The X factor: The evolution and biology of the X chromosome

02/02/06. By Giles Newton

The human X chromosome holds 300 million years' history of the evolution of sex determination in mammals.

In a nutshell

The X chromosome is central to mammalian sex determination; females are XX, males XY.

The human X chromosome has developed gradually through mammalian evolution.

In women, one X chromosome is shut down (X chromosome inactivation).

X inactivation is incomplete; some genes remain active and activity levels vary between women.

At some point in any good conversation, the topic inevitably comes around to sex. If we reproduced asexually – splitting in two like an amoeba or budding like a yeast – we wouldn't need to discuss such matters, nor waste time searching out a partner of the opposite sex. But like the majority of the natural world, humans are driven by the need to bring together males and females to produce the next generation.

Such a union brings together the sex-determining chromosomes: if the X from the mother's egg is matched with an X-carrying sperm, an XX female results; with a Y-carrying sperm, an XY male.

Unlike the 22 other chromosomes we carry, which come in identical pairs and happily meet up and swap DNA, the X and Y are distant relatives. The X is big, the Y tiny, with little in common DNA-wise, apart from two small regions at their tips. Furthermore, a remarkable mechanism of 'X chromosome inactivation' enables females to shut down one of their two Xs and maintain genetic parity with XY males.

Such peculiarities have sparked many debates over the years. If the X and Y started off as two 'normal', non-sex chromosomes back in evolutionary history, as Susumu Ohno proposed in the 1960s, is there an imprint left of their origins? How much DNA do the X and Y actually share in common? And how does the inactivation signal spread along the X – does it have 'boosters' to help it move along the chromosome?

An extra frisson of excitement therefore accompanied the publication in March 2005 of the X chromosome's DNA sequence, coming as it did with the answers to some of these questions and clues to others. For Mark Ross at the Wellcome Trust Sanger Institute, who led the sequencing of this string of 155 million As, Ts, Cs and Gs, the publication was also the culmination of a decade's work and a career's infatuation: "Even when I was at university, I was fascinated by the sex chromosomes," he says. "Fascinated because of their difference from each other and from the other chromosomes. And because at some stage, they evolved from an ordinary pair of chromosomes."

The sequence data revealed that the X chromosome has 1100 or so genes, dramatically more than the Y, with its meagre 76. Even so, Dr Ross points out, the X has relatively fewer genes than one might expect for a chromosome of its size. "There are two possibilities," he speculates. "There may have been an evolutionary selection pressure for genes to move from the X to other chromosomes. Or it may be coincidence: the original chromosomes from which the X was derived may have had relatively few genes."

A tale of X and Y

The story of the X and Y chromosomes begins with an animal that lived some 300 million years ago. This animal was the ancestor of mammals and birds and, it is presumed, it did not use chromosomes to determine its gender. (Quite how it did so is unknown, but it may have used one of the many other mechanisms that exist in nature; in crocodiles, for example, the incubation temperature of the egg determines sex.) When mammals and birds went their separate evolutionary ways, each co-opted a different pair of the ancestral animal's chromosomes to determine sex. In mammals the chromosomes became the X and Y, in birds the W and Z chromosomes.

"As they started off from different pairs of chromosomes, X and Y and W and Z are not similar at all," says Dr Ross. "But we can look at the chicken genome and see the ordinary chromosomes that are related to the mammalian X and Y [the X chromosome is closely related to chicken chromosomes 1, 4 and 12]. Similarly, you can take the bird W and Z and find the equivalents in the human genome."

For mammals, the trigger for this new sex-determination system was the emergence of the Y-chromosome SRY gene – the master switch in male development. Over time, the majority of the Y chromosome DNA lost the ability to swap with the X; without a partner it could use as a template to repair damage, the Y has gradually degenerated to a shrunken stump and lost most of its genes. ("Presumably something similar happened in birds, although we don't know how bird sex determination works yet," says Dr Ross. "In this case, the W has decayed and, interestingly, females are ZW and males are ZZ.")

Evolution has been harsh on the Y chromosome, but the X has not been immune to change over time. Its journey is being charted by Dr Ross and colleagues by comparing the human chromosome sequence with the X chromosome of other mammals. "We don't have the ancestral X chromosome to compare the human X chromosome sequence against, but we can reconstruct it by comparing the human X to other genomes and working out the changes that have occurred and their timings."

The picture that has emerged is of an X chromosome that has gained pieces of DNA at various stages in mammalian evolution, the biggest change being the addition of a large region from another chromosome during the evolution of the placental mammals. Thereafter, alterations have been more subtle, the human X being very similar to those of chimps, dogs, mice and rats.

Such conservatism is probably due to dosage compensation: an upset to the careful balance of X chromosome inactivation, leading to too much or too little gene expression, could have been detrimental indeed. Instead, Dr Ross explains, it appears that chunks of DNA have been acquired by the X and Y chromosomes simultaneously, so there would have been two active copies. "Over time the Y chromosome segment has decayed and the X copies have been recruited into X chromosome inactivation in a gradual fashion."

The off switch?

X chromosome inactivation is a fascinating system, uncovered by Mary Lyon in the 1960s. To prevent females, with their two X chromosomes, getting double the male dose of gene products, one of their X chromosomes is shut down early in development of the female embryo.

In mice, the inactivating signal is triggered by RNA from the Xist gene, the heart of the 'X inactivation centre' on the chromosome. Initially, RNA is produced from both X chromosomes in the female, but when the choice of which X to inactivate is made – the choice being random in placental mammals, but always the paternal X in marsupials – Xist expression is shut down on the active chromosome. This is controlled by the Tisx gene, a mirror image not just in name: the Tisx gene overlaps with Xist and is transcribed in the opposite direction.

On the X destined for inactivation, Xist RNA spreads along and coats the chromosome – its localisation being helped by BRCA1, the breast and ovarian cancer tumour suppressor – recruiting proteins that compact the DNA-chromatin structure (see Mother's pride). Some proteins modify the tails of the histone proteins by methylating, demethylating, deacetylating and ubiquitinylating particular amino acids. (The Polycomb group proteins Eed/Enx, for example, methylate a lysine residue of histone H3.)

Other changes, such as DNA methylation and the recruitment of a variant histone called macroH2A, help make sure that the chromosome in a highly compacted, silenced state. Although a human X chromosome is silenced, the mechanism may not be the same as in the mouse. This may explain surprising differences in the extent of X inactivation between the two species.

It was originally thought that this silenced X would be completely inert but, using the genome data from the Sanger Institute, Hunt Willard (Duke University) and Laura Carrel (Pennsylvania State College of Medicine) showed that this was far from true. Examining the activity of hundreds of genes on the inactive X, they found that about 15 per cent 'escape' from inactivation; even more surprisingly, another 10 per cent of supposedly inactive genes are silent in some women but active in others. In fact, each of the 40 women they studied had a unique pattern of gene activity.

"The extent of escape from inactivation was unexpected," says Dr Ross. "In mice, inactivation is more complete."

So what are the consequences of this escape from X inactivation? It is too early to tell, but it may well underlie the effects of some sex chromosome anomalies. Some XXXY are more severely affected than XXY, perhaps because of the extra activity from escaping genes. Conversely, people with an X chromosome alone (XO – Turner syndrome) may be affected because these genes, perhaps genes in some regions shared by the X and Y, are required in double doses.

Even in XX females, the 200-300 genes that are potentially more active than in males may be crucial to gender differences, sex-specific traits and complex diseases. Perhaps, as Hunt Willard suggested, there may not be one human genome but two – male and female.

As for the mechanism of escape, could it relate to the absence of X inactivation 'boosters'? In 1998, Mary Lyon suggested that a certain class of repeated DNA called LINE1 repeats, known to be relatively frequent on the X, might be act as way stations to boost the signal. "When we looked at the sequence, the LINE1 distribution fits with this idea," says Dr Ross. "The concentration is very high around the X chromosome inactivation centre and neighbouring regions. As you go into regions where more genes are escaping, LINE1s are less frequent. It fits, although it's not conclusive as yet."

"We want to know the differences between regions of X chromosome inactivation and regions of escape. DO the genes that escape form domains, or do they escape on a gene-by-gene basis? And what are the DNA and chromatin modifications [epigenetics] in those regions? Only when we understand these issues will we be able to address the way-station hypothesis properly."

The X factor

Mutations in X chromosome genes usually affect males only, females having a back-up X (but acting as carriers).

Male-specific haemophilia, for example, was recognised, though not named, in ancient times: the Talmud, a collection of Jewish Rabbinical writings from the second century CE, stated that male babies did not have to be circumcised if two brothers had already died from the procedure. The Arab physician Albucasis, who lived in the 12th century, wrote of a family whose males died of bleeding after minor injuries. And in the 18th century there are physicians' descriptions of similar conditions and of colour-blindness, affecting boys in families and apparently passed on through females.

Altogether, more than 300 X-linked genetic diseases have been identified, and many of the genes involved have been found: "We know of about 40 cases where the sequence contributed in some way to the discovery of the affected gene," says Mark Ross. "Even so, there are still a large number of diseases associated with the X where the molecular basis is unknown."

One big surprise was the discovery of nearly 100 'cancer/testis antigen genes'. "It surprised us that there were so many, particularly because there are relatively few elsewhere in the genome," says Dr Ross. "We know relatively little about these genes. They are usually expressed in the testes, and perhaps they confer some kind of male advantage. And we know that they become activated in certain types of tumour – in fact, they have been suggested as possible targets for cancer immunotherapy."

References

Ross MT et al. The DNA sequence of the human X chromosome. Nature 2005;434(7031):325–37. Abstract

Carrel L, Willard HF. X-inactivation profile reveals extensive variability in X-linked gene expression in females. Nature 2005;434(7031):400–4. Abstract