Metagenomic insights into the roles of Proteobacteria in the gastrointestinal microbiomes of healthy dogs and cats

Abstract

Interests in the impact of the gastrointestinal microbiota on health and wellbeing have extended from humans to that of companion animals. While relatively fewer studies to date have examined canine and feline gut microbiomes, analysis of the metagenomic DNA from fecal communities using next-generation sequencing technologies have provided insights into the microbes that are present, their function, and potential to contribute to overall host nutrition and health. As carnivores, healthy dogs and cats possess fecal microbiomes that reflect the generally higher concentrations of protein and fat in their diets, relative to omnivores and herbivores. The phyla Firmicutes and Bacteroidetes are highly abundant, and Fusobacteria, Actinobacteria, and Proteobacteria also feature prominently. Proteobacteria is the most diverse bacterial phylum and commonly features in the fecal microbiota of healthy dogs and cats, although its reputation is often sullied as its members include a number of well-known opportunistic pathogens, such as Escherichia coli, Salmonella, and Campylobacter, which may impact the health of the host and its owner. Furthermore, in other host species, high abundances of Proteobacteria have been associated with dysbiosis in hosts with metabolic or inflammatory disorders. In this review, we seek to gain further insight into the prevalence and roles of the Proteobacteria within the gastrointestinal microbiomes of healthy dogs and cats. We draw upon the growing number of metagenomic DNA sequence-based studies which now allow us take a culture-independent approach to examine the functions that this more minor, yet important, group contribute to normal microbiome function.

Keywords: Proteobacteria; 16S rRNA gene; canine; fecal microbiome; feline; metagenome.

Figure 1

Common metagenomic DNA‐based analyses to define the community composition and function of the microbial communities in the dog and cat GI tract. Metagenomic DNA is extracted from a GI content sample, usually freshly voided fecal material, which contains microbial community members. The microbial community composition (see lower left box) is most commonly determined by amplifying variable regions of the 16S rRNA gene and sequencing the resulting amplicons. Similar 16S rRNA sequences are grouped into Operational Taxonomic Units (OTUs), which can be compared to specialized 16S rRNA sequence‐based taxonomic databases (e.g., RDP, Greengenes, SILVA) to assign taxonomic identities. The community can be described in terms of the relative abundance of the taxa present, and/or their phylogenetic relationships. To enable the potential function of the microbial community to be explored (see lower right box), metagenomic DNA is directly shotgun sequenced. The functional potential of the community can be determined by comparing the sequences to reference genomes, gene catalogs, or functional databases (e.g., SEED, KEGG, and COG). This allows the community to be described in terms of the relative abundances of its genes and pathways. More recently, inferences of community function may be made from 16S rRNA‐based taxonomic profiles using reference genome information implemented in software such as PICRUSt. Moreover, community composition may be deduced from shotgun sequenced DNA by capturing the 16S rRNA gene sequence reads, with classification using dedicated 16S rRNA databases

Figure 2

Functional profiles of representative healthy dog (Swanson et al., 2011), cat (Tun et al., 2012), and kitten (Young et al., 2016) fecal microbiota, compared to profiles from the subset of Proteobacteria‐associated reads within these datasets. Metagenomic shotgun sequence data were analyzed in MEGAN6 CE. The relative abundances of sequence hits for each SEED subsystem are shown as percentages with the length of the blue bar indicating their relative magnitude within the dataset, apart from the “No hits” category. Fold changes (FC) for Proteobacteria relative to microbiota abundances are shown, where fold increases are represented with green bars, and fold decreases (negative values) are represented by red bars.

Figure 3

Network of the most highly correlated taxa (blue circles) and genes (purple squares) among the fecal microbiomes of 17‐week‐old kittens fed canned and kibbled diets. Kitten fecal microbiome sequence data were previously annotated in MG‐RAST (Meyer et al., 2008) for taxonomy and COG and KO functions as described (Young et al., 2016). Gene and taxon abundance network generated by sparse partial least squares regression using the spls function in the mixOmics package (Lê Cao & González, 2009) for R, using a correlation cut off of > |0.9|. The networks were viewed in Cytoscape 3.5.1 (Shannon et al., 2003)