A - C - G - T: I Can Make A Honey Bee!

Science warning:  this post contains actual science.  You have been warned...

The Lego bricks of life!
(Wikimedia Commons, Public Domain)

A few weeks ago, I wrote a blog post about drones.  And, despite the cheeky science warning (repeated here!), regular reader Sophie asked this:

"If a male drone is from an unfertilised egg, I assume he is always haploid. However, he must be XY to be male, so where does the extra sex chromosome come from? The egg would presumably always be X, as the queen must be XX. It makes sense that the male would just be Xo because then all the sperm would be X, which accounts for all fertilised eggs developing into females. Can the male spontaneously develop a Y chromosome somehow? Or does Xo present as male in bees?"

It's a great question.  My initial answers were:  Firstly - yes, drones are haploid. Secondly, the sex determination mechanism in bees (and other hymenoptera) isn't done by a dedicated sex chromosome in the way that it is with mammals - so there is no Y chromosome. 
Thirdly, yes - Xo does indeed present as male in bees. But, it isn't quite as straightforward as saying that "XX is female and Xo is male", even though that is what appears to be going on.

So, if it isn't straightforward, then what is actually going on?  Time to delve right down into the cellular nucleus, and explore the strange world of honey bee genetics...

Firstly, a reminder about the basics in humans.  All our genes are encoded in sequences of DNA.  And the DNA (which is basically a very long molecule) is bundled up in separate packages called chromosomes.  When we started off as a single fertilised egg (which is just one cell), that egg contained 46 chromosomes (occasionally, someone will have 47, which leads to conditions such as Down syndrome and Klinefelter syndrome).  As we grow, cells get copied over and over again, and each new cell gets its own copy of the 46 chromosomes.

23 of the chromosomes came from our mother (from the unfertilised egg) and the other 23 from our father (from the sperm).  And they're not just random - each maternal chromosome has an equivalent paternal chromosome, so they form pairs.  The chromosomes in each pair are similar but not the same - that is to say, if you take a maternal/paternal pair, they will control the same set of genes, but may have different versions of each gene.  So, if we look at chromosome pair 15, we find the gene for eye colour.  The maternal version might be the gene for brown eyes, whereas the paternal version might be the gene for blue eyes.

These two different versions of the same gene are called alleles.  And we'll come back to them later on...

One pair of chromosomes in humans is different - pair 23 is the sex chromosomes.  The one inherited from the mother looks just the same as the other chromosomes, and is referred to as the "X" chromosome.  The one inherited from the father comes in two versions.  Version 1 is also an "X" chromosome (and the father inherited it from his mother).  Version 2 is much smaller, a stubby little chromosome that contains far fewer genes.  This is the "Y" chromosome.  And the simple rule is that if you get two X chromosomes (i.e. you are XX) then you are female, whereas if you get an X and a Y (you are XY) then you are male.  interestingly, pair 23 is the one pair of chromosomes where a gene on one (the X) may not have a match on the other (if it's Y).  For example, the gene that determines colour-blindness is carried only on the X chromosome.

But we digress - this is supposed to be about bees!

There are two big differences in honey bees.  Firstly, there are only 16 pairs of chromosomes.  And secondly, only female bees get their chromosomes in pairs (i.e. a set from both parents for a total of 32).  Male bees only have 16 unpaired chromosomes - all from their mother.  Bee eggs are completely capable of developing into adult bees even if they are unfertilised - and they will always be male in this case.  This process of being able to create children without fertilisation is known as parthenogenesis.  And (male) bees produced through parthenogenesis have half as many chromosomes, so they are referred to as being "haploid".  Female bees, with the full complement of 32 chromosomes, are "diploid".

Now, one of the consequences of this is that there is no Y chromosome.  It's still the case that female bees are XX, but male bees only have one (of each) X chromosome - this is denoted as "Xo" (the "o" denoting "missing" or "absent").

So, one X makes a bee male, two XX makes her female, right?  Well, this is what was thought until 2003, when researchers from the University of California found out it doesn't quite work like that.  They found that sex determination isn't actually done by chromosome count - there's a specific gene that controls it.  The gene is on the 3rd chromosome, and is called the "csd" or "complementary sex-determination" gene.  Now, you might assume that this gene comes in two versions - male and female.  But that would mean that a haploid (Xo) bee could have the "female" version in its one copy of chromosome 3 - in other words, a bee from an unfertilised egg could become female.  But that never happens.  So, what's actually going on?

It comes back to the alleles that we mentioned earlier.  Remember the genes for eye colour in humans?  These come in a number of different versions (blue, brown, green and others).  Each different version of the gene is known as a different allele.  And genes with multiple alleles are referred to as polymorphic - meaning that they come in different forms (poly=multiple, morphic=forms).

Now, the clue to how this works is in the name of the gene - "complementary sex-determination" gene.  Why complementary?  Simply, that different alleles of csd complement each other - in other words, they work together.  When a bee has two different - complementary - versions of csd, she becomes female.  If a bee has only one version of csd, then he will be male.

So, a quick recap - if a bee is Xo, it can only have one version of csd, so he must be male, and that's exactly what we observe.  And if a bee is XX, and the two versions of csd are different (this is known as being heterozygous), then we get a female.  There are around 20-25 different alleles of csd, so most of the time this is what happens.

But - hang on - it must be possible, even though unlikely, that sometimes an XX (diploid) bee, will have two identical copies of the csd gene?  Indeed it is - in this case the gene is homozygous.  But what happens then?  Well, csd has a secondary effect - when two identical alleles are present, it changes the relative amounts of certain compounds in the skin of the larva when it hatches.  This makes the larva smell different from the others, for around the first three days of its life.  In fact, from a worker bee's perspective, it makes the larva smell delicious.  So she eats it.

This means that adult bees with two identical csd alleles are never present in the hive.  But if a csd-homozygous larva continues to be fed (and isn't eaten!) then it's possible for it to grow into an adult bee.  But what does it become?  Some scientists wanted to find out, so they hatched some csd-homozygous larvae in a laboratory, and fed them.  After three days (when they stopped smelling like a tasty treat) they were returned to the hive where the workers continued to feed them.  And then their cells were capped with wax (by the workers, as normal), the larvae underwent metamorphosis and later emerged as adult bees...  which turned out to be...

Male!  In other words, they are diploid (XX like female workers) but are actually drones.  Having one version (but two copies) of the csd gene has the same effect as having only one copy - i.e. the bee is a male drone.  But there are subtle differences.  They are larger, heavier and have smaller testes.  That last point is important - it means that the drones are, effectively, sterile.  It is theoretically possible to inseminate a queen with diploid (XX) drone sperm to produced triploid (XXX - wowzers!) bees.  But in practice it hasn't been achieved due the the extremely low sperm count of the diploid drones.

So there you have it - a haploid (Xo) bee will be a normal drone.  A diploid (XX), csd-heterozygous bee will be a worker.  And a diploid (XX), csd-homozygous bee will theoretically grow up to be a fat, sterile drone (but will actually be eaten by his sisters just after he hatches).

I'll leave you with one final fun fact.  We've said that there are around 20-25 different alleles of the csd gene.  Interestingly, a queen bee will mate with roughly 20-25 males before she starts laying.  Co-incidence?  I suspect not - it seems that bee mating behaviour has evolved to try to maximise the number of different csd genes, and therefore minimise or eliminate the production of diploid drones in the hive.  Isn't that rather clever?