A fine-scale spatial analysis of fungal communities on tropical tree bark unveils the epiphytic rhizosphere in orchids.
Abstract: Approximately 10% of vascular plants are epiphytes and, even though this has long been ignored in past research, are able to interact with a variety of fungi, including mycorrhizal taxa. However, the structure of fungal communities on bark, as well as their relationship with epiphytic plants, is largely unknown. To fill this gap, we conducted environmental metabarcoding of the ITS-2 region to understand the spatial structure of fungal communities of the bark of tropical trees, with a focus on epiphytic orchid mycorrhizal fungi, and tested the influence of root proximity. For all guilds, including orchid mycorrhizal fungi, fungal communities were more similar when spatially close on bark (i.e. they displayed positive spatial autocorrelation). They also showed distance decay of similarity with respect to epiphytic roots, meaning that their composition on bark increasingly differed, compared to roots, with distance from roots. We first showed that all of the investigated fungal guilds exhibited spatial structure at very small scales. This spatial structure was influenced by the roots of epiphytic plants, suggesting the existence of an epiphytic rhizosphere. Finally, we showed that orchid mycorrhizal fungi were aggregated around them, possibly as a result of reciprocal influence between the mycorrhizal partners.
Summary (3 min read)
- The authors hypothesized that (i) as described in soils, these communities have no random distribution on the bark .
- Due to the ability of many fungi to colonize plant roots, (ii) their distribution should be modulated by the distance to roots of vascular epiphytes.
- Particularly, (iii) communities of OMF should be aggregated around their orchid hosts.
2.1 Study area
- The elevation provides frequent fogs throughout the year, and the humidity is around 80%, even in the dry season.
- The climate of the region is humid subtropical mesothermic, with temperatures ranging from 17 to 23°C and annual rainfall averaging 1300 mm (Rolim & Ribeiro, 2001) .
- This forest is characterized by medium to large trees, and a high diversity of orchid species, the majority of which are epiphytic (Lana et al., 2018) .
2.2 Bark and root sampling
- Two trees belonging to Siparuna sp. (Siparunaceae; tree 1) and Himathanthus sucuuba (Apocynaceae; tree 2) were selected in February 2015 and February 2016 (95 m away from each other) respectively because they had epiphytic orchids growing on their lower trunk, namely Isochilus linearis and Epidendrum armeniacum.
- Bark was also collected under each root sample.
- All samples were frozen at -20°C within few hours in the nearby field laboratory of the Serra do Brigadeiro State Park headquarters for downstream molecular analyses.
- Two additional thin sections of orchid roots surrounding each sampled piece were collected to check for mycorrhizal fungal colonization on the following day under the microscope and all, without exception, displayed hyphal coils in at least one of each inspection section.
2.3 High-throughput sequencing of fungal communities
- Tagging system negative controls were performed at this step (Hornung et al., 2019; Zinger et al., 2019) , i.e., pairs of barcoded primers were intentionally omitted in the final sequencing to control for cross-contamination.
- Plate designs were randomized in order to avoid possible cross-contamination leading to misinterpretation in subsequent spatial analysis.
- After visualization on gel, the positive amplicons were purified with NucleoMag® NGS Clean-up and Size Select (Macherey-Nagel, GmbH & Co KG.), quantified by fluorescence with Qubit TM dsDNA High-Sensitivity (Invitrogen TM ), and pooled in equimolar ratios prior to library preparation and 2x250 bp paired-end sequencing on an Illumina MiSeq platform at Fasteris (Geneva, Switzerland).
- Three positive controls (mock community) and three negative controls (ultrapure water) were used per PCR trial , resulting in a total of 36 positive and 36 negative controls in total.
2.5 Fungal functional guilds
- OTUs found in at least one orchid root sample were considered as endophytes.
- Among them, those Basidiomycota belonging to Tulasnellaceae, Ceratobasidiaceae (Veldre et al., 2013) , Serendipitaceae (Weiß et al., 2016) , and Atractiellales (Kottke et al., 2010) were recognized as orchid mycorrhizal fungi (OMF) (Dearnaley et al., 2012) .
- Besides, trophic guilds were assigned to all OTUs using FunGuild (Zanne et al., 2019) : the authors chose to keep those which were either exclusively saprotrophs, symbiotrophs, plant pathogens, or lichenized fungi.
- For the remaining OTUs, guilds provided by FunGuild were validated based on the author's expertise.
- The OMF were kept in a separate category despite their saprotrophic and symbiotrophic ability (Dearnaley et al., 2012; Selosse & Martos, 2014) .
2.6 Statistical analyses
- As the similarities between samples are not independent of one another, coefficients of the binomial GLM were obtained using a leave-one-out Jackknife procedure as described in (Millar et al., 2011) .
- The significance of the distance decay of similarity was tested using a permutational Mantel test (Spearman method, 9999 permutations; Anderson et al., 2013) , while the significance of the distance decay of richness was assessed by ANOVA (F-test).
3.1 Roots and bark harbored distinct, but partially overlapping fungal communities
- Among the 31 OMF OTUs that were found, encompassing the four OMF families (see 2.5, Fig. 3 ), only five OTUs were shared between the two trees after rarefaction (Table S4 ) and only one OTU (Tulasnellaceae, TUL-1) when considering the roots only (Table S4 , S5).
- The sharing of OMF between grids was not statistically different to that of other fungi, meaning that the trees harbored different fungal communities overall.
- On grid 2, where two orchid species co-exist, OMF OTUs belonging to Ceratobasidiaceae (CER-1) and Serendipitaceae (SER-1) were shared between the two species when they were spatially close (Table S5 , Fig. S1 ).
3.2 All fungal communities were spatially structured
- Spatial autocorrelation of single OTUs showed that only OMF on grid 2 tend to be more frequently spatially clustered than other fungi (Table S6 ).
- OMF families showed vertical stratification on grid 2 that covered a greater height on the tree (1.7 m), whereas this pattern was not obvious on grid 1 (covering 0.7 m only; Fig. S9 ).
3.3 Epiphytic roots influenced all fungal communities
- The Jaccard similarity between roots and bark fungal compositions significantly decreased with increasing distance from the roots for the whole fungal community on both grids (Fig. 5 ).
- This was also observed for endophytes on both grids, for non-OMF symbiotrophs on grid 1 only, and for OMF, plant pathogens and saprotrophs on grid 2 only (Table 1, Fig. 5 ; see also Fig. S10 and Table S8 for details).
- The distance decay of bark fungal richness showed contrasting results with either non-significant or opposite results between grids (Fig. S12 -14, Table S9 ).
- By comparing the density distribution of OMF versus endophytes (distance from roots beyond which 80% of the occurrences of a given OTU are limited), the OMF were not statistically closer to roots than other endophytes (Wilcox tests, W = 370, p = 0.423 and W = 1142, p = 0.397 for grid 1 and 2, respectively).
4.1 Features of bark fungal communities compared to the soil's
- This thinness allowed us to exhaustively sample fungal communities at a given position.
- Whether these communities are spatially structured or are either homogeneously or randomly distributed remained an open question, which the authors investigated in this study.
4.5 Fungal communities could modulate epiphytic plant population dynamics
- Here, the OMF were more spatially clustered than any other fungi (Table S6 ), reflected in the vertical stratification on grid 2 (Fig. S9 ), which suggests that they could strongly constrain orchid seed germination.
- In soil, it has also been proposed that the patchiness of orchid individuals (Jacquemyn et al., 2007) could be due to that of their mycorrhizal partners (Jacquemyn et al., 2012) .
4.6 Conclusion and perspectives
- The authors observed a vertical niche differentiation for OMF communities, but not for other fungal guilds, probably because their sampling design was not appropriate to investigate such vertical gradients.
- Yet, a possible trend for lower vertical than horizontal structure was observed.
Did you find this useful? Give us your feedback
Related Papers (5)
Frequently Asked Questions (1)
Q1. What are the contributions in "A fine-scale spatial analysis of fungal communities on tropical tree bark unveils the epiphytic rhizosphere in orchids" ?
HAL this paper is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not.