Novel Laccaria Species From Juglandaceae Forest in Panama With Notes on Their Ecology

Since 2013 there have been 22 new species of Laccaria described worldwide. Only three of these represent species from the neotropics. In Panama, Laccaria is abundant in monodominant Oreomunnea mexicana (Juglandaceae) forests based on sporocarps and environmental sequencing of roots. This study uses a combination of morphological and phylogenetic evidence to document up to seven species of Laccaria from these forests, one previously described, three described as new, and three requiring more data before being formally described. Molecular data used for phylogenetic analysis include the nuclear ribosomal ITS and 28S regions, along with TEF1 and RPB2. Laccaria stellata, has previously been reported from O. mexicana cloud forests of Panama. Laccaria dallingii sp. nov., L. nitrophila sp. nov., and L. fortunensis sp. nov. are described as new based on morphology and phylogenetic analysis of multiple collections. A taxon referred to as “PAN sp3” is resolved sister to L. stellata. Phylogenetic analysis also resolved two separate clades of Panamanian Laccaria as sister to L. roseoalbescens, a species previously described from Mexico. These three taxa are not described in this paper as there is too little material from which to make effective morphological descriptions even though their placement in phylogenetic analysis identify them as being unique. Ecologically, all described species except for L. fortunensis were amplified from O. mexicana ectomycorrhizal root tips. L. nitrophila was one of the most recovered species from the roots of O. mexicana in a previous study, and it has been shown to respond positively to long term nitrogen addition. Our results expand the knowledge of Laccaria diversity for Central America and highlight that at least some species of Laccaria are nitrophilic in neotropical Juglandaceae forests as well as in temperate forests.

The relatively low number of newly described Laccaria species in the Neotropics compared to Asia may be indicative of the difficulty of morphological species recognition within Laccaria, as cryptic species are considered the norm (Vincenot et al., 2017;Wilson et al., 2017b). Thus, the use of DNA sequence data is seen as critical for the recognition of Laccaria species. Montoya et al. (2015) described L. roseoalbescens from Quercus dominated forest in Mexico using a combination of morphological data and nuclear ribosomal internal transcribed spacer region (ITS) and the 5 end of the nuclear ribosomal large subunit (28S).
The following year, Popa et al. (2016) described L. stellata from Quercus/Oreomunnea dominated forest in Panama based on morphology and ITS sequence data. Lastly, Ramos et al. (2017) combined morphology and ITS and 28S sequence data to identify and describe L. squarrosa from forests of endangered Fagus grandifolia var. mexicana in Mexico. Mueller and Halling (1995) surveyed the macrofungi of Costa Rican Quercus montane forests. Their study documented seven species of Laccaria: L. amethystina, L. gomezii, L. laccata, L. major nom prov., L. ohiensis, L. proxima, and L. trichodermophora (see Table 1 in Mueller and Halling, 1995). Although these species have yet to be thoroughly studied using molecular data, this brings the total number of neotropical and subtropical Laccaria species to ten.
In montane tropical forest of western Panama, the ECM host tree Oreomunnea mexicana (Juglandaceae) forms monodominant stands and associates with a highly diverse community of ECM fungi (Corrales et al., 2016). Several inventories employing environmental sequencing of roots in these forests indicate that Laccaria is a dominant genus of the fungal community (Corrales et al., 2016(Corrales et al., , 2017Corrales and Ovrebo, 2020). However, despite all the sequencing efforts, much of the reported Laccaria diversity from these forests remains unidentified and it is apparent that this region has a number of undescribed species. Based on previous research in neotropical monodominant forest of Guyana, as many as 70% of the ECM fungal species in a newly explored area may be undescribed (Henkel et al., 2012). We have observed this same pattern in the Oreomunnea monodominant forest where many ECM fungal species appear to be new (Corrales and Ovrebo, 2020).
This study adds to the known diversity of neotropical species of Laccaria by using a combination of ITS, 28S, RPB2 and EF1α sequence data, along with macro-and micro-morphological characters. Phylogenetic analysis used the most comprehensive dataset available for each molecular marker. In addition, ITS sequences of species of Laccaria growing on root tips of neotropical plant hosts were evaluated to better estimate the diversity of neotropical Laccaria, and evaluate how much diversity remains undescribed.

MATERIALS AND METHODS
Specimens were collected in a primary lower montane forest (1000-1400 m.a.s.l) at the Fortuna Forest Reserve in western Panama. The mean annual temperature at the reserve ranges from 19 to 22 • C and the annual rainfall varies from ca. 5800-9000 mm (Cavelier et al., 1996;Andersen et al., 2012). Trips to the study area were made between April 2012 and May 2015 to collect fleshy macrofungi of Agaricomycetes and Pezizomycotina. Specimens were collected in forests dominated by the ECM host tree Oreomunnea mexicana with occasional Quercus species. Other ECM host species, i.e., Coccoloba spp. and Alfaroa costaricensis, were also present in the study area at low abundances (Corrales et al., 2016(Corrales et al., , 2017 Corrales et al., 2016 for detail information about the sites). Additionally, specimens were collected in a long-term N addition experiment in the Honda watershed that has been ongoing since 2006 (Corre et al., 2010;Corrales et al., 2017).
Macromorphological features like color, size, shape and surface features of the pileus, lamellae and stipes were recorded immediately after returning from the field. Sporocarps were then dried for later analysis of micromorphological features. Colors were described using alphanumeric codes from the Methuen Handbook of Color (Kornerup and Wanscher, 1993) in the format of page number, column letter, row number.
When possible, micromorphological measurements were taken from multiple samples. Measurements separated by an "×" indicate length by width aspects, respectively. Individual extremes, or outliers, from this range are given in parentheses. For each description the symbol "x" describes the range of specimen means. For basidiospores, the Q value indicates a long axis and short axis ratio although this is not necessarily a length and width ratio because of the difficulty of observing globose spores in a profile dimension. Basidiospore length and width measurements do not include ornamentation (echinulae). An important diagnostic feature of Laccaria species is the length and the width at the base of the basidiospore ornamentation. Species with an average spine length (x) less than 1.5 µm are described as having echinulate spores. Spores with mean spine length x > 1.5 µm are described as echinate.
Total DNA genomic content was extracted from fresh sporocarp tissue stored in CTAB 2X using REDExtract-N-Amp Plant PCR Kit following the manufacturer's instructions (Sigma-Aldrich). The following primers were used for PCR and cycle sequencing: ITS1F (Gardes and Bruns, 1993) and ITS4 (White et al., 1990) for the ITS region; LR0R and LR5 (Vilgalys and Hester, 1990) for 28S region domains 1-3; and fRPB2-6F (Liu et al., 1999) and bRPB2-7R (Matheny, 2005) for the RPB2 regions 5-7. PCR conditions consisted of 95 • C for 1 min, and then 35 cycles of 95 • C for 30 s, annealing temperature for 30 s to 1 min, and 72 • C for 1 min, with a final extension time at 72 • C for 10 min. Annealing temperatures were 50 • C for ITS and 58 • C for RPB2. Also EF1-983F, EF1-2218R, and EFcf (Rehner and Buckley, 2005) were used for the EF1α. PCR for EF1α was performed using a touchdown PCR procedure. PCR conditions consisted of 94 • C for 2 min, then the initial annealing temperature was 66 • C, and was subsequently incrementally reduced by 1 • C per cycle over the next 9 cycles, then 36 cycles of 94 • C for 30 s, 56 • C for 30 s, 72 • C for 1 min, with a final extension time at 72 • C for 10 min (Rehner and Buckley, 2005). PCR amplicons were visualized on 1.5% agarose gel stained with SYBR Green. Positive products were cleaned using ExoSap-IT (Affymetrix, Santa Clara, CA, United States) and sequenced. Root tips from adults, saplings, and seedlings of Oreomunnea mexicana collected from the sampling sites (Corrales et al., 2016) that were processed using the same extraction protocols and ITS primers were included in the analysis to inform the ecology of the species.
Newly generated sequences were edited using Codon Code ALIGNER v.3.5.7 (CodonCode Corporation, Dedham, MA, United States) 1 with generic-level identities for sequences confirmed via BLAST queries of GenBank 2 . Nucleotide datasets were assembled using a combination of new sequences and those derived from GenBank. Datasets were aligned using MUSCLE v.3.8.31 (Edgar, 2004) with default settings, followed by manual alignment using MESQUITE v.2.75 (Maddison and Maddison, 2009).
Systematic datasets were individually assembled for ITS, 28S, RPB2, and EF1α sequences using representative sequences of Laccaria species from all over the world in order to delimit phylogenetic species from the neotropics. Separate maximum likelihood analysis was performed on each gene region to check for systematic conflicts between the different molecular datasets. Once a lack of conflict was determined, all available sequences were put in to a supermatrix for combined analysis. All analyses were implemented on the CIPRES web portal (The CIPRES Portals 2009) 3 . Maximum likelihood bootstrap (ML) analyses were performed using RAxML v.2.2.3 (Stamatakis, 2006). Rapid bootstrapping analysis was performed for total of 1000 bootstrap replicates under default parameters set to GTR + G. This model is sufficient for the purposes of this study as it will produce a reasonable phylogeny for evaluating species relationships in our samples, and approximate their systematic relationship to other Laccaria species. Bootstrap support from maximum likelihood analysis ≥80% is reported on the branches of phylogenies. Maximum likelihood bootstrap support ≥90% is considered as "strong" support. Bayesian Metropolis-coupled Markov chain Monte Carlo (MCMC) analyses were performed using the GTR + G model of evolution in mrbayes v.3.1.2 (Ronquist et al., 2012). A test was performed on the combined dataset to identify the optimal model for assessing nucleotide substitution rate using jModelTest (Darriba et al., 2012). The selected model TIM2 + I + G was used in a maximum likelihood analysis using RAxML-NG (Kozlov et al., 2019). The resulting phylogenetic tree was nigh identical to the one produced using RAxML under GTR + G. All further results discussed reflect those from the latter analysis. The analyses used four chains and sampling every 100th tree for 10 million generations. The burnin to be removed was determined using LogCombiner v.1.8.0 (Drummond and Rambaut, 2007) either by taking the first 10% of the iterations, or, if convergence around a stable average likelihood involved >10% of the iterations, by simply removing this proportion of trees from the analysis. Bayesian posterior probabilities (PP) ≥ 0.95 are reported with MLB support on the branches of the phylogenies, with PP ≥ 0.98 considered "strong" support.

RESULTS
Phylogenetic analysis of combined and individual ITS, 28S, RPB2, and EF1α dataset identify and lend support for multiple species of Laccaria occurring in Panama, Costa Rica, and tropical Mexico. The supermatrix dataset used to phylogenetically delimit species of Laccaria from Panama consisted of sequence data from 152 specimens ( Figure 1A). The supermatrix dataset combines ITS, 28S, RPB2, and EF1α sequences, each contributing 679, 734, 1073, and 1008 base pairs, respectively for a total of 3494 bp in the combined dataset. The number of polymorphic sites in the dataset was 1168/3494 (33%). For the Bayesian analysis, the average effective sample size (ESS) is 2143.37 using a 10% burnin. The number of specimens represented by each sequence region includes 149 for ITS, 110 for 28S, 83 for RPB2, and 58 for EF1α. The results of the separate analyses, as well as a table for all sequence data used and generated from this study can be found in the Supplementary Materials.
The phylogeny from the combined datasets is represented in Figure 1A. This figure identifies 15 putative species-rank lineages of Laccaria from Mexico, Costa Rica, and Panama. Up to seven Laccaria species from Oreomunnea mexicana forests of Fortuna, Panama were resolved by these analyses, with L. stellata as the only one previously reported from these forests. The other species include L. nitrophila sp. nov. (71% ML and 0.86 PP), L. dallingii sp. nov. (100% ML and 1.0 PP), and L. fortunensis sp. nov. (100% and 1.0 PP). These species are formally described within this paper. The other three potential species (L. aff. roseoalbescens 1 and 2 and Laccaria "PAN3") ( Figures 1A,B), are not described or proposed in this study due to the few observed specimens and limited material available to assess their morphological and ecological features.
Another ITS dataset consisting of 108 ITS sequences was created to phylogenetically identify Laccaria species occurring in ECM roots of Oreomunnea mexicana. This dataset contained ITS sequence data from 16 root tip samples. All 16 ECM sequences resolved with at least one sequence from a vouchered specimen of Laccaria ( Figure 1B). The majority of root tips FIGURE 1 | (A) Maximum likelihood phylogeny of northern hemisphere Laccaria using combined ITS, 28S, RPB2, and EF1α dataset. (B) Maximum likelihood phylogeny using ITS sequence data to identify species of Laccaria growing on roots of Oreomunnea mexicana. Root tip samples are in green. Numbers above branches represent maximum likelihood bootstrap percentages (in bold) followed by Bayesian posterior probabilities. Gray arrows identify additional species of neotropical Laccaria not fully addressed here or in previous systematic studies.
resolved with either L. dallingii or within the L. roseoalbescens complex. At least two ECM sequences resolved with L. nitrophila, L. stellata, or with a specimen identified as L. aff. ohiensis ( Figure 1B). No ECM root tips were recovered representing L. fortunensis.

Etymology
This species is named in honor of ecologist Professor James W. Dalling who has dedicated a significant part of his career to study the neotropical montane forests of the Fortuna Forest Reserve. Professor Dalling was the Ph.D. advisor to lead author A. Corrales.

Diagnosis
Distinguished from other Panamanian Laccaria by basidiomes that are purple overall and the presence of cheilocystidia.

Habitat and Phenology
Solitary, scattered, gregarious or caespitose, on wood, soil or leaf litter, two collections on moss covered log. Growing in Oreomunnea mexicana monodominant forest and recovered from ECM root tips of O. mexicana. Found fruiting during January -February and July thorough December.

Notes
The new species Laccaria dallingii is clearly the most striking of this study. As part of the L. amethystina complex, it has a deep purple coloration that helps it to stand apart from the other Laccaria species in these forests. As discussed in Vincenot et al. (2017) the L. amethystina complex consists of several, geographically separate but morphologically similar species and several species have been formally described within the group. Thus, it is not surprising to recover a distinct taxon in theses Oreomunnea mexicana monodominant forests. However, there is currently at least two other purple Laccaria occurring in neotropical montane forests, L. gomezii and a taxon awaiting formal designation recovered from Costa Rica and Colombia ( Figure 1A, specimens F1102432, F1112018, and GMM7167). Laccaria gomezii is morphologically distinct and easily distinguished based on its close, decurrent lamellae and subglobose basidiospores. While L. dallingii forms a clade distinct from the aforementioned unidentified Costa Rican and Colombian taxa in the L. amethystina complex (see above), they are still very similar morphologically making it difficult to distinguish them from each other. Based on current distribution data for the two taxa, they can be differentiated by geography, but more survey work throughout the region is needed to confirm this. Further work on the American (North, Central, and South American) members of the complex is needed. Etymology nitr (L., reference to nitrogen), phil (Gr., reference to love), this species prefers soil habitats with high inorganic nitrogen as it was the most recovered OTU in the roots of Oreomunnea mexicana collected from a long-term nitrogen addition experiment at the Fortuna Forest Reserve.

Notes
Laccaria nitrophila has the least amount of statistical phylogenetic support from ML or Bayesian analysis. This is due in part to the over reliance on ITS data, since there are not sequence data available from other markers for more than one specimen. For these ITS sequences there is high variability in coverage among the sequences available as the quality of the ITS sequences obtained from these specimens was low. As a result, many of the sequences needed to be trimmed excessively, making them short and therefore producing inconsistent overlap between sequences for comparison. The sequence overlap in ITS data matrix provided just enough comparison to allow these specimens to be clustered in phylogenetic analysis, but was limited enough that the bootstrap analysis and Bayesian posterior probabilities turned out lower than expected for intraspecific comparisons. Morphologically, L. nitrophila is part of the L. ohiensis complex along with Laccaria stellata F. Popa & S. Y. Castillo. Species in the complex are relatively small, have orange brown translucent striate pilei, distant thick lamellae, and globose basidiospores with prominent echinulae. Because of the slight variations among these characters, field identification of the different species is very difficult. Laccaria stellata which co-occurs with L. nitrophila has red and pink colorations and slightly smaller basidiospores with longer echinulae along with a reported presence of pleuro-and cheilocystidia. However, it is important to note that the descriptions of these cystidia in the original publication by Popa et al. (2016) do not effectively differentiate these cells from basidioles (1). We were able to detect cystida-like cells in the hymenium of L. stellata specimens collected in this study (A. Corralles 576).
Laccaria squarrosa Mandala, Montoya & Ramos and L. roseoalbescens T. J. Baroni, Montoya & Bandala are both described from neotropical forests of Mexico. The former has distinct squarrose hairs along the stipe. This species also has cheilocystida which are not observed in Laccaria nitrophila. Laccaria roseoalbescens is distinctly paler and pinkish, fading to yellow relative to the orange-brown-beige of L. nitrophila. There are also "cystidia-like" sterile elements within the hymenium of L. roseoalbescens.

Diagnosis
Characterized by the rather small basidiomes, tan pileus color, more or less globose basidiospores with short and narrow echinulae, neotropical distribution and sequence data.

Notes
Laccaria fortunensis is characterized by the small, thin-fleshed basidiomes and by the phylogenetic placement. In this study it is distinguished from L. nitrophila by having smaller, finely ornamented basidiospores.
The coloration for Laccaria fortunensis lies somewhere between L. nitrophila and L. roseoalbescens in being paler than the former, but darker brown than the latter. The fine ornamentation on globose basidiospores is diagnostic and not seen in other described Neotropical species of Laccaria.

DISCUSSION
Phylogenetic analysis (Figure 1) resolves up to 15 lineages of Laccaria within the neotropics. These include three species recently described using phylogenetic analysis: L. stellata, L. roseoalbescens, and L. squarrosa (Montoya et al., 2015;Popa et al., 2016;Ramos et al., 2017). Of the remaining twelve species, five of these represent species not recovered in the forests of Fortuna, Panama (gray horizontal arrows in Figure 1A). Many of these represent Laccaria from Costa Rica collected from montane Quercus forests. These species represent systematically distinct lineages. Two of them resolve within the Laccaria bicolor complex, one resolves within the Laccaria amethystina complex, and another two fall within the broadly defined L. ohiensis complex.
There are two sporocarp collections as well as two ECM sequences that resolve within the L. stellata clade (Figures 1A,B). The lack of obvious morphological differences between L. stellata and L. nitrophila makes identifying these sympatric species without molecular data challenging. Similarly, there are no clear morphological features differentiating L. dallingii from others in the L. amethystina clade. This highlights the need for using molecular data to differentiate Laccaria species within complexes.
One of the more interesting results of this study is the observation of a new species complex centered around L. roseoalbescens. This species was originally described from the cloud forests of Mexico (Montoya et al., 2015). In Panama, there are two taxa sister to L. roseoalbescens in the main phylogeny ( Figure 1A). The specimen Ovrebo 5332 is directly sister to L. roseoalbescens, along with Costa Rican specimen GMM6016. These specimens form a clade and are tentatively identified as L. aff. roseoalbescens 1 in the phylogeny, and together with L. roseoalbescens form a clade that is supported with 95% MP and 1.0 PP. Until more specimens and sequence data are available to evaluate this potential species, we are content to leave it undescribed. Sister to this group is another putative species named L. aff. roseoalbescens 2. This is represented by two collections: Corrales 70 and Corrales 274. Unfortunately, these consist of only one and two sporocarps each. As a result, there is insufficient material to properly describe this species morphologically. Altogether this Laccaria roseoalbescens complex is supported with 100% ML and 1.0 PP (Figure 1A).
Based on sequences from root tips, we were able to confirm Oreomunnea mexicana as the host species for L. dallingii, L. nitrophila, and two species in the L. roseoalbescens complex. Laccaria fortunensis was not recovered from O. mexicana root tips due to the lack of root tip inventories in the locality were this species was collected. While we expect L. fortunensis to be associated with the same host plant, ECM associations with Quercus in the region is also a possibility. Laccaria has been reported as a nitrophilic genus due to the positive response to soil nitrogen availability by some of its species in terms of both sporocarp production and colonization of root tips in temperate and boreal ecosystems (Lilleskov et al., 2011;van der Linde et al., 2018). In tropical Oreomunnea mexicana monodominant forest of the Fortuna Forest Reserve, Laccaria has also been reported to be more abundant in sites with high natural soil fertility that in sites with low nitrogen availability (Corrales et al., 2016). Also, in a long term nitrogen addition experiment at the same study site (Corre et al., 2010), the relative abundance of Laccaria increased from 1% in the control plots to 10% in the nitrogen fertilized plots (1 OTU showed a negative response while 5 OTUs responded positively) in an study using highthroughput amplicon sequencing to characterize the ECM fungal communities associated with the roots of O. mexicana individuals (Corrales et al., 2017). L. nitrophila, described in this paper, is the species that was reported by Corrales et al., 2017, as the Laccaria OTU with the most positive response to inorganic nitrogen addition.
Neotropical forests of Central America host a number of Laccaria species that remain to be identified. Many of these are likely to be new, but require additional material to be adequately analyzed and described. While sufficient specimens already exist for some of the suspected species, these specimens require additional study. Altogether, this study suggests that there are as many as 15 species of Laccaria in the neotropical forests of Central America, and it is quite probable that the diversity is even higher. These results increase the number of suspected species in the region by 50% and doubles the number of systematically described Laccaria from this part of the world.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are openly available in GenBank of NCBI under the accession numbers provided in the article.

AUTHOR CONTRIBUTIONS
AC, CO, and AW conceived the idea, performed the research, and collected the data. AW analyzed the molecular data. AW, AC, and CO wrote the first draft of the manuscript. All the authors contributed substantially to the revisions.