Q J Med 2004; 97: 381-382
QJM vol. 97 no. 6 © Association of Physicians 2004; all rights reserved.
Biologic |
How does X mark the spot?
Mammals can be distinguished as male and female by their differing number of X and Y chromosomes. Although sex determination may be dependent on a number of different mechanisms, there has to be a way of avoiding a double dose of active X genes in those with two Xs. This is achieved by a number of differing mechanisms in different groups; in XX C. elegans, genes on both X chromosomes are expressed, but each X is repressed by half, relative to the output of a single male X (Xm). In Drosophila, the solution is to up-regulate genes on the male X instead of down-regulating genes on the female X. In mammals, the double dose of X in females is avoided by imprinting the paternal X to undergo inactivation—it is inactive in placental tissues. Inactivation in the embryo, however, happens at random, resulting in a mosaic pattern that persists in adults (a bit early for a diversion, but did you know that around 1/2000 phenotypically normal individuals are mosaics for non-sex-chromosome-carried genes?). Marsupials always inactivate Xm, although not very efficiently.
The essential point is that a double dose of X genes must be avoided. How is this done? In Man it has been though that X inactivation is first seen in the blastocyst at day 3.5, when it consists of an inner cell mass (ICM) and an outer trophoectoderm. The ICM consists of primitive endoderm cells and a mass of cells that will be the embryo and a core of pleuripotent cells (stem cells). In the classical view, preferential inactivation of Xm occurs in the trophoectoderm and primitive endoderm at 3.5 days; in the embryo, proper X inactivation begins at day 6.5 (the beginning of gastrulation) and is random. A number of recent papers, using different approaches, have shown that Xm inactivation occurs very early, at the two- or four-cell stage (a major problem for cloners). It begins to be clear that it happens in two main phases.
The first and very early event depends on Xist (X-inactivation-specific transcript), a large (17 kb) RNA non-coding molecule that coats the X that is to be inactivated. The coating attracts an enzymically active complex (Ezh2/Eed) that alters histone H3 by methylating lysine 27 and possibly lysine 9. At the 32-cell stage, histone H3K27 is also methylated. Now, as X inactivation is random in the embryo and selective in the extraembryonic tissues, it may be that this randomness depends on some cells in the morula escaping methylation—these cells may be those that form the ICM. But division of blastomeres in the 8-celled zygote is asymmetrical, and this could also account for the development of the inner cells that form the ICM; they would perform differently as a distinctly derived cell population, in lineage terms. Mak et al. (2003)1 have shown an extraordinary thing; it appears that Xm inactivation occurs in all cells in the zygote, both extra-embryonic and in the ICM, and that Xm is then selectively reactivated in the ICM. The process occurs only in the cells of the inner core of pleuripotent epiblast cells (the epiblast is the cell mass that sits on the primary endoderm inside a sphere of trophectoderm) and not in the primary endoderm. Having been reactivated, Xm is then subsequently inactivated at random.
This second phase consists of changes that occur in the DNA of the X chromosome, allowing re-programming of the genome. The change responsible for persistent alteration of gene activity is often considered to be the methylation of critical cystine residues at CpG dinucleotides in the promoter and 5’ region. The pattern of stage- and tissue-specific DNA methylation is established around the time of gastrulation. Once established, the genetic inactivity is maintained clonally, and spontaneous change in status is rare (it will occur in the germ line as the male or female imprint in sperm and eggs). The X chromosome contains a large number of housekeeping genes (which are normally unmethylated), but the CpG islands of most X-linked genes are still methylated on the inactive X chromosome.
Now this is the way in which many things are done in development: a blanket inhibition followed by selectivity in cell activation. It is the way in which Parliamentary bills are drafted; having been involved with the drafting of the Food and Environment Protection Bill, I can paraphrase the process by saying that the lawyers usually suggest something like: Clause one. All pesticides are banned. Clause two. Notwithstanding Clause one ... and so on—a series of exceptions are proposed. This is a device to allow you to prevent the use of anything you hadn’t thought of if it turns up later, but it is difficult to see how selection pressure could work in this way. In fact, many developmental switches depend on the ability to allow differentiated cell populations to develop while maintaining a stem cell population; reactivation of a previously suppressed genetic activity is the basis of differentiation in many tissues and this is, I suppose, another version of the process.
There is more to come in this field, driven in part by an increasing need to know about early events post-fertilization in vitro. But as I have given no party points this month, how about this: the turtle is unusual in vertebrates in having an ‘exoskeleton’ of about 50 bones. The ventral part of his shell (the plastron) is made from neural crest cells by a process akin to the membrane bone ossification seen in the formation of the human skull. The top is made by ribs which grow straight out, pushing through muscle and inducing the surrounding mesenchyme to form bone by the activity of BMPs. We could all do it if we tried.
References
1. Mak W, Nesterova TB, de Napoles M, Appanah R, Yamanaka S, Otte AP, Brockdorf N. Reactivation of the Paternal X chromosome in early mouse embryos. Science 2004; 303:666–9.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||