Just the tip of the iceberg: uncovering a hyperdiverse clade of African Russula (Basidiomycota, Russulales, Russulaceae) species with signs of evolutionary habitat adaptations

Cathrin Manz; Mario Amalfi; Bart Buyck; Felix Hampe; Nourou S. Yorou; Slavomír Adamčík; Meike Piepenbring

doi:10.3897/imafungus.16.140321

Research Article

Just the tip of the iceberg: uncovering a hyperdiverse clade of African Russula (Basidiomycota, Russulales, Russulaceae) species with signs of evolutionary habitat adaptations

Cathrin Manz^‡, Mario Amalfi^§|, Bart Buyck^¶, Felix Hampe^#, Nourou S. Yorou^¤, Slavomír Adamčík^«», Meike Piepenbring^‡

‡ Goethe University, Frankfurt am Main, Germany

§ Botanic Garden Meise, Meise, Belgium

| Fédération Wallonie-Bruxelles, Service Général de l’Enseignement Universitaire et de la Recherche Scientifique, Bruxelles, Belgium

¶ Muséum national d’histoire naturelle, CNRS, Sorbonne Université, Paris, France

# Unaffiliated, Butzbach, Germany

¤ University of Parakou, Parakou, Benin

« Institute of Botany, Plant Science and Biodiversity Center, Slovak Academy of Sciences, Bratislava, Slovakia

» Comenius University in Bratislava, Bratislava, Slovakia

Corresponding author: Cathrin Manz ( c.manz@em.uni-frankfurt.de )

Academic editor: Marc Stadler

This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Citation: Manz C, Amalfi M, Buyck B, Hampe F, Yorou NS, Adamčík S, Piepenbring M (2025) Just the tip of the iceberg: uncovering a hyperdiverse clade of African Russula (Basidiomycota, Russulales, Russulaceae) species with signs of evolutionary habitat adaptations. IMA Fungus 16: e140321. https://doi.org/10.3897/imafungus.16.140321

Abstract

The diversity within the ectomycorrhizal genus Russula (Basidiomycota) in West Africa is largely unexplored. The study area was Benin, where only ten out of the 159 species endemic to tropical Africa have been previously reported. We focused on “Afrovirescentinae”, which is a monophyletic lineage within Russula subgen. Heterophyllidiae sister to subsect. Virescentinae. The phylogenetic placement of this clade was analysed using sequence data from ITS, LSU, mtSSU, tef1, rpb1 and rpb2 regions. Ten “Afrovirescentinae” species are recognised, described and illustrated from Benin. Four of them, R. carmesina, R. hiemisilvae, R. inflata and R. sublaevis, were previously published. Five species, Russula acrialbida sp. nov., R. beenkenii sp. nov., R. coronata sp. nov., R. florae sp. nov. and R. spectabilis sp. nov., are newly described. Species within this group are characterised by densely reticulated spore ornamentation, but they exhibit considerable variation in field appearance and pileipellis structure. In gallery forests, their basidiomata are ephemeral, small and their basidiospores have prominent ornamentation; while in savannah woodlands, the basidiomata are fleshy, large and basidiospores present low ornamentation. We suggest that these morphological traits may represent evolutionary adaptations to a specific environmental condition. We analysed the species richness, ecological range and distribution of the “Afrovirescentinae” clade globally based on data from the UNITE database, estimating a total diversity of 94 species primarily distributed in sub-Saharan Africa, but also in the Neotropics. Four additional previously described species not detected in Benin were assigned to this clade, based on holotype sequencing. Several species are widely distributed across tropical Africa and do not show specificity regarding their associated plant symbionts.

Key words:

Benin, diversity estimation, gallery forests, morphological traits, phylogeny, savannah woodlands

Introduction

Russula species are key components in ectomycorrhizal (= ECM) forests worldwide (Looney et al. 2016; Buyck et al. 2018). The genus Russula is one of the ten most species rich fungal genera and the second most species rich genus of ECM fungi (Bhunjun et al. 2022). Approximately 1,300 Russula species are described so far, but nearly 20,000 Russula OTUs were detected by ITS sequence data in a global dataset for soil fungi (Tedersoo et al. 2022).

Based on the analysis of environmental sequences, savannah appears to be the most diverse biome for Russula species (Adamčík et al. 2019). African Russula species occur in various vegetation types in association with plants belonging to Asteropeiaceae, Dipterocarpaceae, Fabaceae (Detarioideae), Phyllanthaceae or Sarcolaenaceae (Thoen and Bâ 1989; Ducousso et al. 2008; Tedersoo et al. 2011; Sanon et al. 2014). In tropical Africa, Russulaceae is the most frequently found and most abundant ECM lineage (Corrales et al. 2022) and harbours mostly generalists regarding their plant partner (Diédhiou et al. 2010; Tedersoo et al. 2011). The genus Russula is one of the most intensely studied genera of macrofungi in tropical Africa with 159 described species, all of which are strictly endemic to the African continent (see Suppl. material 1, Hennings (1908), overview of species described until 1992 in Buyck (1992), Härkönen et al. (1993), Buyck (1995), Buyck (1999), Härkönen et al. (2003), Buyck (2004), Buyck (2005), Buyck and Sharp (2007), Buyck (2008), Douanla-Meli and Langer (2009), Sanon et al. (2014), Wang et al. (2018), Wang et al. (2019b), Rossi et al. (2020) and Bhunjun et al. (2022)). However, little is known about the diversity of the genus in West Africa and only ten Russula species have been reported from Benin so far (Piepenbring et al. 2020).

The study area in Benin, West Africa, is part of the Guineo-Sudanian transition zone and harbours two major habitat types with ectomycorrhiza forming trees: 1) gallery forests with Berlinia grandiflora (Vahl) Hutch. & Dalziel and Uapaca guineensis Müll.Arg. and 2) Sudanian savannah woodlands with Isoberlina doka Craib & Stapf, Isoberlinia tomentosa (Harms) Craib & Stapf, Monotes kerstingii Gilg and Uapaca togoensis Pax. Gallery forests are characterised by a dense vegetation, water availability throughout the year and frequent flooding events during rainy seasons (Kirchmair 2017), while savannah woodlands consist of an open vegetation of more or less scattered trees and shrubs with soils covered by grasses exposed to sunlight and wind, which are frequently superficially dried out, even during the rainy season. Both habitats are widely distributed in West Africa and harbour only partially overlapping ECM fungal communities which are dominated by species of Russulaceae (Meidl et al. 2021). Several ECM species are reported to have broad distribution areas across sub-Saharan Africa. For example, Cantharellus congolensis Beeli known from gallery forests in Benin (Dramani et al. 2022), was originally described from a lowland tropical rainforest in the Democratic Republic of the Congo (DRC) (Beeli 1928) and is reported from savannah woodlands adjacent to rainforests in Tanzania (Buyck et al. 2013). The distribution area of several ECM species occurring in Sudanian savannah woodlands in Benin extends to the miombo woodlands of the Zambezian biogeographic region (Badou et al. 2018; Han et al. 2018; Aïgnon et al. 2023). Nevertheless, community-based global studies predict at least some endemicity of ECM species in sub-Saharan Africa (Tedersoo et al. 2022).

The present study is focusing on the diversity of an underexplored lineage of mostly African Russula species which is closely related to subsect. Virescentinae Singer (Wang et al. 2019a) and is hereinafter referred to as “Afrovirescentinae”. Species in this lineage are morphologically highly diverse, but share a reticulate spore ornamentation and single-celled pileocystidia. They occur in gallery forests and/or savannah woodlands in Benin. Due to contrasting conditions in these two habitats, we expect corresponding morphological adaptations. These adaptations probably evolved multiple times as convergences and were facilitated by low host specificity of Russula spp. in Africa. Our aim was to define the species concepts and assign correct names to Beninese “Afrovirescentinae” members by morphological and molecular analyses of recently collected basidiomata and historical collections including holotypes. We reconstructed phylogenetic relationships within “Afrovirescentinae” and define the position of the group within subgen. Heterophyllidiae Romagn. based on the analysis of five DNA regions. Environmental nrITS sequence data retrieved by a PlutoF search (https://plutof.ut.ee/ accessed on 29.05.2024) were used to estimate the species diversity of “Afrovirescentinae”, their ecological range and distribution.

Methods

Sampling

In total, 283 specimens of species belonging to the genus Russula were collected in gallery forests and savannah woodlands in Benin during field research in June and July 2021 and 2022. The material was dried with a dehydrator at 40 °C. Molecular sequence data of the barcoding ITS nrDNA region were generated (for method, see below) and allowed a preliminary sorting of the species. Thirty-nine of the specimens from Benin were selected for this study and deposited in the Herbarium Berolinense (B). Additional 21 specimens including eight holotypes from various African countries were loaned and investigated from the Herbaria of the University of Parakou (UNIPAR), the Botanical Garden Meise (BR) and the University of Helsinki (H).

Morphological investigation

Fresh basidiomata were photographed in the field using a Panasonic DMC-TZ81 or an Olympus TG-6 digital camera and macroscopic features were registered in the fresh state with colour codes referring to Kornerup and Wanscher (1978). Reactions to ferrous sulphate (FeSO₄), potassium hydroxide (KOH), sulphovanillin and phenol were tested on intact tissues in the fresh state. The reaction to Guaiac was tested after 8–10 seconds on stipe and lamellar surfaces following Chalange (2014). Spore print colour codes are referring to the scale of Romagnesi (1967). Light microscopic observation and terminology follow the standards proposed by Adamčík et al. (2019) and were carried out using a Nikon eclipse 80i microscope with a Nikon Y-IDT drawing attachment at a magnification of 2,000× for line drawings and a Nikon DS-Fi2 camera for measurements from microscopic pictures. At least three collections per species were measured for microscopic descriptions, if available. The number of measured elements is indicated in brackets after the indications of the sizes for each structure. Scanning electron microscopy (SEM) was carried out using a Hitachi (S 4500) microscope with a magnification of 6,000–10,000 times.

DNA extraction, amplification and sequencing

The genomic DNA of recently collected specimens was extracted using the innuPREP Plant DNA Kit (Analytik Jena, Germany) following the manufacturer’s instructions (protocol 1) using SLS as lysis solution. The genomic DNA of old herbarium specimens was extracted using a method developed in the context of the present study. Some 20–50 mg of dry material was placed in a 2 ml sterile Eppendorf tube. Sterile beads were added and the samples were ground using a Tissue Lyser with 2 cycles à 90 s at 1.8 Hz. 2% of PVP (polyvinylpyrrolidone) were added to the 2× cetyltrimethylammonium bromide (CTAB) buffer before the addition to the samples. The stock solution of the CTAB buffer consisted of 2% CTAB dissolved in an aqueous solution of 10% Tris-HCl 1M pH8, 28% NaCl 5M and 4% EDTA 0.5M pH 8. One ml of 55 °C hot CTAB buffer and 10 µl of proteinase K were added to each sample. The samples were incubated at 55 °C for one hour and at room temperature overnight. The next day, the samples were incubated at 65 °C for one hour. After centrifugation at room temperature at 10,000 rpm for 10 min, the supernatant liquid was transferred to a new sterile Eppendorf tube. An equal volume of isoamyl alcohol / phenol / chloroform 1:25:24 was added, then the tubes were shortly vortexed and then shaken for 2 min. After a centrifugation at room temperature at 8,000 rpm for 10 min, the upper phase was transferred to a new sterile Eppendorf tube. An amount of 2/3 of the recovered volume of -20 °C cold 2-propanol was added to the recovered upper phase, mixed by inversion for approximately 5 min and incubated at -20 °C for 15 min. After a centrifugation at room temperature at 10,000 rpm for 10 min, the supernatant was discarded. 600 µl of 70% ethanol were added to the pellet, the samples were then incubated at -20 °C for 20 min and centrifuged again at room temperature at 10,000 rpm for 2 min. The supernatant was discarded and the pellets were air dried for approx. 1 hour. The pellets were re-suspended with 75 µl 60 °C warm sterile TE-buffer and stored overnight in the fridge at 4 °C.

Six markers were amplified: 1) the internal transcribed spacer region of ribosomal DNA (ITS), comprising the ITS1 and ITS2 spacer regions and the ribosomal gene 5.8S using primers ITS1F and ITS4B (Gardes and Bruns 1993) for recently collected material; for old herbarium specimens, partial sequences were obtained using primers ITS5 and ITS2 (White et al. 1990) for the ITS1 partition or 58A1F (Martin and Rygiewicz 2005) and ITS4 (White et al. 1990) for the ITS2 partition; 2) a part of the ribosomal large subunit 28S region (LSU), using primers LR0R (Cubeta et al. 1991) and LR7 (Vilgalys and Hester 1990); 3) a part of the mitochondrial small subunit rDNA (mtSSU) using primers MS1 and MS2 (White et al. 1990); 4) the translation elongation factor 1-alpha (tef1) using primers tef1F and tef1R (Morehouse et al. 2003); 5) the largest subunit of the RNA polymerase II (rpb1) using primers gRPB1-Af (Stiller and Hall 1997) and fRPB1-Cr (Matheny et al. 2002); 6) the region between the conserved domains 6 and 7 of the second largest subunit of the RNA polymerase II (rpb2) using primers bRPB2-6F (Liu and Hall 2004) and bRPB2-7.1R (Matheny et al. 2005).

PCR products were obtained using a peqSTAR 2× Gradient Thermal Cycler (PEQLAB, Erlangen, Germany) and the VWR Taq DNA Polymerase (VWR, Darmstadt, Germany). The cycling conditions were as follows: For all loci, the initial denaturation was carried out for 4 min at 95 °C and all elongation steps were carried out at 72 °C. ITS & LSU: 35 cycles of denaturation for 45 s at 94 °C, annealing for 30 s at 53 °C and elongation for 60 s, final elongation for 5 min; mtSSU: 35 cycles of denaturation for 45 s at 94 °C, annealing for 45 s at 52 °C, elongation for 60 s, final elongation for 10 min; tef1: 35 cycles of denaturation for 45 s at 94 °C, annealing for 60 s at 52 °C, elongation for 75 s, final elongation for 10 min; rpb1: 20 cycles of denaturation for 45 s at 94 °C, annealing for 30 s at 55 °C (decreasing 0.5 °C each cycle), elongation for 90 s followed by 20 cycles of denaturation for 45 s at 94 °C, annealing for 30 s at 45 °C, elongation for 90 s, final elongation for 10 min; rpb2: 35 cycles of denaturation for 45 s at 94 °C, annealing for 60 s at 58.5 °C, elongation for 60 s, final elongation for 5 min. Successfully amplified products were sent to Microsynth Seqlab (Göttingen, Germany) for purification and sequencing using the same primers as used for PCR.

UNITE search and analysis of ITS sequence data

The initial ITS dataset was based on sequences obtained in this study, few sequences retrieved from GenBank via BLAST search (https://blast.ncbi.nlm.nih.gov/Blast.cgi) and 13 sequences of Heterophyllidiae retrieved from the alignment in Wang et al. (2019a). One representative ITS sequence for each of the ten “Afrovirescentinae” species from Benin was used for a UNITE database search (https://unite.ut.ee/, accessed on 14.05.2024). Two approaches were tested to retrieve the maximum number of sequences belonging to “Afrovirescentinae” from UNITE.

Species hypothesis (SH) approach: For each of the ten analysed species, the massBLASTer search (https://unite.ut.ee/analysis.php, accessed on 14.05.2024) resulted in 30 most similar sequences. All unique SHs at the 1.5% dissimilarity threshold containing at least one sequence from this search were sorted. All sequences belonging to these sorted SHs were retrieved.
Compound cluster (CC) approach: For each of the ten species only the most similar BLAST result was selected. All corresponding SH at the 1.5% dissimilarity threshold were searched in PlutoF (https://plutof.ut.ee/, accessed on 14.05.2024) to identify the corresponding CC. All ten SH were grouped in the CC “ Agaricomycetes | UCL10_011118”. All sequences belonging to this CC were selected for further analysis.

The selected sequences were exported by the clipboard tool in PlutoF (https://plutof.ut.ee/) and were downloaded together with corresponding metadata (columns “Sequence ID”, “Sequence”, “Sampling area.Country”, “Identifiacation.Taxon name”, “Isolation source”, “Parent.Interactions”). The metadata were added to the sequence headers using Microsoft Excel version 2405 (Microsoft Corporation). Through repeated steps of calculating phylogenetic trees and sorting out sequences, the datasets from the two approaches were reduced to only a few sequences for each species level clade in the “Afrovirescentinae” representing unique information about the country of origin or ecology. In the final step, both datasets were combined into a single one used for the calculation of the phylogenetic ITS tree.

Phylogenetic analyses

For the analysis of nr ITS data, sequences were aligned by the online version of the multiple sequence alignment programme MAFFT v. 7 (Katoh et al. 2019), using the Auto-strategy. Sequences of Russula species from subgen. Compactae (Fr.) Bon (De Lange et al. 2021; De Lange et al. 2023) were used as an outgroup. Information on the specimens of which nr ITS sequences were newly generated for the present study are listed in Table 1. The alignment was trimmed and edited in Geneious Prime 2023.2.1 (https://www.geneious.com). A Maximum Likelihood analysis was run using the RAxML-HPC BlackBox (Stamatakis 2014) via the CIPRES Science Gateway (https://www.phylo.org/) with automatically halted bootstrapping. The final trees were edited using TreeGraph 2 (Stöver and Müller 2010) and Adobe Illustrator 27.3.

For the multigene phylogenetic analysis, 79 collections representing 55 potential species clades were included (Table 2). A combined dataset including sequence data of five partial loci (LSU, mtSSU, tef1, rpb1, rpb2) was constructed and used for further phylogenetics analyses. Nucleotide sequences were automatically aligned with the MUSCLE algorithm (Edgar 2004) with default settings, then manually adjusted as necessary with PhyDE® v.0.9971 (Müller et al. 2010). Potentially ambiguously aligned segments were also detected using the Gblocks v.0.91b programme (Castresana 2000) with the following parameter settings: minimum number of sequences for a conserved position = 40 (minimum possible); minimum number of sequences for a flank position = 40 (minimum possible); maximum number of contiguous non-conserved positions = 4 bp, minimum block size = 4 bp and gaps allowed within selected blocks in half of the sequences. The assignment of codon positions was confirmed by translating nucleotide sequences into predicted amino acid sequences using MacClade 4.0 (Maddison and Maddison 2005) and then compared with several annotated Russula sequences available on GenBank. Phylogenetic analyses were performed separately for each individual and concatenated loci using Bayesian Inference (BI) as implemented in MrBayes v.3.2 (Ronquist et al. 2011) and Maximum Likelihood (ML) as implemented in RAxML 7.0.4 (Stamatakis 2006; Stamatakis et al. 2008).

Figure 1.

Phylogenetic Maximum Likelihood tree of Russula subg. Heterophyllidiae, based on concatenated LSU, mtSSU, tef 1, rpb1 and rpb2 sequence data with bootstrap (ML) and Bayesian Inference (BI) values indicated over the branches. Branches in bold are supported by ML ≥ 75 or BI ≥ 0.95; black – with full support by both analyses (ML = 100, BI = 1.00), blue – supported by both ML and BI, grey – supported only by one analysis, asterisk (*) indicates full support. Highlighted species are described and illustrated in this study. Labels of holotypes of newly-described species are written in bold. Habitats (gallery forests or savannah woodlands), sizes of basidiomata (small and thin-fleshed or large and thick-fleshed) and prominence of spore ornamentation (low and high) for species occurring in Benin are indicated by icons.

Table 1.

Download as

CSV

XLSX

Information on the specimens of which nrITS sequences were newly generated for the present study.

Taxon	Voucher no.	Origin	Gen Bank Acc. No.
R. acrialbida	CM-21-083	Benin	PQ150333
R. acrialbida	CM-21-062	Benin	PQ150332
R. acrialbida	CM-21-138	Benin	PQ150341
*R. acrialbida*	CM-21-093	Benin	PQ150334
R. acrialbida	CM-21-054	Benin	PQ150331
R. acrialbida	CM-21-038	Benin	PQ150325
R. acrialbida	CM-21-053	Benin	PQ150330
R. acrialbida	CM-21-049	Benin	PQ150329
R. acrialbida	CM-21-039	Benin	PQ150326
R. acrialbida	CM-21-041	Benin	PQ150327
R. acrialbida	CM-22-215	Benin	PQ150353
R. acrialbida	CM-22-175	Benin	PQ150349
*R. albofloccosa*	Rammeloo6096	Burundi	PQ150375
R. beenkenii	CM-21-132	Benin	PQ150340
R. beenkenii	CM-21-152	Benin	PQ150346
R. beenkenii	CM-22-232	Benin	PQ150355
*R. beenkenii*	CM-21-047	Benin	PQ150328
R. carmesina	CM-21-143	Benin	PQ150342
R. carmesina	CM-21-108	Benin	PQ150336
R. carmesina	SYN5103	Guinea	PQ150380
R. coronata	CM-22-261	Benin	PQ150357
*R. coronata*	CM-22-241	Benin	PQ150363
R. florae	CM-21-145	Benin	PQ150343
*R. florae*	CM-21-098	Benin	PQ150335
R. florae	CM-21-160	Benin	PQ150348
R. florae	CM-21-109	Benin	PQ150337
R. florae	CM-21-121	Benin	PQ150339
R. florae	CM-22-178	Benin	PQ150350
R. florae	CM-22-284	Benin	PQ150359
R. florae	CM-22-185	Benin	PQ150351
R. florae	CM-22-206	Benin	PQ150352
R. hiemisilvae	CM-22-150	Benin	PQ150345
*R. hiemisilvae*	1028	Tanzania	PQ150381
R. inflata	CM-22-231	Benin	PQ150362
R. inflata	CM-21-128	Benin	PQ150360
R. inflata	CM-21-144	Benin	PQ150361
R. inflata	Degreef90/9	DRC	PQ150373
R. inflata	Schreurs1738	DRC	PQ150370
R. inflata	Schreurs1415	DRC	PQ150369
R. inflata	Thoen5209	DRC	PQ150371
R. inflata	Thoen7647	Senegal	PQ150372
*R. inflata*	Buyck1652	DRC	PQ150366
R. inflata	Buyck1560	DRC	PQ150365
R. inflata	Buyck1653	DRC	PQ150367
*R. intricata*	Schreurs1778	DRC	PQ150368
R. mollicula nom. prov.	CM-22-244	Benin	PQ150356
*R. roseoalba*	GF3024	DRC	PQ150377
*R. roseovelata*	Schmitz-Levecq102	DRC	PQ150374
*R. roseoviolacea*	Schreurs1638	DRC	PQ150378
R. sp.	OAB1010	Guinea	PQ150382
R. sp.	OAB1004	Guinea	PQ150383
R. sp.	SYN5095	Guinea	PQ150384
R. sp.	Buyck1660	DRC	PQ150376
R. spectabilis	CM-21-116	Benin	PQ150338
*R. spectabilis*	CM-21-154	Benin	PQ150347
R. sublaevis	CM-21-148	Benin	PQ150344
R. sublaevis	CM-22-281	Benin	PQ150358
R. sublaevis	CM-22-219	Benin	PQ150354
R. sublaevis	ADK3317	Benin	PQ150364
*R. sublaevis*	Schreurs985	DRC	PQ150379

Table 2.

Download as

CSV

XLSX

Information on the DNA sequences used to reconstruct the phylogenetic tree based on concatenated LSU, mtSSU, tef1, rpb1 and rpb2 sequence data. Holotype collections are labelled with “HT”.

Taxon	Voucher no.	Origin	LSU	mtSSU	tef1	rpb1	rpb2	References
R. aureoviridis	H16082612	China	MK881920	MK882048	MN617846	–	–	Das et al. (2017)
*R. acrialbida*	CM-21-083	Benin	PQ152097	PQ157909	PQ179296	PQ203937	PQ203916	this study
*R. acrialbida*	CM-21-062	Benin	PQ152098	PQ157910	PQ179297	PQ203936	PQ203918	this study
*R. acrialbida*	CM-21-138	Benin	PQ152099	PQ157907	PQ179294	PQ203935	PQ203919	this study
R. acrialbida HT	CM-21-093	Benin	PQ152100	PQ157908	PQ179295	PQ203934	PQ203920	this study
*R. acrialbida*	CM-21-054	Benin	PQ152101	PQ157911	PQ179298	PQ203933	PQ203917	this study
R. aeruginea	MSCZ2017100110	China	MK881944	MK882072	MT085583	MT085539	–	unpublished
R. aff. brunneoannulata	BB 06.029	Madagascar	KU237452	KU237295	KU237887	KU237602	KU237738	Buyck et al. (2018)
R. aff. chloroides	FH12273	Belgium	KT933876	–	–	KT957386	KT933947	Looney et al. (2016)
R. aff. crustosa	BB 06.616	Canada	KU237461	KU237305	KU237896	KU237612	KU237747	Buyck et al. (2018)
R. aff. laurocerasi	BB 06.610	Canada	KU237458	KU237302	KU237893	KU237609	KU237744	Buyck et al. (2018)
R. aff. madecassensis	BB 06.255	Madagascar	KU237475	KU237319	KU237906	KU237624	KU237761	Buyck et al. (2018)
R. aff. virescens	BB 09.021	New Caledonia	KU237582	KU237430	KU238009	KU237723	KU237868	Buyck et al. (2018)
R. albidogrisea	TLH18053005	China	MN839562	MN839610	MT085609	MT085560	MT085637	unpublished
R. amoenolens	BB 08.675	Italy	KU237562	KU237410	–	KU237706	KU237848	Buyck et al. (2018)
R. aureoviridis	H17071311	China	MN839561	MN839609	MT085608	MT085542	MT085636	unpublished
*R. beenkenii*	CM-21-132	Benin	PQ152093	PQ157915	PQ179302	PQ203928	PQ203903	this study
*R. beenkenii*	CM-21-152	Benin	PQ152094	PQ157913	PQ179300	–	PQ203902	this study
*R. beenkenii*	CM-22-232	Benin	PQ152095	PQ157929	PQ179317	PQ203929	PQ203905	this study
R. beenkenii HT	CM-21-047	Benin	PQ152096	PQ157914	PQ179301	PQ203927	PQ203904	this study
R. callainomarginis	GDGM79715	China	MN839582	MN839632	MT085602	MT085535	MT085659	Song et al. (2022)
*R. carmesina*	CM-21-143	Benin	PQ152084	PQ157925	PQ179309	PQ203938	PQ203923	this study
*R. carmesina*	CM-21-108	Benin	PQ152085	PQ157922	PQ179308	PQ203926	PQ203922	this study
R. cf. annulata	BB 06.048	Madagascar	KU237470	KU237314	KU237902	–	KU237756
R. cf. illota	BB 06.380	Mexico	KU237464	KU237308	KU237898	KU237615	KU237750	Buyck et al. (2018)
R. cf. pseudocarmesina	BB 06.030	Madagascar	KU237453	KU237296	–	KU237603	KU237739	Buyck et al. (2018)
R. cf. roseoalba	BB 06.105	Madagascar	KU237472	KU237316	–	KU237621	KU237758	Buyck et al. (2018)
R. cf. vesca	BB 06.525	Mexico	KU237465	KU237309	KU237899	KU237616	KU237751	Buyck et al. (2018)
R. chloroides	BB 07.209	Slovakia	KU237559	KU237407	KU237990	KU237703	KU237845	Buyck et al. (2018)
*R. coronata*	CM-22-261	Benin	PQ152088	PQ157931	PQ179320	PQ203939	PQ203924	this study
R. coronata HT	CM-22-241	Benin	PQ152089	PQ157934	PQ179321	–	PQ203925	this study
R. crustosa	BPL265	USA	KT933826	–	–	KT957338	KT933898	Looney et al. (2016)
R. crustosa	BPL251	USA	KT933822	–	–	KT957334	KT933894	Looney et al. (2016)
R. elastica	BB 06.009	Madagascar	KU237451	KU237294	–	KU237601	KU237737	Buyck et al. (2018)
R. flavobrunnea var. violaceotincta	BB 06.050	Madagascar	KU237468	KU237312	KU237901	KU237619	KU237754	Buyck et al. (2018)
*R. florae*	CM-21-145	Benin	PQ152077	PQ157926	PQ179314	PQ203948	–	this study
R. florae HT	CM-21-098	Benin	PQ152078	PQ157921	PQ179311	PQ203946	PQ203906	this study
*R. florae*	CM-21-160	Benin	PQ152079	PQ157927	PQ179313	PQ203945	PQ203909	this study
*R. florae*	CM-21-109	Benin	PQ152080	PQ157923	PQ179310	PQ203947	PQ203907	this study
*R. florae*	CM-21-121	Benin	PQ152081	PQ157924	PQ179312	PQ203949	PQ203908	this study
R. floriformis subsp. floriformis HT	Corrales943	Colombia	MT023729	MT039861	MT024552	–	MT021752	Vera et al. (2021)
R. floriformis subsp. symphoniae HT	Corrales591	Panama	MT023730	MT039862	MT024553	–	MT021753	Vera et al. (2021)
R. grisea	BB 07.184	Slovakia	KU237509	KU237355	KU237939	KU237659	KU237795	Buyck et al. (2018)
R. herrerae	BB 06.532	Mexico	KU237486	KU237330	KU237915	KU237635	KU237772	Buyck et al. (2018)
*R. hiemisilvae*	CM-22-150	Benin	PQ152090	PQ157918	PQ179305	PQ203940	PQ203910	this study
R. ilicis	MF 00.300	Italy	KU237595	KU237443	KU238021	KU237733	KU237880	Buyck et al. (2018)
*R. inflata*	CM-22-231	Benin	PQ152086	PQ157933	PQ179318	PQ203944	PQ203899	this study
*R. inflata*	CM-21-128	Benin	PQ152082	PQ157919	PQ179306	PQ203942	PQ203900	this study
*R. inflata*	CM-21-144	Benin	PQ152083	PQ157920	PQ179307	PQ203943	PQ203901	this study
R. ionochlora	BB 07.338	Slovakia	KU237508	KU237354	KU237938	KU237658	KU237794	Buyck et al. (2018)
R. langei	BB 07.792	France	KU237510	KU237356	KU237940	KU237660	KU237796	Buyck et al. (2018)
R. madecassensis	BB 06.146	Madagascar	KU237456	KU237300	KU237891	KU237607	KU237742	Buyck et al. (2018)
R. maguanensis	XHW4765	China	MH714537	OQ586179	MH939983	OQ579058	MH939989	Wang et al. (2019a)
R. mariae	BB 07.038	USA	KU237538	KU237384	KU237968	KU237687	KU237824	Buyck et al. (2018)
R. medullata	BB 07.252	Slovakia	KU237546	KU237392	KU237976	KU237693	KU237832	Buyck et al. (2018)
R. mollicula nom. prov.	CM-22-244	Benin	PQ152087	PQ157930	PQ179319	PQ203941	PQ203921	this study
R. mustelina	FH12226	Germany	KT933866	–	–	KT957376	KT933937	Looney et al. (2016)
R. mustelina	1176/S. Adamcik 09.88	Slovakia	KU237596	KU237444	KU238022	–	KU237881	Buyck et al. (2018)
R. oleifera	BB 98.024	Tanzania	KU237490	KU237334	KU237919	–	KU237776	Buyck et al. (2018)
R. ornaticeps	BB 06.530	Mexico	KU237466	KU237310	–	KU237617	KU237752	Buyck et al. (2018)
R. pectinatoides	BB 06.538	Mexico	KU237462	KU237306	–	KU237613	KU237748	Buyck et al. (2018)
R. prolifica	BB 06.161	Madagascar	KU237455	KU237299	KU237890	KU237606	KU237741	Buyck et al. (2018)
R. pseudociliata	BB 08.061	Madagascar	KU237537	KU237383	KU237967	KU237686	KU237823	Buyck et al. (2018)
R. pulverulenta	BB 05.160	USA	KU237563	KU237411	–	KU237707	KU237849	Buyck et al. (2018)
R. redolens	BPL141	USA	KT933808	–	–	KT957321	KT933879	Looney et al. (2016)
R. sp.	EDC 12 061	Cameroon	KR364201	–	–	KR364468	KR364338	De Crop et al. (2017)
*R. spectabilis*	CM-21-116	Benin	PQ152091	PQ157916	PQ179303	PQ203950	PQ203914	this study
R. spectabilis HT	CM-21-154	Benin	PQ152092	PQ157917	PQ179304	PQ203951	PQ203915	this study
*R. sublaevis*	CM-21-148	Benin	PQ152102	PQ157912	PQ179299	PQ203930	PQ203911	this study
*R. sublaevis*	CM-22-281	Benin	PQ152103	PQ157932	PQ179316	PQ203932	PQ203913	this study
*R. sublaevis*	CM-22-219	Benin	PQ152104	PQ157928	PQ179315	PQ203931	PQ203912	this study
R. substriata	XHW4766	China	OQ359148	OQ371394	MH939987	OQ359996	MH939993	Wang et al. (2019a), Buyck et al. (2023)
R. vesca	B17100103	China	MK881939	MK882067	MN617851	MT085493	–	unpublished
R. vesca	K17092601	China	MN839566	MN839614	MT085613	MT085494	MT085641	unpublished
R. vesca	BPL284	USA	KT933839	–	–	KT957351	KT933910	Looney et al. (2016)
R. violeipes	BB 07.273	Slovakia	KU237534	KU237380	KU237964	KU237683	KU237820	Buyck et al. (2018)
R. virescens	HJB9989	Belgium	DQ422014	–	–	–	DQ421955	unpublished
R. xanthovirens	H15060611	China	MN839560	MN839608	MT085607	MT085501	MT085635	Song et al. (2018)
R. xanthovirens	B17091630	China	MK881928	MK882056	MT085570	–	–	Song et al. (2018)

Russula species from subgen. Brevipedum Buyck & V. Hofstetter (Russula herrerae A. Kong, A. Montoya & Estrada strain 239/BB 06.532, Russula chloroides (Krombh.) Bres. strain 572/BB 07.209 and voucher FH12273 and Russula callainomarginis J.F. Liang & J. Song voucher GDGM79715) were designated as outgroups. Models of evolution for BI were estimated using the Akaike Information Criterion (AIC) as implemented in ModelTest 3.7 (Posada and Crandall 1998). In order to facilitate the data partitioning by codon position, the three clade specific, confidently alignable introns, present in the tef1 partition, in the rpb1 partition and the one at the end of the rpb2 partition were excised and analysed as a distinct partition.

The dataset was subdivided into eleven partitions: (LSU) (mtSSU) (rpb1 codons 1+2) (rpb1 3^rd codon position) (rpb2 codons 1+2) (rpb2 codon 3) (tef1 codons 1+2) (tef 1 codon 3) (tef1 introns) (rpb1 introns) (rpb2 intron).

The best-fit models for each partition were implemented as partition-specific models within partitioned mixed-model analyses of the combined dataset and all parameters were unlinked across partitions. The combined dataset for the Bayesian analysis was implemented with four independent runs, each with four simultaneous independent chains for 10 million generations, starting from random trees and keeping one tree every 1000^th generation.

All trees sampled after convergence (ave. standard deviation of split frequencies < 0.01) and confirmed using Tracer v.1.4 (Rambaut and Drummond 2007) were used to reconstruct a 50% majority-rule consensus tree (BC) and to calculate Bayesian posterior probabilities (PP). PP of each node was estimated based on the frequency at which the node was resolved amongst the sampled trees with the consensus option of 50% majority-rule (Simmons et al. 2004). A probability of 0.95 was considered significant.

Maximum Likelihood (ML) searches were conducted with RAxML involving 1000 replicates under the GTRGAMMAI model, with all model parameters estimated by the programme. In addition, 1000 bootstrap (ML BS) replicates were run with the same GTRGAMMAI model. In order to let RaxML software estimate the parameters of the evolution model separately for each independent locus, we provided an additional alignment partition file to the software. Clades with ML BS values of 75% or greater were considered supported by the data.

Nucleotide sequences are considered to be phylogenetically informative until they reach the substitution saturation; especially in coding sequences, saturation will be more pronounced in the rapidly evolving third codon position. At this point, it is no longer possible to deduce whether an observed similarity between a pair of sequences results from their common ancestry or whether this has occurred by chance (Jeffroy et al. 2006). To detect the possible bias from substitution saturation, we tested the first, second and the third codon position of the coding region studied (tef1, rpb1 and rpb2) as well as the non-coding loci (LSU, mtSSU, tef1, rpb1 and rpb2 introns) by using Xia’s test (Xia et al. 2003; Xia and Lemey 2009), as implemented in DAMBE (Xia and Xie 2001). As the critical index substitution saturation (Iss.c) is based on simulation results, there is a problem with more than 32 species. To circumvent this problem, DAMBE was used to randomly sample subsets of 4, 8, 16 and 32 OTUs multiple times and perform the test for each subset to see if substitution saturation exists for these subsets of sequences. In order to confirm the results of the Xia’s method, we also plotted transitions and transversions at the first, second and third codon positions against Tamura-Nei genetic distances with the aid of the DAMBE package, with an asymptotic relationship indicating the presence of saturation.

Before combining the data partitions, topological incongruence between the datasets was assessed using 1000 replicates of ML BS under the same models described above, on each locus separately. Paired trees were examined for conflicts only involving nodes with ML BS > 75% (Mason-Gamer and Kellogg 1996; Lutzoni et al. 2004; Reeb et al. 2004) compared with the software compat.py (Kauff and Lutzoni 2002) available at https://www.lutzonilab.net/downloads. A conflict was assumed to be significant when two different relationships for the same set of taxa (one being monophyletic and the other non-monophyletic) were observed in rival trees.

Results

Phylogenetic analyses

The final combined DNA sequence alignment for the multigene phylogeny including 132 new sequences of Russula spp. from Benin resulted in 4,253 characters including gaps. Alignment sizes, summary statistics of sequence data, best-fit models and tests of substitution saturation for each dataset are provided in Table 3.

Table 3.

Download as

CSV

XLSX

Summary of datasets used for phylogenetic inferences.

Datasets
Properties	LSU	mtSSU	tef1	tef1	tef1	rpb1	rpb1	rpb1	rpb2	rpb2	rpb2
Properties	LSU	mtSSU	1^st & 2^nd	3^rd	introns	1^st & 2^nd	3^rd	introns	1^st & 2^nd	3^rd	introns
Alignment size	911	596	448	223	269	325	162	524	477	238	81
Excluded characters	-	32	-	-	-	-	-	16	-	-	11
Model selected	GTR+I+G	GTR+I+G	HKY+I+G	GTR+G	HKY+I+G	GTR+I+G	K80+I+G	GTR+I+G	GTR+I+G	GTR+G	GTR+I
-Likelihood score	4177.5352	2130.6577	1315.8429	2819.5667	3379.1914	1068.9609	2583.7498	3745.0576	1900.6295	4743.6709	1086.4247
Base frequencies
Freq. A =	0.2569	0.3975	0.3000	0.1290	0.2473	0.3273	Equal	0.1714	0.2957	0.2038	0.2258
Freq. C =	0.2143	0.1150	0.2008	0.3705	0.2404	0.2150	Equal	0.2641	0.2369	0.2782	0.2549
Freq. G =	0.2926	0.1673	0.2738	0.2186	0.2035	0.2582	Equal	0.2484	0.2454	0.2734	0.1748
Freq. T =	0.2362	0.3201	0.2253	0.2818	0.3088	0.1995	Equal	0.3161	0.2220	0.2446	0.3445
Proportion
of invariable sites	0.6604	0.6294	0.7879	0	0.1123	0.6222	0.1315	0.3556	0.7447	0	0.2859
Gamma shape	0.5652	0.6717	0.5475	1.9875	2.4166	0.6049	4.2319	0.5721	0.6079	2.3269	-
Test of substitution saturation
Iss	0.155	0.206	0.318	0.505	0.539	0.265	0.465	0.439	0.278	0.518	0.661
Iss.cSym	0.741	0.702	0.696	0.686	0.692	0.684	0.709	0.703	0.700	0.684	0.999
P (Sym)	< 0.0001	< 0.0001	< 0.0001	< 0.0001	<0.001	< 0.0001	< 0.0001	< 0.0001	< 0.0001	< 0.0001	< 0.0001
Iss.cAsym	0.430	0.377	0.502	0.365	0.380	0.354	0.412	0.378	0.374	0.360	0.925
P (Asym)	< 0.0001	< 0.0001	< 0.0001	< 0.0001	< 0.0007	< 0.0001	< 0.0001	< 0.0012	< 0.0001	< 0.0003	< 0.0001

No conflict involving significantly supported nodes was found between the tree topologies obtained for the datasets of individual loci, using the 75% ML BS criterion; the datasets were therefore combined. The test of substitution saturation showed that the observed index of substitution saturation (Iss) was significantly lower than the corresponding Iss.c in every partition, indicating that there was little saturation in our sequences (P < 0.001).

In the ML searches, the combined dataset showed 1,922 distinct patterns with a proportion of gaps and undetermined characters of 17.14%.

The two Bayesian runs converged to stable likelihood values after 605,000 generations. Therefore 9,000 stationary trees (10%) from each analysis were used to compute a 50% majority rule consensus tree and to calculate posterior probabilities (PP). The consensus tree of the BI and the ML tree were congruent as far as the terminal clades or supported lineages are concerned.

The phylogeny showed an overall support for groups within the genus Russula previously recognised at section or subsection rank. The nomenclature of these infrageneric groups was assigned, based on the current taxonomic literature. All samples of tropical African species provisionally recognised as “Afrovirescentinae” clade were placed in a monophyletic lineage as sister to subsect. Virescentinae. Within this clade, sequences of the Beninese collections formed eight species-rank terminal clades and two additional singletons. The identity of three Beninese species was confirmed by similarity with newly-generated ITS sequences from the holotype collections: Russula hiemisilvae Buyck, Russula inflata Buyck and Russula sublaevis (Buyck) Buyck. The first species is represented by a single collection in the multigene phylogeny (Fig. 1), but by multiple collections in the ITS tree (Fig. 2). In addition, based on the holotype sequence, Russula intricata Buyck is recognised as a synonym of R. inflata. The identity of Russula carmesina R. Heim is verified, based on morphology after re-investigating of the holotype collection. Five other Beninese species are described as new in the present study: R. acrialbida sp. nov., R. beenkenii sp. nov., R. coronata sp. nov., R. florae sp. nov. and R. spectabilis sp. nov. One singleton from Benin represents an undescribed species and was labelled as “R. mollicula nom. prov.”. There are seven additional singletons in the multigene phylogeny representing African collections from outside Benin. The “Afrovirescentinae” clade included two already described subsections: Inflatinae Buyck and Pseudoepitheliosinae Buyck. Members of this clade were further grouped into two major clusters. One of them received support by both ML and BI and mostly contained species with small, thin-fleshed basidiomata with a high basidiospore ornamentation of 1.0–2.6 µm occurring in gallery forests, but also species with large and thick-fleshed basidiomata occurring in savannah woodlands. The second major cluster is supported by BI (ML = 67, BI = 0.97) only and includes species occurring in Sudanian savannah woodlands with large, thick-fleshed basidiomata and basidiospores with low basidiospore ornamentation of 0.1–0.5 µm. For Beninese collections, we observed that the species with small basidiomata almost always occur in gallery forests and species with large basidiomata exclusively occur in Sudanian savannah woodlands (Figs 1, 3). One species with basidiomata of intermediate size, R. inflata, occurred in both habitat types.

For the estimation of the species richness, distribution and ecological range of “Afrovirescentinae” species, the UNITE search applying the SH approach led to the analysis of 54 individual SH at 1.5% threshold which included 630 unique ITS sequences. The CC approach resulted in 453 retrieved ITS sequences from CC “Agaricomycetes | UCL10_011118”. The first approach yielded two species level clades, belonging to CC “Russulales | UCL10_026694”, which were not included in the dataset recovered by the CC approach. The second approach resulted in additional 17 SHs and 13 singletons not recovered by the SH approach and belonging to the “Afrovirescentinae” clade. The final alignment of 311 sequences for the ITS phylogeny (Fig. 2) had a length of 860 bp with 723 unique columns. The ITS tree had a topology similar to the multigene phylogeny, but with lower supports over the tree backbone. The “Afrovirescentinae ” clade was well supported as sister to subsect. Virescentinae. Compared to the multigene phylogeny, there was no support for the two major clusters within the “Afrovirescentinae” lineage. “Afrovirescentinae” include 55 supported species level clades and 39 singletons. Thus, we estimate a total diversity of approx. 94 species in the “Afrovirescentinae” clade, ten of which occur in Benin. Eight already described species names (Russula albofloccosa Buyck, R. hiemisilvae, R. inflata, R. intricata, Russula roseoalba Buyck, Russula roseovelata Buyck, Russula roseoviolacea Buyck, R. sublaevis) were recognised within the “Afrovirescentinae” clade by sequences from their holotypes. Russula albofloccosa and R. roseoviolacea were also represented by additional sequences retrieved from the UNITE database and originated from African countries other than Benin. The holotype sequences of R. roseoalba and R. roseovelata remained singletons. Three of the already-published species were confirmed to occur in Benin. Russula carmesina, R. hiemisilvae and R. inflata are newly recorded for West Africa. Russula sublaevis is reported for the first time from Benin. Thirty-seven of the supported species clades represented by more than one sequence are only known from a single country, mostly from Cameroon or Madagascar. Four species described in this study (R. acrialbida, R. coronata, R. florae, R. spectabilis) are only known from Benin. Russula albofloccosa, R. beenkenii, R. hiemisilvae, R. sublaevis and at least five unidentified species only known from sequence data are broadly distributed in sub-Saharan Africa. One terminal clade contains sequences of species originating from the Neotropics that were either identified as Russula brasiliensis Singer, Russula hygrophytica Pegler, Russula panamae Buyck & Ovrebo or Russula puiggarii (Speg.) Singer or remained unidentified. Besides this cluster, there are only two further singleton sequences of non-African origin from soil samples from Mexico and New Zealand.

According to data from ECM root tips, “Afrovirescentinae” species interact mutualistically with trees from the families Asteropeiaceae, Fabaceae (Detarioideae), Phyllanthaceae and Sarcolaenaceae in tropical Africa and Dipterocarpaceae, Fabaceae (Detarioideae, Papilionoideae), Nyctaginaceae and Polygonacae in the Neotropics. Several taxa, especially from Madagascar, are not specific regarding the family of their associated host tree.

Figure 2.

Phylogenetic Maximum Likelihood tree of Russula subgen. Heterophyllidiae, based on nrITS sequence data. Only bootstrap values ≥ 75 are displayed. Species described and illustrated in this study are highlighted with coloured boxes. Additional species level clades are represented by sequences predominantly retrieved from UNITE. Labels of sequences generated for this study are in bold. Font colour indicates the source of the sequence: black – basidioma, red – type material, blue – soil sample, green – mycorrhizal root tip.

Taxonomy

Russula acrialbida Manz, F. Hampe & Yorou, sp. nov.

MycoBank No: 856862

Figs 4, 5, 6

Holotype.

Benin. Collines, Toui, Forêt de Toui-Kilibo, co-ord. 8°36.4'N, 2°38.0'E, alt. 340 m, Sudanian woodland, under Isoberlinia doka, on sandy soil, 06.07.2021, leg. C. Manz, F. Hampe, N. S. Yorou & G. Abohoumbo, CM-21-093 (holotype B 70 0105401, isotype UNIPAR).

Additional material examined.

Benin. Borgou, Wari-Maro, Forêt de Wari-Maro, co-ord. 9°11.1'N, 2°12.8'E, alt. 280 m, Sudanian woodland, under I. doka, on sandy soil, 30.06 2021, leg. C. Manz, F. Hampe, D. Dongnima & S. Badou, CM-21-038 (paratype, B 70 0105402, UNIPAR); ibid. co-ord. 9°07.9'N, 2°07.5'E, alt. 340 m, Sudanian woodland, under I. doka, on sandy soil, 30.06 2021, leg. C. Manz, F. Hampe, D. Dongnima & S. Badou, CM-21-039 (paratype, B 70 0105403, UNIPAR); ibid. CM-21-041 (paratype, B 70 0105404, UNIPAR); ibid. co-ord. 9°11.0'N, 2°12.8'E, alt. 310 m, Sudanian woodland, under I. doka, on sandy soil, 25.06.2022, leg. C. Manz, F. Hampe, S. Sarawi, A. Rühl & D. Dongnima, CM-22-215 (paratype, B 70 0105405, UNIPAR); ibid. N’dail, Forêt de N’Dali, co-ord. 9°45.4'N, 2°40.1'E, alt. 360 m, Sudanian woodland, under I. doka, on sandy soil, 01.07.2021, leg. C. Manz, F. Hampe & G. Abohoumbo, CM-21-049 (paratype, B 70 0105406, UNIPAR); ibid. Kpéssou, Forêt de l’Ouémé Supérieur, co-ord. 09°15.8'N, 002°11.1'E, alt. 330 m, Sudanian woodland, under I. doka, on sandy soil, 02.07.2021, leg. C. Manz, F. Hampe, N. S. Yorou & G. Abohoumbo, CM-21-053 (paratype, B 70 0105407, UNIPAR); ibid. CM-21-054 (paratype, B 70 0105408, UNIPAR); ibid. CM-21-062 (paratype, B 70 0105409, UNIPAR); ibid. co-ord. 9°45.6'N, 2°8.0'E, alt. 320 m, Sudanian woodland, under I. doka, Isoberlinia tomentosa & Uapaca togoensis, 18.07.2021, leg. C. Manz, F. Hampe, N. S. Yorou & G. Abohoumbo, CM-21-138 (paratype, B 70 0105410, UNIPAR); ibid. Collines, Toui, Forêt de Toui, co-ord. 8°37.7'N, 2°35.6'E, alt. 320 m, Sudanian woodland, under I. doka, on sandy soil, 05.07.2021, leg. C. Manz, F. Hampe, N. S. Yorou & I. Oguchina, CM-21-083 (paratype, B 70 0105411, UNIPAR); ibid. Atakora, Natitingou, Kota waterfalls, co-ord. 10°12.7'N, 1°26.6'E, alt. 500 m, Sudanian woodland, under I. tomentosa, on rocky soil, 15.06.2022, leg. C. Manz, F. Hampe, S. Sarawi, A. Rühl & D. Dongnima, CM-22-175 (paratype, B 70 0105412, UNIPAR).

Diagnosis.

One of the most common Russula species in Sudanian woodlands in Benin, characterised by large white basidiomata, pileus surface with fine cream-coloured patches, burning acrid taste, small spores with a nearly smooth surface, a trichodermal pileipellis with attenuated hyphal terminations embedded in a gelatinous matrix and cystidia rapidly staining black in sulphovanillin, occurring in savannah woodlands. Differs from R. roseovelata by the absence of large detaching areolae on the pileus surface.

Description.

Growth habit: basidiomata solitary or in groups of up to ten. Pileus: mostly large to very large, rarely medium-sized, 35–155 mm in diam., when young, hemispherical, apically truncated or sometimes even slightly depressed, with margins touching the stipe, in shape similar to a matchstick head, later expanding plane, centrally depressed; margin involuted and slightly remaining so even when mature, distinctly tuberculate-striate up to 10–15 mm, frequently radially cracked, regular or slightly undulate; cuticle smooth, dry or slightly greasy when wet, finely areolate, patched towards the pileus margin, peelable up to ¹/₂–³/₄ of the pileus radius, colour white, yellowish-white (4A2), ivory (4B2) or cream (4A3), rarely with a faint pinkish-white (9A2) hue, near the centre also orange-white (5A2), patches near the margin grey orange (5B4-5), apricot yellow (5B6), pompeian yellow (5C6), cocoa brown (6E6), cognac brown (6E7) or rusty brown (6E8) on white background. Lamellae: 3–8 mm wide, 8–11 lamellae present along 1 cm near the pileus margin, narrowly adnate, first white, with maturity yellow white (4A2) to cream (4A3) coloured, with frequent furcations, especially near the stipe attachment, anastomoses absent, lamellulae usually absent, only few observed in one collection, edges entire, concolourous. Stipe: 55–110 × 15–35 mm, cylindrical or slightly tapering towards the base, frequently bumpy or with 2–4 irregular depressions or constrictions corresponding to the distinct chambers inside, smooth to slightly rugose or with slight longitudinal ridges, annulus absent, white; cottony stuffed, cavernate with 3–5 distinct chambers. Context: 3–10 mm thick at half pileus radius, white, parts damaged by insects turning brownish-orange, young firm, with maturity brittle, taste burning acrid after 2–3 seconds, odour inconspicuous or sometimes slightly fruity; macrochemical reactions: guaiac after 8–10 seconds strongly positive (+++) on both stipe and lamellae surfaces, FeSO₄ strong, deeply orange, sulphovanillin negative, sometimes bluish, KOH yellow on stipe surface and context, but negative on pileus surface, phenol negative. Spore print: cream (IIb-IIc).

Spores: (5.6–)6.2–6.6–7.1(–7.9) × (5.1–)5.4–5.7–6.1(–6.4) µm (n = 90), Q = (1.02–)1.1–1.15–1.21(–1.31), globose, subglobose to broadly ellipsoid, surface almost smooth, ornamentation very inconspicuous, composed of elements hardly visible under light microscope, ornamentation approx. 0.1 µm high as estimated by SEM, very dense weakly amyloid lines and warts forming a complete reticulum as observed by SEM, density of the individual elements not quantifiable by light microscopy; suprahilar plage small, inamyloid, covered by minute wrinkles visible only by SEM. Basidia: (31.5–)35.5–40–44.5(–52) × (7–)8–8.5–9(–10) µm (n = 61), clavate to broadly clavate, 4-spored; basidiola approx. 5–7 µm wide, cylindrical to clavate. Hymenial cystidia: on lamellae sides (56–)63–77–91(–117) × (7.5–)9–10–11(–14) µm (n = 60), moderately numerous, 1,200–1,400 cystidia/mm², predominantly clavate, sometimes subcylindrical, frequently with a slightly curved base or slightly flexuose, originating in subhymenium and somewhat protruding over basidia, thin-walled, apically obtuse, usually with a 1.5–11 µm long appendage; heteromorphous contents dense, crystalline, dispersed over the entire cell, rapidly turning to dark black in sulphovanillin. Hymenial cystidia near the lamellae edges distinctly shorter and slightly narrower, (31.5–)40.5–49–57.5(–76) × (6.5–)8–9–10(–12) µm (n = 60), cylindrical to broadly clavate, occasionally with a 1–7 µm long appendage; heteromorphous contents similar to the one in hymenial cystidia on lamellae sides. Lamellae edges: fertile, with equal representation of cystidia, basidia, basidiola and marginal cells. Marginal cells: (13.5–)16.5–19.5–23(–29.5) × (2.5–)3.5–4.5–5(–7) µm (n = 62), not well differentiated, cylindrical to subclavate, sometimes slightly bent or with a secondary septum, similar to basidiola, optically empty, thin-walled. Pileipellis: orthochromatic in Cresyl blue, sharply delimited from the underlying context, 230–300 µm deep; suprapellis a trichoderm, 50–80 µm deep, composed of erect hyphal terminations embedded in a gelatinous matrix; gradually passing to a 170–220 µm deep, strongly gelatinised subpellis of more or less parallel, moderately dense, intricate, 2.5–3.5 µm wide hyphae and abundant cystidioid hyphae. Acid resistant incrustations absent. Hyphal terminations: near the pileus margin composed of (1–)2–4 unbranched cells, originating in intricate hyphae of the subpellis, thin-walled or with slightly thickened walls (up to 0.5 µm), terminal cells (10.5–)23.5–33.5–43.5(–57) × (2.5–)3–3.5–4(–5) µm (n = 90), mainly attenuated, less frequently cylindrical, apically obtuse; subterminal cells mainly shorter, 3.5–7 µm wide, cylindrical or ellipsoid. Hyphal terminations near the pileus centre of (1–)2–5 unbranched cells, thin-walled, terminal cells of similar dimensions compared to the ones near the pileus margin (11.5–)17.5–28.5–39.5(–57.5) × 2.5–3.5–4(–5) µm (n = 91), cylindrical, attenuated or subulate, often more distinctly base-inflated, sometimes apically acute; subterminal cells distinctly shorter, 3–7 µm wide, ellipsoid to ovoid, forming chains before branching. Pileocystidia: near the pileus margin (23.5–)32–42.5–53(–86.5) × (5.5–)7.5–9–10.5(–14) µm (n = 60), one-celled, one two-celled cystidium observed in one collection, broadly clavate, fusiform or lanceolate, originating in the suprapellis, thin-walled, apically mainly obtuse, sometimes acute, usually occasionally with a 2.5–8 µm long appendage; heteromorphous contents dense, crystalline or banded, rapidly turning to black in sulphovanillin. Pileocystidia near the pileus centre distinctly shorter, (11.5–)20–32.5–45(–62) × (4.5–)6.5–8.5–10(–13.5) µm (n = 71), globose, ovoid, ellipsoid, fusiform, clavate or lanceolate; contents similar to the one of pileocystidia near the pileus margin. Context: with dispersed, but distinct cystidioid hyphae, approx. 6–9.5 µm wide, sparsely septate, branched, contents dispersed or sometimes almost homogenous; oleiferous hyphae absent.

Etymology.

Referring to the strongly acrid taste and white colour of the basidiomata.

Distribution and ecology.

Only known from Sudanian woodlands dominated by Isoberlinia doka in Benin.

Notes.

Russula acrialbida has spores with extremely low and fine ornamentation which is also present in R. roseovelata, a species with a pinkish pileipellis covered by large detaching rose-beige areolae (Fig. 7) and a pileus margin without striations (Buyck 1994). Microscopically, the latter species differs by slightly larger spores, hymenial cystidia that are up to 200 µm long and the presence of capitate terminal cells in the pileipellis with intracellular refringent transversal bands (Buyck 1994). Specimens very similar to R. acrialbida are illustrated and described in local African field guides as R. cf. roseovelata or R. roseovelata from Zambian miombo woodlands (Härkönen et al. 2015; Niemelä et al. 2021) or R. aff. roseovelata from Tanzania (Härkönen et al. 2003). It is likely that these records either represent R. acrialbida or a closely-related species, but probably not R. roseovelata s. str. because of the differences apparent in the field pictures. During our fieldwork in Benin, we encountered a man from the nearby village Wari-Maro collecting R. acrialbida as an edible fungus. According to him, the species would lose its acrid taste after soaking in an aqueous solution of baking soda overnight with subsequent boiling. Further studies are needed to confirm the suitability of the species for consumption as edible fungus. A similar-looking species from Tanzania identified as R. roseovelata is reported as edible after parboiling (Chelela et al. 2015). Russula albofloccosa, Russula ochrocephala Buyck and Russula terrena Buyck & Sharp are tropical African species with pileus colours similar to R. acrialbida. Russula albofloccosa is known from Burundi and DRC and can be distinguished from R. acrialbida by smaller, much more fragile basidiomata, a weak reaction to FeSO₄, a mild to weakly acrid taste and larger more elliptical spores with a more prominent ornamentation (Buyck 1994). Russula ochrocephala is known from DRC and Senegal and differs by spores with a more prominent ornamentation and an amyloid suprahilar plage (Buyck 1997). Russula terrena is known from Zimbabwe and differs by its smaller basidiomata, weak FeSO₄ reaction, more prominent spore ornamentation and cystidia that are insensitive to sulphovanillin (Buyck and Sharp 2007). To our knowledge, R. acrialbida is the only species with burning acrid taste in Russula sect. Heterophyllae Fr.

Figure 3.

Habitats with occurrence of ECM fungi in Benin A, C, E gallery forests with Berlinia grandiflora and Uapaca guineensis, B, D, F Sudanain savannah woodlands with Isoberlinia doka, Isoberlinia tomentosa, Uapaca togoensis and Monotes kerstingii.

Figure 4.

Russula acrialbida A–F field pictures of basidiomata: A CM-21-093, holotype, B CM-21-062, C CM-21-053, D CM-22-215, E CM-21-054, F CM-21-083; G, H scanning electron microscopical pictures of basidiospores (both CM-21-093, holotype). Scale bars: 5 cm (A–E); 2.5 cm (F); 3 µm (G, H).

Figure 5.

Hymenial elements of Russula acrialbida (holotype, CM-21-093) A basidia and basidiola, B hymenial cystidia near the lamellar edges, C marginal cells, D hymenial cystidia, E spores as seen in Melzer’s reagent. Scale bar:10 μm, but only 5 μm for spores.

Figure 6.

Elements of the pileipellis of Russula acrialbida (holotype, CM-21-093) A, B elements near the pileus centre: A hyphal terminations, B pileocystidia; C, D elements near the pileus margin: C hyphal terminations, D pileocystidia. Scale bar: 10 μm.