Novel Pharmacologies of Metformin in the Gut: insights from a washout/redosing paradigm in type 2 diabetes

Program: Abstracts - Orals, Featured Poster Presentations, and Posters
Session: SAT 824-833-GI Regulatory Peptides
Bench to Bedside
Saturday, June 15, 2013: 1:45 PM-3:45 PM
Expo Halls ABC (Moscone Center)

Poster Board SAT-832
Antonella Napolitano*1, Sam Miller2, Stephanie Van Horn3, Alan Kwan3, Ganesh Sathe3, George Livi3, Deepak Rajpal3, Aaron Spivak3, Jim Brown3, Darren Robertson3, David Baker3, Andy Nicholls3 and Derek Nunez3
1GlaxoSmithKline, Cambridge, United Kingdom, 2GlaxoSmithKline, Cambridge, 3GlaxoSmithKline
Metformin has pleiotropic effects in type 2 diabetes (T2D), including decreased hepatic glucose production and increased peripheral glucose utilization. However, intravenous delivery of metformin to achieve comparable plasma concentrations to efficacious oral doses, does not improve glucose control.We have focused on the gut as a potential target organ of metformin because metformin alters the secretion of the gut hormones, and/or the absorption of bile acids. The redistribution of nutrients and bile acids may also alter the populations of bacteria in the gut, which could also affect metabolic control.

Methods:  In a controlled clinical study, we monitored various metabolic variables  and the gut microbiome in 14 T2D patients over the course of four separate visits  when metformin was withdrawn and re-instated, using pre-defined changes in fasting blood glucose (FBG)  levels to determine when the subjects should be investigated: ( 1) at baseline, 2) 7 days after withdrawing metformin, 3) when mean FBG increased by 25% , and 4) when mean FBG was back to baseline values after restarting metformin.

Results: Bile acids.  The total serum bile acid pool increased significantly after the withdrawal of metformin (mainly primary bile acids) (p=0.0004 ).  Analysis of concentration indicated that the largest quantitative changes were in AUC (4-8 h) for chenodeoxycholic acid. A circadian rhythm of bile acid secretion was observed. Faecal concentrations of the bile acids (mostly the secondary bile acids, lithocholic and deoxycholic acids) were greater when subjects were on metformin, although the differences were not statistically significant.Incretins.  Metformin withdrawal significantly reduced AUC (0-12 h) of total and active GLP-1, with little change in GIP or PYY, indicating that metformin may increase the secretion of GLP-1 and inhibit the degrading enzyme, dipeptidyl peptidase-IV (DPP-IV).  Faecal Microbiome analysis. We used 16S rRNA sequencing to identify bacterial species in the gut of each subject at the four timepoints. The total number of microbial species was not significantly influenced by metformin, but Veillonella and Eubacterium increased when off metformin, while Asaccharobacter and Acetanaerobacterium decreased, although these associations were not significant after correction for the total number of species detected in the study. Further analyses are underway investigating microbial gene and pathway level associations. 

Conclusion: We have shown that metformin has novel pharmacologies in the gut in T2D, including modulation of bile acid reabsorption from the gut lumen, secretion of gut hormones and alteration of the faecal microbiome. Understanding the potential novel effects of metformin will be important to identify potential novel approaches to control blood glucose and obesity.

Nothing to Disclose: AN, SM, SV, AK, GS, GL, DR, AS, JB, DR, DB, AN, DN

*Please take note of The Endocrine Society's News Embargo Policy at

Sources of Research Support: Study entirely funded by GlaxoSmithKline