Reprogramming Kidney Cells to Treat Type 1 Diabetes

Heya Desai
10 min readApr 5, 2021


I didn’t recognize the benefits and opportunities that would be automatically unlocked when I was younger, but I am so grateful that I’ve gotten to witness and experience healthcare innovations in the diabetes space first hand.

Though every product has not been beneficial for me, I am thankful that I’ve had the opportunity to test what might improve my well-being.

Growing up, I wasn’t exactly appreciative of what now seems like a gift of knowledge. I didn’t recognize how I could make the most of what I knew and what I could learn to improve my own life, and down the road, potentially millions of others.

Right now in my head, I’m envisioning myself looking at the situation from a distance from an optimistic perspective and just moving along with the timeline, it seems kind of magical.

Want to read this story later? Save it in Journal.

To give you some sort of a visual, picture a kid in awe.


We’ve come far as the world and at a stage that was unimaginable when insulin was discovered by Frederick G banting in 1922. #reppingcanadiansamirite

Looking back, it is unbelievable that we’ve been able to progress from original insulin…


to smart insulin pens among the other developments for the treatment of the disease.


When I was first diagnosed in 2012, I was using the one time use, disposable injections along with insulin in small glass vials. As time went on, I tried different pens including the Lily HumaPen and now the Lily KwikPen.

The difference between disposable pens and discardable injections is that pens contain the insulin already, so I would use a 4mm needle with them, and I’d be ready to go. I can re-use the same pen until it is empty, making the entire process a lot more convenient.

If you want to learn more about my experiences, check out this article.

Along with different devices, I’ve also tried different types of insulin with the intention of finding what combination of medications helps me avoid glucose fluctuations. For instance, Tresiba — a long-acting basal insulin does a better job of preventing overnight hypoglycemia (lower blood sugar than normal) since it lowers blood sugar levels in a more flatter, hence more predictable way.

Furthermore, insulin pumps are small computerized devices that deliver short acting (humalog) doses of insulin constantly at a basal rate set by the physician and at meal time where the dosage is programmed by you.

Below, is the MiniMed640G. It is the closest we have come to having an artificial pancreas available. The ability to predict and prevent low glucose levels in type 1 diabetics and the rebound hyperglycemia attacks gives the system its name.

Wearable insulin pump, CGM kit, therapy management software — Source

These were only some of the milestones and highlights for diabetes technology for the past decade.

Problems with Existing Approaches (Non-cell based)

Nevertheless, there are some problems that still persist. What’s common about all of these treatment methods is that they are external machines and mostly systems involving multiple components and as small as these patches might get, there are no approaches in use that provide closed-loop control (no human input required) of the therapeutic dosage of insulin according to the glucose sensing that’s done.

The insulin pumps that have been closely able to mimic an artificial pancreas do require a decent amount of user involvement and can possibly inhibit them from freely doing the activities they want to.

Calibrating the sensor, disconnecting and reconnecting them when you shower or go for a swim are responsibilities that have to be fulfilled regularly, and the risks associated with forgetting them can be severe.

If I had a pump and was to workout, and forgot to reduce my post-workout basal, I’d be at higher risk for hypoglycemia since exercise makes insulin work more effectively and lowers blood sugar.

These gaps contribute to one of the reasons why I’ve been leveraging my experiences and have been focusing on the potential for development when we shift the efforts from just treating this condition to simplifying and improving patients’ lives. This is really when capitalizing on what you know comes in.

Overview of Diabetes & Hyperglycemia

Diabetes, both type one and two currently affect more than 415 million people worldwide. Depending on when it is diagnosed and how it is managed, the clinical complications can vary. For a type 1 diabetic specifically, maintaining a normal blood sugar level is essential to avoid complications that can include heart and blood vessel disease, nerve damage and even foot damage. Sustained hyperglycemia can initiate pathological cascades.

Fortunately, I have not personally experienced any of these complications, however whenever I do get cuts and blisters, my parents put extra emphasis on not letting them go untreated because they can eventually become infections.

And what this shows us is that going the extra mile to mitigate high blood sugar when it is in the individual’s control, is in their best interests.

Yet, factors such as having an infection, or being ill or simply being stressed can lead to high blood sugar, so occasional bad days do become inevitable despite regular exercise and maintaining a healthy diet.

This is why it is crucial for diabetics to optimize to stay healthy in every aspect whether that be doing what it takes to avoid a cough or cold and reflecting on their lifestyles every once in a while to see if anything can be ameliorated.

Stabilizing Blood Sugar Levels Autonomously

Along my research journey, I came across a research paper entitled β-cell–mimetic designer cells provide closed-loop glycemic control”. This was one of the first studies I had read proposing a solution with glucose-sensor components, that too using synthetic biology — a field I’ve become passionate about and one that has applications ranging from renewable fuels to biosensors.

The synthetic biologists implanted the designer kidney cells they re-engineered from the original HEK 293 cells to mimic the functions of pancreatic beta cells (β-cells) in mice with t1d.

Immunofluorescent HEK 293 Cells — Source

This solution was able to stabilize their blood sugars successfully and overall restored glucose and blood insulin homeostasis which implies a self-regulating process through which an organism can maintain a steady state.

Self-sufficient insulin expression in t1d mice — Source

It is important to note that the solution I will be discussing is not the only study involving biology intersections that has been done to resolve this issue. However, it does address several issues that have arised with past research endeavours.

In the research paper, HEK 293 cells are also compared to other candidates for cell-based diabetes therapies, but there are several limitations that show HEK-β cells to have a more promising future.

Challenges with various cell-based diabetes therapies

  • Human islets are low in supply and difficult to maintain in culture and treatment success does remain unpredictable.
  • When compared to the HEK-β cells, they had lower insulin secretion capacity in vitro and post-prandial (after one eats) metabolism was only restored in 2 of 4 mice
  • Induced pluripotent stem cells differentiated into β-like cells leverage growth factors, mixtures of chemicals as well as rational programming of synthetic-lineage control networks.
  • They may provide the missing link to genetically programme somatic cells into autologous cell phenotypes for regenerative medicine (Saxena, P. et al.).
  • But, these processes are relatively expensive and possibly incompatible at a large scale. They also require a long period of time to mature into β-like cells post-implantation.
  • Artificial synthetic beta cells are shown to be eventually killed off by the body after being recognized as beta cells.

I want to do a walkthrough for this solution, so you can understand how synthetic biology can provide a hands-off solution for treating T1D, but if you want to learn more I’d recommend giving the research paper a need and Professor Glass’ other work — the leader of the JCVI Synthetic Biology group.

Essentially, by minimally engineering human HEK 293 kidney cells, we can create mimetic designer b-cells as an alternative to the β-cells that stop synthesizing and secreting insulin in the pancreas of a type 1 diabetic.

To provide an overview of what this solution consists of and enable, here are the general highlights from the research paper.

  • Uses glucose-sensor components evolved in native β-cells
  • Takes advantage of robust parental cell lines with a track record in biopharmaceutical manufacturing
  • Shows glucose-induced insulin release performance comparable to that of β-cell lines and human islets
  • Rational programming of designer cells enables straightforward fine- tuning of performance parameters and provides
  • Provides flexibility to couple glucose sensing to the production of other therapeutic proteins such as GLP-1 required for T2D therapy

But first, let’s break that down a bit, do a bit of review and understand what exactly this means.

Breaking down the solution

Pancreas: An organ that is located in the abdomen and has an endocrine function that regulates blood sugar, as well as an exocrine function that helps with digestion. During the process of digestion, our pancreas produces enzymes that break down sugar, fats and starches.

β-cells: Found in pancreatic islets (groups of cells in the pancreas). They synthesize and secrete insulin as well as amelyn which is another peptide hormone found in the pancreas

HEK 293 cells: Line of cells derived from human embryonic kidney cells that are grown in tissue culture. Because of their reliable growth among other factors, they have been used in cell biology research for years.

Going back to T1D, β-cells are destroyed by the T cells of the immune system. They are insulin producing cells, hence needed to detect glucose levels and secondly to secrete insulin to process the carbs we consume and regulate our blood sugar levels.

How does the glucose-induced excitation-transcription coupling system work?

To achieve glucose responsiveness, a synthetic circuit was created that couples glycolysis — a metabolic pathway converting glucose into pyruvate- induced calcium entry into the system in kidney cells. The excitation-transcription coupling component means signalling by ion channels to the nucleus. Changes in intracellular Ca2+ concentrations (calcium ions) due to stimuli has a role in regulating cellular processes, including this one.

Flexibility was provided to couple glucose sensing to the production of other therapeutic proteins as well such as GLP-1 which is known for being able to suppress postprandial glucagon release and inducing the expansion of B-cell mass that secretes insulin.

GLP-1 Molecular Visualization

It is glucagon like peptide and is produced in our L-cells of the distal ileum and colon and released as a response to any nutrient intestine, so glucose is a nutrient that upregulates the transcription of the gene that encodes GLP-1.

The strategy was to evaluate the components managing glucose sensing in native B. cells using a screening approach.

This revealed CAV1.3 the calcium channel, voltage-dependent, L type, alpha 1D subunit protein as something that can program extrapancreatic, meaning outside of the pancreas cells like the embryonic kidney cells to be able to profile in a sense blood-glucose concentrations in the body in an accurate and dynamic way.

A dynamic mathematical was also used which allowed them to build a quantitative representation of the system and make in-vivo (studies testing inside an organism) behaviour predictions. This model also predicted that the implant would improve glucose tolerance in t1d mice.


This is the closed loop quality of this solution. It demonstrates that there is potential for being able to stabilize glucose levels without the hassle of several blood sugar checks and injections when both are addressed.

It’s like killing 2 birds with one stone!

This system would measure the output which is the glucose level and feed it back to the input. The errors with respect to the desired output is determined and a signal goes to the controller. This component controls the amount of input — being insulin according to the desired response for stabilizing.

The presence of feedback compensates for the disturbance and improves the accuracy of the system.

By recreating the natural processes performed by functioning pancreas, insulin would be distributed in response to fluctuations, and eliminate the considerable involvement from patients otherwise.

According to Jacob Brogan, an editor at the Washington Post who is also a type 1 diabetic and has read and actually spoke to John Glass himself, this work is still in its earliest stages and as expected, we won’t likely see results any time soon. Considering that it took 15 years to come up with the idea making beta cells out of stem cells to producing them, for any feasible solution there is a long journey ahead.

The autoimmune response of a diabetic’s body to synthetic and mimetic cells, as well as modified insulin, may continue to be the largest risk and obstacle preventing success.

Regardless, synthetic biologists are continuing to accelerate healthcare innovation and these are all indications that there is potential for those impacted by the condition to be able to live their lives without any extra burden.

We have the experts, we have the technology, we have the skills, we have the understanding and we have the goal.

If we continue to have the willingness to disrupt and solve this and try again when we don’t succeed, at the end of the day, this is how we’ll know we are going in the right direction.

Connect with me!

Thank you for reading! If you enjoyed my article or want to learn more, subscribe to my monthly newsletter to keep up with my progress in this area and to receive an insight on everything I’ve been up too! In the meanwhile, let’s connect on linkedin.

📝 Save this story in Journal.