Top Reddit comment:

Matching the MHCs of two different individuals in a research setting is not very difficult. In fact, it's accepted you can probably even get away with just a 2/3 match or so between donor and recipient. Likewise, using immunosuppressants is also standard fare.

So why is this a Nature paper?

While the promise of stem cell therapies is often sold as individuals each getting their own cells turned into stem cells, modified, and reimplanted to cure them of a disease, this would be prohibitively expensive in practice. What has been proposed to combat this is a system called haplobanks--cell repositories filled with cell lines who's haplotype (self vs non-self signals, the MHC in this study's case) covers a large portion of the population. Then when you get sick, you can go to the doctor, get yourself sequenced, and then have stem cells which match you well on hand that can be used therapeutically right away, without the cost or time of reprogramming. (This might work better in other countries than America for diversity reasons, but that's another discussion.)

That's the flashy thing about this from a scientific perspective: it's taking cells from one individual to another like you might in that hypothetical clinical context.

200+ Reddit comments:

https://reddit.com/r/science/comments/56sd68/stem_cells_from_one_monkey_successfully/