In a lab in South San Francisco, someone is watching a nerve fiber that should be falling apart. It isn't. That stubborn refusal to disintegrate is, more or less, the entire thesis of Nura Bio. Cut an axon, bruise a brain, flood the body with chemotherapy, and a molecular switch flips on inside the nerve telling it to self-destruct. Nura Bio's job is to keep that switch off.
This is a biotech with one unfashionable conviction: that the best time to treat a neurological disease is before the wreckage. Most of the field chases symptoms or clears debris after neurons are already gone. Nura Bio wants to get there earlier, at the moment a nerve decides whether to live.
Nerves have a self-destruct button
For more than a century, scientists had noticed that when a nerve is severed, the part downstream of the cut doesn't just sit there. It actively dismantles itself in a tidy, programmed way. The phenomenon got a name - Wallerian degeneration - back in 1850, when Augustus Waller cut frog nerves and watched what happened. For a long time it was treated as a curiosity rather than a target.
Then came SARM1, an enzyme that turns out to be the on-switch for that whole demolition process. When a nerve is injured or stressed, SARM1 wakes up and starts chewing through NAD+, a molecule axons need to stay alive. Run the NAD+ down far enough and the axon collapses. SARM1 isn't a side effect of nerve damage. It is closer to the cause.
Axon degeneration is the early, shared first domino across a startling range of neurological diseases - and SARM1 is the finger that tips it.
Here is the inconvenient, and rather useful, part. That same first domino shows up in conditions that otherwise have nothing in common: ALS, multiple sclerosis, traumatic brain injury, the burning nerve pain that follows chemotherapy, glaucoma. Different diseases, same early failure. Block SARM1, the thinking goes, and you might protect nerves across all of them. It is the kind of idea that sounds too neat to be real, which is precisely why it took serious science to make it credible.
A venture firm, two scientists, and a hunch
Nura Bio was conceived in 2018 by The Column Group, a San Francisco venture firm with a habit of building companies around hard biology rather than quick wins. The scientific founders - Marc Freeman and Steven McKnight - brought the academic firepower on axon biology and metabolism. The bet was specific: that SARM1 could be drugged with a small molecule you could take by mouth, and that the molecule could cross the blood-brain barrier, the wall that keeps most promising neuro drugs out of the brain.
Brain-penetrant, oral, and aimed at a target nobody had successfully drugged. None of those three are easy. All three at once is the sort of thing that either fails quietly or rewrites a treatment category. The company stayed mostly in stealth, published its mechanism work in journals like Neuron and Cell Reports, and let the data do the talking before the press releases did.
Nura Bio published the structural and mechanistic basis for inhibiting SARM1 before it made much noise commercially. Science first, narrative second - a refreshing order of operations.
How a hunch became a clinical company
NB-4746, in one capsule
NB-4746 is the lead. It is described as a first-in-class, oral, brain-penetrant SARM1 inhibitor - which is a lot of hyphens for what amounts to a pill that reaches the brain and tells nerves to stop dismantling themselves. In preclinical disease models it protected axons across ALS, MS, traumatic brain injury and chemotherapy-induced peripheral neuropathy. Then it cleared a Phase 1 trial in healthy volunteers with good tolerability and no serious adverse events.
A clean Phase 1 is not a cure. It is permission to keep going. But for a mechanism that lived in academic papers a few years ago, getting a brain-penetrant inhibitor safely into humans is the part where the idea stops being interesting and starts being a drug.
Behind the lead candidate sits the rest of the machinery: a structure-based discovery engine, primary neuronal models, imaging, genetic screens, and biomarker platforms that track axonal loss. There's also a second front - neuroimmune work aimed at restoring the conversation between neurons and the glial cells that are supposed to look after them. The axon-intrinsic switch is the headline; the neuron-glia relationship is the sequel.
Money follows mechanism
Investors tend to be skeptical of platform stories that promise to treat everything. Nura Bio's cap table suggests the skeptics were persuaded. The company raised $73 million in 2020, then added $68 million in 2024 to bring total Series A financing past $140 million. The round was led by founding investor The Column Group, with Samsara BioCapital and Euclidean Capital continuing and Sanofi Ventures joining as a new strategic backer.
Series A, in two acts
When the 2024 financing closed, the company also made a telling internal move: it promoted Shilpa Sambashivan, a co-founder and its chief scientist, into the CEO seat. In an industry that often parachutes in an outside executive at this stage, handing the wheel to the person who knows the molecule best is a quiet statement about what Nura Bio thinks it's selling. Not a pitch. A mechanism.
Keep the wiring intact
Nura Bio's stated aim is plain enough: discover and develop neuroprotective medicines that preserve neuronal integrity. The company's own tagline is less clinical and, frankly, harder to forget - a world where the mind is free to soar and the feet are free to run. For anyone who has watched a nervous system fail, that line is not marketing. It's a job description.
The competition is real. Disarm Therapeutics chased a similar axon-protection idea before Eli Lilly bought it, and the broader neurodegeneration field, including companies like Denali, is crowded with people trying to crack the brain. Nura Bio's wager is that getting to the axon switch early, with one well-behaved oral molecule, beats arriving late to clean up.
The nerve that didn't fall apart
Go back to that lab bench, and the nerve fiber that should have disintegrated but didn't. That image is the whole company in miniature. If NB-4746 does in patients what it does in models, the implication is not one new drug for one disease. It's a different default - a world where nerve damage doesn't automatically mean nerve loss.
That's still an if. Phase 1 buys the right to ask the question in patients, not the answer. But Nura Bio has done the unglamorous work: turned a 175-year-old observation into a target, the target into a molecule, and the molecule into something safe enough to keep testing. The switch is real. So is the company built to flip it off. What happens next is what the next trial is for.