Novel Associations of Empagliflozin on the Gut Microbiome and Metabolome in Type 2 Diabetes
https://academic.oup.com/jcem/article/107/10/e4246/6604095?login=false
"The beneficial effects of sodium/glucose cotransporter-2 inhibitors (SGLT2is) in individuals with type 2 diabetes (T2DM) at high risk for cardiovascular disease (CVD) is well documented (1). In addition, their use in individuals with both diabetic and nondiabetic chronic kidney disease to attenuate disease progression is extensively validated (1). However, much less is known about the mechanisms behind these beneficial effects, including whether these agents can affect gut microbiome and metabolomics.
The article by Deng et al (2) reports on the association of cardiovascular benefits of empagliflozin with gut microbiota and plasma metabolites in individuals with T2DM. This trial included 76 treatment-naïve adults with diet-controlled T2DM and risk factors for CVD randomized in an open label, 2-arm clinical trial for 3 months. Patients were treated with either empagliflozin or metformin and the authors reported the following: (1) significant and similar reductions in glycated hemoglobin levels and improvements in glucose metabolism in the 2 groups with only empagliflozin improving CVD risk factors; (2) empagliflozin significantly reshaped the gut microbiota after 1 month of treatment and this effect was maintained until the completion of the study period; (3) empagliflozin elevated levels of sphingomyelin and demonstrated reduced levels of cis-aconitate, uric acid, and glycochenodeoxycholate; (4) empagliflozin increased favorable short-chain fatty acid–producing bacteria such as species from Roseburia, Eubacterium, and Faecalibacterium and reduced harmful bacteria such as Escherichia-Shigella, Bilophila, and Hungatella.
This study reaffirms further metabolic benefits of SGLT2is compared with metformin, particularly in relation to elevated levels of hematocrit and adiponectin, and reduced levels of blood pressure and uric acid observed only in individuals treated with empagliflozin.
While there have been previous studies investigating the effect of metformin on gut microbiota (3), there have been few studies in the literature regarding the effect of SGLT2is on the gut microbiome. In fact, the limited evidence in the literature did not demonstrate a difference in the composition of the gut microbiome with the use of dapagliflozin compared with gliclazide in those with T2DM (4). This is in contrast to the findings of the study by Deng et al, which found empagliflozin to increase beneficial levels of short-chain fatty acid–producing bacteria such as species from Roseburia, Eubacterium, and Faecalibacterium, and reduced those of several potentially harmful bacteria including Escherichia-Shigella, Bilophila, and Hungatella compared with the microbiome in individuals with T2DM treated with metformin. The inconsistent results between these 2 randomized controlled trials may be attributable to the limitations associated with selection of intervention, as some studies have demonstrated empagliflozin to be more effective than dapagliflozin at reducing glycated hemoglobin (5), patient cohorts, or analysis of the gut microbiome. Replication of these studies in larger numbers will help with validating such discrepancies.
SGLT2is by virtue of their mechanism of action of sodium and glucose inhibition in the proximal convoluted tubules cause glycosuria and natriuresis (1), resulting in a change in the biochemical milieu. The intestinal microbiota can be very adaptable to changes in the biochemical milieu. It was therefore interesting to observe the changes in the gut microbiome and plasma metabolites occurring early after the initiation of empagliflozin. In fact, the gut microbiome changes occurred as early as 4 weeks and were consistently stable over the 8- and 12-week study period. This highlights the importance of serial measurements to reflect potential early beneficial changes with intervention and to provide more consistent data. One limitation, however, is the study period was only 3 months, and we do not know whether these metabolome and microbiome changes lasted beyond the treatment period or whether they returned to baseline. A consideration for future design would be to sample even more frequently and perhaps even earlier at 2 weeks to determine whether changes occurred earlier and ensure follow-up over at least a 12-month period.
The inhibition of sodium/glucose cotransporter-2 reduces plasma glucose levels, thereby promoting lipolysis in adipose tissue, resulting in ketone generation. A feature of SGLT2i-treated individuals is the increase in circulating levels of ketone bodies (6). Ketone bodies or β-hydroxybutyrate, beyond acting as a carrier of energy from adipocytes to peripheral tissues, have important cellular signaling roles (7). It has been previously demonstrated that empagliflozin via increased serum β-hydroxybutyrate levels and reduced serum insulin significantly suppressed the NOD-like receptors (NLR) family, pyrin domain–containing (NLRP3) inflammasome activation, and subsequent secretion of interleukin-1β in human macrophages, thus reducing the pathogenic effect on T2DM and CVD (7). These effects were further verified ex vivo (7). The study by Deng et al also demonstrates a consistent increase in serum β-hydroxybutyrate levels in the empagliflozin group but is not statistically significant. The authors noted that after the 3-month treatment period with empagliflozin, there were increased levels of metabolites such as fatty acids, fatty acyls, organic acids, and phosphosphingolipids but reduced levels of metabolites such as amino acids and uric acid. Furthermore, there was no significant correlation between serum β-hydroxybutyrate levels and the differential metabolites in those treated with empagliflozin. This is an evolving area of interest and further study into the role of serum β-hydroxybutyrate levels to ameliorate inflammation would provide some insight into the cardioprotective role of SGLT2is.
The unanswered question is whether the metabolic effects of SGLT2i treatment result in an improved gut microbiome and plasma metabolite profile, or whether the metabolic effects themselves are driven by changes in the microbiome and metabolome. The authors postulate it may be the former and that modulation of the microbiome and metabolome may be another mechanism of action of SGLT2is in addition to the inhibition of sodium/glucose cotransporter-2. As pointed out by the authors, this inference would require the detailed investigation into further mechanisms of action of SGLT2is. This is pertinent, as human microbiome investigations are moving from bacterial description and investigation to an understanding of the mechanisms of action, paving the way for improved clinical intervention (8). Perhaps future studies could consider a focus on the gut microbiota’s metabolic activity as a better reflection of treatment intervention than the traditional microbiota richness or diversity (9), as understanding the metabolic capabilities of the gut microbiota is necessary in explaining their role in health and disease. Further areas for consideration are whether different types of SGLT2is result in similar or different changes in the microbiome and metabolome.
The authors are to be commended for this important addition to an ever-increasing knowledge on the benefits of SGLT2is, now particularly describing the potential novel benefits on the gut microbiota and plasma metabolites. However, the robustness of these associations needs to be tested in larger cohorts, over longer durations, and in replicated studies. We look forward to future studies that clarify the issues outlined above."