Draft genome sequence of a monokaryotic model brown-rot fungus Postia (Rhodonia) placenta SB12

Author(s): Gaskell, Jill; Kersten, Phil; Larrondo, Luis F; Canessa, Paulo; Martinez, Diego; Hibbett, David; Schmoll, Monika; Kubicek, Christian P; Martinez, Angel T; Yadav, Jagjit; Master, Emma; Magnuson, Jon Karl; Yaver, Debbie; Berka, Randy; Lail, Kathleen; Chen, Cindy; LaButti, Kurt; Nolan, Matt; Lipzen, Anna; Aerts, Andrea; Riley, Robert; Barry, Kerrie; Henrissat, Bernard; Blanchette, Robert; Grigoriev, Igor V; Cullen, Dan

We report the genome of Postia (Rhodonia) placenta MAD-SB12, a homokaryotic wood decay fungus (Basidiomycota, Polyporales). Intensively studied as a representative brown rot decayer, the gene complement is consistent with the rapid depolymerization of cellulose but not lignin.

Experimental design, materials and methods
Common inhabitants of forest litter and decaying wood, brown-rot fungi play a key role in carbon cycling. These Basidiomycetes rapidly depolymerize cellulose while leaving the bulk of lignin as a modified residue. The preponderance of evidence supports oxidative mechanisms involving diffusible hydroxyl radicals, but much uncertainty remains. To examine the system more closely, a dikaryotic isolate of the brownrot fungus, Postia placenta (which is also classified in the genus Rhodonia [2]), was previously sequenced [3]. The genome has been used for phylogenomic comparisons and for analyses of transcriptomes and secretomes, but investigations are hampered by allelic variation [4][5][6][7][8][9][10][11][12][13].
Addressing this problem, single basidiospores were collected from the fruiting dikaryon strain Mad-698 by inverting agar plates containing malt extract medium. The basidiospores were eluted from the lids with sterile water and, after streaking onto agar, individual germinating basidiospores were transferred to new plates. The monokaryotic condition was confirmed by PCR amplification and direct sequencing of genes encoding a glycosyl transferase family 66, and representatives of glycoside hydrolase families 55 and 1 [14].
The genome of P. placenta MAD-SB12 was sequenced using a combination of platforms: 454 (Roche), Illumina, and Sanger. Firstly, Illumina reads obtained from 300 bp insert size library sequenced in 2 × 72 bp format were assembled using Velvet [15], followed by shredding the velvet assemblies into~1000 bp fragments. Then, these fragments were assembled with 454 Titanium standard and 2.8 kb insert size paired-end reads as well as Sanger fosmids using Newbler (2.5internal-10Apr08-1) (Roche). The 42.5 Mbp genome assembly consisted of 549 scaffolds and 1446 contigs (scaffold N50 and L50 were 8  and 2.1 Mbp, respectively). Secretion signals were predicted in 1047 sequences. Assembly and general annotation features are summarized in Table 1.

Data description
Consistent with the degradative potential of brown rot fungi, no ligninolytic peroxidases, cellulose binding modules, or members of glycoside hydrolase (GH) families 6 and 7 were detected in the P. placenta SB12 genome (Table 2). Among the brown rot fungi, potential cellulases included representatives of glycoside hydrolase (GH) families GH5, GH45 and GH12. However, like the GH7s in Laetiporus sulphureus, none of the brown rot catalytic domains are associated with a family 1 cellulose binding module (CBM1), and their activity on crystalline cellulose is therefore suspect. In the white rot fungi, these exocellobiohydrolases and endoglucanases are typically fused to family CBM1 domains ( Table 2). A total of 326 P. placenta SB12 genes encode carbohydrate active enzymes (CAZys), of which 144 are glycoside hydrolases [14].
To recognize single haplotypes within the dikaryon, BLASTN alignments of putative alleles plus 500 bp of upstream regions were used to delete 4996 allelic variants. This resulted in 12,227 total gene predictions [3], an estimate similar to the actual number of haplotypes shown here in the monokaryon (12,541). However, a substantial number of genes involved in lignocellulose degradation were not captured by the computational approach. For example, dikaryotic P. placenta MAD-698 was predicted to encode only 243 CAZys including 129 GHs [3]. Glycosyl transferases were particularly underestimated in the dikaryon, as were 15 GHs and several oxidoreductases ( Table 2). Among the latter, alcohol oxidase genes (AA3_3) are particularly important as evidence suggests their peroxide-generating activity may be directly related to the generation of small molecular weight oxidants via Fenton chemistry [16].