Thread by @pjie2

Here's the latest preprint from my lab. It's taken us a good few years to develop the methods to do this one properly, but I'm really pleased with the story we were able to assemble.. So, with your indulgence, here's the Twitter version of the paper. (1/n)

https://twitter.com/sexgenomics/status/1132138229384093698

The story (as with all research) goes back a long way, to some curious observations in the early 90s. At this time, a couple of different groups independently identified mutant strains of mice with a biased sex ratio.
ncbi.nlm.nih.gov/pubmed/2055460
ncbi.nlm.nih.gov/pubmed/7516747
(2/n)

Biased sex ratios are profoundly weird from a genetic point of view - there are powerful natural selective forces that normal act to enforce a 50:50 ratio of male and female offspring.
(3/n)

Not least of these is simple Mendelian genetics. The father has XY chromosomes, and the X and Y chromosomes should segregate evenly, to produce equal numbers of X- and Y-bearing sperm. So this situation seems to contradict the First Law of Mendelian Genetics.
(4/n)

Before anyone else says it: You Do Not Talk About Mendelian Genetics
(4.5/n)

Another funky thing about these mice is that their sperm are somewhat deformed. Mouse sperm have a much more interesting shape than human sperm: asymmetrical, with a smoothly curved "hook" at the front.
researchgate.net/profile/France…
(5/n)

In contrast, these males had flattened sperm, with short or broken hooks, almost like they'd run headlong into a brick wall (compare top left and top right panels here).
encrypted-tbn0.gstatic.com/images?q=tbn:A…
(6/n)

The last thing to notice at this point was that the males always passed on the mutation to their sons - i.e. it was tracking with the Y chromosome. It turns out that they have a deletion of part of the Y, which in turn leads to the abnormal sperm shape and the sex skew.
(7/n)

This is where I came in to the story, in the mid 2000s. We measured the gene expression in the testes of the mutant males and showed that when the Y chromosome gets deleted, hundreds of genes on the X chromosome start to be expressed more strongly.
ncbi.nlm.nih.gov/pubmed/16087683
(8/n)

That is, the X and Y chromosomes are at war. There are genes on the X that promote the production of female offspring (i.e. transmission of the X chromosome), and genes on the Y that switch these off and restore a normal sex ratio.
(9/n)

In the next few years, Julie Cocquet (NIMR, Burgoyne lab) showed in a lovely series of papers that there are duelling gene families Slx and Sly that act as a "thermostat" controlling sex ratio in mice.
ncbi.nlm.nih.gov/pubmed/19918361
ncbi.nlm.nih.gov/pubmed/20739462
ncbi.nlm.nih.gov/pubmed/23028340
(10/n)

Knock down Sly, you get wonky sperm and excess female offspring.
Knock down Slx, you get wonky sperm and excess male offspring.
Knock down both, you restore normal sperm and a normal sex ratio.
These genes seem to be purely "selfish", acting only in their own interests!
(11/n)

This warfare between these genes has had significant consequences - it sets up an "arms race" between the X and Y, that drives massive increases in the copy number of each of these genes, like the US and USSR stockpiling warheads.
(12/n)

Other work from a whole host of other groups has shown how this conflict has played out throughout evolution - with consequences for genome structure, mouse population genetics and ecology.
(13/n)

Which brings us to the current study. Despite over two decades of work in these animals, we couldn't say *how* the sex ratio skew actually happens. We know it's caused by genes on the sex chromosomes, but what do they actually do to the sperm?
(14/n)

The first thing we did was to look at the fertilisation process. When we allow the mice to breed normally, we see the same skew whether we look at live-born pups, in mid-gestation, or in very early embryos. That shows the skew has nothing to do with embryo survival.
(15/n)

In these same experiments, we also found that the mother's genotype doesn't matter, and that sperm age doesn't matter.
(16/n)

However, if we perform IVF, and put the sperm directly on top of the eggs to let them fertilise, the skew entirely vanishes. This tells us two things. First, X and Y sperm must be produced in equal numbers. Second, the X sperm must be better at getting to the egg.
(17/n)

That second bit is pretty wild - people have been trying to find functional differences between X and Y sperm for decades, in all sorts of different species. And, despite many claims, they have always failed.
(18/n)

In the usual run of things, there is NO difference between X and Y sperm other than DNA content - the Y is smaller than the X. They look the same, swim the same, have all the same surface molecules - there's no way to tell them apart.
(19/n)

However in this particular set of mutant mice with the Y chromosome deletion, there *is* a difference. So we then set out to explore what that difference might actually be.
(20/n)

We knew from the early work that the Y-deleted males had abnormally distorted sperm - so we asked whether the Y sperm were more severely distorted than the X sperm.
(21/n)

Just developing the tools to answer that question took 3 years, since we had to invent new ways of measuring and comparing 2D shapes, and then work out how to separately analyse X and Y sperm. Methodology papers are here:
ncbi.nlm.nih.gov/pubmed/30753283
ncbi.nlm.nih.gov/pubmed/30717218
(22/n)

It's fiddly - first you image the sperm at high magnification. Then you take the slide away and stain it using FISH hybridisation to label the X and Y cells - doing this destroys the sperm morphology. The, re-image the same cells and match up the before/after pictures.
(23/n)

However, we were able to show that on two different genetic backgrounds, either deletion of the Y chromosome or knockdown of the Sly gene causes sperm to be distorted, and in all cases the Y sperm are more severely affected than the X sperm.
(24/n)

So, the sex ratio distorter genes are acting "spitefully" - they actually lower fertility by distorting sperm shape, but because they distort the Y bearing sperm more severely, the X sperm have a relative advantage in fertilisation.
(25/n)

You don't have to run faster than the bear, just faster than your opponent.
(26/n)

However, one piece of the jigsaw was still missing - we know the Y sperm are more distorted, but we also know that if you put the sperm directly on top of the egg, they can fertilise just fine. We had to check sperm motility (swimming ability) as well.
(27/n)

This brought us to another speed bump, since there are many very sophisticated ways to look at sperm movement, none of which can tell you the difference between X and Y sperm.
(28/n)

Eventually, we settled on the simplest possible experiment: a sperm race. We deposited sperm at the bottom of a tube, let them swim up, and collected the sperm in successive layers from top to bottom of the tube to enrich for fast swimmers vs slow swimmers.
(29/n)

Then, stain the cells to work out which are X vs Y (this kills them, otherwise we'd stain them before the race), and count them. And we did indeed find that in our Y-deleted males, but not in control males, the X sperm are faster than the Y sperm.
(30/n)

However, to prove this beyond all doubt we had to count a LOT of cells - many replicates, six different fractions of fast/slow cells, normal vs mutant animals, other controls.... over 90,000 cells in total.
(31/n)

So, many congratulations to @Mooseenthusiast who did most of the lab work, to @bmattsk who wrote the image analysis software, to Julie Cocquet and colleagues who provided sperm samples from Sly knockdown males, and to everyone else involved.
(32/n)

We can now say we have a very nearly complete story, tracing from the original mutation (Y deletion) to the transcriptional consequences (X gene upregulation), to the physiological consequences (spiteful sperm distortion and motility loss) to the output (skewed sex ratio).
(33/n)

There is of course more work to do - as always - not least of which is to find out if we can transfer this to other species! The world does not need more female mice, but it could do with more dairy cattle and female pigs. However, I'm really proud of how far we've got.
(end/n)

Postscript: I have naturally elided over all the false starts and slaughtered hypotheses we left along the way like so many rusting AT-ATs. Let's just say we spent a good four or five years iterating this cartoon...

@threadreaderapp unroll