A Rare Gene Fault Is Linked to Less Belly Fat and Less Diabetes

Inside your liver cells, two proteins lock together and sit on the machinery that burns fat. Their job is thrift. When fuel runs short the clamp loosens, a transcription factor slips through into the nucleus, and the cell sets about building mitochondria and the acidic recycling sacs called lysosomes, all of it geared to break down whatever energy happens to be lying about. Roughly one person in 7,000 carries a broken copy of one side of that clamp, and a new analysis of 1,032,116 people suggests they are better off for it.

The gene is FNIP1, and until now nobody had tied it to metabolic health in the general population. Carriers of a single damaged copy tend to have less fat packed around their organs, less fat inside the liver, a higher proportion of lean mass and lower blood sugar.

Getting there took an odd choice of yardstick. Rather than chase genes for obesity or diabetes head on, the Regeneron-led team went after a cheap and rather old-fashioned blood measure, the ratio of triglycerides to the high-density lipoprotein that most of us call good cholesterol. It is the sort of number that sits unremarked near the bottom of a routine lipid panel. As a readout of how a body is handling energy, though, it turns out to be a rather good one, tracking with visceral fat, liver fat, insulin resistance and, in these cohorts, with later cases of type 2 diabetes, heart attack and cirrhosis. Other genetic studies of insulin resistance have leaned on the same proxy.

A Metabolic Brake, Loosened From Birth

Sequence enough people and even the rarest variants start to surface. Across 11 cohorts on three continents, 59 genes emerged whose rare coding variants shifted that ratio, 23 of them encoding targets for drugs already approved or in clinical development, and the whole set held its shape when the team ran the analysis again in 114,942 participants from the All of Us Research Program.

FNIP1 was the standout. Ultra-rare variants that truncate the protein, 86 distinct ones scattered across the gene, were associated with a ratio about 0.5 standard deviations below average, among the largest effects in the entire scan. Follow the carriers into their medical records and the pattern holds: less fat in the liver, lower long-term blood sugar, a lower ratio of belly fat to hip and thigh fat, and around 60% lower odds of a composite outcome covering coronary artery disease, type 2 diabetes, fatty liver disease and cirrhosis. That last figure comes from 227,636 cases set against 265,114 controls, and its confidence interval is wide.

“The higher this ratio is, the higher the risk of metabolic disease,” says Luca Lotta, a human geneticist at the Regeneron Genetics Center in Tarrytown, New York, and one of the study’s senior authors, quoted in Nature’s news coverage of the paper. Kari North, a genetic epidemiologist at UTHealth Houston School of Public Health who was not involved in the work, told Nature the arithmetic of the search is the striking part: “They studied a million people to find this very, very rare variant.”

What FNIP1 does is hold a leash. The protein it makes latches onto another called folliculin, and the pair hold a set of transcription factors out of the nucleus, keeping the genes for mitochondria and lysosomes quiet and the cell in energy-saving mode. Silence FNIP1 in cultured human liver cells (the team used a small interfering RNA, knocking the message down by more than 90%) and the lysosomal and fat-breakdown genes light up. That is the leash slackening. It is at least a plausible drug target, too, since liver-targeted RNA therapies already reach patients for other genes, so silencing FNIP1 in liver cells alone, rather than everywhere at once, might sidestep the trouble of losing the pathway body-wide.

What the Mice Would Not Do

Plausible, not proven. When the team knocked out the mouse version of FNIP1 in liver on its own, the animals put on weight on a high-fat, high-fructose diet exactly as the controls did; only knocking out folliculin, or FNIP1 together with its sister gene, protected them. The researchers put that down to differences between the species, which is a fair argument to make. Nature’s news write-up compressed the whole business into the team doing the same thing in mice and it working, which is not quite what the mouse experiments show.

The Price of Losing Folliculin

There are other reasons for restraint. Interfering with this pathway is no free lunch in people: those born with one broken copy of folliculin, which causes Birt-Hogg-Dubé syndrome, run an 18-fold higher risk of a collapsed lung and a 9-fold higher risk of kidney cancer, and children who inherit two broken copies of FNIP1 develop a syndrome of immune failure and thickened heart muscle. The mouse work, meanwhile, used male animals only, a limit the authors flag themselves.

Whether any of this becomes a medicine is a separate question, and genetics alone will not answer it. The human evidence is associational: carriers were born with their variants and have lived alongside them for decades, which is not the same thing as a drug given to an adult for a year or two. Regeneron funded the work, many of the authors hold company stock or options, and five are named on pending patent applications covering FNIP1 and folliculin genetics. Worth knowing, that.

Lotta’s own reading of it is evolutionary: a brake on fat burning made obvious sense when calories were scarce and running short was a real way to die. Whether releasing that brake deliberately, in one organ, in a grown adult, does anything like what being born with it already loosened appears to do is the experiment nobody has run yet.

See also: What Is Psyncretin: The AI-Designed Weight-Loss Molecule Your Muscles Would Manufacture

Frequently Asked Questions