Questions arise about disease-in-a-dish model for some diseases

We have written a fair amount recently about taking tissue samples from people with genetic diseases and reprograming those cells to become induced Pluripotent Stem Cells (iPSC). The goal is to mature those cells into the impacted tissue, say dopamine-producing neurons in Parkinson’s disease, in the dish. Then you can see how they behave compared to normal neurons and perhaps test potential drugs on the cells to correct the defect. But what happens if the cells in the dish stop behaving like the cells in the patient?

Since all your cells have all your genes, getting any one cell to behave as desired is all about turning on the right genes and turning off the others. Parts of DNA that are not actually genes, small tags that get added to or taken away from the outside of DNA over time, handle much of the silencing of genes. A team at The Scripps Research Institute and the University of California at San Diego have found that with stem cells grown in the lab, in some cases, those silencing molecules can lose their function over time. This leaves some genes turned on in the disease model that would normally be turned off in the patient.

The work was published in the May 4 issue of the journal Cell Stem Cell. A Scripps press release on the work quotes the lead author Louise Laurent sounding a cautionary note:

“Our results show that human pluripotent stem cells change during expansion and differentiation in ways that are not easily detected, but that have important implications in using these cells for basic and clinical research.”

This issue seems to be most significant in diseases defined by a single gene from one parent, rather than the usual two genes, with one from each parent having a voice in the cell’s function. Since women have two X chromosomes, normally one is inactivated, but in stem cells in a dish sometimes the second X gets reactivated. So, in those rare X-linked diseases, a disease-in-a-dish model may not mimic the real world disease. We also have a number of genes that scientists describe as “imprinted,” meaning they are only expressed from the chromosome of one parent. The same problem could arise in any disease caused by a mutation in an imprinted gene.

This all requires a caveat. This research has not shown that the cells in a dish actually function differently over time. It may turn out that these changes in the on/off switches don’t impact the usefulness of these disease models. But it does create another layer of complication in getting to our goal. For example, the team found that some of the genetic changes seemed to be linked to the cell culture conditions in which they were grown. So, simply paying careful attention to how the cells are grown might minimize this bump in the path to progress in understanding disease and finding therapies.

CIRM Funding: RT1-01108, TR1-01250, RN2-00931-1

D.G.