Molecular and morphological analyses confirm Rhizopogon verii as a widely distributed ectomycorrhizal false truffle in Europe, and its presence in South America

The genus Rhizopogon includes species with hypogeous or subepigeus habit, forming ectomycorrhizae with naturally occurring or planted pines (Pinaceae). Species of the genus Rhizopogon can be distinguished easily from the other hypogeous basidiomycetes by their lacunose gleba without columella and their smooth elliptical spores; however, the limit between species is not always easy to establish. Rhizopogon luteolus, the type species of the genus, has been considered one of the species that are more abundant in Europe, as well as it has been cited in pine plantation of North and South America, different parts of Africa, Australia, and New Zealand. However, in this study, based on molecular analyses of the ITS nuclear ribosomal DNA (nrDNA) sequences (19 new sequences; 37 sequences from GenBank/UNITE, including those from type specimens), we prove that many GenBank sequences under R. luteolus were misidentified and correspond to Rhizopogon verii, a species described from Tunisia. Also, we confirm that basidiomes and ectomycorrhizae recently collected in Germany under Pinus sylvestris, as well as specimens from South of Brazil under Pinus taeda belong to R. verii. Thanks to the numerous ectomycorrhizal tips collected in Germany, a complete description of R. verii/P. sylvestris ectomycorrhiza is provided. Moreover, since in this paper the presence of R. verii in South America is here reported for the first time, a short description of basidiomes collected in Brazil, compared with collections located in different European herbaria, is included.


Introduction
The species of the genus Rhizopogon Fr. belong to the order Boletales and suborder Suillineae in the Agaricomycetidae (Binder and Hibbett 2006). The genus is represented with over 100 species distributed worldwide (Smith and Zeller 1966;Martín 1996;Martín and García 2009). All species produce hypogeous or semi-hypogeous basidiomes and form ectomycorrhizae (EcM) with members of the Pinaceae (Pinus, Pseudotsuga, and Tsuga). Rhizopogon species are easy to cultivate in pure culture (Molina and Trappe 1994;Brundrett et al. 1996); thus, some were frequently applied to study physiology, morphology, or ecology of its ectomycorrhizae in the agroforestry systems (Smith and Zeller 1966;Hung and Trappe 1983;Chu-Chou and Grace 1984;Miller 1986;Molina et al. 1997;Beiler et al. 2010). Zeller and Dodge (1918) were the first authors to present a worldwide monograph of Rhizopogon. Later, Smith and Zeller (1966) produced the first modern account of the genus to North America including a total of 137 taxa, in which 128 were new for science. Since this paper, the Pacific Northwestern USA has been considered the greatest area of diversity of the genus (Hosford 1975;Molina et al. 1997;Grubisha et al. 2002), as well as other parts of the USA (Harrison and Smith 1968;Miller 1986). However, in posterior systematic studies undertaken in several part of the world, authors described new species in Mexico Guzmán 1971, Cázares et al. 1992), Tunisia (Pacioni 1984a), China (Liu 1985), Japan (Mujic et al. 2014), and different countries of Europe (Pacioni 1984b, Martín 1996, Martín and Calonge 2001; as well as new records, such as those of Mexico and Caribean countries (Hosford and Trappe 1980), Italy (Montecchi and Sarasini 2000), Japan (Hosford and Trappe 1988) and Spain (Martín and Calonge 2006), showing that the knowledge of the genus is not yet complete.
Nowadays, systematics and taxonomy of Rhizopogon have been under profound changes, mainly due to the use of molecular tools, specially using sequence-based analyses of the nuclear rDNA regions (nuc-ssu, nuc-lsu, ITS) and also mitochondrial genes (atp6, mt-lsu) (Grubisha 1998;Martín et al. 1998;Grubisha et al. 2002;Kretzer et al. 2003;Grubisha et al. 2005;Binder and Hibbett 2006;Martín and García 2009). According to Grubisha et al. (2002), the species are distributed in five subgenera: Amylopogon, Rhizopogon, Roseoli, Versicolores, and Villosuli. The species of the subgenus Rhizopogon have shown a combination of features, such as a simple peridium completely covered by rhizomorphs. Rhizopogon luteolus is the type species of the subgenus and it has been considered widely distributed in the Northern hemisphere.
Rhizopogon verii Pacioni (Pacioni 1984a) was described from Tunisia under Pinus pinaster. However, studies related to the systematic and distributions of R. verii are limited to only few collections from Italy, Spain, and Tunisia (Martín 1996). From other continents, R. verii has not been cited yet. Recent collections on an abandoned coal mine area near Crinitz (Brandenburg, Germany) on Pinus sylvestris could fit with R. verii, as well as the specimens collected during a survey of hypogeous fungi in State of Rio Grande do Sul (Brazil) growing under Pinus taeda.
Thus, with the opportunity to study new fresh specimens, the main objective of this paper was to clearly identify the specimens from Germany and Brazil using molecular analyses of ITS nrDNA sequences. This has allowed us also to confirm the presence of R. verii in these countries, as well as the EcM of R. verii on P. sylvestris. A detailed description is provided, both to the basidiomes and the EcM formed by R. verii/ P. sylvestris. Moreover, information to R. verii worldwide distribution in different native and pine plantation areas is provided.

Materials and methods
Specimens from Brazil were collected during mycological trips in the State of Rio Grande do Sul, close to the "Estação Ecológica do TAIM" in a sandy dune near to mature trees of P. taeda. In Germany, fresh basidiomes and soil cores to collect ectomycorrhizal tips were taken from the abandoned coal mine area along the side road toward Schlabendorfer See near the village Crinitz; the area is represented by a ca. 30-year-old P. sylvestris plantation established on silicate sandy neosol with shallow organic layer and poor understory vegetation. Data of new specimens and ectomycorrhiza collected for this paper are included in Table 1.

Morphological analyses
Fresh basidiomata were collected and analyzed macro-and microscopically following previously described methods (Miller and Miller 1988;Martín 1996), and compared with R. verii collections located at AQUI herbarium, including the type, as well as collections in BCN herbarium. Color codes followed Munsell Soil Color Charts (2009). Presentation of basidiospore data follows the methodology proposed by Tulloss et al. (1992), slightly modified by Wartchow (2012) and Wartchow et al. (2012). Abbreviations include L(W) =average basidiospore length (width), Q = the length to width ratio range as determined from all measured basidiospores, and Q m = the Q value averaged from all basidiospores measured. Herbarium abbreviations follow those of the online version of Thiers [continuously updated]. Specimens are deposited in UFRN, URM and LJF herbaria.
Soil was gently washed from ectomycorrhizae (EcM) under binocular using forceps and brush, and subsequently EcM were stored in 2 % glutaraldehyde in 0.1 M sodium cacodylate buffer (pH 7.2) at room temperature. For semi-thin sections of mycorrhizae, six washes (10 min each) in 0.1 M sodium cacodylate buffer were performed. Samples were postfixed in 1 % osmium tetroxide in the same buffer for 1 h in the dark under room temperature. After six washes with distilled water, samples were dehydrated in acetone (25, 50, 70, and 95 %, for 15 min each) and three times in 100 % acetone for 1 h. The mycorrhizal tips were embedded in Spurr's plastic (Spurr 1969) and sectioned with a diamond knife on an Ultracut Reichert Ultramicrotome (W. Reichert-LABTAC, Wolfratshausen, Germany). The sections (0.5 μm thin) were stained with crystal violet. Twenty mycorrhizal tips were investigated by the use of a light microscope (Axioscop 50, Zeiss, Oberkochen, Germany).

DNA extraction, amplification, and sequencing
Total genomic DNA was extracted from the gleba of airdried basidiomes or from stored ectomycorrhizal root tips (5-10 tips from the same cluster per extraction) by using a Plant DNeasy Mini Kit (Qiagen, Hilden, Germany). Extracted DNA was resuspended in pre-warmed, sterile Milli-Q water to the approximate final concentration of 100 ng μl −1 and kept at −80°C. Primer pair ITS1F (Gardes & Bruns 1993) and ITS4 (White et al. 1990) was used for PCR amplification of the complete nuclear ITS region. Amplification reactions were performed in a PE 9700 DNA thermocycler, with an annealing temperature of 55°C. Negative controls, lacking fungal DNA, were run for each experiment to check for any contamination. Amplified DNA was separated and analyzed as described in Grebenc et al. (2009).
Amplified DNA fragments were first separated and purified from the agarose gel using the Wizard SV Gel and PCR Clean-Up System (Promega Corporation, Madison, WI, USA) and sent to Macrogen Korea (Seoul, Korea) for sequencing. Sequencher 5.1 (Gene Codes Corporations, Ann Arbor, MI, USA) was used to identify the consensus sequence from the two strands of each isolate.

Molecular analyses
Preliminary identification of the new sequences obtained were done through UNITE database (http://unite.ut.ee) species hypothesis (SH) search (Kõljalg et al. 2013). The PlutoF multiple sequence alignments obtained in UNITE were merged and manually adjusted using Se-Al v.2.0a11 (Rambaut 2002). The sequence AF062933 of Rhizopogon succosus A.H. Sm. was chosen as outgroup, since it is one of the few sequences available of subgen. Roseoli Fr. with voucher collection, excluding the sequences of R. luteolus and R. verii.
Analyses were conducted using parsimony and Bayesian inference. In the parsimony analyses, nucleotide characters were treated as unordered and all changes were equally weighted; gaps were treated as missing data. Searches for most parsimonious (MP) trees were performed using a two-stage strategy with PAUP* v.4.0b10 (Swofford 2002). First, the analyses involved 10,000 replicates with stepwise random taxon addition, tree bisection-reconnection (TBR) branch swapping saving no more than 10 trees per replicate, and MULTREES option off. The second round of analyses was performed on all trees in memory with the same settings except the MULTREES option on. Both stages were conducted to completion or until one million trees were found. Relative support for clades was inferred by nonparametric bootstrapping (Felsenstein 1985) as implemented in PAUP* using 500 pseudoreplicates, each with 20 random sequence addition cycles, TBR branch swapping, and MULTREES option off (DeBry and Olmstead 2000). To the Bayesian analyses, the program MrModeltest v.2.3 (Nylander 2004) was used to determine the model of sequence evolution that fits best the dataset. The Hasegawa-Kishino-Yano (Hasegawa et al. 1985) of DNA substitution, with rate variation among nucleotides following a discrete gamma distribution (HKY + G), was selected as the best-fit by both the hierarchical likelihood ratio test (hLRT) and Akaike information criterion (AIC). Bayesian phylogenetic inferences were performed using MrBayes v.3.2.2 (Ronquist et al. 2012) run on the CIPRES Science Gateway (Miller et al. 2010). Two runs starting from random trees were carried out using the HKY + G substitution model. All model parameters were treated as unknown variables with uniform prior probabilities and were estimated as part of the analysis together with tree topologies. Metropolis-coupled Markov chain Monte Carlo algorithm was used with eight simultaneous chains for each run, set at two million generations, and sampled every 1000 generations. Of the 40,002 trees obtained, the first 25 % were discarded as burn-in; the 50 % majority-rule consensus tree and the Bayesian posterior probabilities (PP) were obtained in MrBayes from the remaining 30, 002 trees.

Molecular analyses
The matrix contained the 19 sequences obtained in this study ( Table 1) and sequences of the species hypothesis groups SH5_008910 and SH5_008911 obtained through UNITE search (Table 2: SH5_008910, clade A and C; SH5_008911, clade B). After manual adjustment, the matrix had 749 characters, 95 of them variable and 23 parsimony-informative that produced >1,000,000 MP trees, 107 steps in length. There was a consistency index of 0.953 and a retention index of 0.941. The harmonic mean of the estimated marginal likelihoods from the Bayesian analysis was −ln = 1727.88. The MP and Bayesian analyses produced trees of identical topology (Fig. 1), representing the Bayesian Majority Rule Consensus tree with the PP and Bootstrap values on the branches.
Including R. succosus as outgroup, sequences are distributed in three highly supported clades. The clade A (bs = 92 %, pp = 1.0) grouped three sequences from Japan and South Korea, collected under Pinus densiflora and Pinus thunbergii from unidentified collections (both basidiomata and ECM). The clade B (bs = 93 %, pp = 1.0) included three sequences, two from Estonia and the sequence from the neotype of R. luteolus from Uppsala (Sweden) [designated in Martín (1996)], all under Pinus species; this R. luteolus clade is the sister group of the clade C (bs = 84 %, pp = 1.0) that grouped 47 sequences, including the sequence of the type of R. verii, a species described from Tunisia under P. pinaster, eight sequences identified as R. luteolus collected under different Pinus species (mainly P. pinaster and P. sylvestris), from Europe and New Zealand, and many sequences from
Specimens examined: The data of the specimens examined are included in Table 1. Brazilian collections are located at the herbarium of the Universidade Federal do Rio Grande do Norte of (UFRN) and Universidade Federal de Pernambuco (URM), and the German specimens at the Mycotheca and Herbarium of Slovenian Forestry University (LJF).

Discussion
Rhizopogon verii was originally described from Tunisia (Pacioni 1984a), under P. pinaster. Since Pacioni's discovery, little was published related to these species; however, in the past few years, new samples were collected in Spain (Martín 1996). With the data obtained in our study, a general distribution of R. verii is shown, both in natural and planted pine forests, and it is expected that this species can be found in other countries where pine plantations were established using European seedling material.
Based on morphological data, Martín (1996) considered that in Europe, the wider distributed Rhizopogon species with luteolus-type peridium was R. luteolus. However, the present study combining basidiomata and EcM anatomy, together with molecular analyses, shows that the species R. verii is well defined and commonly present in Europe. Distribution of R. verii on other continents is fairly unknown, but the ecology of known sites indicates several similarities. Mineral and sandy soil requirements (Table 1) with Pinaceae for Rhizopogon were recorded for Europe and Africa . Similarly the Brazilian specimens were gathered on sandy soil in natural sand dune plots, at the base of a P. taeda site in the Campos Sulinos (or Pampa) biome, especially covered by open grassy formations used as natural pastures (Overbeck et al. 2007;Fiaschi and Pirani 2009). In a floristic study for Brazil, Porto and Dillenburg (1986) reported that the indigenous vegetation is composed by members of Bignoniaceae, Cactaceae, Euphorbiaceae, Fabaceae (subfam. Caesalpinoideae and Faboideae), Myrtaceae, Nyctaginaceae, Rubiaceae, and Sapotaceae, but the presence of exotic tree species, such as Pinus or Eucalyptus, was required. Rhizopogon data from the tropical and subtropical region are rarely available, limiting the knowledge about the identification and phylogenetic placement of those fungi. In Brazil, the genus was introduced through seedlings of exotic Pinus spp. (Sulzbacher et al. 2013) in the southern and southeast region (Giachini et al. 2000;Baseia and Milanez 2002;Giachini et al. 2004;Sobestiansky 2005;Cortez et al. 2011). Neves and Capelari (2007)  The studied R. verii Brazilian basidiomata covered different developmental stages (Fig. 2a-c), thus providing Fig. 6 Anatomy of Rhizopogon verii ectomycorrhiza emanating elements. a Cystidia in the form of a normal hypha but twisted (type L). b Rhizomorphs differentiated with thick central hyphae and complete septa (type E). Scale bars represent 20 μm (a) and 10 μm (b) more information related to the basidiome morphology. The Brazilian collection exhibits basidiomata very similar to that illustrated by Pacioni (1988): subglobose to irregular basidiomata, covered by reddish yellow rhizomorphs and with an olive brown gleba. Rhizopogon verii shared several features with the widespread R. luteolus (Martín 1996), for example, the basidiome shapes, rhizomorphs covering the whole peridium surface, also the shape and size of basidiospores and the luteolus-type peridium. However, R. luteolus has a clavate to cylindrical basidia, with thin wall, and R. verii has mostly lageniform basidia, with a thick-walled ventricose base up to 1.5 μm diam. as described in Martín (1996). This morphological distinction between R. luteolus and R. verii is well supported by phylogenetic species delimitation using nrDNA ITS spacer molecular characterization which separated these two morphological groups in two distinct terminal clades (Fig. 1).
T h e c o m p r e h e n s i v e d e s c r i p t i o n o f R . v e r i i ectomycorrhiza on P. sylvestris is provided for the first time. In comparison to other described ectomycorrhizae from the genus Rhizopogon (www.deemy.de; Mohan et al. 1993), ectomycorrhiza of R. verii can be easily distinguished at least by the plectenchymatous mantle type B bearing some inflated cells at proximal end next to septae, the presence of rhizomorphs type E, cystidia type L, and loose hyphal net covering ectomycorrhiza. On the other hand, R. luteolus ectomycorrhiza showed type E of outer mantle layers, no emanating hyphae of cystidia and type F rhizomorphs (Uhl 1988). A more distant related R. roseolus ectomycorrhiza showed outer mantle type C and distinct reddish to whitish color of vital ECM tips  and Rhizopogon melanogastroides with the same mantle type but ectomycorrhiza color similar yellowish to R. verii . Morphological ECM characters of R. verii fit well to some previous observations by Agerer (2006) who noted that Rhizopogon has one of the most advanced rhizomorphtype structure (the boletoid rhizomorphs).
The combination of molecular analysis and morphological analyses of sporocarps and ectomycorrhiza supports the separation of R. verii from other Rhizopogon species. We also confirmed that R. verii has global distribution, most likely to originate from Europe but being introduced to all continents, with most recent discovery in South America, namely from pine plantation in Brazil and any further exploitations of the species globally would contribute valuable information to its distribution and ecology. As mentioned, in Tedersoo et al. (2010) and Tedersoo and Smith (2013), some lineages, such as the genus Rhizopogon, are restricted to this host family, hence explaining their geographical distribution.