Takahashi K and Yamanaka S. "Induction of Pluripotent Stem Cells from Mouse Embryonic and Adult Fibroblast Cultures by Defined Factors." Cell 126, 2006: 663-676.
Introduction:
ES cells are pluripotent: they come from mammalian embryo and can make cells from all 3 germ layers.
- super useful for treating diseases
- but ethically controversial (can we use human embryos for this purpose?)
- and physically difficult if implanted tissue is rejected by host
possible solution? get pluripotent cells from host themselves!
- take normal body cell (somatic cell) DNA and stick it in an oocyte or fuse the cell with an ES cell
- the other cell contents of ES cells/oocyte have been shown to contain some factors that make somatic cells pluripotent!
- we know some factors that maintain pluripotency, but maybe these factors also INDUCE pluripotency
what do we know about factors that maintain pluripotency:
- Oct3/4, Nanog handle maintenance
- Stat3, E-Ras, c-myc, Klf4, Beta-catenin are highly expressed in tumors--they also help maintain long term ES cell phenotype and proliferation
Results:
We tried out 24 genes that could have been the factors that induce pluripotency.
- experiment: if gene X induces pluripotency, cell will be resistant to G418 (a molecule that inhibits protein synthesis).
- ergo, if we see cell resistant to G418, then the gene that we unregulated in the cell is a factor that induces pluripotency.
step 1: in mouse, knock out the gene Fbx15. This gene is super important for maintaining pluripotency in mouse development.
- ES cell with knocked out Fbx15 resist G418
- somatic cells with knocked out Fbx15 cannot resist G418
step 2: one by one, insert each of the 24 candidate genes into these knockout embryonic mice cells
- no resistance observed
- ergo, these genes cannot induce pluripotency by themselves
step 3: upregulate all 24 genes in these knockout cells
- get lots of resistant colonies!
- some of these look very similar to ES cells (morphology, proliferation traits, gene expression markers, etc)
step 4: upregulate all but 1 gene in these knockout cells
- found 10 genes that, once you did NOT upregulate them in the cells, you did not get resistant colonies
- ergo, these are super important factors that you can't NOT put in in order to induce pluripotency
step 5: upregulate all 10 special genes in knockout cells
- get lots of resistant colonies!
- many of these look similar to ES cells
step 6: upregulate all except 1 of these 10 special genes in knockout cells
- not including either Oct3/4 or Klf4 resulted in no resistant colonies
- not including Sox2 resulted in very very few resistant colonies
- not including c-myc resulted in weird looking resistant colonies
- not including any of the others produced all resistant colonies, so the others were not as important as the above 4.
step 7: upregulate only those 4 super special genes in knockout cells
- get same result as step 5
- culture and confirm that these are iPSC (induced pluripotent stem cells)
step 8: upregulate pairs and triplets of these 4 super special genes in knockout cells
- no two of them could form any resistant colonies
- 2 triplets produced a few colonies but they did not survive further culturing
- 2 other triplets produced more colonies but looked weird (different from ES or previously determined iPSC)
step 9: do gene expression analysis of iPSCs induced with the various combos
- the 4 combo and the 10 combo cells, both are similar to ES cell expression profiles but not exactly the same
- the 3 combo cells were very very different
step 10: try to make teratomas with these various combo-formed iPSCs
- there was inconsistent data: some of the 10 combos and the 4 combos made teratomas with cell types of all 3 layers, but some of the same combos could only form 2 or 1 of the germ layers
- so conclude the majority of the 10-combo and 4-combo cells are pluripotent, but not all
- tumors from 3-combo cells did not differentiate = not pluripotent!
- similar results when trying to form embryoid bodies in culture as opposed to forming teratomas in vivo.
step 11: introduce the 4 combo gene into mouse tail fibroblasts (somatic cells) and then inject these cells into blastocyst
- were able to observe that these injected cells helped form some of the germ layers and baby mice were actually born from these blastocysts that received injections!
step 12: compare gene expression levels of these 4 factors with protein expression levels btw iPSC and ES
- saw that while some of these genes were higher or lower in iPSC cells than in normal ES cells, the protein levels (Western blot!) were about the same!
step 13: try to grow iPSC without them differentiating in culture
- they always differentiated unless they were provided "feeder cells" in the same culture
Discussion:
- Oct3/4, Sox2, Nanog are essential for maintaining pluripotency
- Oct 3/4 and Sox 2 are essential for MAKING iPSCs
- Nanog is not important for that
- c-Myc Klf4 are also essential
- c-Myc upregulates genes for proliferation and transformation
- it affects some histone modifying enzymes (histone acetyltransferase, for example)
- there are a LOT (upt to 25000) of sites for c-Myc binding in mammal genome
- this is way more than what we'd guess for Oct3/4 or Sox2 binding sites
- it could be that c-Myc causes global histone acetylation, causing a lot of the genome to open up, so that Oct3/4 and Sox2 can find all their target binding sites
- what about Klf4? represses p53
- okay, what does p53 do? It suppresses Nanog during differentiation
- so if you repress p53, you enable Nanog, which should normally NOT be active for differentiation.
- this might contribute to making the iPSC or at least ES-like cell phenotype
- Klf4 activates p21 which suppresses proliferation, and c-Myc suppresses p21. This opposites relationship of c-Myc and Klf4 might be important (in other words, we're just guessing)
- one important question: which cells of the tissue given these four factors are becoming iPSCs?
- only a small portion of cells treated with the 4 factors become iPSCs
- maybe it's the progenitor/stem cells that already exist in tissue that are kinda multipotent but not pluripotent that transform into pluripotent cells
- the frequency doesn't change when we try this out with bone marrow, which should have a high percentage of progenitor/stem cells to begin with to change
- so it can't be those cells..
- maybe getting the right expression level of each factor in the cells is important
- experimental evidence: just a 50% increase or decrease in Oct3/4 proteins in an ES cell causes it to differentiate and lose pluripotency
- we know our iPSC clones overexpress RNA levels but their protein levels of the 4 factors are just right
- but these cells must be able to regulate that, b/c high high levels are necessary to become ES-cells but in order to stay ES-like, too much of the 4 factors is badddd
- they might need some chromosomal alterations too to stay ES-like
- this may be spontaneous or induced by some of the 4 factors
- where the retrovirus brings in the transgenes to overexpress the 4 factors also matters: could have impacted the expression of any native genes depending on how the transgene got shoved into the genome
- another question: are these 4 factors also important when we're trying to reprogram somatic cells by fusion with ES cells or plucking out nuclei and putting them in oocytes?
- the precise roles of Klf4 and c-Myc are also confusing and vague. they aren't essential for mouse development before the egg implants. c-Myc isn't detectable in oocytes at all. Hmm??
- well, related proteins, L-myc and Klf17 and Klf7 do exist. maybe Klf4 and C-myc's real properties are being supplanted by these relatives in wildtype development
- some other questions that this paper brings up...
- still unsure if these 4 factors can make pluripotent cells out of human somatic cells.
- testing/experimental process is going to require super specific culture environments
- but this is all really cool in the search for the tools to control pluripotency, and one day we might be able to make pluripotent cells from a patient's somatic cells.
No comments:
Post a Comment