The Leucine Trick That Boosts Your Mitochondria Also Shielded Lung Tumor Cells in the Lab

I keep a tub of leucine-spiked protein powder next to the coffee, and until this week I’d have told you the amino acid had exactly one job: telling my muscles to rebuild after a long run. So when the headline said leucine “powers up your cells”, I had the reaction the supplement aisle is counting on. More power? Pour it in. Then I read the paper behind the headline, and the lab that found the power boost had also watched the same biology shield lung tumor cells from a drug designed to starve them.

The work comes from Thorsten Hoppe’s group at the University of Cologne, published in Nature Cell Biology. Your mitochondria, the little furnaces that turn food into usable energy, are wrapped in an outer membrane studded with proteins, and some of the most important ones work as loading docks, because most of the proteins a mitochondrion runs on are made outside it and have to be hauled in. Cells constantly tag those surface proteins for the shredder and swap in fresh ones. Give worms or human cells leucine for 3 hours, though, and the shredding slows, the docks pile up, and respiration climbs. Isoleucine and valine, leucine’s two chemical siblings, didn’t do it.

Wait, why would more of a nutrient make a cell destroy less of its own hardware? I had to sit with that one. The answer runs through a sensor called GCN2, which normally raises the alarm when amino acids run short. Knock GCN2 down and the docks stabilize on their own, and once it’s gone, leucine has nothing left to add. Leucine also pulled a protein called SEL1L off the mitochondria, a helper in tagging those docks for destruction, while leaving the cell’s main SEL1L pool alone. With fewer tags, fewer docks got hauled away. A cell swimming in leucine reads it as a sign of plenty and stops cannibalizing its own engines for parts. None of this needed mTOR, the famous growth switch leucine is already known for: blocking the worm version of mTOR didn’t stop it. So one molecule turns out to work two separate levers.

The team also checked that the extra burn came from the docks and not from some side effect. When they knocked down TOMM40, a core piece of the gate proteins pass through on their way in, worms lost leucine’s boost to maximal respiration, and human cells given an import blocker called MitoBlock-12 didn’t get the boost either. Spared docks meant more cargo coming in, and more cargo meant a hotter furnace. It’s a satisfyingly physical chain, and I’ll admit I enjoyed following it right up until the researchers pointed it at cancer.

They took that same import blocker to three human non-small cell lung cancer lines. Two of them, H2030 and H1666, carried higher internal levels of branched-chain amino acids, leucine’s family, and less destruction tagging on their mitochondrial docks. MitoBlock-12 “substantially inhibited the proliferation” of the lower-BCAA line, H1437, and of the human kidney-derived cells used throughout the study. The two high-BCAA lines? It “only mildly affected” them. The trick that gives a cell more horsepower also looks like armor for a tumor cell facing a drug meant to choke its supply line.

That sent me back to older work, because leucine and tumors already had history. In a 2014 study at the University of Texas at Austin, Stephen Hursting’s lab implanted mouse pancreatic cancer cells into mice and fed some of them a diet that was 5 percent leucine. In lean mice on a 30 percent calorie-restricted diet, tumors averaged 104 cubic millimeters with leucine against 46 without, more than double. In overweight mice it was 367 against 230. In the lean mice the leucine group showed more mTOR signaling, and the authors closed by urging “caution regarding the clinical use of leucine supplementation for purposes of lean muscle enhancement.”

The wider record on branched-chain amino acids and cancer is a mess, and a 2023 review in Frontiers in Oncology says so plainly: “the role of BCAAs in cancer remains controversial.” That same review notes that leucine “can significantly promote cell growth by activating the mTORC1 pathway.” The Cologne result adds a second route that skips mTOR entirely, one by which a tumor rich in leucine’s family can keep its mitochondrial import machinery intact.

Even in healthy animals, the Cologne team’s data look like a dial, not an on-switch. Crank GCN2 up in ordinary worms and they laid fewer eggs. Knock it down in worms carrying a mutation that mimics a human disorder of BCAA processing and their broods shrank. Push the system too far either way and the animal pays, which is not a biology that rewards “more.”

None of this has been tested in people. Nobody has shown that a scoop of leucine changes anyone’s cancer risk, and leucine’s muscle-building case rests on separate evidence. Still, worm and cell data are enough to puncture the pitch that extra mitochondrial horsepower comes free, because tumor cells run on mitochondria too, and in these experiments the tumor lines richest in leucine’s family were the ones that shrugged off the drug.

So the tub stays, but it’s getting demoted. Leucine in ordinary food, the eggs and chicken and lentils, isn’t what anyone is questioning here. What I won’t do is stack isolated leucine on top of an already high-protein diet chasing cellular horsepower; I’ll take mine at the dose a plate delivers and let the hunger sensor keep doing its job.

Sources

1. Nature Cell Biology: Li et al., “Leucine inhibits degradation of outer mitochondrial membrane proteins to adapt mitochondrial respiration” (2025)
2. Cancer & Metabolism: Liu et al., “Leucine supplementation differentially enhances pancreatic cancer growth in lean and overweight mice” (2014)
3. Frontiers in Oncology: Xu et al., “Branched-chain amino acids catabolism and cancer progression: focus on therapeutic interventions” (2023)
4. ScienceDaily: “Leucine does more than build muscle. It powers up your cells” (2026)