Chemical biology 21: stem cells and regenerative medicine

These are my notes from lecture 21 of Harvard’s Chemistry 101: Chemical Biology Towards Precision Medicine course, taught by Dr. Stuart Schreiber on November 24, 2015.

The goal of regenerative medicine is to treat diseases of “cellular deficiency”. The hope is that either embryonic stem cells and iPS cells can be differentiated into cell types of interest, or that somatic cells can be directly transdifferentiated into other cell types of interest. It is also hoped that small molecules will soon make it possible to do reprogramming and differentiation without introduction of any exogenous DNA.

Here are some examples of diseases that can be viewed as diseases of cellular deficiency:

Like most new therapeutic modalities, the hype to science ratio for regenerative medicine has been very high.

One of the earliest clinical attempts to implement regenerative medicine was for Parkinson disease, by grafting human embryonic neural tissue into the substantia nigra of PD patients. I previously discussed this history here; despite isolated early reports suggestive of success [Sawle 1992, Piccini 1999], it was eventually determined that the therapy had been ineffective [Olanow 2003].

Another early attempt was to replace a trachea scarred by chronic tuberculosis. A doctor used a donor trachea, treated it chemically to kill all donor cells while retaining the structural scaffold, then coated it in bone marrow stem cells from the recipient patient’s hip, grew the cells for 4 days, and then transplanted it. For many years this success story was marred by accusations of scientific misconduct against the doctor, but a couple of months ago he was cleared of all charges.

One oft-cited example is the Berlin patient. The Berlin patient was an AIDS patient who had leukemia and needed radiation and bone marrow transplantation. The doctor managed to find an HLA-matched bone marrow donor who was also homozygous for the CCR5 Δ32 frameshift variant, which eliminates a receptor for HIV-1 and confers greatly reduced susceptibility to HIV infection. To this day, the Berlin patient may be the only person ever cured of HIV. This inspired clinical trials to use zinc finger nucleases to delete CCR5 in patient cells and then do autologous stem cell transplantation, as well as the development of small molecules targeting CCR5. However, since then, several AIDS patients have undergone transplantation of bone marrow from CCR5 null donors, and none have responded like the Berlin patient did — none have been cured. So it is possible that the Berlin patient was just extremely lucky and that this success is not reproducible.

John Gurdon won half of the 2012 Nobel Prize in Physiology or Medicine (shared with Yamanaka for iPS reprogramming) for somatic cell nuclear transfer, the first method of cloning an organism. He started by transferring a nucleus from a tadpole somatic cell into an enucleated frog egg. For decades, people dismissed this discovery as being specific to frogs, rather than generalizable across life. The significance of his work was finally accepted when the sheep Dolly was successfully cloned using the same method Gurdon had used. Dolly had many health problems and only lived 7 years, which is short for a sheep. This proved that cloning was possible, but also that there were many problems yet to be resolved.

The feasibility of cloning by somatic cell nuclear transfer proved that the reprogramming factors that determine cell fate were present in the cytoplasm. What were they? One obvious candidate would be transcription factors, particularly master regulator transcription factors. Shinya Yamanaka tried every combination of master regulator transcription factors until he found the minimum set of 4 genes that had to be expressed in order to make a differentiated cell revert to pluripotency. This is most commonly done with fibroblasts from skin punches, but the efficiency (proportion of treated cells that acquire pluripotency) is ~100-fold higher with keratinocytes from plucked hair follicles [Aasen 2008]. Early work used integrating retroviruses, but then it was found that non-integrating viruses such as Sendai could also transfer the required genes, and then it was found that just transfecting the mRNA would work, and then that chemically modified, extra stable RNA worked even better. The newest frontier is using small molecules [reviewed in Firestone & Chen 2010]. Early protocols used several modulators of DNA methylation and histone modification, such as aza-cytidine and valproic acid, to increase the efficiency of existing DNA- or RNA-based approaches. A couple of years ago there was a report that the reprogramming could be achieved with only small molecules and no nucleic acids [Hou 2013]. Making reprogramming easier and more efficient, and interrogating whether the resulting cells really possess embryonic-like properties, is an ongoing challenge [reviewed in Krupalnik & Hanna 2014].

Transplantation of pancreatic beta cells is sometimes used as a therapy for type 1 diabetes. However, there are not enough donors to go around, and recipients have to take immunosuppressants for the rest of their lives. So autologous transplantation is an appealing possibility. Doug Melton recently developed a protocol for differentiating beta cells from human ES cells in culture [Pagliuca 2014]. In vivo, ES cells become pancreatic beta cells through many separate steps of differentiation. To make a simpler cell culture protocol, it was desirable to find small molecules that could jump some of the intermediate steps. To do this, they put reporters (GFP or RFP) under the promoters of transcription factors that define certain cell states, and screened for molecules that would turn them on. They originally figured out a way to get to pancreatic progenitors using two small molecules [Borowiak 2009, Chen 2009], but it took many years to figure out a protocol to get all the way to beta cells [Pagliuca 2014]. Another interesting approach is transdifferentiation of acinar cells into beta cells [Li 2014].

Therapeutic transdifferentiation is an appealing approach to get cancer stem cells to differentiate so that they can be killed.

Shortly after protocols for making and differentiating iPS became available, they became a popular tool for modeling particular diseases in the lab — see for instance ALS [Dimos 2008], Rett syndrome [Marchetto 2010], Alzheimer disease [Israel 2012], and many many more. Yet making any stem cell-based regenerative therapy safe and effective has so far proven elusive, and this continues to be an area of great challenges, including for chemists to make small molecules to achieve targeted differentiation.

Regenerative medicine is also an area of many ongoing controversies. Recently in the news, NIH announced it was considering a “possible policy revision” on human-animal chimeras to grow organs. The background here is that pigs have organs of a similar size to humans, and pig organs can actually be transplanted into humans in some cases, but with a need for intense immunosuppression to even temporarily stave off rejection. A new idea is to modify pig embryos to be unable to generate certain organs, and then inject into the embryo human stem cells so that specific organ will be human and could be transplanted (hopefully) without rejection. This idea, of pigs with human organs, has proven controversial.