Land use change can decrease animal biodiversity, but its impact on their gut microbiome is not well understood. Because land use change can result in reduced food variety, it may alter animal diet and consequently the colonization of their gut by microbes via different food items. We sought to quantify the effect of land use change on bird microbiomes in a Costa Rican landscape that contained a diversity of habitat types ranging from pristine forests to agriculture. In this landscape, we collected fresh fecal samples (n = 346) from six common species of insectivorous birds (clay-coloured thrush, Turdus grayi; Swainson’s thrush, Catharus ustulatus; orange-billed nightingale-thrush, Catharus aurantiirostris; yellow warbler, Dendroica petechia; rufous-capped warbler, Basileurerus rufifrons; and buff-throated saltator, Saltator maximus) at 24 sites across a land use gradient. Using these samples, we characterized gut microbial composition and insect diet by DNA metabarcoding. We extracted DNA from each sample, amplified bacterial 16S rRNA gene and arthropod CO1 gene for diet via PCR, and assigned taxonomy to OTUs using the Greengenes database on QIIME software.
Results/Conclusions
We identified a total of 3,333 bacterial OTUs and 333 arthropod OTUs. The most dominant bacterial phyla across the six species of birds included, Proteobacteria, Firmicutes, and Bacteroidetes. Diversity indices showed no significant difference among habitat types or bird species identity, suggesting that land use change or host identity may not affect gut microbial diversity. However, PERMANOVA analysis revealed a significant association between bird species identity and microbial species composition (R2 = 0.11). Ordination (NMDS) indicated clustering of gut microbial communities by bird species and by bird family. In addition, when we did NMDS and PERMANOVA separately for individual bird species, we found significant clustering of microbial communities by habitat type in the yellow warbler and the clay-coloured thrush. These results suggest that the effect of land use change on microbial communities may be host-specific and dependent on habitat type. The Mantel test indicated a significant relationship between diet composition and microbial communities (Mantel statistic r = 0.19), suggesting that an increase in diet dissimilarity is correlated to an increase in microbiome dissimilarity. AIC values from a generalized linear model suggest host species is the strongest explanatory variable of the gut microbiome, followed by diet. Overall, our data suggest that host species identity, diet and habitat type influence gut microbial composition.