Inside a laboratory at the University of Tokyo, miniature intestines are dying. Not real intestines, exactly, but something close enough to matter: tiny spheres of human gut tissue, grown from stem cells, each one pulsing with the same cellular machinery that lines the bowel of anyone reading this.
The researchers flood them with a protein called tumor necrosis factor, or TNF, at concentrations found in patients whose Crohn’s disease is spiralling out of control, and they watch. The cells collapse into apoptosis, the orderly self-destruction sequence the body uses to eliminate damaged tissue. Which, in this context, is precisely the problem. And then, quietly tucked among roughly 3,500 candidate compounds the team was screening for any sign of protection, something unexpected showed up. A natural extract from licorice root. Thirty times sweeter than sugar. Used in candy and traditional Chinese medicine for centuries.
That compound, glycyrrhizin, has now emerged as a credible candidate for treating inflammatory bowel disease, in findings published today in Stem Cell Reports by Yu Takahashi at the University of Tokyo and colleagues. The results are perhaps as notable for what they reveal about the screening method as for the compound itself.
Why Standard Lab Models Were Missing the Signal
IBD affects an estimated 4 million people worldwide, and the numbers are rising, particularly in countries that have relatively recently adopted westernised diets and lifestyles. Crohn’s disease, one of the two main forms, is characterised by patchy inflammation that can strike anywhere along the digestive tract, destroying the epithelial lining that separates the body from the bacterial chaos of the gut. The condition’s go-to biological treatments, drugs that neutralise TNF (the very protein Takahashi’s team used to model the disease), work well for some patients. A substantial proportion, though, lose their response over time, partly because their immune systems begin generating antibodies against the drugs themselves. There’s an obvious gap here, and filling it has proved harder than it looks.
Part of the difficulty is arguably methodological. Drug screening for gut diseases has relied heavily on cancer-derived intestinal cell lines, particularly a line called Caco-2, which has been a workhorse of gut biology for decades. When Takahashi’s team tested those cells against TNF, nothing much happened. The cells barely registered the protein. Human stem cell-derived intestinal organoids, by contrast, responded dramatically, dying at TNF concentrations found in actual Crohn’s patients. The implication is somewhat uncomfortable: conventional cell lines may have been quietly filtering out exactly the compounds that would work in the real intestinal epithelium, because they don’t actually behave like real intestinal epithelium.
This mattered a great deal when it came to glycyrrhizin. The compound protected organoid-derived cells dose-dependently and quite specifically. It had no effect on cell death driven by anything other than TNF, no effect on cell growth in the absence of the inflammatory stimulus, and crucially, it did nothing at all in L929 cells, a murine fibroblast line often used as a reference in cytotoxicity assays. Had the screen been run on conventional lines from the start, glycyrrhizin would likely have been missed entirely.
The Molecular Twist in the Candy
What glycyrrhizin is actually doing inside these cells turns out to be more mechanistically interesting than the initial screening hit might suggest. TNF can kill cells in more than one way. The “cleaner” route is apoptosis, the programmed death pathway mediated by a cascade of enzymes called caspases. The messier alternative is necroptosis, a form of cellular rupture that spills inflammatory signals into surrounding tissue and tends to amplify the immune response rather than resolve it. Organoid-derived gut cells, it turns out, primarily undergo apoptosis when TNF strikes (unlike L929 cells, which prefer necroptosis, which is why glycyrrhizin couldn’t help them). Glycyrrhizin intercepts the apoptotic cascade at a point downstream of caspase-8, blocking the cleavage of proteins that execute the death program without touching the signalling immediately upstream of it.
There’s a further wrinkle worth noting. Glycyrrhizin contains within its structure a smaller molecule, glycyrrhetinic acid, which is released when intestinal bacteria break down the parent compound during digestion. Glycyrrhetinic acid, the team found, does essentially nothing to protect organoid cells from TNF. The intact glycoside structure appears necessary for the anti-apoptotic function. This is, in a sense, a delivery problem: if gut bacteria convert glycyrrhizin to glycyrrhetinic acid before it reaches the epithelial cells it’s supposed to protect, the therapy may lose much of its potency in practice. Formulation strategies that slow that conversion, perhaps by targeting delivery to specific sections of the gut, could matter a lot in any eventual clinical application.
Evidence Beyond the Organoid
The team tested the compound in mice with colitis induced by dextran sodium sulfate, a chemical irritant that damages the gut lining in a pattern resembling IBD. Mice receiving oral glycyrrhizin showed measurably less intestinal damage on histological examination, reduced cell death markers in the epithelium, and a reversal of colon shortening, a rough physical indicator of how severely inflamed the bowel has become. Weight loss from the colitis trended toward recovery in treated animals, though that difference didn’t reach statistical significance. The in vivo results are reasonably consistent with the organoid data, which is the kind of concordance you want if you’re trying to argue that the organoid system reflects real biology rather than a laboratory artefact.
The organoids used throughout the study consist solely of epithelial cells. Real intestinal tissue, of course, contains fibroblasts, immune cells, blood vessels, and a resident microbiome that influences epithelial function in ways still being worked out. Whether glycyrrhizin would behave similarly in a more complete intestinal environment remains to be seen, and the researchers acknowledge this openly as a limitation of the model. Testing the compound in organoids derived from Crohn’s disease patients themselves, rather than from healthy tissue exposed to TNF, would also help establish how clinically relevant these findings actually are.
None of that is damning. Organoid-based drug discovery is still a relatively young approach, and the platform built by Takahashi and colleagues (developed over several years specifically to reduce the cost and technical burden of large-scale organoid screening) represents a meaningful step toward making it practically viable. The fact that their screen turned up a compound with a known safety profile, established regulatory precedent as a food ingredient, and some prior evidence of anti-inflammatory activity in cellular and animal systems is, at minimum, useful. Whether glycyrrhizin will survive clinical trials is a different question; the list of things that work beautifully in mice and organoids before quietly failing in humans is long and sobering. But the method that found it, arguably, is the part of this story worth watching most closely.
DOI: 10.1016/j.stemcr.2026.102891
Frequently Asked Questions
Could eating licorice actually help with Crohn’s disease?
Almost certainly not in the amounts found in licorice candy. The compound responsible, glycyrrhizin, does appear to reduce intestinal cell death in both lab models and mice at specific doses, but there’s a significant catch: gut bacteria convert it into a related molecule, glycyrrhetinic acid, that doesn’t seem to work. Whether you could deliver enough intact glycyrrhizin to the gut lining through ordinary food is far from clear, and the research is still nowhere near human trials. Any therapeutic use would likely require a purpose-designed formulation, not a trip to the sweet shop.
Why do existing Crohn’s drugs stop working for so many patients?
The main biological drugs for Crohn’s work by neutralising a protein called TNF, which drives the inflammatory cascade. Over time, a substantial share of patients develop antibodies against those drugs, effectively disabling them before they can act. Glycyrrhizin works differently: rather than blocking TNF in the bloodstream, it appears to protect the gut lining cells directly, interrupting the death cascade triggered when TNF reaches epithelial cells. If it works in humans, it could potentially complement existing treatments rather than replace them.
What are organoids, and why do they matter for drug discovery?
Organoids are tiny three-dimensional structures grown from stem cells that self-organise into tissue resembling a real organ, in this case the intestinal lining. They respond to inflammatory signals in ways that standard cancer-derived lab cell lines apparently do not, which matters enormously for drug screening: the Tokyo team found that conventional cell lines didn’t even register the TNF concentrations that devastate organoid tissue, meaning potentially useful compounds would have been filtered out of any screen that relied on them. Organoid-based platforms are still technically demanding and expensive, but that balance is shifting.
Is glycyrrhizin safe enough to test in humans?
Glycyrrhizin has a relatively well-established safety profile as a food ingredient, and the World Health Organisation has set a no-observed-adverse-effect level based on clinical data. At higher doses it can cause problems including raised blood pressure, so it’s not without risk. The researchers note that further studies will be needed to assess therapeutic efficacy and individual variation in humans before any clinical use could be considered.
ScienceBlog.com has no paywalls, no sponsored content, and no agenda beyond getting the science right. Every story here is written to inform, not to impress an advertiser or push a point of view.
Good science journalism takes time — reading the papers, checking the claims, finding researchers who can put findings in context. We do that work because we think it matters.
If you find this site useful, consider supporting it with a donation. Even a few dollars a month helps keep the coverage independent and free for everyone.
