Nutrition and hormones: how blood sugar also moves estrogen and cortisol

A group around Stephen Plymate incubated a human liver cell line with insulin for 72 hours. SHBG production fell by around 28 percent, and insulin additionally blocked the SHBG raising effect of estradiol. The authors note themselves that insulin also increased the cell number, which makes the reading harder. For you this means: the direction is clear in the lab, but it comes from a cell line in a petri dish and not from your everyday life.

David Selva and colleagues gave glucose and fructose to transgenic mice carrying the human SHBG gene, and to liver cells. Both sugars lowered SHBG production in a dose dependent way, and they did so through fat synthesis in the liver: the palmitate that forms there can downregulate HNF-4alpha, one of the most important regulators of the SHBG gene. For you this could mean: SHBG may say more about the liver than about the gonad. So far that has been shown in liver cells and in the mouse model, with glucose and fructose about equally strong. Whether fructose plays a role of its own beyond that in humans is not yet answered.

Eric Ding and colleagues compared 359 women with newly diagnosed type 2 diabetes against 359 controls and repeated the analysis in 170 men. In the highest SHBG quartile, diabetes risk in the women was clearly lower than in the lowest, with an odds ratio of 0.09. That was a case control analysis inside running cohorts. For you this means: a low SHBG can be an early hint of a disturbance in glucose metabolism, often before fasting blood sugar stands out. How it compares directly against established markers is not settled by this.

John Perry and colleagues used a gene variant near the SHBG gene as a natural experiment across 86,138 people. The SHBG raising allele went along with a lower risk of type 2 diabetes, and the observed effect matched the expected one almost exactly. For you this means: a mere correlation turns into a working hypothesis about cause that is worth taking seriously. Two limitations belong with it. The effect of the gene variant is small, the odds ratio was 0.94 per allele. And the same paper found no association of the variant with insulin secretion or insulin resistance. Through which route SHBG exerts its influence here is therefore still open.

John Nestler and Daniela Jakubowicz gave metformin to eleven of 24 overweight women with PCOS for four to eight weeks, the others received placebo. On metformin the insulin area under the curve nearly halved, LH and free testosterone fell clearly, and SHBG rose from 0.8 to 2.3 micrograms per deciliter. For you this means: when the insulin signal comes down, both ends of the chain can move together.

Kate Marsh and colleagues compared a low glycemic diet against a conventional healthy diet with identical macronutrients in overweight women with PCOS. At similar weight loss, insulin sensitivity improved more in the low glycemic group, and more women reported a more regular cycle, 95 against 63 percent. Two limitations belong with it: 49 percent dropped out in both groups, so the numbers apply only to those who saw it through. And the strongest improvement was seen in the women who were also prescribed metformin. The diet alone therefore does not carry the result. For you this means: how fast your blood sugar rises can move the cycle too.

Fozia Ahmed and colleagues examined aromatase in subcutaneous fat of men with obesity or type 2 diabetes. It was more strongly expressed and correlated with measures of obesity, high glucose values and insulin resistance, and in the tissue experiment estradiol lowered insulin dependent glucose uptake. For you this means: the conversion plant runs harder with more fat mass, and its product acts locally.

Giovanni Corona and colleagues evaluated 24 papers that had measured weight loss and sex hormones in men. On a calorie reduced diet, total testosterone rose by 2.87 nmol/L, after bariatric surgery by 8.73 nmol/L, and in both cases estradiol fell. The authors note themselves that only few randomized studies were available and that their results conflicted with one another. For you this means: fat mass itself can shape the hormone situation.

Philip Cryer summarized at which value the body answers falling blood sugar. The thresholds are well reproducible. They are movable as well, but that has been studied mainly in people with insulin dependent diabetes: after repeated lows, the threshold for noticing them can slide downward. Whether the same holds for metabolically healthy people with frequent afternoon dips has, to my knowledge, not been measured directly. For you one thing is interesting: cortisol can rise before you notice anything at all.

Patrick Wyatt, Sarah Berry and colleagues had 1,070 healthy people wear sensors and log more than 80,000 meals. The dip 2 to 3 hours after eating predicted later eating behavior better than the height of the peak: more hunger and more energy taken in over 24 hours. Part of the context is that the paper comes largely from a company that sells sensor supported nutrition programs. For you this means: the interesting number sits in the valley, not only on the summit.

Jonathan Purnell and colleagues measured the cortisol production rate over 24 hours and compared it with fat in the abdominal cavity, subcutaneous fat and insulin sensitivity. A higher cortisol production went along with more intra abdominal fat and lower insulin sensitivity, but not with subcutaneous fat. For you this means: the deep fat depot appears to have a relationship of its own with the stress axis.

Elissa Epel and colleagues put women with high and low waist to hip ratios under controlled stress on four days in a row. The women with a central fat pattern released more cortisol, and the lean ones among them had still not got used to it by day three. For you this means: this pattern does not hang on weight alone.

Samantha Ervin and Matthew Redinbo tested 35 bacterial beta-glucuronidases from the gut for whether they can release conjugated estrone and estradiol again. Certain enzyme classes could do it, the reactivation could be inhibited, and the matching inhibitor still did not reduce tumour development in the mouse model. For you this means: the estrobolome exists in a biochemically demonstrable way, a clinical benefit from it has not been shown so far.

Barry Goldin and colleagues studied 20 premenopausal women four times over more than a year. The vegetarian women ate 28 instead of 12 grams of fiber, excreted more estrogen in stool, had lower plasma levels and lower beta-glucuronidase activity. For you this means: here fiber, gut enzyme and blood value hang together in one study for the first time.

David Rose and colleagues raised fiber intake from around 15 to 30 grams a day, once with wheat bran, once with oat bran, once with corn bran. After two months, estrone and estradiol were lower only in the wheat bran group, not on oat or corn bran. There was no control group. That is a study result in 62 women and not a claim about the effect of a food, and there is no authorised health claim for it. For you this means: fiber is not one single substance, the type of fiber could have a say.

Cynthia Thomson and colleagues gave 130 women who were taking tamoxifen either DIM, a breakdown product from cruciferous vegetables, or placebo in a double blind design. Tamoxifen is a prescription only medicine used in breast cancer aftercare. The ratio of estrogen metabolites in urine shifted clearly and SHBG rose by 25 nmol/L. The primary endpoint did not move: breast density changed neither on mammography nor on MRI.

And now the finding that outweighs everything else. Blood levels of the tamoxifen metabolites fell clearly, among them endoxifen, the metabolite through which the effect of tamoxifen comes about. The authors write themselves that it is open whether this weakens the protection tamoxifen provides.

Anne Loucks and Jean Thuma set energy availability precisely over five days in menstruating women and then drew blood every ten minutes for 24 hours. At 30 kilocalories per kilogram of fat free mass per day, LH pulsatility stayed unchanged, below that the pulse frequency fell. For context: energy availability was set there through supervised exercise, in 29 sedentary young women, over five days. It does not transfer one to one to intermittent fasting. For you this means: the response is not a slide, it has a threshold, and cortisol moved along with it.

Stephen Spaulding and colleagues put participants on randomized 800 kilocalorie diets with differing carbohydrate content. Without carbohydrates, T3 fell by 47 percent over two weeks, with at least 50 grams it stayed unchanged at the same calorie count. For you this means: what the calories are made of might have a say in thyroid conversion.