Harvard researchers identify gene that could unlock insulin production in pancreatic cells

Scientists at Harvard Medical School have discovered a gene, ALDH3B2, that acts as a brake on insulin production in pancreatic cells. Its suppression could enable the reprogramming of duct cells into insulin-producing beta-like cells, paving the way for new diabetes treatments.

Harvard Medical School researchers have identified a gene that appears to act as a brake on the pancreas’s ability to make insulin, raising the possibility of a future treatment approach for diabetes that starts with a patient’s own cells. In work published in the journal Diabetes and discussed in a preprint indexed by PubMed, the team found that disabling ALDH3B2 could reprogramme human pancreatic duct cells into beta-like cells capable of making and releasing insulin.

The finding matters because beta cells, which sit in the pancreatic islets, are the body’s natural insulin factories. Their loss or failure drives diabetes: in type 1, the immune system destroys them; in type 2, they gradually become less effective and can decline in number. Current options, including insulin therapy, glucose monitoring and donor-cell transplantation, help manage the disease but do not fully restore the body’s own insulin-producing capacity.

To search for a way around those limits, the Harvard group used a genome-wide CRISPR screen, testing more than 19,000 genes in human pancreatic duct cells. The researchers were looking for genes that keep those cells locked into their original identity. ALDH3B2 emerged as a key candidate. When the gene was switched off, the duct cells shut down markers of their usual state and turned on beta-cell genes, including genes involved in insulin production and processing.

The conversion was not trivial. The cells did not leap straight from duct cells to mature beta cells; instead, they passed through a more immature progenitor-like state before taking on insulin-producing features. The researchers also reported epigenetic changes, including reduced DNA methylation around the insulin gene, which made it easier for the cells to maintain their new identity. In the experiments, the share of duct cells showing this transformation rose from less than 1% to about 8.5% after ALDH3B2 was removed.

The team then transplanted the reprogrammed human cells into diabetic mice. According to the study, the cells survived, released human insulin in response to glucose and helped bring blood sugar levels close to normal over a six-week period. The researchers also tested DEAB, a broad inhibitor of aldehyde dehydrogenase enzymes, and saw a similar effect, hinting that a drug-based approach might one day replace permanent gene editing. But the authors and subsequent coverage of the work emphasise that this is still early-stage research. Much more laboratory work, followed by animal studies and carefully controlled human trials, will be needed before the approach can be considered for treatment.