Objective
“Alpha cells” are pancreatic cells that make the hormone, glucagon. They are abundant in people with T1D, and, in theory, could serve as a reservoir for new beta cells. It is well documented that rodent alpha cells can convert into beta cells, but there are no genetic lineage-tracing tools to assess the possibility of “alpha-to-beta transdifferentiation” in human alpha cells. Most investigators think this is an unlikely path to human beta cell regeneration.
Our group has developed human beta cell regenerative DYRK1A inhibitor drugs, such as harmine, with or without the presence of a GLP1 Receptor Agonist (GLP1, exendin-4, semaglutide/Ozempic). These cause a dramatic increase in human beta cell numbers and reverse diabetes. Most investigators, including ourselves, had assumed that these drugs act exclusively by causing human beta cells to proliferate, thereby increasing in numbers.
Recently, I have performed single RNA cell sequencing studies on harmine-treated human islets. They show that harmine causes a significant increase in the abundance of a subset of human pancreatic alpha cells called “Cycling Alpha Cells”, and these cells appear to be converting into beta cells.
Proving alpha-to-beta transdifferentiation in rodents is accomplished by making special transgenic mice, and performing alpha cell lineage tracing. Developing transgenic humans is of course impossible, but other means of studying transdifferentiation is possible. Therefore, in this study, I aim to sort and collect pure alpha cells and generate pseudoislets that would allow for a longer-term drug treatment that will ultimately be crucial to understand if and when alpha cells convert to beta cells. In addition, I will develop a genetically engineered human stem cell (hESC) line that will allow my group and other scientists to trace the lineage trajectory of human pancreatic alpha cells. This would be a novel a critical tool that the beta cell regenerative field currently lacks. More specifically, this hESC line will be instrumental to unequivocally confirm or refute the alpha-to-beta cell transdifferentiation phenomenon that seems to occur with DYRK1A inhibitor drug treatment.
If this can be shown, it will be very good news for people with longstanding T1D, whose numbers of beta cells are markedly reduced, because all people with T1D have an abundant reservoir of pancreatic alpha cells, so replacing beta cells from existing alpha cells could be game-changing.
Background Rationale
Beta cell regenerative drugs can provide a scalable and affordable treatment approach for millions of people with T1D. We and others showed that small molecule inhibitors of DYRK1A such as harmine is able to induce human beta cell replication at a rate of 2% in human cadaveric islets. More recently, we demonstrated that addition of a GLP1 Receptor Agonists (GLP1RAs) such as exendin-4 or semaglutide/Ozempic or a TGF beta inhibitor increases this percentage to 5-8%. Further, our in vivo studies in immunodeficient mice that were transplanted with a small number of human islets showed that continuous infusion of harmine +/- exendin-4 for three months resulted in 300% and 700% increase in human beta cell mass, respectively. Importantly, we received FDA Investigational New Drug (IND) approval for a human Phase 1A single rising dose study with harmine. This study also is registered with ClinicalTrials.gov (NCT#05526430) and funded by the NIH. Collectively, these results indicate that harmine is a valid beta cell regenerative drug that can be given to humans at doses that are safe and well tolerated.
However, the exact mechanism of action of harmine (+/- GLP1RAs) remains elusive. With this goal in mind, I performed single cell transcriptomics on cadaveric human islets that were treated with these regenerative drugs for four days. My results suggest that a subset of alpha cells has the potential to transdifferentiate into beta cells when exposed to harmine +/- GLP1RAs and may serve as a potential progenitor reservoir for beta cells. Human alpha-to-beta cell transdifferentiation has never been reported or studied in the context of beta cell regenerative drugs. In addition, unequivocally demonstrating the occurrence of alpha-to-beta cell transdifferentiation in the cadaveric human islet system is impossible. Therefore, I will first sort and collect pure alpha cells and generate pseudoislets that will allow for a longer-term drug treatment that will ultimately be crucial to understand if and when alpha cells convert to beta cells.
I will also engineer a hESC where alpha cells will be labelled with Green Fluorescent Protein after differentiation of hESCs to islet-like structures. This design will help me pinpoint if any of the GFP+ alpha cells would co-express Insulin under basal or drug-treated conditions. This in turn will elucidate conclusively if alpha-to-beta cell transdifferentiation occurs under these conditions.
Description of Project
All forms of diabetes result from inadequate number of insulin-producing pancreatic beta cells. As a result, beta cell replacement studies, such as whole pancreas transplantation, pancreatic islet transplantation and transplantation of human stem cell-derived beta cells, are currently in use or in development for the treatment of Type 1 diabetes (T1D). As a simpler, less expensive, and scalable alternative, our group has focused on developing drugs that are able to restore the endogenous beta cell mass in people with T1D. In this effort, we have shown that a simple, inexpensive pill containing a drug we identified, called “harmine”, is able to induce expansion of human beta cells derived from human organ donors. We have also demonstrated that simple addition of a harmine to other drugs such as Semaglutide, Ozempic etc, can even further increase human beta cell proliferation. Remarkably, harmine alone or with a GLP1RA (Semaglutide, Ozempic, etc) increases human beta cell mass by 300-700%: enough to restore beta cell mass to normal, and completely reverse diabetes.
This application focuses on understanding exactly how this beta cell expansion occurs. We initially assumed that it resulted from drug-induced proliferation of human beta cells. We now know that the actual beta cell proliferation rates cannot explain the 300-700% increase in human beta cell mass after drug treatment.
Recent single cell RNA sequencing studies in my lab suggest that the new beta cells formed in response to harmine are derived from alpha cells by “transdifferentiation”. This is an attractive phenomenon, because all people with T1D have ample supplies of alpha cells in their pancreas: converting some of them to beta cells would allow everyone with T1D – even those with longstanding T1D and very limited numbers of beta cells – to “fill up their beta cell tank”.
The big question now, is: “is it really possible to convert alpha cells into beta cells?” I will use three parallel approaches to tackle this question. First, I will expand the number of my initial single cell sequencing experiments to provide more confidence. In addition, I will perform drug treatment for 48 and 96 hours. Second, I will collect pure alpha cell populations from human cadaveric islets and test their potential for beta cell conversion after regenerative drug exposure.
And third, to provide the ultimate proof for cell type conversion I will use a molecular technique called “lineage tracing”. Human stem cell technology has advanced to the point that relatively mature human islets can be generated from human embryonic stem cells (hESCs). This provides an opportunity to permanently “mark” human alpha cells and follow them over time. Accordingly, in this application I will engineer a hESC line in which alpha cells will be permanently marked with a Green Fluorescent Protein. This will allow me to follow the ultimate fate of alpha cells, both in tissue culture, and in mice transplanted with hESC-derived islets, to unequivocally define if cells that were once alpha cells, are able to become beta cells.
Collectively, these studies will be instrumental in shedding light on the lineage dynamics in human islets, and more specifically on the alpha-to-beta cell conversion. This will be an important advance: even in people with long-standing T1D, with few remaining beta cells, there are ample supplies of alpha cells that could be converted to beta cells. These studies will ultimately provide a better understanding of the mechanism(s) of action of harmine and related beta cell regenerative drugs, hastening the development of this next generation drug therapies for T1D.
Anticipated Outcome
I anticipate that my attempts to sort and collect pure alpha cells will be successful as my preliminary data present in this application strongly suggest it. In addition, I anticipate that I will not face any hurdles with alpha cell pseudoislet generation as it has been reported before. Further, I expect that the engineering of a pancreatic alpha cell lineage tracing hESC line will be successful, without any major drawbacks. I also expect no setbacks in differentiating the hESCs to human islet-like structures. My collaborator, Dr. Chen, is a world-renowned expert on human stem cell engineering and hESC-derived human islets. After obtaining islet-like structures following a multistep differentiation protocol, and documenting the ability to lineage trace and label alpha cells, I will transplant them into the kidney capsules of immunodeficient mice and continuously infuse the mice via Alzet pumps with harmine (+/- exendin-4) both of which are regularly performed experiments at the Human Islet and Adenovirus Core (HIAC) at Mount Sinai on daily basis.
In this system, cells that are, or ever were, alpha cells are labeled with the green cell marker, Green Fluorescent Protein. After developing these tools and model systems, under basal conditions where no drug treatment is performed, I anticipate that all alpha cells will be green, and no beta cells will be green. In response to drug treatment, alpha cells will remain green, and some percentage of beta cells will also be permanently marked green, indicating that they were once alpha cells.
Collectively, I expect these anticipated results to signify an important advance in the field of human beta cell regenerative drug development, as it will help us conclusively determine if alpha-to-beta cell transdifferentiation can take place upon harmine (+/- exendin-4).
Relevance to T1D
This is a highly relevant study to T1D research. In broad terms, it investigates the potential of human alpha-to-beta cell transdifferentiation upon beta cell regenerative drug treatment. Further, it also aims to understand other possible fate conversion events that may be happening in human islets with regenerative drug treatment. T1D results from the insufficient amount of functional beta cells. Therefore, therapeutic approaches that aim to increase the beta cell mass in humans with T1D presents an important avenue to pursue. Beta cell regenerative drugs, exemplified by harmine (with or without the addition of exendin-4) have been extensively studied and shown to be robust inducers of beta cell mass in the past decade. Our studies showed that the increase in the human beta mass was also accompanied by reversal of diabetes. Nonetheless, the detailed mechanisms of action of these drugs remains incompletely defined.
Collectively, the proposed study provides: 1) unique, essential and novel tools for human diabetes researchers; 2) the opportunity to define or refute for the first time the existence of “human alpha-to-beta transdifferentiation”; and, 3) surprising and welcome good news for all people with T1D: abundant alpha cells can be converted to beta cells, even in the rare case where there are no beta cells remaining.