This week's Nature has an article summarizing the sequencing of the human X chromosome by Ross et al. (that should be Ross ET AL.!!!; see the author list at the end). There is an impressive wealth of quantitative and genetic detail here, but I'm not going to reiterate it. Mostly, I want to outline the evolutionary story.

And it really is an obligate evolutionary story they're telling. A paper on the sequence of a chromosome is not just a recitation of As, Gs, Cs, and Ts—it is about extensive analyses, comparisons of genomes from different species, reconstructions of past translocations, inversions, and mutations, and about the logical and mathematical modeling of the history of transformations that produced a particular arrangement of genes. What we have in the X chromosome is a text that shows the smudges and strike-outs and rearrangements of hundreds of millions of years of editing.

Approximately 300 million years ago, our therian ancestors possessed a pair of ordinary autosomes, or non-sex chromosomes, that were recruited into a chromosomal mechanism of sex determination. How do we know this? We can use synteny analysis, or the comparison of regions of similarity in chromosomes from different species, to reconstruct the organization of ancient genomes. The structure of the X chromosome is more highly conserved than that of other chromosomes, for reasons I'll get to in a moment, and so we can reconstruct in a fair amount of detail the rearrangements that have occurred within it over evolutionary history.

X and Y were initially homomorphic chromosomes, or identical in appearance, with the same arrangement of genes. So what happened to Y? Why has it become such a pathetic, empty shadow of X? In the process of becoming a sex determinant, one chromosome acquired a specific gene, SRY, that is responsible for initiating the embryological process that leads to maleness. As the designated male chromosome, you don't want the SRY factor to be jumping ship to the X chromosome, and you don't want crossover events between SRY and non-homologous regions of the X chromosome, so recombination is suppressed in the SRY region. The evolution of recombination suppressors is not at all difficult—a simple inversion on one chromosome can do it—so this is not a difficult thing for evolutionary processes to do.

The initial recombination suppression encompassed just SRY and perhaps a few adjacent genes. Additional rearrangements occurred, however, which widened the area of suppression. Analysis of the degree of divergence of regions of the Y chromosome from homologous regions of X reveals that these rearrangements occurred in 5 steps, generating 5 regions of the Y that represent evolutionary strata, with the oldest chunk exhibiting the greatest degree of divergence from the X, and the youngest exhibiting the least, because it was still capable of recombining with the X chromosome a mere 30 million years ago.

There has been a progressive expansion of this suppression of recombination over our evolutionary history, but it is not complete. The Y chromosome still retains a few regions, the pseudoautosomal regions (PAR), that retain homology with comparable regions on the X. These are the areas where X and Y in males hold hands when they pair up in that complicated cellular square dance called meiosis. We can see history even here. The PAR at the tip of the short arm is relatively ancient, and has been maintained by obligatory recombination events, but the one at the end of the long arm is new to us humans. Chimpanzees lack it, so it was acquired sometime after our divergence from their lineage—it was created by a translocation of a short piece of the X chromosome to the Y.

So the Y chromosome was progressively isolated, with a few exceptions, from recombining with the X chromosome. Why would that lead to its degeneration?

Recombination is a kind of repair mechanism; not a clever one, but one that can have long term effects. Imagine a very stupid mechanic who maintains two cars by constantly swapping pieces from one to another. If one has a broken carburetor, at least one car will function. If one has a broken carburetor and the other has a broken alternator, every once in a while in his manic swapping, one of the cars will end up with both functioning components, and the other both broken ones: that means there will be at least one car he can send off the lot. Note that this is not a viable business model for a human garage, but it works well enough in a ruthless sort of way for biology. While it produces junk cars as well as good ones, most of the junkers blow up as they're leaving the lot, so no one survives to complain, and all anyone sees actually traveling down the road are the lucky products of random exchanges, and it's only those lucky drivers who go on to make more cars.

The Y chromosome allows no recombination, and is always lonely and alone in the cell. There is no source of spare parts. As damage occurs—duplication errors, cosmic rays, bad splices—the Y becomes increasingly raddled and decrepit, and the individuals bearing it become ever more reliant on the functionality of the genes present on the X chromosome. The gradual erosion of the Y shows us what the fate of our other chromosomes would be, if there weren't recombination and selection working to keep them up. There but for the grace of natural selection go I.

But wait! What's all this chatter about the Y chromosome in an article about X? X and Y are sibling chromosomes, and one of the ways we learn about the evolutionary history of X is by comparing it to X chromosomes in other species, and to its homolog in the male. They have to be considered hand in hand. The paper also discusses an avian autosome, chicken chromosome 1, exhibits similarity to the human X, and is partially homologous. All of the bits and pieces of the various genomes interlock and inform one another in interesting ways. A good science paper on X is going to have to explore Y and the autosomes, so don't assume that the X in the title means the authors are in a figurative straightjacket.

Another misconception that I often see is the idea that the X chromosome is the 'female' chromosome, while Y is 'male'. Avoid that bias. We males also have an X chromosome, and are as dependent on it as are females, and maybe more. The Y chromosome bears almost none of the genes that will be selectively activated in males as they develop. Most of those genes are scattered throughout the genome and are present in both sexes (yes, ladies, you have most of the ingredients to build testicles, just as we gentlemen also contain the complete secret recipe for ovaries in our cells). In fact, the X chromosome is a favored place to stash male-specific genes.

One point made in this paper is that the X chromosome is relatively enriched in genes of a particular class, the cancer-testis (CT) antigen group. These are genes that have anomalous alleles that are expressed certain cancers (about the name: these are not genes that cause cancer. They are genes that, when damaged, can't protect us against cancer), and are also expressed solely or predominantly in the healthy testis. The investigators found 32 CT genes with a particular motif on the X chromosome, while only 4 others are present on the autosomes. They estimate that 10% of the genes on the X chromosome of the CT antigen type, and are therefore expressed almost exclusively in the testis!

I just want to take a moment to say thanks, Mom, for giving me the precious gift of these testicular proteins. I couldn't have made them without you.

Why should male-specific proteins be preferentially archived on the X chromosome? There are sound mathematical reasons for that, derived from principles of population genetics. Males only have one X chromosome, so any allele present on it is going to be expressed in the male. A gene that confers an advantage to males will be most effective and more likely to reach fixation (or ubiquitous presence) more rapidly if it is on the X chromosome. A similar allele on an autosome will be expressed relatively rarely in males, only when both parents donate an identical copy, and so will be hidden from selection.

One other detail about the X chromosome that isn't completely resolved but promises interesting research in the future: dosage compensation. Gene dosage is important—changes in the quantities of certain gene products can cause serious aberrations. Down Syndrome, for instance is caused by having an extra chromosome 21. There are no anomalous genes present, nothing wrong in the code, just an overdose of some genes. Lacking any autosome so that the individual has only one copy is uniformly lethal in the embryo. The X chromosome is special—males only have one copy (we should be dead, victims of an underdose!) and females have twice as many copies as we males do (they should be dead of an overdose, or have some syndrome named after them!). The way levels of expression are equalized in the two sexes is that the output of genes on the X is doubled, to give us males a chance, and one X chromosome is inactivated in females, so that they don't OD on X genes. All human beings are adjusted to having just one fully functional X chromosome, no matter whether you are male or female.

How this inactivation is accomplished is only sketched out at this point. Both X chromosomes have a gene called XIST that produces RNA that coats the DNA of the chromosome, sheathing it in a layer that inactivates most of the genes on it. There is a competition of an undetermined nature in early development: both X chromosomes express XIST, but eventually one wins out and shuts off the other. The chromosome with the active XIST gene then produces more XIST transcript, which again in some undetermined manner, affects only the chromosome producing it. XIST creeps over the chromosome from an inactivation center called XIC, eventually engulfing the whole thing.

These XIST coat expansions center on a particular region. Various experiments suggest that one region is not enough—there must be other loci scattered throughout the chromosome that can also initiate the creeping XIST. One fascinating observation is that the X chromosome is particularly rich in a form of parasitic, junk DNA called L1, and that the L1 sequences are particularly prominent in regions that are inactivated, and more scarce in regions are active. The observation is only a correlation at this point, and there are conflicting data in the literature, but it does raise the interesting possibility that XIST has coopted the L1 sequence as a signal for inactivation.

Way back at the beginning of this article, I mentioned that the organization of the X chromosome was highly conserved, and the reason involves dosage compensation. The levels of expression of the genes on the X chromosome are calibrated to maintain parity between males and females. There is no comparable process for the genes on the autosomes, because both genders have two copies of each. What this means is that dosage compensation is a barrier to translocation: genes cannot be readily shuffled from an autosome to the X chromosome, or vice versa, because they will then experience a different mode of regulation. Unlike the autosomes, the X chromosome is therefore resistant to these kinds of juggling of fragments, and so tends to preserve its original complement of genes.

This kind of research is impossible to do and impossible to interpret except in terms of evolution. There's much more to this work than dumb technical hacking, grinding samples through a sequencer, and transcribing a series of letters. Throughout, there is extensive comparative analysis, cross-checking the data with information from other species and trying to infer patterns of change, and evaluation of the observations with respect to predictions from theory. And the result is a kind of molecular/genetic archaeology, an elegant reconstruction of our history.

I counted 282 authors on this paper. Impressive, huh?

Ross MT, Grafham DV, Coffey AJ, Scherer S, McLay K, Muzny D, Platzer M, Howell GR, Burrows C, Bird CP, Frankish A, Lovell FL, Howe KL, Ashurst JL, Fulton RS, Sudbrak R, Wen G, Jones MC, Hurles ME, Andrews TD, Scott CE, Searle S, Ramser J, Whittaker A, Deadman R, Carter NP, Hunt SE, Chen R, Cree A, Gunaratne P, Havlak P, Hodgson A, Metzker ML, Richards S, Scott G, Steffen D, Sodergren E, Wheeler DA, Worley KC, Ainscough R, Ambrose KD, Ansari-Lari MA, Aradhya S, Ashwell RI, Babbage AK, Bagguley CL, Ballabio A, Banerjee R, Barker GE, Barlow KF, Barrett IP, Bates KN, Beare DM, Beasley H, Beasley O, Beck A, Bethel G, Blechschmidt K, Brady N, Bray-Allen S, Bridgeman AM, Brown AJ, Brown MJ, Bonnin D, Bruford EA, Buhay C, Burch P, Burford D, Burgess J, Burrill W, Burton J, Bye JM, Carder C, Carrel L, Chako J, Chapman JC, Chavez D, Chen E, Chen G, Chen Y, Chen Z, Chinault C, Ciccodicola A, Clark SY, Clarke G, Clee CM, Clegg S, Clerc-Blankenburg K, Clifford K, Cobley V, Cole CG, Conquer JS, Corby N, Connor RE, David R, Davies J, Davis C, Davis J, Delgado O, Deshazo D, Dhami P, Ding Y, Dinh H, Dodsworth S, Draper H, Dugan-Rocha S, Dunham A, Dunn M, Durbin KJ, Dutta I, Eades T, Ellwood M, Emery-Cohen A, Errington H, Evans KL, Faulkner L, Francis F, Frankland J, Fraser AE, Galgoczy P, Gilbert J, Gill R, Glockner G, Gregory SG, Gribble S, Griffiths C, Grocock R, Gu Y, Gwilliam R, Hamilton C, Hart EA, Hawes A, Heath PD, Heitmann K, Hennig S, Hernandez J, Hinzmann B, Ho S, Hoffs M, Howden PJ, Huckle EJ, Hume J, Hunt PJ, Hunt AR, Isherwood J, Jacob L, Johnson D, Jones S, de Jong PJ, Joseph SS, Keenan S, Kelly S, Kershaw JK, Khan Z, Kioschis P, Klages S, Knights AJ, Kosiura A, Kovar-Smith C, Laird GK, Langford C, Lawlor S, Leversha M, Lewis L, Liu W, Lloyd C, Lloyd DM, Loulseged H, Loveland JE, Lovell JD, Lozado R, Lu J, Lyne R, Ma J, Maheshwari M, Matthews LH, McDowall J, McLaren S, McMurray A, Meidl P, Meitinger T, Milne S, Miner G, Mistry SL, Morgan M, Morris S, Muller I, Mullikin JC, Nguyen N, Nordsiek G, Nyakatura G, O'dell CN, Okwuonu G, Palmer S, Pandian R, Parker D, Parrish J, Pasternak S, Patel D, Pearce AV, Pearson DM, Pelan SE, Perez L, Porter KM, Ramsey Y, Reichwald K, Rhodes S, Ridler KA, Schlessinger D, Schueler MG, Sehra HK, Shaw-Smith C, Shen H, Sheridan EM, Shownkeen R, Skuce CD, Smith ML, Sotheran EC, Steingruber HE, Steward CA, Storey R, Swann RM, Swarbreck D, Tabor PE, Taudien S, Taylor T, Teague B, Thomas K, Thorpe A, Timms K, Tracey A, Trevanion S, Tromans AC, d'Urso M, Verduzco D, Villasana D, Waldron L, Wall M, Wang Q, Warren J, Warry GL, Wei X, West A, Whitehead SL, Whiteley MN, Wilkinson JE, Willey DL, Williams G, Williams L, Williamson A, Williamson H, Wilming L, Woodmansey RL, Wray PW, Yen J, Zhang J, Zhou J, Zoghbi H, Zorilla S, Buck D, Reinhardt R, Poustka A, Rosenthal A, Lehrach H, Meindl A, Minx PJ, Hillier LW, Willard HF, Wilson RK, Waterston RH, Rice CM, Vaudin M, Coulson A, Nelson DL, Weinstock G, Sulston JE, Durbin R, Hubbard T, Gibbs RA, Beck S, Rogers J, Bentley DR. (2005) The DNA sequence of the human X chromosome. Nature 434:325-337.