Identifying the core microbial community in the gut of fungus‐growing termites
Gut microbes play a crucial role in decomposing lignocellulose to fuel termite societies, with protists in the lower termites and prokaryotes in the higher termites providing these services. However, a single basal subfamily of the higher termites, the Macrotermitinae, also domesticated a plant biom...
Saved in:
| Published in | Molecular ecology Vol. 23; no. 18; pp. 4631 - 4644 |
|---|---|
| Main Authors | , , , , , , , , , |
| Format | Journal Article |
| Language | English |
| Published |
England
Blackwell Scientific Publications
01.09.2014
Blackwell Publishing Ltd |
| Subjects | |
| Online Access | Get full text |
Cover
Loading…
| Summary: | Gut microbes play a crucial role in decomposing lignocellulose to fuel termite societies, with protists in the lower termites and prokaryotes in the higher termites providing these services. However, a single basal subfamily of the higher termites, the Macrotermitinae, also domesticated a plant biomass‐degrading fungus (Termitomyces), and how this symbiont acquisition has affected the fungus‐growing termite gut microbiota has remained unclear. The objective of our study was to compare the intestinal bacterial communities of five genera (nine species) of fungus‐growing termites to establish whether or not an ancestral core microbiota has been maintained and characterizes extant lineages. Using 454‐pyrosequencing of the 16S rRNA gene, we show that gut communities have representatives of 26 bacterial phyla and are dominated by Firmicutes, Bacteroidetes, Spirochaetes, Proteobacteria and Synergistetes. A set of 42 genus‐level taxa was present in all termite species and accounted for 56–68% of the species‐specific reads. Gut communities of termites from the same genus were more similar than distantly related species, suggesting that phylogenetic ancestry matters, possibly in connection with specific termite genus‐level ecological niches. Finally, we show that gut communities of fungus‐growing termites are similar to cockroaches, both at the bacterial phylum level and in a comparison of the core Macrotermitinae taxa abundances with representative cockroach, lower termite and higher nonfungus‐growing termites. These results suggest that the obligate association with Termitomyces has forced the bacterial gut communities of the fungus‐growing termites towards a relatively uniform composition with higher similarity to their omnivorous relatives than to more closely related termites. |
|---|---|
| Bibliography: | http://dx.doi.org/10.1111/mec.12874 Table S1 Relative abundance of the sequences identified and their taxonomic placement (down to genus) are presented for all taxa identified (separate excel file: Otani_TableS1.xlsx).Table S2 Genus-level taxon contributions to difference between termite gut communities observed in the PCoA analyses (Fig. 2, main paper). Estimates of principal components PC1-PC9 in total, sequence abundances, and taxonomic levels (to genus) are presented for all taxa identified (separate excel file: Otani_TableS2.xlsx).Table S3 The number of genus-level taxa shared (top) and proportions of communities shared (bottom) in all possible combinations for the nine communities.Table S4 Two-sample t-test analyses of the relative abundances of taxa assigned to the seven bacterial phyla in fungus-growing termites (this study), cockroaches, lower and higher nonfungus-growing termites (Dietrich et al. 2014). Only reads classified to the genus- or subgenus levels were included. Fisher's tests of combined p-values in each combination in the bottom. FGT: fungus-growing termites, HT: higher nonfungus-growing termites, LT: lower termites; significant P-values after Bonferroni correction in bold.Fig. S1 Rarefaction curves of sequence depth generated with R (R Core Team 2013). The curves represent the nine analysed termite gut samples, and each curve shows the number of genus-level taxa as a function of the number of sequenced reads after filtering.Fig. S2 PCoA plots for pairwise combinations of the first three principal components visualising community dissimilarities (UniFrac analysis).Fig. S3 PCoA and NMDS plots of Bray-Curtis distances between communities including quality-filtered and classified reads, after running classification-based analysis.Fig. S4 PCoA and NMDS plots of Bray-Curtis distances between communities including all quality-filtered reads (including the unclassified), after running an OTU-cluster analysis at 97% sequence similarity. Danish National Research Foundation Centre of Excellence - No. DNRF57 The Danish Council for Independent Research | Natural Sciences University of Copenhagen Max Planck Society ark:/67375/WNG-N15JWPRJ-M ArticleID:MEC12874 istex:488E1A8BEF2D3FBBF46A52A18C463A6CF0C234F3 ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
| ISSN: | 0962-1083 1365-294X |
| DOI: | 10.1111/mec.12874 |