« POPULATION FALLACIES: PART 2 | Gene Expression Front Page | Don't be too smart »
June 27, 2003

Implications of chromosomal sex determination

Razib's recent entry on the erosion of the Y chromosome got me thinking: what advantages does this method of sex determination have that allow it to persist despite the ongoing problem of Y chromosome deterioration? A little research suggests that it may help us mammals make larger evolutionary jumps than would be possible for other classes of vertebrates, which use other methods of determining sex (birds also use chromosomes for sex determination, but more on them later). To begin with, see here for a discussion of a recent Y-chromosome mutation that may have led to the creation of modern humans as a separate species. The paper also goes into some detail on how mutations of the Y chromosome provide a mechanism for evolutionary saltation that overcomes some of the objections raised by the gradualists.
This has interesting consequences. It implies that mammalian and bird evolution could follow the model of punctuated equilibrium while other creatures might be (mostly) stuck with gradual evolution.
Birds, however, have homogametic males (where humans and mammals in general have heterogametic males). As a result the risks associated with mutations to the non-recombining sex chromosome fall first on the females of the species, and consequently the costs are higher. Birds would seem therefore less able to make large evolutionary leaps, and I would guess that the amount of genetic variation throughout the class of birds (10,000 species or so) is less than that found in mammals (4,000 species). Examination of the phenotypes seems to support this idea.
Another question this raises is whether creatures must be able to manifest some minimal rate of species change or else risk extinction. There are a number of features that a species can have that allow for rapid mutation and selection:
- a large number of descendants (which also implies rapid elimination of bad variations)
- a short time interval between successive generations
- imperfect DNA replication mechanisms to increase various sorts of error rates
- the Y chromosome mechanism described above

Since mammals and birds don't do well in the first two categories, and can't afford the third feature given their investment in their young, I would guess that the development of their method of sex determination was a necessary precursor to the high-K reproductive strategies they use. Without it, the speciation or evolution rate for a high-K species would fall too low, and we would not have seen the great bursts of adaptive radiation that the mammals have shown.

Posted by bbartlog at 11:52 AM

Sorry to be a boring old gradualist again, but I don't understand the argument. Why do you think evolution would be slower if females are the heterogametic sex? Maybe females would then have more variable fitness, but so what? This would only be a problem if the average loss of fitness were so great that the population went into irreversible decline. There is no sign of this in birds. Judged by the number of species, they seem to evolve at least as fast as mammals. (NB the adaptive radiations of birds in the Galapagos and Hawaii.) Admittedly, birds seem rather uniform in morphology, compared to mammals, but I suspect this is largely due to their specialisation for flight. If you look at bird behaviour, rather than morphology, it is very varied - e.g. all the different kinds of nest building and courtship behaviour.

I read some of the paper you linked to, and I noticed that the author is a psychiatrist. I think he may have been spending too much time with his patients.

Posted by: David B at June 28, 2003 11:37 AM

The variety of birds suggests to me a class of animals that speciates more easily but with smaller changes overall. I will have to write something else to explain why I think the rate of speciation is not necessarily the same as the overall rate of genetic change. Of course flight also allows for easier colonization of islands, which contributes to the greater number of bird species. Anyway, I am by no means sure of my claim - I will be interested to see the data when and if we start wholesale sequencing of other species.

I will use HF and HM for heterogametic females and males respectively:
As for HF leading to less dramatic changes: more variable fitness is not desirable for females because they can't derive much extra benefit at the high end of the curve, and suffer just as much at the low end (as males). A male with very high fitness might be twenty times as successful as an average one, whereas for females of any high-K species that's virtually impossible. In animals with HM it is possible to make what I would call evolutionary gambits that are initially expressed almost exclusively on the male side (in humans colorblindness and hemophilia are good examples). These variations, if they are defects, would reduce the overall reproductive fitness of a HF population immediately (by reducing the number of fit females), while a HM population would scarcely be affected (reducing the number of fit males is much less of a problem). If they were advantageous they could still spread rapidly in either population. So generally, the cost to a HM population of keeping a bunch of oddball sex chromosome mutations of questionable immediate utility around is less. This would seem to provide a more fertile ground for large genetic changes.

As for the author of the paper being a psychiatrist - I don't know enough hardcore genetics to evaluate his claims critically, but they sounded reasonable to me. His credentials are irrelevant to me (INTP that I am ;-) )

Posted by: bbartlog at June 28, 2003 12:55 PM

Thanks for explaining the HF/HM point. It's an interesting idea - basically, that if you are going to experiment with new genes, it's best to concentrate them on the sex with the greatest variance in reproductive success. This is a bit reminiscent of G. C. Williams's 'lottery ticket' theory of the advantage of sexual reproduction.

It could help explain why males tend to be the heterogametic sex, though this is by no means a universal rule. Birds, snakes, butterflies and moths are HF. But mammals and most diptera (flies) are HM, and according to M. J. D. White (Cytology and Evolution) the majority of organisms that have a heterogametic system of sex determination are HM.

However, I'm not convinced it would be a major factor in affecting the rate of evolution. For one thing, it is only relevant where genes are recessive, and these are usually harmful. Also, it only applies to one chromosome out of many. (E.g. in man, one out of 46 in the genome, assuming that the Y chromosome is mainly junk.) If there were really a big evolutionary advantage in exposing recessive genes to selection, then most organisms would be haploid, or at least haplodiploid (as in bees and ants), which is obviously not the case.

Posted by: David B at June 29, 2003 03:09 AM

Yes. I agree that because the area of effect is limited to the sex chromosomes, the theory is rather tenuous. Thanks for the information on other HM and FM creatures - I knew there were some others, but concentrated birds and mammals because of their other similarities.

Posted by: bbartlog at June 29, 2003 06:28 AM

There is a book on the evolution of sex determination systems by J. J. Bull, (about 1983), but I haven't read it.

It occurs to me that another way of concentrating 'experimental' changes in males would be to make them 'sex limited', i.e. expressed in males but latent in females, like the antlers of deer. But I haven't thought through whether there would be any way for this to work in a 'selfish gene' context.

Posted by: David B at June 30, 2003 02:46 AM