Perhaps one of the most exciting frontiers of medicine is the exploration of the human gut flora, or the microbiome. While it remains with potential as a target for future therapeutics, defining a healthy microbiome and pathologic variant remains a daunting task. In a pioneering study to further characterize healthy variation, the Human Microbiome Project Consortium in 2012 presented the microbiota of 242 healthy Western subjects from 18 different anatomic sites, including the distal gut, and while significant interindividual variation existed, metabolic pathways remained stable [1, 2]. As indicated by Arumugam et al., Bacteroidetes phyla dominate the gut microbiome. In their analysis, 33 samples across several nations formed three distinct clusters: Bacteroidetes, Prevotella, and Ruminococcus [3]. Generally, the gut microbiome is shared among family members with variation present in each individual’s microbial community [4]. Diet does have a critical impact [4–6]. As an example, in a comparison of microbiome in children from Europe and from rural Africa, the microbiome of African children displayed significantly increased bacterial richness and diversity, leading the authors to conclude that gut microbiome coevolved with their polysaccharide-rich diet [6].

Obesity is associated with a significant decrease in microbiome diversity and richness [5]. In a study of obese and non-obese Danish individuals, obese individuals shared an “inflammatory” phenotype, having a higher prevalence of Bacteroides and R. gnavus, which have been associated with Inflammatory Bowel Disease [6–8]. Patients characterized as having a low gene count had a greater prevalence of obesity, insulin resistance, fatty liver, and low-grade inflammation than patients with a high gene count. Low gene count patients were also more prone to gaining more weight over time [5, 9]. In another study by Vrieze et al., male humans with metabolic syndrome were the recipients of either autologous microbiota or that of lean donors. Six weeks after small intestinal infusion, recipients from lean donors experienced increased insulin sensitivity, along with increased levels of butyrate-producing microbiota [10]. Perhaps one of the most telling studies was performed by Ridaura et al. who transplanted fecal microbiota from adult twin pairs discordant for obesity into germ-free mice. Not only were these obesity-associated phenotypes transmissible, cohousing with lean co-twin’s microbial recipient prevented increasing body mass and adverse metabolic outcomes [11]. Diet had an integral role, and when mice were fed a low-fat, high-fiber diet, they were protected against the effects of obese microbiota [12].

Correlations between shifts in the microbiome and obesity have been observed, although causality remains to be discerned. In general, one of the noteworthy observations has been the increased ratio of the phylum Firmicutes to the phylum Bacteroidetes seen with obesity [13–15]. This concept was initially proposed by Turnbaugh et al. who also demonstrated that obese microbiome has perhaps an increased capacity to harvest energy from the diet [16]. Further support for the Firmicutes to Bacteroidetes ratio was lent after the observation that this ratio is decreased with percentage reduction in body weight [13]. However, this observation has been met with some controversy, as other studies have failed to identify the difference in this ratio between obese and lean patients [17, 18]. Perhaps controversy in this regard lies in the fact that these are entire phylum changes, and more specific family and species levels changes should be pursed to characterize pathogen with phenotypic changes. For example, Zhang et al. assessed contributions of host genetics and diet in altering the microbiome in mice. Sixty-five species-level phylotypes were correlated with differences induced by diet, with diet explaining 57 % of the total structural variation in gut microbiome. Genetic mutation accounted for 12 %. Barrier-protecting Bifidobacterium species were nearly absent in all animals, with an observed increase in Desulfovibrionaceae, a sulfate-reducing and endotoxin-producing bacteria, in all animals with impaired glucose tolerance [19].

Not surprisingly, T2DM in humans is associated with microbial shifts as well. In a study by Qin et al., a protocol for a metagenome-wide association study was developed to study 345 Chinese individuals. Patients with T2DM portrayed moderate intestinal dysbiosis as indicated by a decrease in butyrate-producing Roseburia intestinalis and F. prausnitzii, along with an increase in opportunistic pathogens including Bacteroides caccae, Clostridiales, and Escherichia coli. Lean patients in this study exhibited an enriched population of butyrate-producing bacteria [20]. Similarly, Karlsson et al. studied the gut meganome in 145 European women with normal, impaired, or diabetic glucose control. This group similarly displayed increases in Clostridiales species, and decreases in Roseburia intestinalis and F. prausnitzii [21, 22]. They also illustrated increased Proteobacteria, increased expression of microbial genes involved in oxidative stress, and decreased genes involved in vitamin synthesis [9, 21, 22]. Another protective bacterium identified is Akkermansia municiphila, a mucin-degrading bacterium, negatively associated with type 1 and type 2 diabetes mellitus [20], while Bacteroidetes and Prevotella are increased in proportion to a decrease in Firmicutes and Clostridia [21–23].

Circulating lipopolysaccharide (LPS) is an integral and early step in the development of insulin resistance and diabetes having been observed in mice and humans [24–26]. Mice fed a high fat diet had two to three times increased levels of circulating LPS [26]. CD-14 mutant mice, which are resistant to LPS, are also resistant to high-fat diet-induced insulin resistance [26, 27]. Cani et al. administered antibiotics to mice on a high-fat diet, demonstrating that changes in the gut microbiome are responsible for endotoxemia and the subsequent inflammatory cascade, correlating strongly with intestinal permeability [27]. In patients with diabetes mellitus, LPS levels are significantly elevated, leading to increased secretion of pro-inflammatory cytokines [4, 28]. In mice fed a high-fat diet for 3 months, LPS was elevated in the presence of diabetes, and related this to increased gut permeability in the ileum and cecum [29]. Obese patients and rodents exhibit increased richness of LPS-producing bacteria, notably increased Enterobacteriacaea and Desulfovibrionaceae, corresponding with increased circulating LPS [30]. In obese human adolescents undergoing an exercise and diet program, fecal Enterobacteriaceae was significantly decreased [31, 32]. Once systemic, LPS binds LPS-binding protein and is recognized by CD14 and TLR4 [33, 34]. This in turn leads to a pro-inflammatory response and production of inflammatory cytokines potentially linking the microbiome to the diabetic phenotype [35].

Following the gastric bypass, there is an increase in richness and diversity of the gut microbiome associated with white adipose tissue genes [36]. In a study comparing nine individuals, with three in each of normal weight, morbidly obese, and post-gastric bypass groups, Firmicutes was dominant in the first two groups, but decreased in post-gastric bypass individuals with a proportional increase in Gammaproteobacteria [15]. Akkermansia was also increased in the post-gastric bypass population. Graessler et al. compared patients before and 3 months following gastric bypass, having also observed a reduction in Firmicutes, with an increase in Proteobacteria. However, they also observed a decrease in F. prausnitzii correlating directly with fasting blood glucose levels [37]. Furet et al. observed an increase in F. Prausnitzii correlating with improved circulating CRP and IL6 in 30 patients with T2DM following gastric bypass. In an interesting and telling study performed by Liou et al., shifts observed following RYGB are conserved between humans, mice and rats, observing an increase in Gammaproteobacteria (Escherichia) and Verrucomicrobia (Akkermansia). Transferring gut microbiota from mice having undergone RYGB to germ-free mice resulted in weight loss and decreased fat mass relative to recipients of microbiota from mice following sham surgery [38].

In a study by Monte et al., 15 morbidly obese patients with T2DM underwent RYGB, with blood samples collected on day of surgery and 180 days after surgery. Interestingly, systemic inflammation as measured by multiple cytokines, including MCP-1 and CRP, and bacterial LPS were significantly reduced following surgery, possibly linking this improvement in systemic inflammation as a mechanism underlying metabolic improvement [39].

Certainly, further characterization of the microbiome is required; more important may be the role species have in altering gut physiology translating to systemic changes, particularly in diabetes. Much of this may involve gaining a stronger understanding of gut immunology at the enterocyte level given the known systemic inflammation that characterizes white adipose tissue in T2DM.