On this page
Every cell in your body started from one fertilised egg, and for a long time biologists thought development ran in one direction only. A skin cell stays a skin cell. Shinya Yamanaka showed that this is not a law of nature. With four genes, his lab turned ordinary connective tissue cells back into stem cells. This post tells that story, works through the counting problem his team faced, and draws out what students can copy from his way of working.
What did Yamanaka win the Nobel Prize for?
The 2012 Nobel press release awarded the prize jointly to John B. Gurdon and Shinya Yamanaka "for the discovery that mature cells can be reprogrammed to become pluripotent". Pluripotent stem cells are immature cells that can develop into all cell types of the body.
The two halves of the prize fit together. In 1962 Gurdon replaced the nucleus of a frog's egg cell with a nucleus from a tadpole's intestinal cell, and the egg developed into a normal tadpole. That showed the DNA of a specialised cell still holds all the instructions. But it needed a pipette and an egg. Yamanaka asked whether an intact mature cell could be turned back on its own.
How did the iPS cell experiment work?
The answer appeared in Cell in 2006. In Takahashi and Yamanaka's paper, the authors report inducing pluripotent stem cells from mouse embryonic or adult fibroblasts "by introducing four factors, Oct3/4, Sox2, c-Myc, and Klf4". They named the cells iPS (induced pluripotent stem) cells. The cells looked and grew like embryonic stem cells, expressed embryonic stem cell marker genes, formed tumours containing tissues from all three germ layers when transplanted into mice, and contributed to mouse embryonic development.
A year later the same group reported, in a 2007 Cell paper, that the same four factors produced iPS cells from adult human dermal fibroblasts. Those human cells could differentiate into cell types of all three germ layers.
How do you find four genes among twenty-four?
In his Nobel biography, Yamanaka explains that by 2004 his lab had collected 24 candidate genes that might induce pluripotency. They built a test: mouse cells that survive an antibiotic only if they switch on a gene active in embryonic stem cells. Added one at a time, no candidate produced resistant colonies. Added all 24 together, a few colonies appeared. His student Kazutoshi Takahashi then found that four of the factors were essential.
Here is why pooling first was smart. Suppose you guessed that exactly four genes were needed and tested every group of four.
The number of ways to choose 4 genes from 24 is C(24, 4) = (24 × 23 × 22 × 21) / (4 × 3 × 2 × 1).
The top is 255,024 and the bottom is 24, so C(24, 4) = 10,626 separate experiments.
If you did not know how many genes were needed, the number of possible subsets would be 2^24 = 16,777,216.
Testing all 24 at once answers a cheaper question first: is the answer somewhere in this pool at all? Once that is yes, you can narrow down. The biography does not list every narrowing step, so treat this arithmetic as an illustration of the scale of the problem, not a record of the lab notebook. Our post on permutations and combinations shows how to compute C(n, r) quickly.
Where did a "failed surgeon" fit in?
Yamanaka's biography is unusually frank. He trained as an orthopaedic surgeon, found his surgical skills were slower than other residents', and was moved by patients with diseases that surgery could not fix. He switched to a PhD in pharmacology. Looking for postdoctoral positions, he writes that "a failed surgeon with little experience in molecular biology had a hard time finding a position" until a lab in San Francisco took him. Back in Japan he struggled for funding, cared for mice himself and nearly returned to medicine. A new position at the Nara Institute of Science and Technology in 1999 gave him his own lab.
Has the discovery reached patients?
Carefully, yes. In a 2017 report in the New England Journal of Medicine, a Japanese team that included Yamanaka transplanted a sheet of retinal cells grown from a patient's own iPS cells into one patient with age-related macular degeneration. At one year the sheet was intact and vision had neither improved nor worsened. That is a feasibility result, not a cure, and it is a good example of how medicine moves in small, checked steps.
What can a student take from Yamanaka's way of working?
Change direction when the evidence says so. He moved from surgery to basic research because of what he saw on the wards.
Design the test before the search. The antibiotic-survival test made it possible to screen many genes quickly.
Ask the cheap question first. Pooling all 24 genes checked whether the answer was there before hunting for it.
Report what you saw, including the limits. The 2017 paper says plainly that vision did not improve.
If terms like transcription factor or germ layer are still fuzzy, the Fermi Concept Library covers every concept in biology, so you can fill the gap before you read on. Our guide to the cell cycle, mitosis and meiosis covers the cell division these stem cells depend on.
Next step: write out C(24, 3) and C(24, 5) by hand and see how fast the number of experiments grows.
