Mycorrhiza (2016) 26:1–17 DOI 10.1007/s00572-015-0641-8 ORIGINAL PAPER Variation in ectomycorrhizal fungal communities associated with Oreomunnea mexicana (Juglandaceae) in a Neotropical montane forest Adriana Corrales1 & A. Elizabeth Arnold2 & Astrid Ferrer1 & Benjamin L. Turner3 & James W. Dalling1,3 Received: 7 October 2014 /Accepted: 8 April 2015 /Published online: 5 May 2015 # Springer-Verlag Berlin Heidelberg 2015 Abstract Neotropical montane forests are often dominated Cortinarius, represented by 14 species and previously report- by ectomycorrhizal (EM) tree species, yet the diversity of their ed to extract nitrogen from organic sources under low nitrogen EM fungal communities remains poorly explored. In lower availability, was found only in low fertility/high rainfall sites. montane forests in western Panama, the EM tree species Phylogenetic diversity analyses of Russula revealed greater Oreomunnea mexicana (Juglandaceae) forms locally dense evolutionary distance among taxa found on sites with con- populations in forest otherwise characterized by trees that trasting fertility and rainfall than was expected by chance, form arbuscular mycorrhizal (AM) associations. The objective suggesting that environmental differences among sites may of this study was to compare the composition of EM fungal be important in structuring EM fungal communities. More communities associated with Oreomunnea adults, saplings, research is needed to evaluate whether EM fungal taxa asso- and seedlings across sites differing in soil fertility and the ciated with Oreomunnea form mycorrhizal networks that amount and seasonality of rainfall. Analysis of fungal nrITS might account for local dominance of this tree species in oth- DNA (nuclear ribosomal internal transcribed spacers) re- erwise diverse forest communities. vealed 115 EM fungi taxa from 234 EM root tips collected from adults, saplings, and seedlings in four sites. EM fungal Keywords Beta diversity . Community structure . Fortuna communities were equally species-rich and diverse across Forest Reserve . Mycorrhizal networks . Russula Oreomunnea developmental stages and sites, regardless of (Russulaceae) soil conditions or rainfall patterns. However, ordination anal- ysis revealed high compositional turnover between low and high fertility/rainfall sites located ca. 6 km apart. The EM Introduction fungal community was dominated by Russula (ca. 36 taxa). Nutrient uptake and transfer via mycorrhizal associations strong- ly influences the growth and survival of most plant species in Electronic supplementary material The online version of this article nearly all of earth’s most species-rich and threatened terrestrial (doi:10.1007/s00572-015-0641-8) contains supplementary material, which is available to authorized users. biomes (Smith and Read 2008; Bonfante and Genre 2010). In tropical forests, trees predominantly form associations with * Adriana Corrales arbuscular mycorrhizal (AM) fungi (Glomeromycota) (Janos [email protected] 1983; St John and Uhl 1983; Béreau and Garbaye 1994; Onguene and Kuyper 2001;StJohn1980; McGuire 2008). 1 Department of Plant Biology, University of Illinois at However, forests dominated by tree species that associate with Urbana-Champaign, Urbana-Champaign, IL 61801, USA ectomycorrhizal (EM) fungi, especially Basidiomycota, have 2 School of Plant Sciences and Department of Ecology been recognized in all major tropical regions (Becker 1983; and Evolutionary Biology, The University of Arizona, Connell and Lowman 1989;Hartetal.1989;Henkel2003). Tucson, AZ 85721, USA Ectomycorrhizal plants in lowland tropical forests belong mostly 3 Smithsonian Tropical Research Institute, Apartado Postal to the Dipterocarpaceae and Fabaceae (primarily a narrow group 0843–03092, Republic of Panama of Caesalpinioideae), whereas Fagales (including members of 2 Mycorrhiza (2016) 26:1–17 the Juglandaceae, Betulaceae, and Fagaceae) frequently occur in Oreomunnea forms dominant stands on several distinct soil montane sites (Itoh 1995; Conway and Alexander 1992;Hart types that are distributed over a scale of only a few kilometers. et al. 1989;Henkel2003;Morrisetal.2008). In some cases, These soils are derived from contrasting parent materials and these EM species grow in Bmonodominant^ forests, wherein a occur in areas that differ in the seasonality and quantity of single tree species accounts for more than 50 % of canopy trees annual rainfall (Andersen et al. 2010), making this system in a stand (Connell and Lowman 1989). Why these unique for the study of EM fungal ecology. Preliminary field monodominant forests persist in otherwise diverse plant commu- surveys of fungal fruiting bodies indicated that diverse com- nities is not fully understood (Peh et al. 2011). munities of EM fungi associate with Oreomunnea in these Mast fruiting, low rates of disturbance, high tolerance of stands (A. Corrales et al. unpublished data). shade by seedlings, slow litter decomposition, and escape from In this first characterization of the EM fungal community herbivory have been proposed as mechanisms to explain tropical associated with Oreomunnea, we used data generated from root monodominance (reviewed by Peh et al. 2011). Strikingly, a tips of seedlings, saplings, and adult trees across this landscape common feature of many monodominant tree species in tropical to test four predictions. First, we predicted that infection fre- forests is the formation of EM associations (Malloch et al. 1980; quency of EM fungi would be lower in more fertile soils, con- Connell and Lowman 1989;Henkel2003). In temperate forests, sistent with the general view that benefits of EM fungi depend natural isotope abundance and radio-isotopic labeling experi- on soil conditions (Treseder 2004). Second, we predicted that ments have shown that some EM tree species can develop EM the diversity, composition, and phylogenetic diversity of EM networks, where hyphal connections transfer water, carbon, and fungi would vary with soil fertility. Third, we expected to see nutrients from adult to juvenile plants (Simard et al. 1997; (a) commonalities in EM fungal communities shared across Plamboeck et al. 2007; Booth and Hoeksema 2010;see seedling, sapling, and adult life stages of Oreomunnea,andthat Simard et al. 2012 for review). In tropical forests, direct evidence (b) community similarity among developmental stages would of resource transfer among individuals is currently lacking, but be particularly strong in the lowest fertility soils, where selection decreased survival and growth of seedlings when isolated from for EM networks or particularly beneficial symbionts would neighboring plants is consistent with EM network effects likely be strongest. Fourth, we expected lower phylogenetic (Onguene and Kuyper 2002;McGuire2007). diversity of EM fungi in high-fertility sites, reflecting lower Ectomycorrhizal networks may increase survival of conspe- colonization rates and consequently lower community diversity. cific seedlings in a spatially structured fashion, disproportion- ately increasing their abundance near adult trees (Onguene and Kuyper 2002; Henkel 2003;McGuire2007; Teste et al. 2009; Methods Booth and Hoeksema 2010) in a manner consistent with posi- tive plant–soil feedbacks (reviewed by Bever et al. 2012). In The study focused on stands of Oreomunnea mexicana turn, the presence and strength of plant–soil feedback depends (Juglandaceae; hereafter, Oreomunnea) in three watersheds on the functional traits and taxonomic composition of the EM in a primary lower montane forest (1000–1400 m.a.s.l.) in fungal community (e.g., Dickie et al. 2002;O’Brien et al. 2010; the Fortuna Forest Reserve in western Panama (Fig. 1;here- Kennedy et al. 2012). Determinants of EM fungal community after, Fortuna; 8°45′N, 82°15′W). Oreomunnea is a mid- composition remain poorly understood in tropical forests. For elevational canopy tree distributed from southern Mexico to example, there is conflicting evidence regarding host specificity western Panama at 900–2600 m.a.s.l. (Stone 1972). It pro- in tropical EM fungal communities (e.g., for evidence of host duces ca. 100 mg, wind-dispersed fruits, which can generate preference, see Tedersoo et al. 2008 , 2010a and Morris et al. high-density seedling patches in the understory (Table 1). 2009; for evidence of low host specificity, see Diédhiou et al. Oreomunnea is locally dominant at some of our study sites, 2010; Tedersoo et al. 2011; and Smith et al. 2011, 2013). accounting for up to 70 % of individuals and stand basal area Similarly, the influence of soil type on EM fungal community at the Honda watershed (A. Corrales unpublished data). composition remains unresolved, in part because EM fungal Dominance by Oreomunnea is not directly related to particu- communities associated with the same host species have rarely lar functional traits such as leaf chemistry (i.e., nitrogen (N) been studied across a range of soil conditions. and phosphorus (P)) and wood density, which are close to Here, we examine EM fungal communities associated with community averages for the area (K. Heineman unpublished Oreomunnea mexicana (Standl.) J.-F. Leroy, a widely distrib- data). However, in contrast to almost all co-occurring tree uted neotropical tree in the walnut family (Juglandaceae), and species at Fortuna, Oreomunnea forms EM associations. EM one of the few examples of a monodominant EM species in status was reported from Mexican populations of the Neotropics. In montane forests in western Panama, Oreomunnea (Quist et al. 1999)