New Findings About What Metformin Really Does

As many of my readers know, there is no requirement that the companies that sell pharmaceutical drugs provide an accurate explanation of what it is that their drugs do or of how they do it. All that they have to prove is that the drug has an impact on some measurable phenomenon. The company may claim that a drug functions using a mechanism that is later proven to be untrue. This has been the case with the SSRI drugs which it turned out actually work by remodelling the nerves in the hippocampus, NOT by changing levels of serotonin.

Metformin, which has been used for decades, is another drug whose effect is well understood--it lowers blood sugar and reduces the amount of insulin needed to lower blood sugar. This has been interpreted to mean that it lowers insulin resistance.

But new findings are calling this into question, as we discover that metformin may actually be stimulating insulin release or blocking the liver's release of glucose rather than impacting insulin resistant cell receptors.

The first finding is one I stumbled over recently, one which seems to have gone unnoticed by the medical press. It is that metformin appears to boost GLP-1 levels. GLP-1 is an incretin hormone secreted in the gut which stimulates the beta cell to secrete insulin in the presence of high blood sugars. GLP-1 may also lower glucagon production at the same time. While Byetta and Januvia are higly promoted as being incretin drugs, some little known research suggests that metformin may also raise the level of GLP-1 in the body.

Enhanced secretion of glucagon-like peptide 1 by biguanide compounds. Yasuda N et al. Biochem Biophys Res Commun. 2002 Nov 15;298(5):779-84.

This was old news, but it may partially explain some of the stomach symptoms people experience with metformin. GLP-1 stops or slows stomach emptying and that often causes nausea--a side effect many people experience with metformin.

Meanwhile, some brand new findings are making metformin's function even more intriguing.

A mouse study published on May 15, 2009 suggests that Metformin lowers blood sugar by directly stimulating a gene in the liver which is not responding to insulin the way it should. When this gene is stimulated, the liver stops producing glucose. So rather than improving insulin sensitivity, what metformin may actually be doing is bypassing a broken insulin signaling system and doing the job itself that insulin should have done--making the liver stop secreting glucose. In this case metformin is not increasing insulin sensitivity, it is replacing insulin.

Metformin and Insulin Suppress Hepatic Gluconeogenesis through Phosphorylation of CREB Binding Protein Ling He et al,, Cell Volume 137, Issue 4, 635-646, 15 May 2009. doi:10.1016/j.cell.2009.03.016

You'll find this study explained in layman's language in today's edition of Diabetes in Control:

New Information on how Metformin works.

Metformin may also prevent beta cell death. Another new study published in the May 2009 issue of Diabetologia examined pancreases of autopsied organ donors with and without Type 2 diabetes and found increased evidence of beta cell apoptosis (cell suicide) in the beta cells of people with Type 2. This reconfirms the finding of an earlier pancreas autopsy study.

What is interesting, though, is that this study also found that

Metformin ameliorated autophagy alterations in diabetic beta cells and beta cells exposed to NEFA [nonesterified fatty acides], a process associated with normalisation of LAMP2 expression.

This suggests that metformin might work to prevent beta cell death due to apoptosis.

Autophagy in human type 2 diabetes pancreatic beta cells M. Massini et al. Diabetologia Volume 52, Number 6 / June, 2009, DOI 10.1007/s00125-009-1347-2.Pages 1083-1086

I was sent the full text of this article by a subscriber and it turns out that they are not talking about the effect of metformin on the pancreas when taken before death. Instead what they did here was rescue still living cells from dead organ donors, most of them dead of heart attack, separate the beta cells and culture them. Then they infused the cultures with the fats and a solution of metformin. The metformin prevented some genetic expression in the presence of the fat, but it is difficult to know whether this has much relevance to what goes on in your living pancreas. Or to what extent, being bathed in fats reflects what damages islets in diabetes.

Still it is interesting to see people attempt to find out the impact of metformin on gene expression, which is what is going on in both these studies.

Anecdotal Musings about Metformin

I am particularly interested in metformin of late, because I stopped taking it for a year and noticed that while my blood sugars stayed pretty much constant, I started gaining weight eating the diet on which I had, until then, successfully maintained a significant weight loss since 2003.

I went back on metformin in early April and, sure enough, seven weeks later my weight has dropped some 6 lbs while I have been eating the identical diet--and insulin doses--on which I had gained 6 lbs the previous month.

I am not insulin resistant. I usually use a tiny dose of insulin--2-3 units to cover up to about 60 g of carbs. (I don't eat 60 grams of carbs very often, but if I do I can cover it with 3 units.) When I am taking metformin my insulin dose might drop from 3 units to 2 units, so while that represents a large percentage drop, it but a very tiny drop in the actual amount of insulin used and the kind of dose an insulin sensitive person might use.

But metformin does appear to do something that makes it extremely hard for my body to gain weight. When I first started injecting insulin a few years ago, when I was still taking 1500 mg a day of metformin, I actually LOST weight using insulin. As soon as I stopped metformin I started packing on weight whenever I used insulin, even if my blood sugars were completely flat.

One thing I noticed was that metformin also changes the way my body gains and loses the weight associated with glycogen when I cut my carbs to a ketogenic level, (a topic discussed in detail HERE). If I am not using metformin and start eating a very low carb, ketogenic diet, I will drop 3 or 4 lbs of water weight within days. If I am taking metformin, I won't. On the other end, if I go over the ketogenic boundary when I am taking metformin I won't pack on that sudden 3 lbs water weight gain.

My guess is that metformin somehow interferes with the creation and/or burning of glycogen which may make the body more likely to burn fats instead of stored glycogen. I have been told that some body builders use metformin to hasten the process of getting into a ketogenic state.

Another observation which repeats what I found in the past is that while I am taking metformin, I do not need to take my blood pressure medication to keep my blood pressure in the normal range. When I stop metformin, I do.

The worst side effect I experience with metformin is exhaustion. While taking it I tend to drop off to sleep around 9:30 whereas when I stopped metformin I was much more energetic and always stayed up to 11. My muscles tire more easily too. These side effects were a major reason I stopped taking it. It also gives me a heart-burn like pain (without reflux) which is annoying though my doctor did tests which suggest I don't have an ulcer, though that is what it feels like.

However, gaining 6 lbs in one month eating about 1600 calories a day is also annoying, and I have been battling weight gain all year since I stopped the metformin. I spent more than 6 months of the last 12 eating a very low carb diet and controlling calories and still found myself at the highest weight I'd been in 6 years. So for now, I'm back to taking the metformin.

Once I get my weight stabilized I'm going to experiment to see if I can find the very lowest dose that is still effective for weight purposes. In the past, I found that I had to take 1500 mg to see any impact on my blood sugars, and then it only lowered them about 10-20 mg/dl per meal (or reduced the insulin dose by that 1 unit.)

I am hoping that less might still help prevent weight gain, but this new study that suggests that metformin is working by stimulating a gene that stops the production of glucose makes me wonder if I won't see weight prevention effects until I see that slight lowering of post-meal blood sugars that suggests that my liver's glucose production has been stopped.

What's your experience with metformin and weight gain, loss, or maintenance? Does your experience cast any light on any of these new findings?