Lignin-degrading enzyme activities and degradation products. Table 1 shows that Lac and MnP activities, but not LiP activity, were found in a crude extract obtained from fungal mycelia grown in Kirk medium with the addition of veratryl alcohol and oxygen flashing. The activities of Lac and MnP were 0.102 and 0.992 fkat/mg, respectively. Interestingly, according to the results of GC-MS analysis of the catabolites derived from degradation of the lignin model dimer (1-(4-ethoxy-3-methoxyphenyl)-2-(2,6-dimethoxyphenoxy)-1,3-propanediol) by I. obliquus (Supplementary Fig. 1), this fungus was able to cleave the O-C4 linkage of the lignin model dimer to form derivative I (1-(4-ethoxy-3-methoyphenyl)glycerol) (Fig. 2). Derivative I was then oxidized at the Cα position to form derivative II (1-(4-ethocy-3-methoyphenyl)-2,3-dihydroxypropane-1-one) (Fig. 2). This indicates that I. obliquus has the ability to degrade non-phenolic units of lignin, even though LiP activity is absent.
Table 1
Specific activity of lignin-degrading enzymes in I. obliquus.
Type of enzyme
|
Specific activity (fkat/mg)
|
Lac
|
0.102
|
MnP
|
0.992
|
LiP
|
0.000
|
Note: Lac, laccase; MnP, manganese peroxidase; LiP, lignin peroxidase. |
Genome features. The statistics of the assembled, predicted, and annotated genome of I. obliquus are summarized in Table 2. The genome of I. obliquus was successfully sequenced using a next-generation sequencing strategy on the Illumina MiSeq platform and assembled into 15,755 contigs with N50 length of 9,032 bp. The length of contigs ranged from 200 bp to 199,051 bp with an average length of 2,694.8 bp. Overall, the I. obliquus genome generated 42.5 Mbp nucleotides with 47.6% GC content. This genome size is within the typical range size of genomes in basidiomycetes19–23, 30, and somewhat similar with that of L. edodes (41.8 Mbp)22. L. edodes is a white rot fungus (order Agaricales in the phylum Basidiomycota) and is widely known as an edible mushroom due to its high nutrition and medicinal effect22. Furthermore, genome assembly of I. obliquus contained two genes for rRNAs and 136 genes for tRNAs (Table 2). These 136 tRNAs were found to correspond to the full set of 20 amino acids. A total of 21,203 protein coding genes were detected from this genome assembly. The number of protein coding genes in the I. obliquus genome is within the range for fungi, around 11,000-20,00030.
Table 2
Assembly, prediction, and annotation statistics for the I. obliquus genome.
Characteristic
|
Statistic
|
Number of genome contigs
|
15,755
|
N50 length (bp)
|
9,032
|
Shortest contig length (bp)
|
200
|
Longest contig length (bp)
|
199,051
|
Average contig length (bp)
|
2,694.8
|
Total number of nucleotides in genomic contig (bp)
|
42,456,479
|
GC content (%)
|
47.6
|
rRNA genes
|
2
|
tRNA genes
|
136
|
Number of protein coding genes
|
21,203
|
Number of genes annotated by nt database
|
8,280
|
Number of genes annotated by Swiss-prot database
|
4,188
|
Number of genes annotated by Pfam
|
16,190
|
Number of genes annotated by GO
|
11,450
|
Number of genes annotated by KEGG
|
5,277
|
Number of genes annotated by PHI-base
|
1,621
|
Note: N50 length, the minimum contig length needed to cover 50% of the genome; bp, base pair; GC, guanine-cytosine; tRNA, transfer RNA; rRNA, ribosomal RNA. |
The predicted protein coding genes were further blasted in the non-redundant nucleotide (nt) database, Swiss-prot database, Pfam database, and Kyoto Encyclopedia of Genes and Genomes (KEGG) database. Among the 21,203 predicted protein coding genes, 8,280 and 4,188 genes had significant similarity to those documented in the nt and Swiss-prot databases, respectively (Table 2). The result of the Pfam search clarified that 16,190 genes have structural domains (Table 2 and Supplementary Fig. 2). This result revealed that the I. obliquus genome is enriched in three conserved domains, including protein kinase domain (PF00069.24; 464 genes), protein tyrosine kinase (PF07714.16; 390 genes), and the AAA ATPase domain (PF13191.5; 388 genes). Moreover, the predicted protein coding genes were also assigned to Gene ontology (GO) terms to gain functional information. A total of 11,450 genes were included in this annotation (Table 2). Among them, 7,936, 2,032, and 1,482 genes were mapped to molecular function (MF), biological process (BP), and cellular component (CC), respectively (Supplementary Fig. 3).
A total of 5,277 genes were documented in the KEGG database and were significantly included in the biosynthesis of secondary metabolites (map01110; 257 genes) (Supplementary Fig. 4). The result of KEGG pathway analysis indicated that several pathways in the biosynthesis of secondary metabolites are known to be related to the pathways for medicinal compound biosynthesis. This information confirms previous studies, which showed that I. obliquus is used as a traditional medicine in West Siberia to treat several diseases, such as heart, liver, and stomach diseases, tuberculosis, and diabetes31. In addition, the pathway of lanosterol biosynthesis was also found in the present study (Supplementary Table 1), which is associated with six genes of terpenoid backbone biosynthesis (map00900; Supplementary Fig. 5A) and three genes of steroid biosynthesis (map00100; Supplementary Fig. 5B). The pathways of terpenoid backbone biosynthesis seem to be distributed only in the mevalonate (MVA) pathway. A similar result was also found in other basidiomycetes such as G. lucidum20. The putative pathway of lanosterol biosynthesis is shown in Supplementary Fig. 5C. Lanosterol is a class of chemical compounds that have beneficial properties for human health, such as an antitumor effect32. On the other hand, lanosterol is also considered as an important intermediate in the synthesis of inotodiol and trametenolic acid through hydroxylation and oxidation reactions32. These chemical compounds are present in I. obliquus and are known to have anti-inflammatory, anticancer, and antitumor effects3,32.
Genes related to wood degradation. From the functional annotation results, a total of 134 genes were detected in the I. obliquus genome that are potentially involved in the degradation of wood chemical components (Table 3). These genes were typical of white-rot fungi, with the highest number of genes for lignin degradation observed. These genes consist of 36 candidate cellulase genes, 35 candidate hemicellulase genes, 16 candidate pectinase genes, 37 candidate lignin modifying enzyme genes, and 10 candidate lignin degrading auxiliary enzyme genes, respectively. Annotation of these genes in nt, Swiss-prot, Pfam, GO, and KEGG databases is shown in Supplementary Table 2. All findings indicate that I. obliquus has enzyme encoding genes that can potentially target all chemical components of wood cell walls, including cellulose, hemicellulose, pectin, and lignin, reflecting its ability as a white-rot fungus. Possession of the huge variety of enzyme encoding genes in this fungus may be related to its parasitic nature, where its survival depends on degraded wood chemical components in the host as the primary carbon source of nutrients for growth during colonization, as observed for the pathogenic white-rot fungus H. irregulare19,33.
Table 3
Candidate genes involved in degradation of wood chemical components by I. obliquus.
Class
|
Putative enzyme
|
Number of genes
|
Cellulase
|
Endoglucanase
|
9
|
Exoglucanase
|
9
|
β-Glucanase
|
15
|
Cellobiose dehydrogenase
|
3
|
Total
|
36
|
Hemicellulase
|
β-Xylosidase
|
6
|
Endo-1,4-β-xylanase
|
7
|
4-O-Methyl-glucuronoyl methylesterase
|
1
|
Acetylxylan esterase
|
2
|
α-Xylosidase
|
2
|
α-Fucosidase
|
2
|
β-Mannosidase
|
6
|
Endo-β-mannanase
|
1
|
α-Galactosidase
|
3
|
α-L-Arabinofuranosidase
|
3
|
Arabinogalactan endo-β-1,4-galactanase
|
1
|
Feruloyl esterase
|
1
|
Total
|
35
|
Pectinase
|
Pectinesterase
|
1
|
Pectate lyase
|
4
|
Polygalacturonase
|
1
|
Endopolygalacturonase
|
2
|
Exoplygalacturonase
|
4
|
α-L-Rhamnosidase
|
2
|
Rhamnogalacturonan acetylesterase
|
1
|
Arabinan endo-1,5-α-L-arabinosidase
|
1
|
Total
|
16
|
Lignin-modifying enzyme
|
Laccase
|
14
|
Manganese peroxidase
|
14
|
Versatile peroxidase
|
7
|
Dye-decolorizing peroxidase
|
2
|
|
Total
|
37
|
Lignin-degrading auxiliary enzyme
|
Glucose oxidase
|
3
|
Alcohol oxidase
|
2
|
Aldehyde oxidase
|
5
|
|
Total
|
10
|
Total
|
134
|
As found in the above results, I. obliquus has the highest number of genes for lignin degradation. A total of 47 genes were related to lignin degradation and were categorized into two main classes (Table 3). The first contains 39 candidate lignin modifying enzyme genes consisting of 14 candidate lac genes, 14 candidate MnP genes, seven candidate VP genes, and two candidate dye-decolorizing peroxidase (DyP) genes. The second contains 10 candidate lignin degrading auxiliary enzyme genes consisting of three candidate glucose oxidase genes, two candidate alcohol oxidase genes, and five candidate aldehyde oxidase genes. These results show that Lac is an enzyme with the highest number of genes present in the I. obliquus genome. This enzyme generally has a larger number of genes than heme-containing peroxidases in white-rot fungi, except for P. chrysosporium20–22, 33. Hence, it is considered that laccase plays an important role in lignin degradation in I. obliquus. In plant pathogenic fungi, laccase also plays an important role in detoxification of phenolic compounds involved in plant host defense33 and melanin production34. Melanin is a natural pigment, which is also produced by I. obliquus to protect its cells in sclerotia4.
The second highest number of genes in the I. obliquus genome are MnP genes. MnP is a member of heme-containing peroxidases that can oxidize the phenolic units of lignin and appears to have the largest number of genes compared to other heme-containing peroxidases in white-rot fungi like L. edodes22. A consistent result for LiP was found in this study, where a gene encoding LiP was not detected in this fungal genome. Interestingly, VP was detected in the fungal genome, although the number of genes is lower than that of other putative enzymes. Moreover, genes encoding DyP were also found in the I. obliquus genome. Even though the ligninolytic activity of this enzyme is prominent in bacteria10, it is also detected in some fungi such as Termitomyces albuminosus35, and Auricularia auricula-judae36. In A. auricula-judae, DyP is known to be capable of degrading non-phenolic lignin units. This is because this enzyme has tyrosine (Tyr337) and tryptophan (Trp377) residues that can be used to participate in long-range electron transfer similar to LiP36,37. These residues were also present in the deduced amino acid sequence of DyP genes in I. obliquus. Therefore, DyP of I. obliquus may also have the ability to degrade non-phenolic lignin units (Supplementary Fig. 6).
As mentioned above, heme-containing peroxidases cannot function without the presence of the second class of lignin degrading enzymes, i.e., lignin degrading auxiliary enzymes10. Genes encoding glucose oxidase, alcohol oxidase, and aldehyde oxidase were detected in the I. obliquus genome. These enzymes have a role in providing the H2O2 necessary for heme-containing peroxidase activities to accomplish the lignin degradation process10. In addition, the enzymes are also known to potentially support the Fenton reaction through the generation of extracellular H2O2 together with cellobiose dehydrogenase, which is included in the cellulase class of enzymes38. These facts indicate that oxidase, peroxidase, and free radical generation are considered to be key components for lignin degradation by I. obliquus, and suggest that the lignin degrading enzymes of I. obliquus potentially target the entire lignin molecule, including phenolic and non-phenolic moieties.
Molecular characterization of IO-Px. As is well known, VP is a superior among other lignin-degrading enzymes and enables the oxidization of all lignin units. Therefore, molecular characterization of VP from I. obliquus is needed to clarify its role in the degradation of non-phenolic lignin unit by this fungus. One of the seven candidate VP genes was successfully cloned in the vector pMD20-T using the TA cloning method, referred as IO-Px. This gene contained 1,078 nucleotides encoding 347 deduced amino acids. Additionally, SignalP analysis indicated the presence of a 20 amino acid signal peptide (Fig. 3). These results suggest that IO-Px is a typical secreted protein, which is consistent with the fact that it is an extracellular fungal enzyme.
A comparison of the deduced amino acid sequence between IO-Px with other deduced amino acid sequences of basidiomycete peroxidases is shown in Fig. 3. The result showed that IO-Px has conserved heme pocket residues, i.e., Arg43, Phe46, His47, His172, Phe189, and Asp219. It is known that the heme (prosthetic group) of heme-containing peroxidases is involved in H2O2 reaction. To investigate the steric orientation of each amino acid and to clarify the catalytic properties of IO-Px, molecular modeling of IO-Px protein was performed by sequence homology using a model server and built using the templates from PDB entries, 1MnP, 2BOQ, and Lga1 for MnP (Fig. 4A), VP (Fig. 4B), and LiP (Fig. 4C), respectively. Fig. 4D shows that His47 and His172 were found as the distal and proximal histidine residues of heme, respectively. Meanwhile, Arg43, Phe46, Phe186, and Asp219 are the residues near the distal and proximal histidines (distal pocket). Pease et al. (1989) reported that heme-containing peroxidases have two histidine residues that are proposed to be essential for their activity39. The proximal histidine residue is predicted to be the axial ligand of the heme iron, while the distal histidine residue participates in peroxide cleavage39. The obtained result revealed that IO-Px has the residues essential for peroxidase activity. In addition, Fig. 3 also shows that the deduced heme pocket residues of IO-Px are conserved among other genes encoding typical MnP (PC-MnP1), typical VP (B-VP, PE-VPL1, PE-VPL2, PE-VPL3, PE-VPS1, and PSA-VP), putative VP (PO-MnP1, PO-MnP2, and PP-MnP5), and typical LiP (PC-LiPA). These results demonstrated that IO-Px is a member of heme-containing peroxidases.
On the other hand, IO-Px also has three acidic residues located in front of the internal heme propionate, i.e., Glu36, Glu40, and Asp178 (Figs. 3 and 4A, 4B). These residues are important to the formation of a Mn-binding site15,26. As is well known, the Mn-binding site is stabilized by three acidic residues (tri-carboxylates of two glutamate (Glu) and one aspartate (Asp) residues), one internal heme propionate, and two water molecules in an octahedral molecular geometry13,17. At the Mn-binding site, Mn2+ is oxidized into Mn3+, where it is chelated by organic acids secreted by fungi, such as oxalate, glyoxalate, and lactate17. The resultant Mn3+-organic acid complex acts as a diffusible redox mediator to oxidize the phenolic units of lignin5. The identified acidic residues in IO-Px also showed homology with typical MnP from P. chrysosporium (PC-MnP1) and VP from P. eryngii (PE-VPL1, PE-VPL2, and PE-VPL3). Therefore, the obtained results indicate that IO-Px has catalytic properties analogous to MnP. Unexpectedly, IO-Px did not have an exposed tryptophan residue (Typ171 for LiP and Typ164 for VP, Fig. 3). This residue is a characteristic component for LiP and is located on the protein surface15,26. The absence of this residue in IO-Px indicates that the enzyme does not exhibit LiP catalytic activity. Furthermore, phylogenetic analysis was performed using 57 deduced amino acid sequences of basidiomycete peroxidases including IO-Px in order to investigate evolutionary relationships between IO-Px and other heme-containing peroxidases (Fig. 5). The result showed that IO-Px was located in the clade of VP of Fomitiporia mediterranea and MnPs from three different species, i.e. Pyrrhoderma noxium, F. mediterranea and Sanghuangporus baumii.