Analytical and Laboratory

Researchers Find an Injection Alternative for People With Type 1 Diabetes Using a Computer Model

26 September 2017

The hypoxia-sensitive vesicles are loaded with glucose oxidase and insulin. Source: MITThe hypoxia-sensitive vesicles are loaded with glucose oxidase and insulin. Source: MIT

People with Type 1 diabetes have to check their blood glucose a few times per day and inject themselves with insulin to keep their blood sugar levels within a healthy range. Researchers have been searching for a better alternative for a long time. They have discovered one alternative: insulin that is engineered to linger in the bloodstream and become active only when it is needed, like right after a meal.

One obstacle in the way of developing this insulin is that it is difficult to know how these drugs might behave without testing them on animals. MIT researchers have created a computer model that should streamline the development process. Their new model can predict how glucose-responsive insulin (GRI) will affect patients’ blood sugar, based on chemical traits like how quickly the GRI becomes activated in the presence of glucose.

"The concept of GRI has been a longstanding goal of the diabetes field," says Michael Strano, the Carbon P. Dubbs Professor of Chemical Engineering at MIT. "If done correctly, you could make it so that diabetics could take an occasional dose and never have to worry about their blood sugar."

The new model allows researchers to identify several strong GRI candidates that they are planning to test in animals.

Recently, scientists have experimented with a few types of GRIs. Strano and colleagues outlined some of the progress that has been made and laid out a series of specific synthetic approaches that could make GRIs a practical reality. These include using mathematical models of the human body to predict how GRIs behave in patients, which makes it easier to design and test these drugs.

The MIT team used that type of modeling to analyze insulin that is modified so it can interact with glucose. The insulin has molecules called PBA attached to it and the PBA molecules can bind to glucose, which activates the insulin.

Other GRI approaches that scientists have tried involve insulin embedded into hydrogels that release the drug when they find glucose, and the insulin-carrying particles, made from polymers that degrade, bind to the glucose. In these cases, it is important to know just how strongly the glucose will interact with the GRI and how quickly the insulin will start acting.

The MIT research team devised equations that describe the behavior of PBA-modified insulin based on parameters, like how strongly glucose binds to the GRI and how rapidly the insulin is activated. They combine these equations with existing models of how glucose and insulin behave in different compartments of the body, like blood vessels, muscle, and fatty tissue.

"We started by thinking about the GRI as a set of equations," Strano says. "The result is the first rational design for the GRI."

Because of these results, researchers can enter specific GRI traits and model how the GRI will behave in the human body over 24 hours, with meals consumed at certain intervals through the day. The model predicts how much blood sugar will spike after meals, the strength of the triggered insulin response and the resulting blood sugar level.

The researchers incorporated the American Diabetes Associations’ (ADA) recommended blood sugar limits into their model. This allowed them to determine which GRI parameters produce blood sugar control within the suggested guidelines.

This model is specific to one category of GRI, but the researchers plan to apply this approach to develop similar models for other types of GRIs.

Strano believes that other researchers, particularly medicinal chemists, will use this new model to guide the development of new GRI candidates. The MIT team is perusing several candidates that were predicted by the model and plans to work with Michael Weiss, a professor of biochemistry at Case Western Reserve University, to test them in mice.

Researchers have confidence that this approach could be extended to other types of drugs that would respond to changes in physiological conditions. For example, anticoagulants are activated when blood clotting proteins become elevated.

"We could envision a future where that's the norm for all therapeutics: We could ask our drugs to modulate their potency based on our immediate, instantaneous need in real time," Strano says. "That's pie-in-the-sky at this point, but the starting point of this concept is a model for their design."

A paper on this research was published in Nature Chemistry.

To contact the author of this article, email

Powered by CR4, the Engineering Community

Discussion – 0 comments

By posting a comment you confirm that you have read and accept our Posting Rules and Terms of Use.
Engineering Newsletter Signup
Get the Engineering360
Stay up to date on:
Our flagship newsletter covers all the technologies engineers need for new product development across disciplines and industries.

Upcoming Events