The thing that has quietly bothered me about the whole GLP-1 gold rush is the muscle. Everyone I know on semaglutide talks about the pounds. Almost nobody talks about what those pounds are made of, and the honest answer, the one you will not find in the glossy before-and-after marketing, is that a real chunk of it isn’t fat at all.

So a Berkeley lab going the other direction caught me. Body weight, as senior author Anders Näär puts it, “responds to two levers: taking in fewer calories, or spending more energy. GLP-1s work almost entirely on the first, so we went after the second.” And the compound they reached for to pull that second lever was first cooked up in the 1970s and left on a shelf, which is a strange place to go looking for the future.

On August 21, UC Berkeley researchers published in Science Advances that an old compound called TOFA, 5-tetradecyloxy-2-furoic acid, did something worth stopping for in obese mice. The animals, fed a high-fat diet into obesity, burned up to 18 percent more energy with no change in physical activity or body temperature. They lost weight that came almost entirely from fat, with no significant loss of lean muscle mass and, notably, no drop in how much they ate. Their insulin sensitivity improved, their triglycerides came down, and their fatty liver got better. Read that list again: this is not an appetite drug.

ENERGY BURNED
18 percentmore, in obese mice
TOFA raised energy expenditure with no change in activity or body temperature. Source: Lee et al., Science Advances, 2026

And that is why the muscle matters. When you lose weight by eating less, your body does not politely take it all from your fat stores. A review in Circulation and a network meta-analysis of body-composition trials put roughly 25 to 40 percent of the weight lost on these drugs in the lean-mass column. Muscle. The engine you use to stand up from a chair at 75. The eat-less lever works, and it takes the muscle along for the ride.

MUSCLE IN THE WEIGHT LOST (percent)
of weight lost on GLP-1 drugs2540
Share of GLP-1 weight loss that comes off as lean mass, not fat. Source: Circulation, 2025; body-composition network meta-analysis

So how does one small molecule both stop you from building fat and make you burn more of it? Those sound like opposite jobs, and the answer is where the biology got its hooks in me.

TOFA blocks acetyl-CoA carboxylase, or ACC, the enzyme your body uses to spin sugar and spare carbons into new fatty acids. Throttle that, and the liver makes less new fat and cholesterol. That much has been known for decades, and it is exactly why TOFA sat around as a research-grade ACC inhibitor after the 1970s. But this compound also partially switches on two nuclear receptors, PPAR-alpha and PPAR-delta, which are the cell’s “start burning lipids for fuel” switches. So it clamps the fat-building tap with one hand while flipping the fat-burning switch with the other. First author Justin Lee said TOFA “appears to engage a coordinated metabolic response” and “is not simply blocking lipid synthesis.” That coordination is, I think, why the muscle stays: you are not starving the body into eating itself, you are telling it to spend more of what it already carries.

I want to say the unglamorous parts out loud, because the coverage around this has been very “beyond GLP-1” and very light on “in mice, so far.” This is a mouse study, and metabolism is precisely the field where gorgeous rodent results go to die in humans. Every number above lives in a mouse on a high-fat diet. TOFA has never been given to a person for weight loss.

There is also a specific ghost in this drug’s closet. The ACC-inhibitor idea has been to the clinic before and washed out: earlier compounds in this class reached mid-stage human trials and were never approved, largely because they raised triglycerides, which is itself a cardiovascular red flag (a 2025 meta-analysis pegged the hypertriglyceridemia risk on ACC-inhibitor monotherapy at more than tenfold over placebo). The Berkeley team’s whole pitch is that TOFA is the exception, that it lowers triglycerides instead of raising them. That would be a genuine distinction if it holds. But “if it holds in humans” is carrying enormous weight in that sentence, and the evidence that it holds is, right now, a mouse.

Then there is the money, which I always want named. The researchers did not just publish and walk away. Näär, Lee, and a colleague, Prabha Ibrahim, co-founded a startup, ReRx Therapeutics, and hold equity in it. That is disclosed, to their credit, and it is completely normal for academic metabolism labs. It is also a reason to read the enthusiasm through a slightly narrowed eye. The people telling you this old molecule is the promising new thing also own a piece of the promise. The core work was funded by UC Berkeley discretionary money with help from the UCSF Liver Center and a University of Michigan phenotyping core, not, at this stage, by a pharma sponsor.

So what actually changes today? For anyone standing in a pharmacy, nothing. There is no pill, no trial to join, no timeline that means anything yet. What this really is, honestly, is a strong scientific argument that the energy-out lever is worth pulling, and a reminder that the appetite-suppression model everyone is racing to sell is only half the equation, the half that happens to be easiest to package and sell.

Would I take an untested fifty-year-old ACC inhibitor because a mouse did well on it? No, and not only because it isn’t available. I have watched too many clean mouse-metabolism stories evaporate at the human trial. But the muscle point lands on me personally: if I were on a GLP-1 right now, I would not sit around waiting for TOFA to rescue my lean mass sometime in the 2030s. I would be lifting something heavy twice a week and eating the protein tonight, because that is the muscle-preserving move that already works in people, and it does not need a Series A.

Sources

  1. Science Advances – Lee, Näär et al., “A multi-functional oral small molecule targeting energy and lipid metabolism to treat obesity and related metabolic disorders” (Aug. 21, 2026)
  2. Berkeley News – “A promising new weight loss and diabetes treatment helps burn fat while keeping muscle” (UC Berkeley press release, Aug. 21, 2026)
  3. ScienceDaily – “Experimental compound helps burn fat without muscle loss” (Aug. 23, 2026)
  4. Circulation – “Muscle Mass and GLP-1 Receptor Agonists: Adaptive or Maladaptive Response to Weight Loss?”
  5. ScienceDirect – “Effect of GLP-1 receptor agonists and co-agonists on body composition: systematic review and network meta-analysis”
  6. Pharmaceuticals – “Acetyl-CoA Carboxylase Inhibitors for Nonalcoholic Fatty Liver Disease: A Systematic Review and Meta-Analysis of Randomized Controlled Trials” (2025)