plant Biol. 2011, 14: 400-406.
Wang W: The SWI/SNF family of ATP-dependent chromatin remodelers: similar mechanisms for diverse functions. Curr Top Microbiol Immunol. 2003, 274: 143-169.
Lu Y, Su C, Mao X, Raniga PP, Liu H, et al: Efg1-mediated recruitment of NuA4 to promoters is required for hypha-specific Swi/Snf binding and activation in Candida albicans. Mol Biol Cell. 2008, 19: 4260-4272.
Mannhaupt G, Montrone C, Haase D, Mewes HW, Aign V, et al: What's in the genome of a filamentous fungus? Analysis of the Neurospora genome sequence. Nucleic Acids Res. 2003, 31: 1944-1954.
Kwon SJ, Jin HC, Lee S, Nam MH, Chung JH, et al: GDSL lipase-like 1 regulates systemic resistance associated with ethylene signaling in Arabidopsis. Plant J. 2009, 58: 235-245.
Sanseverino W, Roma G, De Simone M, Faino L, Melito S, et al: PRGdb: a bioinformatics platform for plant resistance gene analysis. Nucleic Acids Res. 2010, 38: D814-D821.
Pan Q, Wendel J, Fluhr R: Divergent evolution of plant NBS-LRR resistance gene homologues in dicot and cereal genomes. J Mol Evol. 2000, 50: 203-213.
Eckardt NA, Innes R: Resistance rodeo: rounding up the full complement of Arabidopsis NBS-LRR genes. Plant Cell. 2003, 15: 806-807.
Taler D, Galperin M, Benjamin I, Cohen Y, Kenigsbuch D: Plant eR genes that encode photorespiratory enzymes confer resistance against disease. Plant Cell. 2004, 16: 172-184.
Carvalho CM, Santos AA, Pires SR, Rocha CS, Saraiva DI, et al: Regulated nuclear trafficking of rpL10A mediated by NIK1 represents a defense strategy of plant cells against virus. PLoS Pathog. 2008, 4: e1000247-
Santos AA, Carvalho CM, Florentino LH, Ramos HJ, Fontes EP: Conserved threonine residues within the A-loop of the receptor NIK differentially regulate the kinase function required for antiviral signaling. PLoS One. 2009, 4: e5781-
Miya A, Albert P, Shinya T, Desaki Y, Ichimura K, et al: CERK1, a LysM receptor kinase, is essential for chitin elicitor signaling in Arabidopsis. Proc Natl Acad Sci U S A. 2007, 104: 19613-19618.
Gimenez-Ibanez S, Hann DR, Ntoukakis V, Petutschnig E, Lipka V, et al: AvrPtoB targets the LysM receptor kinase CERK1 to promote bacterial virulence on plants. Curr Biol. 2009, 19: 423-429.
Kepes F, Schekman R: The yeast SEC53 gene encodes phosphomannomutase. J Biol Chem. 1988, 263: 9155-9161.
Smith DJ, Cooper M, DeTiani M, Losberger C, Payton MA: The Candida albicans PMM1 gene encoding phosphomannomutase complements a Saccharomyces cerevisiae sec 53–6 mutation. Curr Genet. 1992, 22: 501-503.
Qian W, Yu C, Qin H, Liu X, Zhang A, et al: Molecular and functional analysis of phosphomannomutase (PMM) from higher plants and genetic evidence for the involvement of PMM in ascorbic acid biosynthesis in Arabidopsis and Nicotiana benthamiana. Plant J. 2007, 49: 399-413.
Sellick CA, Reece RJ: Contribution of amino acid side chains to sugar binding specificity in a galactokinase, Gal1p, and a transcriptional inducer, Gal3p. J Biol Chem. 2006, 281: 17150-17155.
Kaplan CP, Tugal HB, Baker A: Isolation of a cDNA encoding an Arabidopsis galactokinase by functional expression in yeast. Plant Mol Biol. 1997, 34: 497-506.
Kim PG, Cho HG, Park K: A scaffold analysis tool using mate-pair information in genome sequencing. J Biomed Biotechnol. 2008, 2008: 675-741.
Zerbino DR, Birney E: Velvet: algorithms for de novo short read assembly using de Bruijn graphs. Genome Res. 2008, 18: 821-829.
Langmead B, Trapnell C, Pop M, Salzberg SL: Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol. 2009, 10: R25-
Pruitt KD, Tatusova T, Maglott DR: NCBI reference sequences (RefSeq): a curated non-redundant sequence database of genomes, transcripts and proteins. Nucleic Acids Res. 2007, 35: D61-D65.
Kanehisa M, Goto S, Furumichi M, Tanabe M, Hirakawa M: KEGG for representation and analysis of molecular networks involving diseases and drugs. Nucleic Acids Res. 2010, 38: D355-D360.
Magrane M, Consortium U: UniProt Knowledgebase: a hub of integrated protein data. 2011, Oxford: Database, bar009-
Lum G, Min XJ: FunSecKB: the Fungal Secretome KnowledgeBase. 2011, Oxford: Database, bar001-
Winnenburg R, Urban M, Beacham A, Baldwin TK, Holland S, et al: PHI-base update: additions to the pathogen host interaction database. Nucleic Acids Res. 2008, 36: D572-D576.
We acknowledge the genomic data provided by JGI (Joint Genome Institute,http://www.jgi.doe.gov/ ), including the data of Populus trichocarp (v1.1,http://genome.jgi-psf.org/poplar/poplar.home.html ) and Melampsora laricis-populina (the stain “98AG31”, v1.0,http://genome.jgi.doe.gov/Mellp1/Mellp1.home.html). We also acknowledge the genomic data and analysis tools provided by the Broad Institute of Harvard and MIT (http://www.broadinstitute.org) for Bortytis cinerea (the strain “B05.10”,http://www.broadinstitute.org/annotation/genome/botrytis_cinerea), Sclerotinia sclerotiorum (the strain “1980”,http://www.broadinstitute.org/annotation/genome/sclerotinia_sclerotiorum/MultiHome.html ), Magnaporthe grisea (the strain “70-15”,http://www.broadinstitute.org/annotation/genome/magnaporthe_grisea/GenomesIndex.html), Fusarium graminearum (the strain “PH-1”,http://www.broadinstitute.org/annotation/genome/fusarium_group/MultiHome.html)), Schizosaccharomyces pombe (the stain “972 h-”,http://www.broadinstitute.org/annotation/genome/schizosaccharomyces_group/Downloads.html), and Saccharomyces cerevisiae (the stain “RM11-1a”,http://www.broadinstitute.org/annotation/genome/saccharomyces_cerevisiae. ), and Argo Genome Browser (v1.0.31,http://www.broadinstitute.org/annotation/argo/). We thank Lei Zhang at the Institute of Plant Physiology and Ecology, Chinese Academy of Science, China, for implementing the scaffolding algorithm, and the Shanghai Supercomputer Center of China for computing technical support and services. This work was supported by grants from National Basic Research Program of China (2009CB19100), The Doctorate Fellowship Foundation of Nanjing Forestry University, The Science and Technology Commission of Shanghai Municipality (10QA1400600), The Graduate Research Innovation Project of Jiangsu Province, The Priority Academic Program Development of Jiangsu Higher Education Institutions, The Changjiang Scholars Award and “One-thousand Person Plan” Award.
The authors declare that they have no competing interest.
SZ, YC, CJ, BT, ZW, SF performed data analysis, SZ, YC, CJ, BT, LZ, XS, BB, TV, MX carried out the experiments at different stages. MW, SZ, MH, RW, YZ conceived and designed the experiments. SZ, YC, CJ, BT, RW wrote the manuscript. All authors read and approved the final manuscript.