How important is gene flow?

There’s a handy article available from the American Journal of Botany that’s caught my eye: Is gene flow the most important evolutionary force in plants? by Norman C. Ellestrand. It opens with a strong statement.

Some scientists consider the word “evolution” to be more or less equivalent with “natural selection” or adaptation. They would, of course, be wrong.

Ellestrand states that biological evolution is the change in allele frequencies in a population over time, and that this is due to four evolutionary forces: mutation, selection, drift, and gene flow. Gene flow is important because even low levels of gene flow can have a large impact, counteracting the other evolutionary forces.

So what is gene flow?

It’s the movement of alleles from one population to another. For example, Schulze et al. published a paper Searching for gene flow from cultivated to wild strawberries in Central Europe. They were looking at cultivated strawberries, which are octoploid and bred to be tasty and looking to see if genes humans had selected were moving into wild populations of diploid strawberries. It’s an important question because if genes are flowing then farms could be contaminating local wildlife. They found it wasn’t happening, even though hybrid berries were possible.

But not all plants are the same. Take carrots for example. Rong et al. looked at gene flow in wild carrot populations (subscription-only access till Nov 2014). They found that gene flow could happen at the scale of a kilometre or more, and that mowing roadside verges helped spread genes. It also happens among alpine plants.

It’s led to a change of opinion about gene flow over the past thirty years. In a press release for the paper, Ellestrand said: “When I first started doing plant paternity studies in the 1980s, our lab assumed that gene flow was limited. But we kept identifying ‘impossible fathers’ that could not be assigned to our study population. Surely, these couldn’t be fathers from outside of our wild radish populations—hundreds of meters away? But after excluding all other possibilities, the improbable turned out to be the answer. And the paradigm of limited gene flow in plants began to crumble.”

Ellestrand now describes gene flow as “idiosyncratic, but often significant.” Self-pollinating plants will not have the gene flow that outcrossing plants do. Wind-pollinated plants can have more gene flow than insect pollinated plants. It isn’t just interesting for its out sake, it has important consequences for biology.

One is that gene flow can act as an ‘evolutionary glue’, as Ellestrand calls it. By swapping alleles around the population between each other, there is an evolutionary unit that it makes sense to call a species. Without gene flow you simply have a group of things that look similar at the moment because they’re under similar selection pressures.

He also argues that populations can evolve gene flow to become units, giving the example of adaption to toxic metals among some plants in Britain as an example of how gene flow can isolate and form new populations.

There are also current concerns. Ellestrand raises the issue of gene flow of material from GMO crops to wild relatives. This is similar to the work done by Chen et al. on transgenic rice. You don’t even need a parent plant to cause this kind of trouble. Ellestrand points to his own research on weeds that were descendants of inter- or intrataxon hybrids.

His conclusion could seem at odds with the rest of the paper starting as it does:

Is gene flow the most important evolutionary force in plants?
That’s a silly question!

He argues that it’s a silly question, because you need many forces and looking at gene flow in isolation ignores the constraints that gene flow flows through.

If it is a silly question, then it’s a silly question that’s worth asking. In his press release he says: “This review paper tells the story of gene flow’s rise to respect among plant evolutionary biologists, a fact that hasn’t yet penetrated biology in general that is still mired in selection/adaptation-only thinking.” To return to the top of the post, I would have been happy conflating evolution with natural selection. I think Ellestrand has made a very good argument that I would be wrong doing that.