Characterizing the avian gut microbiota: membership, driving influences, and potential function

Abstract

Birds represent a diverse and evolutionarily successful lineage, occupying a wide range of niches throughout the world. Like all vertebrates, avians harbor diverse communities of microorganisms within their guts, which collectively fulfill important roles in providing the host with nutrition and protection from pathogens. Although many studies have investigated the role of particular microbes in the guts of avian species, there has been no attempt to unify the results of previous, sequence-based studies to examine the factors that shape the avian gut microbiota as a whole. In this study, we present the first meta-analysis of the avian gut microbiota, using 16S rRNA gene sequences obtained from a range of publicly available clone-library and amplicon pyrosequencing data. We investigate community membership and structure, as well as probe the roles of some of the key biological factors that influence the gut microbiota of other vertebrates, such as host phylogeny, location within the gut, diet, and association with humans. Our results indicate that, across avian studies, the microbiota demonstrates a similar phylum-level composition to that of mammals. Host bird species is the most important factor in determining community composition, although sampling site, diet, and captivity status also contribute. These analyses provide a first integrated look at the composition of the avian microbiota, and serve as a foundation for future studies in this area.

Keywords: 16S rRNA gene; avian; bacteria; bird; meta-analysis; microbiota.

Figure 1

The relative proportion of OTUs represented in each study. OTUs were constructed by calculating average-neighbor distance between aligned 16S rRNA gene sequences in mothur and classified as a cluster of sequences with ≥97% similarity. Taxonomic classification for each OTU was derived from a consensus taxonomic classification of each sequence assigned to the OTU. (Top) Samples from clone-library data. (Bottom) Next-generation sequencing samples obtained from Sequence Read Archive. Top labels identify the study from which sequences were downloaded; bottom labels identify the host bird. Top letters denote studies PRJEB3384 (A), PRJEB1467 (B), PRJNA169064 (C) PRJNA193217 (D), Unno, 2010 (E), Bennet, 2013 (F), and PRJEB1549 (G). Bottom letters denote host organisms duck (H), goose (I), fairy prion (J), and petrel (K).

Figure 2

Unweighted UniFrac distances for within- and between-study comparisons. Distances were calculated by extracting reads classified as Clostridium (top) and Bacteroidetes (bottom) from each sample and constructing neighbor-joining phylogenetic trees based on average-neighbor distances between aligned sequences. Differences between each pair of samples were categorized as being the distance between samples from the same study or from different studies and plotted accordingly (blue = within study, orange = between study). The study “Dewar, 2013” investigated the faecal microbiota from little, king, macaroni, and gentoo penguins. The study “LittlePenguin” investigated the faecal microbiota of little penguins, and “Unno, 2010” the microbiota of a chicken, duck, and goose from a farm.

Figure 3

Constrained Canonical Analysis of community structure based on fitting of metadata factors to the clone-library sequence data. Images represent host (top left), sample site (top right), diet (bottom left), and captivity status (bottom right).

Figure 4

Constrained Canonical Analysis of community structure based on fitting of metadata factors to the short-read sequence data. Images represent host (top left), sample site (top right), diet (bottom left), and captivity status (bottom right). Note that the “herbivore” grouping represents exclusively kakapo.