One of the more difficult aspects of playing around with circadian rhythms is that there is quite a bit of individual variability that can affect outcomes.  People regularly ask me:

• When is the best time to eat?
• When should I exercise?
• How much exercise should I do and what type?
• How should I break up my meals?
• When should I block blue light?

The answer to all of these questions is pretty much the same: it depends.  There are both genetic and behavioral factors that go in to determining those things as well as things you can measure to determine if you should even care.  Believe me, it would be way easier for me to just pop out a cookie cutter response, but there really is no cookie cutter answer that’s correct.

A recent study is a fantastic example of this.  It took a comprehensive look at how melatonin interacts with insulin and blood glucose control.  It identified the general effect of melatonin on insulin secretion in humans as well as how a genetic variant in the melatonin receptor contributes to poor blood glucose control in people who have it.  It also hints at how eating late at night may promote leaky gut.  Let’s venture in to the study with a little background.

Image source

The effects of melatonin on insulin secretion

Epidemiological studies in humans have found an association between melatonin receptor B1(MTNR1B) gene variants and risk for future Type 2 diabetes.  But how is melatonin involved in glycemic control?

Mechanistic studies in mice have identified that melatonin may function to block insulin secretion by the pancreas, potentially as a way to prevent hypoglycemia during the overnight fast.  But the data isn’t completely clear, and melatonin has been shown to both promote and inhibit insulin secretion in rodent models.

To get a clearer picture on the effect of melatonin on glycemic control in humans, researchers performed a comprehensive study looking at mice, human cells in vitro, and an intervention trial comparing the effects of melatonin supplementation in humans homozygous for the MTNR1B risk variant(GG) with those with the wild type(CC).

The results of the study identified some general effects of melatonin on insulin and blood glucose regulation in humans that were independent of genes.  First, melatonin blocks insulin secretion by the pancreas by binding to MTNR1B on islet cells.   Furthermore, melatonin improved insulin sensitivity.

Image source

Melatonin, MTNR1B variant, and insulin secretion

While the effects of melatonin were consistent across both groups, those homozygous for the risk allele saw a greater inhibitory response of melatonin on insulin secretion.  The in vitro data determined that this was due to a greater number of MTNR1B receptors in those with the risk allele versus those with the wild type.

What this translated in to was a doubling of islet MTNR1B expression in those with 1 copy of the risk allele(CG) and a quadrupling of MTNR1B expression in those with 2 copies(GG) when compared to those with 2 copies of the wild type(CC).  These differences allowed melatonin to have a greater inhibitory effect on insulin secretion in people with one or more copies of the risk variant.

It’s important to point out that the intervention trial used supplemental melatonin at 4mg/day for 3 months which drove melatonin levels far beyond what you’d see in endogenously.  However, even at baseline, there was a significant difference between the 2 groups(CC vs GG) in the oral glucose tolerance test as well as their immediate insulin response to glucose challenge.

In those with 2 copies of the risk allele, glucose levels were significantly higher at baseline, 30 min, 60 min, and 90 min of the initial oral glucose tolerance test when compared to controls(CC).  They also saw a delayed insulin response and a higher insulin peak.

What this means…

While there appears to be a consistent time of day effect on insulin response mediated by melatonin in humans, those carrying one or more copy of the risk variant of MTNR1B(CG or GG) are potentially at a greater risk for this spiraling in to pathology later down the road.  As such, higher carbohydrate intake close to bedtime probably isn’t great for humans in general, but likely worse for those carrying the risk allele.

Carrying one of these variants may also impact Type 2 diabetes risk via another mechanism: by increasing intestinal permeability, aka leaky gut.  In a recent blog I discussed a study showing hyperglycemia causing leaky gut in mice.  This effect was due to increased GLUT2 mediated glucose metabolism in epithelial cells and was blocked by insulin.

Therefore, in people carrying a MTNR1B risk variant who have a greater inhibitory effect of melatonin on insulin secretion, spiking blood glucose levels late at night may cause greater levels of intestinal permeability than those with the wild type via increased GLUT2 glucose transport and metabolism.  Over the long term, this can cause prolonged periods of endotoxemia that promote chronic inflammation and pathology consistent with Type 2 diabetes.  Of course, other factors including physical activity, smoking status, caloric load, light exposure, and sleep will also have a significant impact.

For those who have had their genes sequenced via 23andme, you can determine your MTNR1B status by searching your raw data for “rs10830963”.  Those with CG or GG status may want to reconsider carb-heavy late night snacks, particularly if you are sedentary, overweight, or are generally in a calorie surplus.

What it doesn’t mean…

While this study gives us some very interesting ways that eating late at night can contribute to Type 2 diabetes risk, it’s important to look at this from a big picture perspective.  This doesn’t mean that you can live an otherwise crappy lifestyle and stop eating earlier in the day to hedge your bets against it.

It also doesn’t mean that having one or more copies of the risk variant dooms you to develop Type 2 diabetes.  While genetic variants can give us indications as to individual susceptibilities to disease, they are hardly life sentences written in stone.  To date, we have no idea as to the contribution of the MTNR1B variant to Type 2 diabetes.

Disease risk is a composite of genes and environment, and we have yet to tease out which factors play the biggest role in a disease as complicated as Type 2 diabetes.  Ultimately, I believe the best practice is to arm yourself with as many beneficial factors as possible to limit your risk of Type 2 diabetes.  This includes but is not limited to maintaining a healthy weight through regular physical activity and adequate but not excessive calorie consumption, proper sleep, not smoking, limiting alcohol intake, increasing vegetable intake, and limiting processed food consumption.

Conclusion

Multiple lines of evidence indicate that food consumed late at night is more likely to be stored as fat than food eaten earlier in the day.  Data also suggests that people who consume a greater percentage of their food closer to dim light melatonin onset have a greater BMI and are more likely to be obese than those that consume most of their calories earlier.

Another recent study found that people with Type 2 diabetes had substantially better glycemic control when they consumed a divided dinner at 1800h and 2100h versus consuming their entire dinner at 2100h.  A companion study found that healthy young women doing the same saw a similar but smaller effect.  This coupled with the data on the effect of melatonin on insulin secretion indicates that late night feeding may increase your risk for Type 2 diabetes.  Particularly if you have one or more copies of the MTNR1B risk variant.

I think it’s important to state that the determination of risk is multifactorial.  I suspect that a lean, healthy, and active individual is at a lower risk for Type 2 diabetes than someone who is none of those things, even if they have 1 or more copies of the MTNR1B risk variant.  This may be at least partially attributed to higher levels of intestinal permeability in obese individuals when compared to lean ones.

I think intestinal permeability may play a fairly big role over the long term, and calorie restriction has been shown to reduce intestinal permeability in obese women.  However, I have very little faith in the ability of most people to practice calorie restriction for prolonged periods of time, particularly at the levels(800cals/day) indicated in this study.

Eating most of your food earlier in the day or eating your last meal further from bedtime may be a viable tool for most people who simply won’t practice calorie restriction.  It may also simply be good general practice for all humans, particularly if you carry one or more copies of the MTNR1B risk variant.  Further research is needed to determine how much of an effect it may have on reducing Type 2 diabetes risk.