Remote cell reprogramming for diabetes treatment

Since I’m not that long in diabetes business, two new Cell papers from Collombat and Kubicek labs looked quite sensational for me. Both are the products of multi-centered collaborations, and both report regeneration of insulin-producing beta-cells in vivo with small molecules.

As I learned from introductions, reprogramming of pancreatic alpha cells (glucagon-secreting) into beta cells is a sort of a Holy Grail of regenerative medicine for diabetes treatment. Naturally, first attempts to reprogramming were performed with aid of transcription factors. But pretty soon small molecules kicked in. These were kinase inhibitors and chromatin-altering probes from Stuart Schreiber lab, resveratrol (of course!), and peptide hormone betatrophin. OK, the last one doesn’t count, and it’s not a small molecule anyway. What’s unusual about the latest Cell papers, is that they describe reprogramming by small molecules acting pretty high upstream from direct gene regulation [1]. Both papers involve messing with GABAA receptor signaling.

Let’s start with the Kubicek lab paper, which found that common (yet Nobel-winning) malaria drug, artemisinin, can make pancreatic alpha cells to secret insulin. The authors identified artemisinin and its metabolite dehydroartemisinin from a library of 280 existing drugs [2]. After they found that the drugs induce insulin secretion, they identified gephyrin as the most likely target. Then, via electrophysiology and a series of inhibitory tests, they linked gephyrin-mediated activity to GABAA receptor signaling. Known agonists of GABAA, however, didn’t increase insulin secretion as much as artemisinin (after 72 h treatment of cells). The drug then increased mass of beta cells islets in zebrafish, healthy and diabetic mice (while reducing basal glucose level in the last ones). Finally, it altered gene expression in human alpha cells and increased insulin secretion by the islets. Frankly, the figure 7A-C, which is supposed to convince in the last effect, raises some questions as data look cherry-picked from different donors. But authors do address that by briefly mentioning donor-to-donor variability. And it’s not surprising at n = 6 sample size.


The paper from Collombat lab branches from the screening results of the first one. Once researchers noticed that activation of GABA signaling correlates with alpha-to-beta conversion, they thought “why not injecting plain ol’ GABA into mice?” And miraculously this simple idea worked. Just look at the jaw-dropping figures 1B-D! Figure S7C,G (below) can somewhat give you the feeling, but go check out the main paper, you won’t be disappointed.

Figure S7 fragment showing increase in insulin-producing cells from rat pancreas. I assume scale bars, if they were present, would be equal (Ben-Othman et al paper)

Here are the main results: daily injections of GABA at 250 μg/kg over three months convert pancreatic alpha cells into beta. But what’s even more exciting is that the new alpha cells are continuously being produced to compensate for those that were converted into beta! They even caught small fraction of cells in some transitional state, where they secret both glucagon and insulin. I particularly liked the discussion section where authors warn that before you, all excited, rush to inject diabetic patients with GABA think why there’s not enough beta cells in the first place. Yes, it is patient’s immune system that attacks her own beta cells. So before this approach makes into clinic one needs to figure out that autoimmune component of type 1 diabetes.

In a sum we have two great papers with rock-solid mouse data and some exciting preliminary results in human beta cells. Let’s see where it will end up. Regardless of the future success, isn’t it amazing how small, simple, and seemingly well-known molecules like GABA (and artemisinin for that matter) can upturn human cells identity?

[1] OK, authors do not strictly claim reprogramming as the identity of cells doesn’t change completely from alpha to beta, but their secretory activity is definitely flipped.

[2] Side note: check out this sexy acoustic liquid handler they used.


MedChem shortcuts

Simplification of complex structures is one way medicinal chemists avoid lengthy (and risky) synthetic routes to the analogs of natural products. And it’s absolutely rational, if one can get the same pharmacological effect with a simple molecule, why bother making a complex 3D scaffold? One classic med-chem textbook example is simplification of morphine scaffold to pethidine that led eventually to the development of fentanyl.


Continue reading “MedChem shortcuts”

Finding new pockets with fragments

Fragment-based design is one of my favorite approaches in drug discovery. It has everything from very simple conception to sophisticated data analysis. The most importantly, it works as it’s supposed to. So it’s always entertaining to find a paper from the very founders of FBDD, I mean Astex Pharmaceuticals and Harren Jhoti himself. Continue reading “Finding new pockets with fragments”

Synthetic lethality as drug discovery platform

One of the features of drug design in the -omics era is the shift from target- and structure-based to function-based drug discovery, when the active compound is identified simultaneously or before the mechanism of action.

A new report in Nature Chemical Biology describes an interesting blend of small molecule high-throughput screening with genetic screening via synthetic lethality. As one might guess, the approach is dealing with cellular death. Traditional ‘simple’ genetic screen identifies individual genes that are critical for cell survival. The principle of synthetic lethality is somewhat different. Scientists seek gene pairs or networks that are crucial in combination but which could be silenced individually without jeopardizing essential cellular functions. Previously it was applied for the discovery of anticancer therapeutics. This time the team from Harvard Medical School aimed at Staphyllococcus aureus. Continue reading “Synthetic lethality as drug discovery platform”

Antiviral sugar superballs


It’s not too common to encounter a compound as one above in a paper tagged ‘medicinal chemistry’. But the team of Spanish and French chemists seems not to care too much about rule of 5 or any other rules of thumb that constrain their creativity. Especially when the rationale for compound design was:

it is difficult to make compounds of an adequate size and multivalency to mimic natural systems such as viruses

The ligands dubbed ‘superballs’ are targeting dendritic cell-specific intercellular adhesion molecule-3-grabbing non-integrin (DC-SIGN). And the multivalency of ligand-receptor interaction is the key factor of viral infection. Previous state-of-the-art ligands were virus-like particles with 1620 mannoses displayed on their surface. So I guess in this study authors significantly optimized ligand efficiency. Continue reading “Antiviral sugar superballs”

Rational serendipity for drug design

With all modern advances and super-sexy scientific tools we currently have, randomness is still arguably the best source of truly innovative discoveries. And scientists do acknowledge that, so various “accelerated serendipity” techniques are flourishing in drug discovery, organic chemistry, chemical biology, and other fields. Continue reading “Rational serendipity for drug design”