Some new work on speciation and species

ResearchBlogging.org

There is a widespread tendency of biologists to overgeneralise from their study group of organisms to the whole of biology. Sometimes this is because the organisms are model organisms, like Drosophila (the "fruit flies" that have been used in genetics since the beginning).Other times it is because specialists tend to overestimate the generality of their results and domain. The recent trend to finding "speciation genes" is an example. For some time now various researchers like Chung-i Wu and his collaborators have sought speciation genes. These are genes that cause speciation, in a general sense, but the slide appears to be made to the conclusion that there are particular genes in many if not most cases of speciation. I want to consider this now.

Phadnis and Orr have found that a particular gene is both a gene causing sterility between hybrids of two Drosphilid subspecies, and a segregation distorter - that is, the gene causes itself to be differentially copied when gametes are formed. A similar process of meiotic distortion occurs in mice as Mihola et al. show. A great result, but how general? In the commentary accompanying the online advance publication, Asher Mullard writes and quotes this:

"There is no question that in this era of whole-genome sequences and genomic data it is much easier to identify speciation genes than it used to be" says Michael Nachman, who studies mouse genetics at the University of Arizona in Tucson.
With more genes should come greater insight into speciation. Some geneticists wonder whether only particular classes of genes are important in speciation — such as epigenetic genes or segregation distorters — or whether many sorts of genes help to drive species apart.
"What is surprising about the speciation genes that have been identified [so far] is that there is a whole hodgepodge of different kinds of genes with different functions," says Nachman. "I don't think we're going to see [trends] until dozens of genes are identified, and there's just a handful now."

But why think that there should be particular classes of genes that contribute to speciation? Sure, there may be genes that are implicated in Drosophilid speciation, or maybe even in insect speciation, but all that matters in sexual species is that some barrier to reproduction exists. It need not be a particular barrier. Consider this - how many ways are there to impede the flow of traffic? SHould we expect there are only a couple of ways? Witches' hats and workers' signs may be common, but there are sinkholes, burning barriers of demonstrations, collapsed cranes, street parties... and the list could be indefinitely extended. I suspect that speciation is like that - it's a negative property, and oen that can be arrived at in an indefinitely large number of ways. There's some bad thinking going on here.

In a paper that has been justly ignored, I suggested what we need is the concept of "reproductive reach" - the broad class of things that can interbreed. Speciation occurs when reproductive reach is interrupted.

In another development, a cryptic species of dwarf freshwater crocodile in the Congo Basin has been found, not by using 568 bp of a mitochondrial gene, but by comparison of many nuclear and mitochondrial genes. That's a good thing.

O. Mihola, Z. Trachtulec, C. Vlcek, J. C. Schimenti, J. Forejt (2008). A Mouse Speciation Gene Encodes a Meiotic Histone H3 Methyltransferase Science DOI: 10.1126/science.1163601
N. Phadnis, H. A. Orr (2008). A Single Gene Causes Both Male Sterility and Segregation Distortion in Drosophila Hybrids Science DOI: 10.1126/science.1163934
John Wilkins (2007). The dimensions, modes and definitions of species and speciation Biology & Philosophy, 22 (2), 247-266 DOI: 10.1007/s10539-006-9043-9

More like this

I have always found the topic of speciation interesting. I would like to follow it closely but I am not in academia. This means I don't have access to all the journals that the articles are published in. This is very frustrating. I can only hope that the open access movement will win out of the corporate ownership model. It would be nice if I could access the articles online. The closest college is a small state college specializing in teaching. They dont subscribe to journals like Biology and Philosophy.

Right on. With close to 30 definitions of 'species', a speciation gene is going to be a somewhat problematic construct, isn't it? No wonder that it looks like a hodgepodge of genes with different functions.

Gee, I was just going to cite Hey 2001 and Wilkins 2003 on speciation when I recognized your name. I liked the 2003 article.

References
Hey, Jody. 2001. The mind of the species problem. Trends in Ecology & Evolution 16, no. 7 (July 1): 326-329. doi:10.1016/S0169-5347(01)02145-0.

Wilkins, John S. 2003. How to be a chaste species pluralist-realist: the origins of species modes and the synapomorphic species concept. Biology and Philosophy 18, no. 5 (November 1): 621-638. doi:10.1023/A:1026390327482.

Given the wide array of pre- and post-mating isolating mechanisms, it seems that there should be a rather large number of different genetic mechanisms.

By Jim Thomerson (not verified) on 13 Dec 2008 #permalink

Thanks for posting this.

I agree with you that people are making too big an issue of those few examples of "speciation" genes.

The other problem is that scientists like Orr are using these results to claim that speciation is driven by natural selection. In the latest issue of Scientific American he says,

To contemporary biologists, then, the question of whether natural selection drives the origin of species reduces to the question of whether natural selection drives the origin of reproductive isolation.

For much of the 20th century, many evolutionists thought the answer was no, Instead they believed that genetic drift was the critical factor in speciation. One of the most intriguing findings from recent research on the origin of species is that the genetic drift hypothesis is wrong. Rather natural selection plays a major role in speciation.

I think Orr is wrong. I think that the actual reproductive isolation mechanism is most likely due to fixation of alleles in one of the populations by random genetic drift. It's really difficult to imagine how natural selection could drive reproductive isolation. The example he gives is not convincing.

I realize that Orr is a confirmed adapationist but this is going way beyond the data, as far as I'm concerned. What do you think?

If we are talking allopatric speciation, then development of reproductive isolating mechanisms between geographically isolated populations is happenstance, not natural selection for isolating mechanisms.

Sympatric speciation is perhaps a different story. If a population includes two morphs, and members which interbreed between morphs have reduced fitness, then natural selection would reenforce any reproductive isolation mechanism whch might oddur between the morphs.

By Jim Thomerson (not verified) on 13 Dec 2008 #permalink

For a very basic background info on speciation, I recommend this review made of questions and answers. Indeed, the whole site is interesting: evolution theory. Read the questions from 19 to 24. Where in the process are specific speciation genes in action?

By Dr. Betin (not verified) on 13 Dec 2008 #permalink

A quote that is thrown around a lot, although the source and exact wording escape me, goes something like: Belief in sympatric speciation is like the flu; most people catch it at some point, but they soon get over it.

The problem with sympatric speciation is that even when there is divergent selection, genes involved in isolating mechanisms don't become fixed because there is too much gene flow. However, it seems like it usually takes a long time for reproductive isolation to evolve through drift. I think that partially isolated populations with limited gene flow are the ones that most quickly evolve prezygotic isolating mechanisms. Gene flow is low enough that relevant alleles have a chance to become fixed but contact occurs often enough to create selective pressure against hybridization. I seem to recall reading about spadefoot toads where calls were more different at the contact zone between sister species then at opposite ends of their respective distributions. The implication was that contact produced selective pressure against hybridization because the hybrids were less fit. This selection led to call divergence at the contact zone.

Further, speciation is thought to be driven in many cases by small peripheral populations. Gene flow is lower and smaller population sizes means that alleles have a greater chance of being fixed.

Jim Thomerson says,

Sympatric speciation is perhaps a different story. If a population includes two morphs, and members which interbreed between morphs have reduced fitness, then natural selection would reenforce any reproductive isolation mechanism whch might oddur between the morphs.

I'm having trouble understanding this example. I hope you can help me.

In sympatric speciation we have two distinct populations living together. They don't interbreed very often--that's why we can recognize them as distinct populations.

In other words, there is already some sort of genetic isolation mechanism in existence even though the barrier is not absolute, as it would be if they were species.

At some point a mutation occurs in one of the populations. This mutation has no effect on mating within the population, otherwise it would be eliminated by negative selection. However, it does have an effect on matings between populations: those matings are now unproductive.

Remember that matings between the populations are already rare so the new mutation has hardly any effect except to eliminate the rare hybrids.

Please explain how this new mutation is positively selected within the originating population. What is the mechanism that detects the absence of rare hybrids and translates this into increased fitness for individuals that mate within the population?

@ Larry Moran

I'm going to try to answer for JimThompson. I think he is talking about divergent selection. The morphs are not reproductively isolated initially so there is no postulated reproductive isolating mechanism at first. If intermediate stages of the morphs are less fit than either morph then there is divergent selection. So any trait that decreases the probability of hybridization (through some sort of prezygotic mechanism, perhaps behavioural) will have increased fitness. Now, I have my doubts about how often this scenario has occurred in a totally sympatric situation, but that is what I think was meant.

"...even though the barrier is not absolute, as it would be if they were species."

I don't think that most biologists view absolute barriers as a requirement for separate species evolution. Ignoring the messy Canis lycaon vs Eastern Coyote issue, Coyotes (Canis latrans) and Grey Wolves (Canis lupus) are a good example. They almost never interbreed in nature, look different, and are ecologically very different and as far as I know just about everyone accepts them as separate species. They coexist in sympatry (more so in the past before wolves bit the dust in the great plains) but do not interbreed, yet they are perfectly cross fertile. It is examples like these that suggest to me that behavioural isolating mechanisms fixed through natural selection are an important component of speciation in many cases. Intuitively, I think that natural selection can speed up speciation under the right conditions (certain levels of geneflow, lack of fitness of hybrids, etc) by selecting for reproductive isolation more quickly then it would be created by drift in a totally allopatric situation.

What I wrote makes perfect sense to me.:) My example is a generalized example, and I don't recall a real example which fits. I am proposing that the two morphs are members of a single population, and that they freely interbreed. Then, something, unspecified, happens to reduce the fitness of individuals who mate with members of the other morph. Now there is selection for individuals who do not make the "mistake" of mating with the other morph. Over time isolating mechanisms (nature unspecified) would accumulate, and the amount of extramorph matings would diminish to the point that we would recognize the two morphs as reproductiely isolated, i.e. separate species.

So I am suggesting a series of events leading to sympatric speciation, without speculating on root mechanisms of the initial loss of fitness or the accumltation of isolation mechanisms.

My term paper for population genetics (1964) proposed sympatric speciation in a species of fish, with a variable blotched color pattern, inhabiting a mosaic environment of grass beds and rocky shoreline.

By Jim Thomerson (not verified) on 14 Dec 2008 #permalink

To use MattK's example of wolf vs coyote. If two species have very good premating isolation mechanisms, and thus, in nature, do not make mistakes, there should be no selection for post mating mechanisms. This is why hybridization in the lab, or in a disrupted habitat, does not necessarily invalidate the idea that two species are involved. It just means that the usual premating isolating mechanisms can be broken down.

To use a particular example. Steelcolor and spotfin shiners are very similar. It takes a practiced eye to be able to rapidly and accurately separate them out of a collection of minnows. In nature, they spawn under slightly different circumstances. In the lab they will hybridize to produce all sterile males. These males are distinctive (to the practiced eye)and unknown in nature. So there are effective isolating mechanisms, both pre- and post-mating. Is it a reasonable pastime to speculate on which came first, the premating or the postmating?

By Jim Thomrson (not verified) on 14 Dec 2008 #permalink

MattK: I think they keyword you're looking for is parapatry. There's lots of examples of that, probably as a result of relatively recent isolation followed by renewed contact (before and after the last glaciation, for instance). I may be out of touch here, but I'm not aware of any examples of "sympatric" speciation that weren't probably parapatric at some stage. Or rather, that can't be shown not to have been...

Jim Thomrson says.

What I wrote makes perfect sense to me.:) My example is a generalized example, and I don't recall a real example which fits. I am proposing that the two morphs are members of a single population, and that they freely interbreed. Then, something, unspecified, happens to reduce the fitness of individuals who mate with members of the other morph.

Exactly. Individuals with the new mutation have no fitness advantage when mating with their own kind but they are at a disadvantage when mating with individuals from the other population.

Thus, individuals carrying the new mutation are at a slight overall disadvantage relative to the wild type because they have a slightly reduced probability of passing their genes on to the next generation. (Because they don't produce hybrids.)

Now there is selection for individuals who do not make the "mistake" of mating with the other morph.

Why?

Over time isolating mechanisms (nature unspecified) would accumulate, and the amount of extramorph matings would diminish to the point that we would recognize the two morphs as reproductiely isolated, i.e. separate species.

I don't see why it would be advantageous for an individual to not pass on its genes by mating with an individual from another population. Why would such individuals have a fitness advantage over those would could produce viable hybrids?

I can see why the new allele could become fixed by random genetic drift even if it was slightly disadvantageous (nearly neutral) but I can't see how positive natural selection could be the cause of fixation. What am I missing?

Ah parapatry. I'm familiar with the term. It would be hard to know about allopatry and sympatry without knowing about parapatry too. I had to reread my posts to double check that I didn't actually use the term when I was talking about, well, parapatry. Anyway, glad you got it in there for me.

I may be out of touch here, but I'm not aware of any examples of "sympatric" speciation that weren't probably parapatric at some stage.

I think that something like sympatric speciation is hypothesized to be pretty common among herbivorous insects with a high degree of host specialization. Host switching occasionally occurs and it can lead to reproductive isolation esp. if mating occurs on the host plant. The classic example (and the only one that I can think of) is the apple maggot fly. This 'species' is descended from hawthorn maggot fly ancestors that made the switch sometime around Johnny Appleseed times. Now they are reproductively isolated (for the most part anyway). I have seen arguments about whether or not this is really sympatric. This is a question of whether or not geographic configuration is the defining characteristic of the various 'patries or whether they should be differentiated based on processes. Perhaps a little bit of moving of the goal posts from both sides.

MattK says,

I'm going to try to answer for JimThompson. I think he is talking about divergent selection. The morphs are not reproductively isolated initially so there is no postulated reproductive isolating mechanism at first. If intermediate stages of the morphs are less fit than either morph then there is divergent selection.

I don't think the term "divergent selection" means what you think it means. In many cases it could just as easily be replaced by "divergent drift."

In any case, the question before us is what happens when the new "speciation gene" arises in one of two populations that have restricted gene flow but are not species in the sense that hybrids can't be produced.

So any trait that decreases the probability of hybridization (through some sort of prezygotic mechanism, perhaps behavioural) will have increased fitness.

Why? Such individuals now have a slightly lower chance of passing on their alleles to the next generation. That does not sound like your average definition of "increased" fitness.

This is not a really good example to explain my scenario because it involves interspecific hybridizing. This was case of a toad species invading the range of another species. Most of the hybridization was males of the local species hybridizing with females of the invading species. The offspring were all sterile males. There was a fairly large mark and recapture study done. None of the females who hybridized were ever recaptured. This was interpeted as evidence of some kind of linkage between willingness to hybridize and reduced fitness in the new environment. I no longer have a copy of the paper. One author was E. Peter Volpe; time late 50's or early 60's. In my hypothetical example,I was thinking of some sort of drastic result for individuals who hybridized.

Syntopic is a useful term. Syntopic individuals of two species encounter each other and could try to interbreed if they so desired. Sympatric is sometimes defined just to mean overlapping ranges on range maps. A pine tree and a minnow could be sympatric but probably would not be syntopic.

By Jim Thomerson (not verified) on 14 Dec 2008 #permalink

Larry Moran: You're right, my bad. What I meant to say was 'disruptive selection'.

You quoted me: " So any trait that decreases the probability of hybridization (through some sort of prezygotic mechanism, perhaps behavioural) will have increased fitness."
and then said:

Why? Such individuals now have a slightly lower chance of passing on their alleles to the next generation. That does not sound like your average definition of "increased" fitness.

But you did not include what I said in the sentence before:

If intermediate stages of the morphs are less fit than either morph then there is divergent disruptive selection.

To be more explicit, if hybrids between two morphs (perhaps it would be better to use eco-types) are intermediate in character and if the intermediates are at a relative fitness disadvantage as they are with disruptive selection, then there will potentially be selection against hybridization. However, this depends on the cost of reproduction. Females at least tend to have a high cost of reproduction because of the investment required to produce offspring so I would expect a greater fitness consequence to them of hybridization. In a situation where females are 'choosy' (rather than 'floozy') mechanisms that prevent hybridization could still work even if the males will hump (or amplex with or whatever) anything that moves. So, from the perspective of a female that has limited reproductive potential, there is a strong advantage to careful selection of mates. Individuals of many species get only one shot at reproduction (e.g. salmon, many insects, etc) so choosing the best partner (one whose genetic contribution will maximize the fitness of the offspring) is strongly selected for.

Now, two clarifying points about what I'm talking about: 1) I'm not actually advocating for sympatric speciation in the classic sense as a common mode. In the comment you quoted I was specifically trying to articulate what Jim said (and, apparently, in a spectacular double-FAIL, I miss-stated my misinterpretation). 2) I think the commenters, myself included, got sidetracked away from the topic of the post. So I was not talking about these 'speciation genes' specifically.

From your comment to Jim:

I can see why the new allele could become fixed by random genetic drift even if it was slightly disadvantageous (nearly neutral) but I can't see how positive natural selection could be the cause of fixation. What am I missing?

I think that the key is that these genes are segregation distorters. They increase their own fitness at the expense of individual (organism) fitness. So they increase their representation in the population because they are present in >50% of gametes produced by heterozygous individuals. This way their proportion increases even if there is a fitness consequence to their 'host'. As a caveat, my understanding of this issue comes from an example vaguely remembered from my reading The Selfish Gene a few years ago. A quick googling to refresh my memory finds that the t-allele is a segregtion distorter in mice (80% of sperm from male heterozygotes carry the allele) but homozygous male mice are sterile (random google source attributes Lewontin and Dunn 1960). Even with the obvious fitness consequences, t-alleles can become fixed in small breeding groups of mice leading to extinction of the group. So a segregation distorter that induces hybrid inviability could probably become fixed. However, I agree with John Wilkins that this is probably not generalizable to most speciation events. Reversal of incipient speciation does occur and is a conservation concern for a lot of fish. Increased turbidity obscuring sexual cues and leading to melding of previously separate species has been suggested for some cichlids in African Lakes (I forget which one(s)) and cyprinids and catastomids in North America. There is also the Canis lycaon/C.latrans issue that I alluded to in a previous comment. I don't see how the presence of these 'speciation genes' would allow populations that are reproductively isolated even in sympatry to merge back together.

This is just a quick sidebar question I'd like to drop in, as I don't get it. Why must we divide speciation into "sympatric," "parapatric," and "allopatric" types, since the locations of populations really aren't very important to speciation itself, and these groups aren't really telling the whole story. Is it just our nature to try and group things into neat little categories? In short, we do, but should we classify organisms into a hierarchical structures for purposes other than evolutionary ancestry?

Without actually reading through the comments, I thought I'd drop in my two cents.

Having met with some of the "speciation genetics" researchers (Daven Presgraves, Michael Turelli, and a few brief comments with H. Allen Orr), I think that a lot of them realize that their results aren't completely generalizable. However, I do think that theoretical work by Turelli and Orr on the genetics of intrinsic post zygotic isolation gives us a reason to believe that, at least in species which show Haldane's Rule, we should expect to find a genetic component to the reproductive isolation that we could rightly call "speciation genes."

DMI says,

However, I do think that theoretical work by Turelli and Orr on the genetics of intrinsic post zygotic isolation gives us a reason to believe that, at least in species which show Haldane's Rule, we should expect to find a genetic component to the reproductive isolation that we could rightly call "speciation genes."

Nobody is disputing the existence of "speciation genes."

What I'm questioning is how those genes become fixed in a population. It's common to attribute that fixation to natural selection but so far I haven't seen an adequate explanation of how that might work. In most cases, it looks a lot like random genetic drift to me.

It is important to differentiate between 'selection of' and 'selection for.' Coyne and Orr argue that genes underlying reproductive incompatibilities (RI) are likely to have been under selection - that is, the changes underlying incompatibilities were fixed be selection rather than drift.

However, Coyne and Orr are more skeptical of the notion that RI genes were selected to function to prevent hybridization (despite some convincing data they collected). Discussion of selection for RI often revolves around selection for mate discrimination upon secondary contact of previously isolated, diverged populations, which are already somewhat incompatible. This form of selection is known as reinforcement, and is somewhat controversial.