Hymenochaetaceae (Basidiomycota, Hymenochaetales) from the Guineo-Congolian phytochorion: Phylloporia rinoreae sp. nov., an additional undescribed species from the Forest Global Earth Observatory Plot in Gabon

1Département de Biologie, Ecologie et Evolution, Université de Liège, Quai Van Beneden, BE-4000 Liège, Belgium 2Institut de Recherche en Ecologie Tropicale (IRET), BP-13354 Gros Bouquet, Libreville, Gabon 3Mycothèque de l’Université catholique de Louvain (MUCL, BCCM), Earth and Life Institute – Microbiology (ELIM), Université catholique de Louvain, Croix du Sud 2 bte L7.05.06, BE-1348 Louvain-la-Neuve, Belgium *Corresponding author: cony.decock@uclouvain.be REGULAR PAPER

Pursuing an ongoing survey of Hymenochaetaceae in the Guineo-Congolian phytochorion Yombi-yeni et al. 2010Yombi-yeni et al. , 2015Decock et al. 2015;Yombiyeni & Decock 2017), additional material from a Phylloporia species, found growing on twigs of an understorey species of Rinorea Aubl. (Violaceae) in several spots of rain forest in Gabon, could not be identified to any of the described species. They also form a single, distant terminal clade in phylogenetic inferences based on partial 28S DNA sequence data (region including domains D1, D2, and D3).
On this basis, as well as considering its ecological specificities, Phylloporia rinoreae is described as a new species, and illustrated. A key to the species currently reported from the Guineo-Congolian phytochorion is presented.

Specimen's identification and description
The holotype specimen of the new species is deposited at NY. Isotypes are deposited at MUCL and LBV (herbarium acronyms are according to Thiers, continuously updated). Colours are described according to Kornerup & Wanscher (1981). Sections of the basidiomes were incubated for one hour at 40°C in a NaOH 3% solution, then carefully dissected under a stereomicroscope and examined in NaOH 3% solution at room temperature (Decock et al. , 2013. To study the staining reaction of the hyphae and basidiospores, sections of the basidiomes were examined in Melzer's reagent, lactic acid Cotton blue, and KOH 4%. All microscopic measurements were done in Melzer's reagent. In presenting the size range of microscopic elements, 5% of the measurements at each end of the range are given in parentheses when relevant. The following abbreviations are used: ave = arithmetic mean, R = the ratio of length/width of basidiospores, and ave R = arithmetic mean of the ratio R. As a rule, whenever possible, 30 microscopic elements of the basidiomes (pores / hyphae / basidiospores) were measured from each specimen.

Molecular study and phylogenetic analysis
DNA extraction, amplification and sequencing of the 5' end of the nuc DNA 28S gene (region including the D1/D2/D3 domains) were as described in Decock et al. (2007). Primers LR0R and LR6 (Vilgalys & Hester 1990) were used to amplify and to sequence the portion of the 28S gene.
One hundred and forty-nine entries representing 89 taxa or potential species clades were included in the phylogenetic analysis. Materials and sequences used in this study are list-ed in supplementary file 1. The dataset is deposited at Tree-BASE under the accession number S24857.
The methodologies and parameters for running phylogenetic analyses [Bayesian inference as implemented in Mr-Bayes ver. 3.1.2 (Huelsenbeck & Ronquist 2001), Maximum likelihood as implemented in RAxML ver. 7.0.4 (Stamatakis 2006) and Maximum Parsimony as implemented in PAUP* ver. 4.0b10 (Swofford 2003)] are described in Decock et al. (2015) and Yombiyeni et al. (2015) and not repeated in details here. For the Maximum Parsimony, the heuristic search was restricted to keep 20 trees at each search with a predefined tree length ≥ 2550, in order to limit the final number of trees. Inonotus micantissimus, MUCL 52413, a species of the Inonotus clade sensu Wagner & Fischer (2002), was designated as outgroup (Larsson et al. 2006).

Phylogenetic analysis
Within Phylloporia, the length of the 28S fragment ranges from 866 to 884 bps. The final alignment of 149 sequences resulted in 966 positions (514 were constant and 387 were parsimony informative). Using the Akaike Information Criterium (AIC) of MrModeltest ver. 2.3 (Posada & Crandall 1998), the best-fit model for the 28S data set was determined as GTR+I+G with unequal base frequencies (A = 0.2436, C = 0.2831, G = 0.3235, T = 0.2499), a gamma distribution shape parameter of 0.5760 and a proportion of invariable sites of 0.3650. The nucleotide substitution rates estimated according to this model were A/C= 1.17, A/G=10.97, A/T=1.52, C/ G=1.0, C/T=25.52 and G/T = 1.0.
The MP analysis produced 30 most parsimonious trees (2601 steps, consistency index = 0.245, retention index = 0.652). The two Bayesian runs converged to stable likelihood values after 10,990,000 generations (out of a total of 20 × 10 6 generations). The remaining stationary trees from each analysis were used to compute a 50% majority rule consensus tree (BC) and to calculate posterior probabilities. In the ML searches, the 28S alignment had 520 distinct patterns with a proportion of gaps and undetermined characters of 9.97%.
The consensus of the most parsimonious trees, the consensus tree of the BI and the most likelihood tree of ML were congruent as far as the terminal clades are concerned. One of the equally most parsimonious trees, representing the dominant topology, is presented in fig. 1.
The topologies of the trees regarding the recovery and the relative positions of the different genera of Hymenochaetaceae included were identical whatever the phylogenetic methodology, in accordance with previous results (Decock et al. 2013Yombiyeni & Decock 2017). The Phylloporia clade is strongly supported ( fig. 1). The terminal clades received significant support whereas most of the internal clades were barely supported ( fig. 1), making it difficult to infer the interspecific phylogenetic relationships within the genus.

Morphological and ecological characteristics
The voucher specimens of the clade "P. rinoreae" are characterized by seasonal basidiomes, solitary, embracing to pending (thickly button-shaped) at maturity ( fig. 2A-H), applano-convex in section, with a cinnamon to cocoa brown, spongy tomentum covering the pileus. The context is homogeneous, covered by a thin, dense layer of melanized hyphae (so-called black line) subtending the tomentum. The hyphal system is dimitic in the contextual and hymenophoral trama, and the basidiospores are mostly oblong to ellipsoid in face view, the abaxial side straight to faintly incurved, averaging 4.3 × 2.7 µm.
The species was found emerging from small twigs / branches (3-5 mm diam.) of a bushy species of Rinorea (Violaceae), inhabiting the understorey of Guineo-Congolian rain forest.
On the basis of the results of both the phylogenetic inferences and the morphological studies, and considering the autecological parameters, we conclude that these specimens represent a distinct species, which is described below as Phylloporia rinoreae.

DISCUSSION
Phylloporia rinoreae is the seventh Phylloporia described and hitherto only known from the western edge of the Guineo-Congolian phytogeographic region in Gabon , Yombiyeni & Decock 2017.
It could be compared to P. weberiana, considered sensu Ryvarden & Johansen (1980;cf. also Yombiyeni & Decock 2017), also reported in Central Africa from Cameroon and the Democratic Republic of Congo (Ryvarden & Johansen 1980;Hjortstam et al. 1993). Both species share the roughly sulcate pileus and a similar anatomy, with a tomentum lying on a thin black line and, comparatively, a thinner underlying context. Ryvarden & Johansen (1980) described P. weberiana with basidiomes 30-100 mm wide and 10-30 mm thick, therefore much larger than in P. rinoreae, for which the available basidiomes did not exceed 20 mm wide and 7 mm thick. Phylloporia rinoreae and P. weberiana also differ by their number of pores / mm, respectively 9-10(-11)/ mm (or 85-100 µm diam.) and mostly 5-6/mm (Ryvarden & Johansen 1980). Furthermore, P. weberiana has likely a different autecology, its basidiomes emerging preferentially from trunks of small-stemmed trees whereas the basidiomes of P. rinoreae emerge from small (2-5 mm diam.) terminal or subterminal branches or twigs of a bushy plant.
To a lesser degree, locally, P. rinoreae could be compared to P. littoralis. Both species share the basidiome anatomy (a tomentum over a comparatively much thinner context) and autecological features (basidiome emerging from small branches of small bushy plants). Phylloporia rinoreae and P. littoralis differ by their pileus surface, sulcate and brown vs. even and pale corky, presence vs. absence of a black line separating the context from the tomentum, 9-10(-11)/mm vs. (3-)4(-5)/mm, and, as far as it is known, by their habitat and hosts; Phylloporia littoralis is known from Nichallea (Rubiaceae) only, in an open, coastal, and seasonally dry forest (Yombiyeni & Decock 2017) whereas P. rinoreae inhabits buffered, closed rain forest.
In a phylogenetic perspective ( fig. 1), until now, P. rinoreae has no known relative but a putative species repre-sented by one unnamed specimen, which also was collected in the Guineo-Congolian rain forest in Gabon (Ivindo National Park, ~ N 00°06′50″ -E 12°37′43″, GenBank Acc. KJ743283). Additional material is necessary to characterize in more detail this latter species.

SUPPLEMENTARY FILE
One supplementary file is associated to this paper: List of species / specimens used in the phylogenetic analyses (pdf) https://doi.org /10.5091/plecevo.2019.1567.1913