In patients with macular degeneration, photoreceptors in the eye slowly die. Since the body has no source of new cells to replace the ones it lost, vision fades to blindness. If fresh new cells could be added like spackle on drywall, that could help ameliorate macular degeneration and a litany of other conditions like heart failure and liver cirrhosis.

Undifferentiated cells can transform into whatever tissue type is needed to paper over wounds or regions of cell death. Stem cells are sourced from embryonic tissue to make human embryonic stem cells (hESCs) or are made by dedifferentiating mature cells to create induced pluripotent stem cells (iPSCs). Scientists most commonly derive iPSC from skin cells, but blood cells are also a common source.

For stem cells to be therapeutically useful, they have to be healthy. Previous studies done on hESCs found mutations popping up repeatedly in cancer-associated genes like TP53 as well as rearrangements of whole chromosomes (1). A new study published in Nature Genetics profiled the mutational burden of skin- and blood-derived iPSCs from a single human donor to better understand the potential usefulness and risks of using stem cells therapeutically. Giving a patient a highly mutated batch of stem cells might result in failed treatment, or potentially hurt more than it helps if the stem cells carry oncogenic mutations. The authors reported that evolutionary pressures and the rigors of age give iPSCs more underlying mutations than previously suspected (2). 

In the course of routine iPSC lab work, Foad Rouhani, a transplant surgeon at the Wellcome Sanger Institute and University of Cambridge and co-author of the paper, noticed something unusual with one set of cell lines. “There was this one patient that we focused on who had one iPSC line with thousands and thousands of mutations and another line that had maybe a tenth of the number of mutations. That sparked our interest,” said Rouhani.

Rouhani’s group sequenced two populations of skin-derived iPSCs and two populations of blood-derived iPSCs from a 22-year-old healthy man. They performed three different types of sequencing: whole-genome sequencing (WGS), whole-exome sequencing (WES), and high-coverage whole-exome sequencing (hcWES). 

WGS didn’t detect structural genomic changes in either skin- or blood-derived iPSCs, but it revealed a significant mutational burden, with huge variation even within cell lines created from the same tissue and from the same person. One of the skin-derived iPSC lines Rouhani studied carried more than 10,000 mutations, while the other had only 2000 or so. Most of the sequence mutations are associated with those induced by UV-damage to the skin over a lifetime of sun exposure.

This variation from one cell line to the next didn’t arise simply from a patient who tans unevenly. Rouhani's team next used the stem cell genomics and proteomics database HipSci to compare pairs of skin-derived iPSCs to see if banked stem cell lines from other patients also showed variation. 46 percent of pairs of iPSCs shared types and burdens of mutations, but the other 54 percent had huge variations. One pair included a cell line with 900 substitutions, while another cell line from the same patient had 11,000, most of which were also UV-associated mutations.

Detecting a mutation does not mean a functional impact since mutations often create no change in protein sequence or expression. One concern for the researchers was cancer-causing mutations since even healthy cells carry a high level of cancer-associated mutations (3). The team found that “actually, none of the sort of common cancer genes really came up,” said Rouhani. “Except for one.”

That one mutation was in a gene called BCL6 Corpressor (BCOR). Mutations in BCOR appear in a number of different cancers, most commonly in leukemia. The Rouhani group found BCOR  mutations in stem cell lines from their test volunteer. To double check that this wasn’t just a one-off mutation, they also looked for BCOR  mutations in another stem cell bank, the Insignia project. 27% of cell lines from Insignia also carried BCOR  mutations. A significant chunk of the BCOR  mutations found both in the original cell line and in Insignia were predicted to be pathogenic.

These BCOR  mutations arose much more frequently in blood-derived iPSCs than in the more commonly used skin-derived cell lines. To see what consequences mutations in BCOR  had, the group tested the ability of BCOR  mutant stem cells to become neurons over a 27-day period and saw that mutants repeatedly failed to convert into neurons, whereas blood-derived iPSCs with functional BCOR genes differentiated successfully, suggesting that BCOR  mutations impaired stem cells’ abilities to differentiate into other tissues.

Looking back at the genomes of the original blood cells from which blood-derived iPSCs were made, BCOR  mutations were nowhere to be found. Sometime between harvesting cells from patients and creating and culturing iPSCs, mutations appeared. For reasons unclear to the authors, this unknown mutational process seemed most focused on creating mutations in BCOR during the stem cell reprogramming process.

While finding recurring mutations in iPSCs is worrying, this isn’t the end of the road for iPSCs from blood or from skin — far from it, in fact. It’s just a matter of researchers knowing what mutations are present in stem cells before the cells are used for research or put into a patient’s body. 

“We and others have been involved in work to analyze embryonic stem cells or iPS cells by whole genome sequencing and see what mutations are present,” said Florian Merkle, a neurobiologist at the University of Cambridge who was not involved in the study. “As long as you know what mutations are present, and you know that they're not likely to affect your target cell population, I think it's perfectly fine to use.”

According to Merkle, future researchers may prefer stem cell lines derived from blood due to their lack of exposure to UV damage and the ease of procuring blood samples. Or, given that skin-derived lines are already more common, researchers may begin sampling from parts of the body that don’t see as much sunlight.  

“It's not glorious work to do that deep characterization,” of iPSC growth, said Merkle. His team recently found a mutation in one of their key cell lines. “It's in a chromatin associated gene that's likely to affect its behavior and likely to affect its differentiation,” said Merkle. “That mutation was present in the fibroblasts. And this is an example of one of these UV-induced mutations. But we've sequenced it; we're aware of it; and we were able to correct it. And now that line is being very widely used.”

References

1. Jaiswal, S. & Ebert, B. L. Clonal hematopoiesis in human aging and disease. Science  366, (2019). 
2. Rouhani, F. J. et al. Substantial somatic genomic variation and selection for BCOR mutations in human induced pluripotent stem cells. Nature Genetics (2022). doi:10.1038/s41588-022-01147-3 
3. Williams, M. J. et al. Quantification of subclonal selection in cancer from bulk sequencing data. Nature Genetics  50, 895–903 (2018).