X-Women Redux

Nature-Nurture-Noise (9)

In this series of posts we have explored irreducible individuality, even in the face of genetic and environmental uniformity. I have provisionally called the source of such individuality, biological randomness and will continue to do so until we need a term that is more than provisional. It’s certainly possible that biological randomness is where we will wind up in the end, but I’m not willing to commit myself yet. I will not deal at all with external randomness, chancy events such as accidents of various sorts.

It's now time to get specific as to sources of biological randomness. Epigenetic processes are a wellspring of biological randomness, from the earliest stages of development. In my epigenetics book,(https://www.amazon.com/Epigenetics-How-Environment-Shapes-Genes/dp/039334228X, the emphasis was on environmental drivers of epigenetic alterations of the genome, that is, the environment as a deterministic influence on epigenetic processes. I did though devote one chapter (X-Women) to a random epigenetic process, known as X-chromosome inactivation. I have also written a bit about X-chromosome inactivation in the series of posts filed under the rubric Epigenetics and Aging, as it relates to cancer. I will assume no familiarity with either. For those who have read one or the other, there will be some redundancy. But this is a deeper dive.

X-chromosome inactivation: A Primer

Our genomes consist of one pair of sex chromosomes and 22 pairs of autosomes (all chromosomes that are not sex chromosomes). Sex chromosomes are so called because in those animals, including most mammals, in which they occur, biological maleness and femaleness is determined by the nature of this pair. Whether the pair consists of two X-chromosomes, or rather one X-chromosome and one Y-chromosome. If the former, you will produce gametes called ova (eggs); if the latter you will produce sperm*. Ova producers are biological females; sperm producers are biological males. If you could produce both, you would be a hermaphrodite. Biological sex is straightforward, unlike gender, which is a subject for the social sciences, and much more complicated. Depending on your criteria, it is reasonable to claim that humans have several genders; but always, just two biological sexes.

Though biological sex determination is wonderfully simple in mammals like us, it greatly complicates embryonic development for both sexes. The developmental problems arise because of the size disparities of the X- and Y-chromosomes. The X-chromosome is one of the largest chromosomes in humans. The Y-chromosome, on the other hand, is puny, one of the smallest chromosomes in the genome. The X-chromosome is also one of the most gene rich chromosomes; the Y-chromosome, the most gene poor.

The size discrepancy of the sex chromosomes poses a fundamental challenge, when it comes to regulating developmental and physiological processes species wide. The problem is this: females, by virtue of doubling up on the X-chromosome, can potentially produce twice as much as males of whatever the genes on the X-chromosome code for. The puny Y-chromosome can provide little compensation in that regard. This is a problem called gene dosage. Male compensation is achieved by epigenetically inactivating one or the other of the female X-chromosomes, either the one inherited from the mother (maternal X-chromosome) or the one inherited from the father (paternal X-chromosome). The epigenetic mechanism is a complex one; it is sufficient for our purposes to note that it involves a long non-coding RNA called Xist (X-inactive specific transcript), one of my favorite acronyms in all of biology.

In humans, as in most mammals, X-chromosome inactivation is random with respect to parent of origin. If this happened at the zygote stage, all cells in some females would have an active maternal X-chromosome and an inactive paternal X-chromosome; in other females, the opposite would occur. But X-chromosome inactivation begins after the zygote has begun dividing. Though X-chromosome inactivation doesn’t occur in the zygote, it does occur very early in development, when the embryo consists of as few as ten cells. It’s as if, in each of the ten cells there is a coin flip, Heads, the paternal X-chromosome is inactivated; tails, the maternal X-chromosome is inactivated. Or vice versa. These coin flips occur independently in each of the ten cells. All the 30 trillion plus cells in adult humans inherit the result of the X-inactivation coin flip, from the particular one of ten cells from which it derives.

In any population of females, say citizens of the United States, the result would be fifty percent paternal X-inactivation and fifty percent maternal X-inactivation. But for any individual female U.S. citizen, the likelihood that five of ten cells are paternally X-inactivated, and five paternally X-inactivated--the fifty-fifty split--is only about 38%. Just as in ten actual coin flips, you can get three heads and seven tails, or six heads and four tails, so too the X-inactivation in embryonic cells. The deviation from perfect parity is called skew. The initial skew is transmitted from the original X-inactivated cell all the way to the vast population of adult cells derived from it. If six of the original ten cells were paternally X-inactivated, then 60% of the adult cells would be paternally X-inactivated. Random X-inactivation is an important source of biological individuality in and of itself, as evidence in copy cat (CC), who failed to inherit her mother’s X-linked calico coloration. Individual differences in X-inactivation skewing amplifies this individualizing effect.

If the maternal X-chromosome and the paternal X-chromosome were genetically identical, skew would be of no consequence. But maternal and paternal X-chromosomes are never genetically identical. Instead, for many genes on the X-chromosome, the maternal X-chromosome has one version—called an allele—of that gene, while the paternal X-chromosome has a different allele. In those cases, skew can have important consequences, with respect to a host of traits, including diseases.

Remember the Armadillo

Recall from the first post in this series that armadillos are always born as genetically identical (hence, same sex) quadruplets. In recent experiments on female quads, the effects of X-chromosome inactivation on individuality were examined (37940702). In armadillos, as in humans, X-inactivation is random with respect to parent of origin. In armadillos it occurs, on average, at the 25-cell stage, which is before the double twinning event, which occurs when the blastocyst consists of 200-300 cells. These 200-300 cells are then divided into four individuals. Let’s say each consists of 50 cells after double twinning. Assuming each of the quads is a random sample of the X-inactivation skew inherited from the original 25 cells, the odds that all of the four will inherit the same X-inactivation skew are small, due to the small sample size effect (called sampling error in statistics).  

The result is individual differences in X-inactivated cells and therefore individual differences in gene expression among the genetically identical female sibs. These differences in gene expression could affect growth, brain development and the levels of various hormones, among other traits, and would help explain the substantial differences that are present in littermates.

What about human monozygotic twins, which though genetically identical often exhibit discordances in disease susceptibility, including Alzheimer’s and Parkinson’s disease. Here too, one possible source of these individual differences may be differences in X-inactivation skew.