Etchevers:Notebook/Genomics of hNCC/2008/06/26

From OpenWetWare
Jump to navigationJump to search
Genomics of human neural crest cells Main project page
Previous entry      Next entry

Nature of stem cells

While I was changing buffers for the H2 in situ hybridization, I was pondering a question a colleague asked me.

They need to reply to a reviewer who challenged their assertion that they found a (dominant) mutation in a mosaic distribution, within the upper layer of the skin (ectoderm) and, because this person's child, the proband, is heavily affected, within the germ line as well. The reviewer thinks that there must be some implication of the mesoderm as well.

This colleague is searching for such potential implication (with allele-specific PCR) but in the meantime, wondered if it were possible to have a precursor cell that only contributes to ectoderm and to the primordial germ cell lineages. The answer is yes, and here is what I wrote back.

The reviewer is correct in pointing out the potential, even probable implication of mosaicism in mesodermal lineages as well as that [we] have observed in the ectodermal and germ cell compartments of the proband's mother. [We] have carefully re-examined her mesodermally-derived lymphocytes and fibroblasts [without finding/and found] the mutant allele. Nonetheless, in the mouse there is precedent for a common precursor to ectoderm/germ cell lineage in the classic fate map experiments of Lawson and Hage (1994) at pregastrula stages. Before gastrulation, such precursors would be localized in the axial epiblast closest to the extraembryonic tissues, well in front of the future head (L&H's zone I). During and after gastrulation, some cells from this region do give rise to both ectodermal and mesodermal derivatives, but by then, the primordial germ cell (PGC) precursors have already redistributed posteriorly within the epiblast before ingressing through the primitive streak.

At mouse embryonic day (E)6, PGC precursors are distributed all around the rim of the proximal epiblast and are subject to signals from both the extraembryonic ectoderm such as BMP4/BMP8b and the visceral endoderm. Tam and Zhou (1996) and Lopes et al. (2007) have shown that ectoderm-competent distal epiblast after gastrulation can continue to form PGC precursors simply in the presence of the extraembryonic ectoderm and visceral endoderm, while PGC-competent proximal epiblast gives rise to lots of ectoderm, but hardly any mesoderm, when distanced from these inductive influences.

Together, these papers strongly imply that there exists a small zone in the pregastrula stage embryo that normally is competent to give rise to definitive ectoderm and PGCs without implicating the mesoderm. We propose that a mutation occurred before gastrulation in the mother, in a multipotent cell localized in the proximal axial epiblast furthest away from the primitive streak.

If you were the reviewer and recognize this response, I'd be happy to talk it over with you (without telling my colleague who you are).

Meanwhile, I put it in my lab book because
(a) it took me two hours to get the background for that, and
(b) I saw that Wnt3a is distributed in the right spatiotemporal pattern in the proximal epiblast to be implicated in this PGC movement from all around the ring at E6.0-E6.25 (pregastrula) to the posterior end at E7.0 (gastrulation)
(c) Dkk1 antagonizes Wnt3 to set up the head territory - but is also dependent on Wnt3 to get transcribed to begin with.
(d) Canonical Wnt signaling through β-catenin needs to be antagonized to get cells to differentiate. That is why GSK3 inhibitors can lead to increased maintenance of stem cell potential (but not to their increased proliferation, rather to increased tolerance of clonal conditions). Although Ying et al also write that "Single blockade of GSK3 therefore has pleiotropic effects, promoting non-neural differentiation, suppressing neural differentiation and enhancing growth capacity." What's that?

This article equates EMT with ES differentiation in feeder-free culture, using the 6-bromo-indirubin-3'-oxime (BIO) GSK3 inhibitor to which Ying et al preferred CHIR99021 for its better effect on viability.

I had been thinking the other day about the other side of that article, implicating FGF-MAPK activation in growth arrest - and here I had been thinking that FGF2 was a mitogen.

(e) Ying et al write that a combination of FGFR inhibitors leads to enhanced propagation and not much commitment (what there is, is neural). The combination with the GSK inhibitors kept the propagation effect and got rid of the differentiation and cell death (for those cells that can't find what they need to differentiate?).
(f) These inhibitors are PD184352 and SU5402 (acting best on FGFR1), which have been used to augment Sonic hedgehog signaling by blocking activation of JNK and less so, of MEK (to phospho-ERK).
(g) I know that Sonic hedgehog is a survival factor for Patched1-bearing mesenchyme-competent cranial neural crest cells.
More, perhaps, later.


Okay, more thinking:

(h) In Wnt3a-/- mice, Oct3/4 (Pou5f1) is expressed throughout the epiblast, instead of getting excluded from the area that would have become the primitive streak (R Behringer in this big document). Do these embryos have Blimp1+ PGC's?
(i) Think on this: "Undifferentiated murine ES cells are capable of reacting to LiCl and Wnt3a (by activating a TCF/beta-catenin reporter)... but have low endogenous TCF/beta-catenin activity. Oct-3/4, nanog and Wnt11 were able to repress TCF/beta-catenin transcriptional activity. During differentiation, activation of the Wnt/beta-catenin pathway influences formation of mesoderm and cardiomyocytes in a time and dose dependent manner."