Thursday, July 30, 2020

New Type of Diabetes Medication

 A new type of diabetes medication is being developed, and it sounds great. However, although it works in rodents, human cells, and dogs, there has not yet been a clinical trial of the drug, so it won't be available for some time.

The drug has the decidedly unmemorable name SRI-37330, and it works in part by decreasing glucose output from the liver. Glucose levels are controlled primarily by two hormones: insulin, which lowers blood glucose (BG) levels, and glucagon, which raises them. When you have type 2 diabetes, you not only don't produce enough insulin, but you produce too much glucagon. The glucagon stimulates the liver to produce glucose (gluconeogenesis) and release it into the bloodstream, even if your BG level isn't low.

SRI-37330 inhibits glucagon secretion and function, reduces glucose production in the liver, lowers serum triglyceride levels, and also reduces fatty liver. It does this by inhibiting a protein called TXNIP (thioredoxin-inhibiting protein), which has been shown to be elevated in people with diabetes and to have detrimental effects on islet function. The islets in the pancreas are where the beta cells produce insulin.

The drug has no effect on insulin secretion or function. It has no effect on glucose uptake by muscle or white adipose tissue. It has no effect on the secretion of glucose in urine.

SRI-37330 has no effect on glucagon secretion when BG levels are low, which reduces the chance that it could cause hypoglycemia. And it has no effect on body weight. But in obese, diabetic mice, it reduced BG levels significantly within 3 days, and they eventually returned to normal.

All these effects sound wonderful, and even if they don't turn out to be as dramatic in humans as in animals, it would be nice to have another drug with a different mechanism of action, as we're all different, and a drug that works wonderfully for one person might not be so great for another.

The lead author of this paper, Anath Salev, kindly sent me a copy of the full text. An interview with her can be seen here.


















Saturday, July 18, 2020

Metformin and Energy

If you're trying to lose weight, there's a good chance that medical people will tell you that the important factor is that energy coming in is less than energy going out.

This is true.

But most people equate "energy out" with exercise. If you want to lose weight, they suggest that you exercise more. One problem with this advice is that for a lot of people, exercising just increases their appetite and makes them eat more, and so "energy in" goes up along with "energy out" and there's no loss of weight.

But exercise isn't the only part of "energy out." Any physiological process that wastes energy and converts it to heat will mean less energy is available to create fat. One example of this is the so-called biochemical futile cycles in which a chemical reaction converts A to B but then the body just converts B back to A. Because no reaction in the body is 100% efficient, each time the cycle occurs, a little bit of energy is lost as heat.

There are substances called uncoupling agents that increase the wasting of energy, and they were used by patients some years ago. The only problem is that one side effect of these compounds is death, so they weren't very popular.

Another part of "energy out" is any food lost in feces.  For example, if you eat a food that is poorly digested, much of the energy in the food is lost. An example of this is fiber. There's energy in fiber, and some microorganisms can digest it and convert it to useful substances. Humans can't. But sometimes bacteria in the lower bowel digest the fiber and produce substances like butyrate that we can digest.

So just measuring what you eat isn't enough. You need to know how much of what you eat is digested. In animal nutrition this is known as TDN, or total digestible nutrients.

Recently, research has shown that the diabetes drug metformin can increase the amount of sugar that is secreted in feces. This would both lower blood glucose levels and mean fewer calories were metabolized so they couldn't contribute to weight. (In animal nutrition, the goal is usually to get the animal to put on weight as quickly as possible, but the principles are the same.)

When I was diagnosed with type 2 in 1996, no one knew how metformin worked, but I was told it was better than the sulfonylureas (they didn't say why), which were the only other oral drugs available at the time, so I chose it and have been taking it ever since.

Eventually they discovered that metformin causes an increase in a molecule called AMPK, which is an energy sensor, and the recently discovered increased glucose excretion is a second mechanism. I suspect there are others that haven't yet been discovered.

The increased secretion of glucose into stool is analogous to the increased secretion of glucose into urine caused by the drugs called SGLT-2 inhibitors, for example, Invokana and Jardiance. In both cases, you're wasting calories by excreting them rather than converting them to heat.

There's still a lot we don't know about how metformin works, but we do know that it helps a lot of people with type 2 diabetes, and it's relatively inexpensive. On some drug plans, it's free.

Recently, the slow-release form of metformin was shown to contain a substance that increases rates of cancer. So far, only some brands have found this contaminant, but it's possible more will be added to the list when they are tested. If you want to see if your metformin is on the list, you can check here. This list was compiled on July 15, and there might be others added later.


Sunday, July 5, 2020

YMMV revisited

Some time ago, I wrote about "YMMV," or "your mileage may vary," the concept that sometimes what works well for one person doesn't work for another. The diabetes community was aware of that variation, but the evidence was annectotal and no one had studied it formally until a study in an Israeli cohort.

Now someone has done a similar study in mostly midwestern Americans without diabetes. The results are similar. Look at Figure 2A to see how people's blood glucose can vary after the same meal, in this case a bagel with cream cheese. One person's glucose levels went up to 190 and another to only 110. Note that these are people who haven't been diagnosed with diabetes. Whether they're on their way to diabetes is not known.

The researchers took 72 parameters into account in developing their model, which is designed to predict  how someone will react to a particular meal. You'll have to go to the Supplemental Content section to see these parameters, which I won't try to list here. They include species of 23 bacteria in the microbiome as well as usual things like nutrients, BMI, age, and CGM results.

The model is not yet available to the general public.