Chromosomes aren't the only thing that determines the sex of an organism. In other species, environmental cues can determine sex – in reptiles like alligators and turtles, where the temperature an egg is kept at determines whether the offspring is male or female. One of your projects has involved mapping these different approaches to sex determination and their evolutionary back-story in a 'tree of sex'. What has this database shown about how sex determination has evolved in different species?
What's intriguing is that sex – maleness and femaleness – is conserved across so many different organisms; almost all vertebrates, and most invertebrates, have very distinct male and female phenotypes. A lot of the genes that determine sex, at least in animals, are conserved across huge swathes of evolutionary time, hundreds of millions of years in some cases. But the actual trigger that causes those genes to determine sex – if it's a gene on a Y chromosome or a W chromosome, birds have W and Z chromosomes instead of Y and X, or if it's environmental – seems to change very quickly, in evolutionary terms. This presents a conundrum as to how something that's under such strong selection can actually change so quickly.
With the tree of sex, we organized it with the idea of trying to understand how sex determination can change so quickly by trying to understand the evolutionary and ecological pressures that cause sex determination to vary across species.
Although many male and female organisms differ in whether they have XY or XX chromosomes, broadly speaking they have essentially the same genes. Yet male and female phenotypes look and behave quite differently. What factors drive how genes are expressed differently in males and females to create the distinct male and female phenotypes?
That's something we've been trying to understand. In birds, for example, males and females differ by a few dozen genes on the W chromosome, which is similar to the Y but inherited from mother to daughter. But for 10,000 to 15,000 genes in the remainder of the genome, males and females both have them. We're trying to understand how expression levels of these shared genes can be differentially regulated in males and females.
Part of it is hormonal – sex hormones have a huge influence on how genes are expressed. But actually how that process happens – how a gene comes under different regulatory controls for males and females – is an interesting question. We've been trying to figure out how you break down the correlation in expression between males and females for a single shared gene, and how quickly that can happen. We think that there's a different optimum for some genes in one sex versus the other.
We've been talking about the two sex phenotypes, male and female, but actually sex is not that simple. Some organisms, like turkeys, can have more than one variant of a sex phenotype. Tell me about your research in turkeys and what you have discovered about gene expression in the two variants of 'male' in this species.
That was a really fun project that came out of going to a turkey farm up in Yorkshire. When I was looking at two male turkeys, one was clearly much brighter and had a longer snood and larger caruncles than the other. I asked the woman who ran the farm whether the less vivid male was just a juvenile, and she said that they were actually brothers from the same year, but one was subordinate and one was dominant.
The way it works is that brothers in a clutch of turkeys come together in the winter before they reach sexual maturity and have epic battles. The winner adopts the dominant male phenotype, and the losers adopt the subordinate phenotype. The dominant male is the only one that actually ever mates with females, and the subordinate males help to attract females to their dominant brother.
When we looked at the genetics of these two types of male turkey, the results were surprising. They showed that for male biased genes – that is, those expressed more in males that we think are more important for male phenotypes – subordinate males express them a little bit less than dominant males. So they're demasculinised across a couple of thousand genes, which is a pretty big proportion of the whole genome. The effect is quite subtle for any single gene, but together it creates an aggregate effect.
I honestly didn't believe the results at first. I thought the difference between dominant and subordinate males would be limited to five or ten genes that would show a really big difference. We did the data analysis three ways because I thought we had made a mistake. But no matter how we did it, the results always came out the same.
Dear Prof. Mank, thank you very much for this interesting interview.
Interview: Helen Jaques (© AcademiaNet)