A molecular survey of Australian and North American termite genera indicates that vertical inheritance is the primary force shaping termite gut microbiomes

Abstract

Background: Termites and their microbial gut symbionts are major recyclers of lignocellulosic biomass. This important symbiosis is obligate but relatively open and more complex in comparison to other well-known insect symbioses such as the strict vertical transmission of Buchnera in aphids. The relative roles of vertical inheritance and environmental factors such as diet in shaping the termite gut microbiome are not well understood.

Results: The gut microbiomes of 66 specimens representing seven higher and nine lower termite genera collected in Australia and North America were profiled by small subunit (SSU) rRNA amplicon pyrosequencing. These represent the first reported culture-independent gut microbiome data for three higher termite genera: Tenuirostritermes, Drepanotermes, and Gnathamitermes; and two lower termite genera: Marginitermes and Porotermes. Consistent with previous studies, bacteria comprise the largest fraction of termite gut symbionts, of which 11 phylotypes (6 Treponema, 1 Desulfarculus-like, 1 Desulfovibrio, 1 Anaerovorax-like, 1 Sporobacter-like, and 1 Pirellula-like) were widespread occurring in ≥50% of collected specimens. Archaea are generally considered to comprise only a minority of the termite gut microbiota (<3%); however, archaeal relative abundance was substantially higher and variable in a number of specimens including Macrognathotermes, Coptotermes, Schedorhinotermes, Porotermes, and Mastotermes (representing up to 54% of amplicon reads). A ciliate related to Clevelandella was detected in low abundance in Gnathamitermes indicating that protists were either reacquired after protists loss in higher termites or persisted in low numbers across this transition. Phylogenetic analyses of the bacterial communities indicate that vertical inheritance is the primary force shaping termite gut microbiota. The effect of diet is secondary and appears to influence the relative abundance, but not membership, of the gut communities.

Conclusions: Vertical inheritance is the primary force shaping the termite gut microbiome indicating that species are successfully and faithfully passed from one generation to the next via trophallaxis or coprophagy. Changes in relative abundance can occur on shorter time scales and appear to be an adaptive mechanism for dietary fluctuations.

Figure 1

Heatmap showing microbial taxa (mostly genus and family level) with relative abundance ≥0.2% in one or more whole gut samples surveyed in this study. Each row represents a gut sample and each column a microbial taxon with relative abundance indicated by shading according to the legend. Phylum-level designations for the microbial taxa are indicated at the top of the figure, and host sample phylogeny is indicated to the left (family) and right (genus) of the figure.

Figure 2

Prevalence versus relative abundance graph of bacterial OTUs (97% sequence identity) in the surveyed gut samples. OTUs with ≥10% relative abundance or ≥50% prevalence across the 66 termite samples are highlighted in red and labeled with OTU ID and closest matching bacterial genus. Relative abundance was calculated only using samples containing detectable amounts of a given OTU. In instances where the OTU is only found in a single termite genus, the termite genus is also included in the label.

Figure 3

Heatmap showing archaeal OTUs (97% seq id) with ≥0.1% relative abundance in one or more of the surveyed gut samples. Each row represents an OTU and each column a gut sample with relative abundance as a percentage of the total microbial community (including bacteria) indicated by numbers and shading according to the legend. The termite genus for each sample is indicated at the top of the figure, and OTU phylogeny is indicated to the left (phylum) and right (mostly genus) of the figure.

Figure 4

UPGMA tree of unweighted (presence/absence only) Soergel pairwise distances between bacterial profiles showing a high consistency with host phylogeny and low consistency with diet (Additional file 14 : Table S4). The values on interior nodes represent jackknife support values ≥49. Termite host affiliation (family) and presumptive diet are indicated to the right of the tree.

Figure 5

Subtrees of host and bacterial community phylogenetic comparisons showing secondary effect of diet on community structure of polyphagous termite genera. When the relative abundance of bacterial OTUs is taken into account (weighted Soergel), samples cluster according to diet. The values on interior nodes of the COII trees are FastTree local support values and jackknife support values ≥49 on the Soergel UPGMA Carbon isotope values of gut contents are shown in the far right panels.