An Updated Global Species Diversity and Phylogeny in the Forest Pathogenic Genus Heterobasidion (Basidiomycota, Russulales)

Heterobasidion species are amongst the most intensively studied polypores because several species are aggressive white rot pathogens of managed coniferous forests mainly in Europe and North America. In the present study, both morphological and multilocus phylogenetic analyses were carried out on Heterobasidion samples from Asia, Oceania, Europe and North America. Three new taxa were found, i.e., H. armandii, H. subinsulare, and H. subparviporum are from Asia and are described as new species. H. ecrustosum is treated as a synonym of H. insulare. So far, six taxa in the H. annosum species complex are recognized. Heterobasidion abietinum, H. annosum, and H. parviporum occur in Europe, H. irregulare, and H. occidentale in North America, and H. subparviporum in East Asia. The North American H. irregulare was introduced to Italy during the Second World War. Species in the H. annosum complex are pathogens of coniferous trees, except H. subparviporum that seems to be a saprotroph. Ten species are found in the H. insulare species complex, all of them are saprotrophs. The pathogenic species are distributed in Europe and North America; the Asian countries should consider the European and North American species as entry plant quarantine fungi. Parallelly, European countries should consider the American H. occidentale and H. irregulare as entry plant quarantine fungi although the latter species is already in Italy, while North America should treat H. abietinum, H. annosum s.s., and H. parviporum as entry plant quarantine fungi. Eight Heterobasidion species found in the Himalayas suggest that the ancestral Heterobasidion species may have occurred in Asia.

Earlier phylogenetic analyses on the H. annosum complex used sequences of the internal transcribed spacer (ITS) and intergenic spacer (IGS) regions of the nuclear genes, and manganese peroxidase genes, and laccase genes (Maijala et al., 2003;Asiegbu et al., 2004). Later, several attempts were made to resolve the taxonomy of the H. annosum complex or H. insulare complex using multilocus phylogenetic approaches (Johannesson and Stenlid, 2003;Ota et al., 2006;Linzer et al., 2008;Chen et al., 2014). Recently, five species in the H. annosum species complex and eight species in H. insulare species complex were also recognized and confirmed by multilocus phylogenetic approaches, and divided into three groups based on five nuclear genes and two mitochondrial genes, i.e., ITS, the large nuclear ribosomal RNA subunit (nrLSU), the largest subunit of RNA polymerase II (RPB1), the second subunit of RNA polymerase II (RPB2), glyceraldehyde 3-phosphate dehydrogenase (GAPDH), mitochondrial ATP synthase subunit 6 (ATP6), and mitochondrial small subunit rDNA (mtSSU) (Chen et al., 2015).
Several hypotheses on the evolutionary scenarios of the Heterobasidion have been put forward (Otrosina et al., 1993;Ota et al., 2006;Linzer et al., 2008). Dalman et al. (2010) proposed that the H. annosum complex originated in Laurasia, H. annosum s.s./H. irregulare arose in Eurasia, and H. parviporum/H. abietinum/H. occidentale, which occurred in eastern Asia or western North America, emerged between 45 and 60 Ma in the Palaearctic; this conclusion was based on non-coding regions of elongation factor 1-α (EFA), glutathione-S-transferase (GST1), GAPDH, and transcription factor (TF). Recently, based on more species and samples of Heterobasidion and the fossil record, molecular dating suggested that ancestral Heterobasidion species originated in Eurasia occurred mainly during the Early Miocene (Chen et al., 2015;Zhao et al., 2017).
Based on a larger set of Heterobasidion samples from Asia, Oceania, Europe and North America, and using combined RPB1 and RPB2 sequence dataset, a further phylogenetic investigation on the genus is carried out. Four new taxa are detected, and three of them are described and illustrated in the present paper. Moreover, most relevant morphological characteristics of different species of Heterobasidion are compared.

Morphological Studies
The studied specimens and cultures ( Ecology and some macromorphological characters were based on field notes. Anatomy was studied, and measurements and drawings were made from slide preparations stained with Cotton Blue. Drawings were made with the aid of a drawing tube. In presenting the variation in the size of the spores, the 5% of the measurements at each end of the range are shown in parentheses. Basidiospore spine lengths are not included in the measurements. The following abbreviations are used: IKI = Melzer's reagent, IKI− = both non-amyloid and nondextrinoid, IKI+ = amyloid, KOH = 5% potassium hydroxide, CB = Cotton Blue, CB+ = cyanophilous, L = mean spore length (arithmetic average of all spores), W = mean spore width (arithmetic average of all spores), Q = variation in the L/W ratios between the specimens studied, n = number of spores measured from given number of specimens. Color terms are from Petersen (1996).

DNA Extraction, PCR Amplification and Sequencing
The Rapid Plant Genome kit based on acetyl trimethylammonium bromide extraction (Aidlab Biotechnologies Co., Ltd., Beijing, China) was used to  extract genomic DNA from dried fungal specimens and cultures, according to the manufacturer's instructions with some modifications (Chen et al., 2014). The PCR primers for all genes are listed in Table 2. The PCR procedure for nrLSU was as follows: initial denaturation at 94 • C for 1 min, followed by 35 cycles at 94 • C for 30 s, 50 • C for 1 min, 72 • C for 1.5 min, and a final extension at 72 • C for 10 min. The following PCR protocol for GAPDH, and ITS was used: initial denaturation at 95 • C for 3 min, followed by 35 cycles at 94 • C for 40 s, (50 • C for GAPDH, 54 • C for ITS), 72 • C for 1 min, and a final extension at 72 • C for 10 min. The PCR procedure for RPB1 and RPB2 followed Justo and Hibbett (2011) with slight modifications: initial denaturation at 94 • C for 2 min, followed by 10 cycles at 94 • C for 40 s, 60 • C for 40 s, 72 • C for 2 min, then followed by 37 cycles at 94 • C for 45 s, 55 • C for 1.5 min and 72 • C for 2 min, and a final extension at 72 • C for 10 min. PCR products were purified with a Gel Extraction and PCR Purification Combo Kit (Spin-column) in Beijing Genomics Institute, Beijing, China. The purified products were then sequenced on an ABI-3730-XL DNA Analyzer (Applied Biosystems, Foster City, CA, United States) using the same primers as in the original PCR amplifications. All newly generated sequences were deposited at GenBank 1 and listed in Table 1.  GG Liu et al., 1999;Matheny, 2005 fRPB2-7cR CCC ATR GCT TGY TTR CCC AT Liu et al., 1999;Matheny, 2005 a Degeneracr codes: S = G or C, W = A or T, R = A or G, Y = C or T, N = A or T or C or G, D = G or A or T, M = A or C.

Phylogenetic Analysis
Bondarzewia occidentalis Jia J. Chen, B. K. Cui and Y. C. Dai and B. submesenterica Jia J. Chen, B. K. Cui and Y. C. Dai were used as outgroups (Chen et al., 2015). Sequences were aligned with BioEdit (Hall, 1999) and ClustalX (Thompson et al., 1997). Sequence alignments were deposited at TreeBase 2 (submission ID 25908). Maximum parsimony (MP) analysis was applied to singlelocus genealogies for ITS, nrLSU, RPB1, PPB2, and GAPDH, and combination datasets that contained the RPB1-RPB2 sequences. The tree construction procedure was performed in PAUP * version 4.0b10 (Swofford, 2002). All characters were equally weighted, and gaps were treated as missing data. Trees were inferred using the heuristic search option with TBR branch swapping and 1000 random sequence additions. Max-trees were set to 5000, branches of zero length were collapsed, and all parsimonious trees were saved. Clade robustness was assessed using a bootstrap analysis with 1000 replicates (Felsenstein, 1985). Descriptive tree statistics tree length (TL), consistency index (CI), retention index (RI), rescaled consistency index (RCI), and homoplasy index (HI), were calculated for each maximum parsimonious tree generated. Phylogenetic trees were visualized using Treeview (Page, 1996).
MrMODELTEST2.3 (Nylander, 2004) was used to determine the best-fit evolution model for the combined dataset for Bayesian inference (BI). The BI was calculated with MrBayes 3.1.2 (Ronquist and Huelsenbeck, 2003) with a general time reversible model of DNA substitution and an invgamma distribution rate variation across sites. Eight Markov chains were run from random starting tree for 1 M generations of RPB1 and RPB2 dataset, and sampled every 100 generations. The burn-in was set to discard the first 25% of the trees. A majority rule consensus tree of all remaining trees was calculated. Branches that received bootstrap values for MP and Bayesian posterior probabilities (BPP) greater than or equal to 75% (MP) and 0.95 (BPP) were considered as significantly supported.
To determine if the datasets were significantly conflicted, the partition homogeneity test option in PAUP 4.0b was used between the loci in all possible pairwise combinations using 2 https://treebase.org/treebase-web/home.html 1000 replicates and the heuristic general search option. This test randomly shuffles phylogenetically informative sites between two paired loci: if the datasets are compatible, shuffling sites between the loci should not produce summed tree lengths that are significantly greater than those produced by the observed data (Farris et al., 1994;Huelsenbeck et al., 1996).

Molecular Phylogeny
All targeted DNA loci were successfully amplified and sequenced from our Heterobasidion samples and the outgroup species. Partition homogeneity test showed no conflicts for the RPB1 and RPB2 combined loci (P = 0.019, P ≥ 0.01). Therefore, the amino acid sequences from RPB1 and RPB2 were combined into a single sequence set. The combined dataset included sequences from 46 specimens representing 18 species. The dataset had an aligned length of 2505 characters, of which 1796 characters were constant, 671 were variable and parsimony-uninformative, and 38 were parsimony-informative. The maximum parsimony analysis yielded four equally parsimonious tree (TL = 1033, CI = 0.789, HI = 0.925, RI = 0.730, RC = 0.211). The best model for the combined RPB1 + RPB2 estimated and applied in the Bayesian analysis: GTR + I + G, lset nst = 6, rates = invgamma; prset statefreqpr = dirichlet (1,1,1,1). The Bayesian analysis resulted in a topology similar to the MP analysis, with an average standard deviation of split frequencies = 0.006737, and only the MP tree was provided. Both bootstrap values (>50%) and BPPs (≥0.90) were shown at the nodes (Figure 1).
According to the present phylogenetic analyses, Heterobasidion spp. consists of three lineages: (1)

DISCUSSION
The current phylogeny considers that Heterobasidion species belong to three species complexes: the H. annosum F complex (previously treated as the H. annosum S group, Woodward et al., 1998), the H. annosum P complex and the H. insulare complex. The F complex of H. annosum includes four species which are mainly associated to true fir species (Abies Mill., Picea abies (L.) Karst. and Tsuga (Endl.) Carrière; Linzer et al., 2008;Dalman et al., 2010). H. subparviporum is mostly found on Picea in Asia, while H. parviporum is mostly associated to Picea in Europe and H. abietinum to Abies in Europe. H. occidentale is colonizing mostly Tsuga and Abies in western North America. The H. annosum P complex includes two taxa which mostly grow on pines: H. annosum s.s. in Eurasia, H. irregulare in North America (Linzer et al., 2008;Dalman et al., 2010). The H. insulare complex includes ten species which are associated to many species of Pinophyta (Abies, Araucaria Juss., Keteleeria Carr., Larix Mill., Picea, Pinus L., Pseudolarix Gordon, Pseudotsuga Carrière and Tsuga). H. araucariae, a species from Southern Hemisphere, is clustered into H. insulare complex, and is closely related to the species H. insulare and H. subinsulare (Figure 1).
Heterobasidion insulare (=Trametes insularis Murrill) was originally described from Philippines (Murrill, 1908) and its type specimen was collected from fallen log of P. insularis in the Benguet Province, Luzon, Philippines in 1905. In 1962, Mendoza obtained the isolate FPRI-429 from P. insularis in the Mountain Province, Luzon. The Benguet Province and Mountain Province are both located in the Cordillera Administrative Region of Luzon Island (Figure 8). FPRI-429 can thus be considered as the type locality of H. insulare. The present results confirmed that FPRI-429 and representatives of H. ecrustosum are nested in the same lineage; the latter taxon was described from central Japan to Okinawa, and from southern China (Tokuda et al., 2009). We did not find any distinct morphological difference between the H. insulare type and samples of H. ecrustosum. Hence, according to the current phylogeny and morphological studies, H. ecrustosum is treated as a synonym of H. insulare.
Heterobasidion subparviporum is closely related to H. parviporum (Figure 1), and the latter was considered as same as the former according to the mating tests (Dai andKorhonen, 1999, 2003;Dai et al., 2006;Dai, 2012). Although both taxa are compatible in laboratory, they form two distinct lineages in our phylogeny (Figure 1). Morphologically, H. subparviporum differs from H. parviporum by longer cystidia (18-24 µm vs. 13-17 µm) and bigger basidiospores (5−6.5 × 4−5.2 µm vs. 4.2−5 × 3.8−4.2 µm). In addition, H. parviporum is a pathogen on P. abies in Europe, while H. subparviporum seems to be a saprophytic species according to our investigations. Based on the above data, we suggest that this Asian taxon is a new species H. subparviporum. The situation is similar with the European taxa H. parviporum and H. abietinum. These two taxa are partly sexually compatible (Capretti et al., 1990;Stenlid and Karlsson, 1991;Woodward et al., 1998), but they do not produce hybrids in nature. So they have been accepted at the species level (Niemelä and Korhonen, 1998;Otrosina and Garbelotto, 2010;Ryvarden and Melo, 2017).
Heterobasidion irregulare was proposed by Otrosina and Garbelotto (2010), and it was originally described as Polyporus irregularis Underwood on pine log from Auburn, Alabama, eastern United States (Underwood, 1897), although P. irregularis is an illegitimate name because there was earlier a fungus named P. irregularis Pers. (Persoon, 1825). The lectotype (NY730756) of H. irregulare was selected from the type material of P. irregularis Underwood, and the epitype (UC1935442) was selected from stump of P. ponderosa in the Modoc National Forest, California, western United States. However, three isolates Korhonen 05030, Korhonen 05038, and Korhonen 05039 associated to P. ponderosa from Lassen National Forest in California formed another lineage which is closely related to H. irregulare (Figure 1). Hence it is possible that another taxon exists in western North America. We did not have the basidiocarps of isolates Korhonen 05030, Korhonen 05038, and Korhonen 05039, and no information on their ecology. For the time being we treat this possible taxon as H. sp.
A comparison of these three new species and their morphological and/or phylogenetically related species is also provided in Supplementary Appendix 1. The phylogenetic analyses on single loci (ITS, nrLSU, RPB1, RPB2, and GAPDH) were shown in Supplementary Figures 1-5).

CONCLUSION
To date, 15 species are recorded in the genus Heterobasidion, including three new species described in the present study.
Five species, H. abietinum, H. annosum s.s., H. irregulare, H. occidentale, and H. parviporum, distributed in Europe and North America are forest pathogens. Ten Asian taxa are all saprotrophs, and the Asian countries ought to consider these five European and North American species as entry plant quarantine fungi. Parallelly, European countries should consider the American H. occidentale and H. irregulare as entry plant quarantine fungi (although the latter species is already in Italy), while North America should treat H. abietinum, H. annosum s.s. and H. parviporum as entry plant quarantine fungi. Eight Heterobasidion species found in the Himalayas suggest that the ancestral Heterobasidion species may have occurred in Asia, as was proposed also in the previous divergence and biogeographic studies on the genus.