Evidence shows that mice are attracted to their mates based on genetic diversity. This they can somehow tell from the smell of their urine. There is also currently some weak evidence that humans indirectly do the same thing.
(accompanying graphic)

Ancestral and recombinant 16-locus HLA haplotypes in the Hutterites. (1999) Immunogenetics 49:p491

In which human mate selection appears to be determined by genetics afterall. This study was done using a small community (Hutterites) since carefully controlled human mate matching and observation would be unethical – at least without reality TV.

Several years back, experiments were performed whereby a known male mouse of specific genotype (i.e. you have a good sense of what its genetics is all about), was presented with a bevy of different females, again of known genotypes. Under these controlled circumstances, one found that there was a noticeable pattern of mate selection occuring – in other words, there always seemed to be a particular preference between certain strains. Furthermore, if you do the reverse experiment with one female and many males (BTW, a much more difficult experiment as a group of male mice seem to want to only kill each other), you get the same general trend, although subject to the female mouse’s stage of estrus.

Anyhow, if one looks at this deeper, you’ll note that these choices correlated very strongly with genes involved in the mouse’s immune system – specifically sequences coding the major histocompatibility complex genes. This is not altogether surprising, especially if you have a good handle on what your immune system is charged with doing.

Because it just so happens that your immune system is doing a lot. From a relatively conservative estimate, your immune system is capable of recognizing about a billion different things. Good thing too, since this elaborate system is in charge of protecting an animal (such as yourself) against pathogens of all shapes and sizes, so in effect this large repertoire is nature’s way of “covering your ass.” Perhaps even more elegant is the fact that not only does the immune system recognize such a large number of things, it’s also capable of gauging whether that thing is good, bad, or more importantly, worthy of an immune response.

This is all nothing short of amazing. For the sake of analogy, these numbers are akin to you knowing everybody in North, Central and South America (that’s about a billion individuals, people!). And on top of that, you would know them well enough to make the call with regards to liking them or not.

Anyway, the point being that repertoire, or the sheer volume of recognition, counts towards a robust immune defense.

What does this have to do with the genetic basis of love? Well, it turns out that this notion is nicely supported by the mate selection data, since mice preferred to mate with mice who had distinct and different immune system genes. Another way of looking at this, is to suppose that you were wanting to make a baby and also wanted to be purely practical about it. If so, from a biological fitness point of view, it should make perfect sense to want a partner who had different and distinct immune system genes – this way, the baby gets the best of both worlds with a more diversified repertoire.

How the mice are in tune with these genetic differences (without little miniature genetic test kits) is possibly derived from phermone signaling (i.e. smelling pee and stuff), which happens to be nicely supported by the fact that there are obvious siimilarities between the structures of phermone receptors and MHC molecules (that major histocompatibility thing I referred to earlier).

The big question, of course, is whether humans do the same thing. And short of performing similar, but highly unethical experiments with humans (although with shows like the Bachelor and Bachelorette perhaps there now exists a way to skirt this), researchers have instead looked at this phenomenon by studying small communities of similar genetics (i.e. a relatively closed population like in Hutterites communities), and looking at the MHC patterns of stable vs non-stable couples. The paper above essentially does this with results showing that a similar mate selection process is occuring (although perhaps without the pee smelling). Consequently, love does have a genetic factor – sort of takes the romance out of love, doesn’t it?