X-Women Redux (Part 2) Why Women Live Longer

Nature-Nurture-Noise (10)

Women outlive men worldwide, in all cultures and at every socioeconomic level (PMC5789901). This sex difference is not unique to humans; it is the rule among mammals generally (10.1073/pnas.1911999117). By way of explanation, scientists have looked for sex differences that all mammals have in common. One obvious universal sex difference is the gonadal hormones produced: higher testosterone levels in males; higher estrogen levels in females, for example. But these hormonal differences decrease with age and sex-specific mortality doesn’t.

An increasingly popular candidate explanation for sex differences in longevity is the X-chromosome. A cool demonstration that the X-chromosome is a longevity factor comes from experiments conducted on mice. These mice were genetically manipulated to produce both XY and XX males, as well as XX and XY females. In both males and females, the XX versions outlived the XY versions. So having two X chromosomes, in and of itself, irrespective of biological sex, is advantageous (10.1111/acel.12871).

The fact that this sex difference is mammal-wide may have something to do with the fact that the X-chromosome is the most highly conserved mammalian chromosome with respect to both its gene content and gene order (PMC8327908). Evolution has not messed much with the mammalian X-chromosome once it first evolved.

But let’s focus on humans. Over 500 human disorders have been linked to genes on the X-chromosome, in virtually all of which males more commonly suffer (red-green color blindness, Duchenne muscular dystrophy), or males experience more extreme symptoms than females (hemophilia A and B). Most of these disorders are heavily influenced by deleterious recessive mutations. As such, if the maternal chromosome has the mutation but the paternal chromosome has the normal version, called the wild type—or vice versa—the individual will exhibit little or no evidence of a disorder, but will act as a carrier to the next generation. That’s true of autosomes and sex chromosomes alike, but with a twist regarding the latter.

Since human males have only one X-chromosome (inherited from mom), a recessive deleterious mutation cannot be masked by a wild type allele on a corresponding paternal X-chromosome, because there is no paternal X-chromosome. Therefore, the deleterious recessive mutation is not really recessive, it is, in essence, a deleterious dominant mutation and acts accordingly.

Females, on the other hand, though they have only one functional X-chromosome per cell, because the other is epigenetically inactivated, generally have two functional X-chromosomes in a cell population or tissue. Those cells in which the active X-chromosome has the wild type allele can transfer the wild-type protein to adjacent cells in which the active X-chromosome has the deleterious mutation. This works best when the X-inactivation is about 50-50 (maternal/paternal X-chromosomes). There is built in tolerance for some skewing, but if it approaches 75%/25% or beyond, the female is less protected by random X-inactivation. Disorders usually expressed only in males can become manifest in highly skewed females. Hemophilia is one such disorder that results from extreme X-inactivation skewing (10.7759/cureus.11216).

Some females are born highly skewed by chance alone. But skewing also tends to increase with age because of age-related dysregulated maintenance of X-inactivation (26342808). To see how this works it will be helpful to delve a bit deeper into the epigenetics of X-inactivation.

X-inactivation: More Than Xist

In the previous post in this series, I mentioned that the long noncoding RNA called Xist plays a crucial role in X-chromosome inactivation. Though essential for this process, Xist is not sufficient. Two other epigenetic processes are also necessary. Xist recruits enzymes that modify the canonical histones in ways that cause chromatin condensation (transition to heterochromatin. Recall that in the heterochromatic (as opposed to euchromatic) state, gene activity is suppressed. These histone modifications include removal of activating chemical attachments, and emplacement of suppressing chemical attachments. The chromatin DNA also becomes hypermethylated, thereby further suppressing gene activity.

Recall that there are two types of heterochromatin, constitutive heterochromatin and facultative heterochromatin. Suppression of gene activity in constitutive heterochromatin is much more absolute. Most of the inactivated X-chromosome is constitutive heterochromatin. Facultative heterochromatin, on the other hand, can transition into the gene activated euchromatin state. Genes in the facultative heterochromatin of the X-chromosome can thereby escape X-inactivation, depending on the tissue in which the cells reside. There is also a small portion of the X-chromosome that remains euchromatin. As such, the genes there are potentially active in all tissues. These genes are usually involved in basic cell activities and called housekeeping genes. So dosage compensation is not across the board, X-chromosome-wise. Females get a double dose of some of the most important genes. This explains part of the physiological advantage enjoyed by females, even when there is a high degree of skewing. But for most of the X-chromosome, and hence most of the genes on the X-chromosome, only one allele is active; as such, pronounced skewing is associated with poor health (36412098).

It is easiest to understand the effects of skewing on a single gene locus. Again, assume that the allele on one X-chromosome is a recessive deleterious mutant and the one on the other X-chromosome is the wild type. That is, the individual female is heterozygous at this locus on the X-chromosome. The skew is only problematic, for this gene, if the skew favors the chromosome with the mutant allele, such that, say, 90% of the X-inactive chromosomes carry the wild type allele; and hence 90% of the active X-chromosomes have the mutant allele. In that case, the shuffling of the wild type gene product to cells with the mutant gene product is insufficient for good health.

Very few individuals have a 90/10 skew after initial X-inactivation in early development. But skewing tends to increase with age, as is true of much epigenetic dysregulation. This age-related increase in X-inactivation skewing does not necessarily occur across all tissues, but it does occur in those tissues that are easiest to monitor, including the lining of the mouth and blood. What causes the age-related increase in X-inactivation skewing? It could be due to selection process during development, such that cells with active X-chromosomes carrying the wild type allele are more likely to survive over time than cells carrying the inactive X-chromosomes with the same alelle. But if that were true age-related skewing would not be associated with ill health. Quite the opposite.

Instead, it has been recently proposed, age-related skewing of X-linked alleles results from the accumulation of random epimutations, called stochastic epimutations (SEM). An epimutation is analogous to a genetic mutation, a mistake in replication of the epigenetic mark during cell division. Like somatic mutations, epimutations increase with age but at a much faster rate. The epimutations involving DNA methylation are the most widely investigated. Epimutations of methylation affect all chromosomes including the X-chromosomes. Random methylation epimutations of the inactivated X-chromosome mainly have a reactivating effect for mutated alleles, Conversely, random epimutations at alleles on the active X-chromosome can suppress the wild type allele. Epimutations of either sort will exacerbate skewing for that gene.

Random Epimutations in Monozygotic Twins

Genetically identical female mammals are inherently more variable than genetically identical males, because of random X-chromosome inactivation. The cloned calico cat called copycat was a dramatic example of this. Though her mother was a calico, the genes for which are on the X-chromosome, copycat was a tabby, completely missing the orange pigment characteristic of calico (and tortoiseshell) coat coloration. But X-inactivation in copycat occurred completely independently of the X-inactivation that occurred in her biological mother, whereas, in human monozygotic twins or armadillo quads, X-inactivation occurs before twinning. We would expect less variable X-inactivation and therefore less individual variation in monozygotic twins and armadillos of the same litter. But genetically identical armadillos, nonetheless, evidenced pronounced individual differences because of random X-chromosome inactivation. And for monozygotic human twins, females are more different than are genetically identical males (https://doi.org/10.1375/twin.7.1.54). Moreover, a number of traits for which genetically identical male twins differ more than genetically identical female twins are influenced by genes on the X-chromosome.

With age, female monozygotic twins diverge even more, Whether age-related X-inactivation skewing results in more divergence of genetically identical females than genetically identical males remains to be established.

X-chromosome inactivation is an important source of random individual variation. In the next post I will turn to a more general epigenetic source of random individual differences.