Stem Cell Particles Delay Type 1 Diabetes in Mice, UCLA Study Finds
Tiny vesicles secreted by reprogrammed stem cells calmed the immune attack that destroys insulin-producing cells, a Nature Nanotechnology study reports, though human use remains years away.

Researchers at the University of California, Los Angeles have shown that microscopic particles secreted by reprogrammed stem cells can blunt the immune attack behind type 1 diabetes, delaying the disease's onset in mice. The findings, published this month in Nature Nanotechnology, describe an injectable approach built from extracellular vesicles, cell-made packets that normally ferry proteins and genetic material between cells, rather than from living cells themselves.
The team, led by UCLA bioengineering professor Song Li, drew the vesicles from induced pluripotent stem cells, or iPSCs, which are adult cells reprogrammed back into an embryonic-like state. In mice, two injections of the vesicles reduced the destruction of insulin-producing pancreatic cells and increased the number of regulatory T cells, the immune cells that keep autoimmune attacks in check.
The Numbers Behind the Result
According to the underlying research record posted through the National Institutes of Health, which matches the figures reported this week, 55% of mice that received the vesicles packaged inside an injectable hydrogel remained diabetes-free 90 days after treatment began, compared with 30% of mice that received the vesicles alone and a much lower share of untreated controls. The hydrogel, made from hyaluronic acid and a synthetic polymer, releases its cargo gradually over several days after being injected under the skin near the neck, where it can act on nearby lymph nodes.
The researchers also compared vesicles from pluripotent stem cells with those from mesenchymal stem cells, a more commonly studied source in regenerative medicine, and found the pluripotent-derived particles were markedly better at suppressing inflammatory signals and expanding regulatory T cells in both mouse and human immune cells. To produce enough material for testing, the team grew the stem cells as free-floating three-dimensional clusters in gently stirred liquid rather than on flat culture dishes, a change that raised vesicle output 3.4-fold per million cells.
"Nature has designed something amazing," said Li, who also holds an appointment at the UCLA Broad Stem Cell Research Center, describing the vesicles' natural signaling abilities.
An Autoimmune Disease With No Cure
Type 1 diabetes occurs when the immune system mistakenly destroys the insulin-producing beta cells clustered in structures of the pancreas called islets of Langerhans, leaving the body unable to regulate blood sugar without lifelong insulin therapy. The American Diabetes Association reports that more than 2 million Americans live with the condition, including roughly 314,000 children and adolescents, and incidence has been rising in recent years. Unlike type 2 diabetes, it cannot currently be prevented or reversed, and existing immune-suppressing drugs carry broad side effects because they dampen the entire immune system rather than the specific cells attacking the pancreas.
That is the gap the UCLA work targets. Because the therapy relies on cell-free particles rather than living, dividing stem cells, the researchers argue it could avoid some of the safety concerns, such as uncontrolled cell growth, that have complicated other stem cell-based treatments. The vesicles appeared to work partly by lowering a protein called CDK8 inside T cells, nudging them toward a regulatory rather than attacking state, though the authors say the exact molecular pathway, including a candidate role for a family of microRNAs, needs further confirmation.
Years of Groundwork, and Caution Ahead
The published paper builds on groundwork UCLA disclosed publicly years earlier. In 2022, the UCLA Broad Stem Cell Research Center announced that Li had received a two-year, $1.3 million grant from the California Institute for Regenerative Medicine, the state's stem cell funding agency, specifically to develop this extracellular-vesicle approach with collaborators including UCLA diabetes researcher Peter Butler. That funding supported the hydrogel-delivery platform now central to the new results.
Li cautioned that the approach, tested so far only in mice with an induced, antigen-specific form of diabetes, is not close to clinical use and would most plausibly help people who still retain some functioning pancreatic tissue rather than those with long-established disease. "In the early stages, you still have some pancreatic cells left," he said, suggesting the treatment's best use case may be newly diagnosed patients. He also noted a result that surprised the team: "One thing that surprised us was that a local injection could provide protection throughout the body," referring to the finding that regulatory T cells increased not just near the injection site but in distant lymph nodes near the pancreas and spleen.
Outside specialists have not yet weighed in publicly on the peer-reviewed version, and the coverage of the paper's publication emphasized that safety, dosing and effectiveness in humans remain unknown. The authors say their next steps include testing the vesicles in models that more closely resemble spontaneous human type 1 diabetes, refining manufacturing to scale up production beyond laboratory quantities, and probing further how the particles communicate with immune cells at a molecular level before any discussion of human trials becomes realistic.
How It Would Compare to What Already Exists
Only one drug is currently approved to delay type 1 diabetes, and it works differently. Teplizumab, sold as Tzield, won FDA approval in 2022 as the first therapy shown to push back the onset of stage 3 disease, based on a 76-patient trial in which 45% of people who received the drug were diagnosed with diabetes over a median follow-up of 51 months, compared with 72% of those who got a placebo, according to the drug's FDA-approved prescribing information. Teplizumab is a 14-day course of daily intravenous infusions and carries warnings for cytokine release syndrome and serious infections, reflecting its broader suppression of immune activity.
The UCLA vesicle approach, by contrast, is designed as a single injectable dose rather than a two-week infusion regimen, and its authors argue a more targeted immune effect could avoid some of teplizumab's side-effect profile. That comparison remains theoretical for now: the vesicle work has only been tested in mice, while teplizumab has years of real-world patient data behind it. Researchers in the field generally see therapies like both of these as complementary rather than competing, aimed at different stages of a disease that, once it reaches full onset, still has no cure.

Voyager 1 becomes first human-made object to reach one light-day from Earth

Webb telescope data confirm a rare, fully tracked stellar merger in the Andromeda galaxy

Australian physicists send a message through a brick wall using only gravity
