Skip to main content

Can Artificial Cells Take Over for Lost Insulin-Secreting Cells?

Posted on by Dr. Francis Collins

artificial beta cells

Caption: Artificial beta cell, made of a lipid bubble (purple) carrying smaller, insulin-filled vesicles (green). Imaged with cryo-scanning electron microscope (cryo-SEM) and colorized.
Credit: Zhen Gu Lab

People with diabetes have benefited tremendously from advances in monitoring and controlling blood sugar, but they’re still waiting and hoping for a cure. Some of the most exciting possibilities aim to replace the function of the insulin-secreting pancreatic beta cells that is deficient in diabetes. The latest strategy of this kind is called AβCs, short for artificial beta cells.

As you see in the cryo-SEM image above, AβCs are specially designed lipid bubbles, each of which contains hundreds of smaller, ball-like vesicles filled with insulin. The AβCs are engineered to “sense” a rise in blood glucose, triggering biochemical changes in the vesicle and the automatic release of some of its insulin load until blood glucose levels return to normal.

In recent studies of mice with type 1 diabetes, researchers partially supported by NIH found that a single injection of AβCs under the skin could control blood glucose levels for up to five days. With additional optimization and testing, the hope is that people with diabetes may someday be able to receive AβCs through patches that painlessly stick on their skin.

In the new study, a team led by Zhen Gu from the University of North Carolina, Chapel Hill, and North Carolina State University, Raleigh, started with a large, double-layered vesicle that resembles the outer plasma membrane of a beta cell. They packed it full of small vesicles containing the same type of lab-made insulin currently used to treat diabetes, mimicking the insulin-storage granules found inside mature beta cells.

The next challenge was to get the AβCs to work like real beta cells, and that took a great deal of time and ingenuity to add the right cellular parts to do the job. Gu and his colleagues started by borrowing GLUT2, the major glucose transporting protein from a real beta cell, and anchoring the protein to the outer surface of the AβCs. That allowed GLUT2 to bind the glucose molecules circulating outside the bubble-like vesicle and haul them inside.

There, an enzyme converts the sugar to an acid, dropping the pH inside the vesicle. The more acidic conditions change the coating of the inner small vesicles, enabling peptides on their surfaces to “zipper up” with complementary peptides anchored to the outer large vesicle. The zipping action pulls the insulin-filled vesicles to the AβC surface, forcing the membranes of the large and small vesicles to fuse and insulin to spill out.

As glucose levels return to normal outside the cell, glucose concentrations inside the vesicle also drop, raising the pH. The higher pH blocks the fusion of large and small vesicles and shuts down nearly all of the insulin release until needed again.

As reported in the journal Nature Chemical Biology, tests of the AβCs in lab dishes showed they were extremely responsive to changes in outside glucose concentrations for 15 hours [1]. Under high glucose (hyperglycemic) conditions, AβCs quickly began releasing insulin, exactly as they were engineered to do. When glucose concentrations returned to normal levels, the release tailed off to just minimal amounts of insulin.

But would the AβCs also work in the body?  To find out, the researchers injected artificial cells suspended in a water-based gel under the skin of diabetic mice that don’t make insulin. Remarkably, within an hour, the animals’ blood glucose reached a normal level, where it remained for several days.

To add to the challenge, the researchers injected the mice with a glucose solution. After an initial spike, blood glucose levels returned to normal and at a rate comparable to those seen in healthy mice. Encouragingly, the studies also suggest AβCs are non-toxic to the mice. Within four weeks, the AβCs empty and completely biodegrade without appearing to cause an immune response.

With further study, the findings suggest that AβCs might be implanted directly into human skin via an injectable gel and without the immunosuppressive drugs required for the transfer of live cells. Even more promising, they might be delivered through adhesive patches carrying arrays of microneedles that lightly prick the skin without causing any pain. In fact, Gu’s team has already engineered such patches for use in related drug delivery applications [2]. He says the next step is to test the new AβCs in a larger animal model before, hopefully, moving on to a clinical trial in people.


[1] Synthetic beta cells for fusion-mediated dynamic insulin secretion. Chen Z, Wang J, Sun W, Archibong E, Kahkoska AR, Zhang X, Lu Y, Ligler FS, Buse JB, Gu Z. Nat Chem Biol. 2017 Oct 30.

[2] Microneedle-array patches loaded with hypoxia-sensitive vesicles provide fast glucose-responsive insulin delivery. Yu J, Zhang Y, Ye Y, DiSanto R, Sun W, Ranson D, Ligler FS, Buse JB, Gu Z. Proc Natl Acad Sci U S A. 2015 Jul 7;112(27):8260-5.


Diabetes (National Institute of Diabetes and Digestive and Kidney Diseases/NIH)

Zhen Gu Lab (University of North Carolina, Chapel Hill)

Clinical and Translational Science Awards Program (National Center for Advancing Translational Sciences/NIH)

NIH Support: National Center for Advancing Translational Sciences


Leave a Comment