In a paper published in Nature in 2002, Patrik Nosil, Bernard Crespi and Cristina Sandoval showed, using a combination of morphological measurements and mate choice experiments, that parallel evolution of sexual isolation in the walking-stick insect, Timema cristinae, was promoted by divergent selection for host adaptation, suggesting that such adaptation could play an important role in the early stages of speciation. Fourteen years after the paper was published, I spoke to Patrik Nosil about his motivation to carry out this study, memories of field and lab work, and what we have learnt since about the role of host-plant adaptation in speciation.
Citation: Nosil, P., Crespi, B. J., & Sandoval, C. P. (2002). Host-plant adaptation drives the parallel evolution of reproductive isolation. Nature, 417(6887), 440-443.
Date of interview: 23 November 2016 (via Skype)
Hari Sridhar: What was your motivation to do this particular piece of work during your PhD?
Patrik Nosil: This paper came out in 2002. The context was that around the late 90s, even early 2000s, the hypothesis of ‘ecological speciation’ (i.e., that natural selection and adaptation to different environments creates new species) was crystallized and articulated, for example by Dolph Schluter at the University of British Columbia and several other people. Before then it could be argued that speciation research had not really been a core focus of evolutionary biology for a while. This idea that natural selection would contribute to the formation of new species – of course, it’s an old idea; it was part of Darwin’s hypothesis for how species form – but really, it was in the late 90s, that the idea was given this name – ‘ecological speciation’. In turn, some of the core predictions of the hypothesis were articulated, and a few papers on the topic were published. So, myself, I was an undergraduate at the time doing my bachelor’s degree. I was getting quite interested in evolutionary biology. I started to think about what I wanted to do for my PhD research. Ecological speciation was an area that, at that time, that was emerging and becoming exciting, but there wasn’t very much published yet. The ideas were out there and there was a lot of circumstantial evidence, but focused studies trying to test the role of natural selection in speciation were still quite limited. So, at this point, it made sense to me that perhaps this was a good area to go into for my PhD. And really, my whole thesis revolved around the topic of ecological speciation. The paper in Nature tested for ecological speciation, and to sum up, the scientific climate at that time was that we really needed data testing the role of natural selection in speciation.
HS: How did you choose to work with Bernard Crespi and on this particular study system?
PN: I was in the Canadian educational system, where, at least back then, in the 2000s, there were quite a few doctoral fellowships that were available. And, if you were competitive for one of these doctoral fellowships, based on your undergraduate marks and research and reference letters etc, often you’ve had quite a bit of flexibility to choose a lab, because you would have your own salary paid for; the professor wouldn’t have to pay you. But back then, you pretty much had to go to a Canadian lab; since then it’s changed, actually, and the Canadian government is much more flexible about you taking those fellowships abroad. So, back then, I was restricted to labs in Canada if I received the fellowship. Of course, there are many excellent scientists and labs in Canada, so this was not a problem. So I started contacting most of the labs, not just the ones that worked on speciation, but basically top labs in evolutionary biology. When I contacted Bernie he had contact with the woman in California who had developed the stick insect system I ended up working on. The species in the Nature paper, Timema cristinae, is actually named after her – who is also an author on the paper – Cristina Sandoval. She had already done her PhD on the same species in California feeding on these different host plants. And so he introduced me to Cris. In fact, even before my PhD, I went down to California and Cris showed me the study system and then we started to plan to study speciation. There were populations on different host plants; they could be adapted to these different host plants. And we saw we could test whether the insects would mate with one another, a key feature of being a member of the same species. These kinds of practicalities meant that the system could be used to test predictions about species formation. In turn, Bernie was a very good evolutionary biologist and he was a great supervisor. However, the connection through Cris was also critical to deciding the PhD project.
HS: Had Cristina Sandoval done any work before this on speciation in this system?
PN: Some, but not much was published. She and Bernie had discussed doing so and they even had written a grant proposal to do so, but it was not funded. They’d built a phylogenetic tree of the different species in this genus, and analyzed how they shift between host plants and how they change color patterns – this kind of thing; more in a macro evolutionary framework than in terms of experimental tests of speciation. They had certainly thought about speciation, but they didn’t have someone to do the work. And so, I came at an opportune moment where, not just in this system but worldwide, people were thinking about ecological speciation. And, in the system, Bernie and Cris had been thinking about it, but there wasn’t anybody – e.g., a student – to do it. Also, Cris was starting to have many work duties outside of academia. She’s the manager of an ecological reserve in California. So I came at a good moment where I could actually go in and do the work that Cris and Bernie had discussed before, but no one was actually doing it.
HS: Before your PhD, you already had a number of publications on different topics? Did you do all that during your undergraduate years?
PN: Mostly, but not completely. I started those studies around the end of my undergraduate degree at the University of Victoria. The lead author on most of those papers was Tom Reimchen. He was a professor that I found particularly inspiring, and he was, really, my first academic mentor. I started working with him. And, the reason that we were able to get a number of papers published fairly rapidly is that he had a lot of data that was unpublished, which he allowed me organize, analyze and explore. The data stemmed from freshwater fish populations in the Haida Gwaii archipelago. After my undergraduate degree, but before my PhD, I actually worked for Tom for one year. I worked with Tom on these stickleback data sets – some that involved decades of data – that concerned interactions between predators and prey, and between parasites and hosts. So, my background before I started the speciation research was in predator-prey evolutionary ecology.
HS: When did you start you PhD and make your initial visit to California to check out the system?
PN: I started my PhD in the fall of 2000, but I had already gone down in the spring of 2000 to meet Cris and start some preliminary work on the system. I had done some of the kinds of mating trials that ended up being key in the published paper, testing if insects would copulate with one another, if they were from different host plants and not the same host plant. And, long story short, those preliminary data were promising, but I didn’t have enough pairs of populations to have a solid publication. I think, at that time, I tested three or four pairs of populations, and in the final paper there are 28 pairs. So the work in 2000, laid the groundwork per se. I started my PhD in the fall of 2000, with those data in hand, which I analyzed and we saw the results and we discussed with Bernie and other people. In 2001, I went back and really did the same thing as in 2000 but writ large – these 28 pair-wise comparisons instead of the three or four. That gave the mating results that are in the paper published in Nature. Then through the summer and the fall of 2001, I collected the molecular sequence data, and then everything got put together into a manuscript, and then the paper was published in 2002.
HS: Was the spring of 2000, when you started this work, the first time you saw walking stick insects?
PN: Yes. I’d worked mostly on fish and aquatic systems before this. So, it was a switch to working on a terrestrial system and in these habitats in California. I had never seen this particular group of insects at all. In fact, you cannot easily see them. Part of the story is that they’re camouflaged on their host plants. So, if you don’t know where to go, and which bushes to search on, then you’re unlikely to see them by accident. These insects are smaller than other stick insects, and really well-hidden.
HS: When you say quite small what do you mean?
PN: They’re between one and three centimeters or four centimeters. The females of the largest species can get quite large – a few centimeters – but they’re not as large as many tropical forms, which people might be more familiar with.
HS: In the spring of 2000 did you visit the field site where you did work for this paper, the Santa Ynez mountains?
PN: Yes. That first year I stayed at the ecological reserve where Cris was working. She hosted me, and gave me a bit of lab space. In fact, I also had lab space from her former PhD supervisor, whose name is John Endler. He was a professor in Santa Barbara. John gave me a bit of space in the back of the lab where I could have the insects and run my mating trials and do this kind of thing. That’s when I first started going up to the field sites and collecting the insects and starting to do the work.
HS: Was this lab space in the university at Santa?
PN: Yes. It was at the University of California, Santa Barbara, again, thanks to John Endler at the time. In more recent times, Todd Oakley at the same institution has kindly offered lab space.
HS: Could you give us a sense of the fieldwork your daily routine when you were doing this work?
PN: Yes, sure. Overall, this early work of mine was quite basic and inexpensive. Really, all it involved was, you drive up to the field sites, only 10-20 kilometers from town, – 20-30 minute drive. Some of the sites you have to walk into a bit, but usually it’s not too intensive. I now work on many more remote areas in California, but back then and to start with, it was often on Forest Service land near Santa Barbara. As I noted previously, when the stick insects are sitting on their hosts they are camouflaged and extremely difficult to see. So, basically, we ‘beat’ on the bushes to collect them. You have a stick, you put the net underneath a branch where you think there may be insects’ rests, and you beat the bushes, and then the insects, of course, fall into your net. Then you search in the net get the ones that you need. There were different parts to the project, but, basically most of it is collecting the insects and then bringing them back to the lab. Often, I wanted to make sure that they were virgin. So, I was focused on the ones that were not adults yet – immatures – because you knew they hadn’t mated yet. And then, I would raise them in the lab by feeding them plants. And then, when they were adults I would do the mating trials, where I put them in a petridish and I would observe them, and you can see if they copulate or not (and when, as I timed the trials). In brief, collect the insects in the field, raise them, do mating trials.
HS: The distance to your field site that you just mentioned, was that from Santa Barbara?
PN: Yes, from the ecological reserve, which is down on the beach. You have to go up into the mountains a bit to start getting to the host plants and these insects. They don’t exist at lower elevations. So, you’re driving from town sea-level up into the Santa Ynez mountains, getting up a little bit into the Chaparral habitat, is what it’s called, and then from there searching by foot.
HS: Did you bring the animals back to the lab in the university to do the experiments?
PN: Yes.
HS: Was all the phylogenetic lab work also done in a lab in Santa Barbara?
PN: No. This was done in Vancouver. I’d been doing my PhD at Simon Fraser University with Bernie. What I would do is, I would preserve some of the samples in California, either in alcohol or silica gel, which dries out the tissue, and then I would take the preserved tissue back to Vancouver. Then, in Bernie’s lab in Vancouver, I would do the DNA extractions and the PCR reactions to amplify the DNA and prepare the samples for DNA sequencing. So, the molecular work was done in Bernie’s lab in Vancouver, in Burnaby.
HS: Do you continue to work in these sites today?
PN: Yes. It’s expanded a lot but those core sites are still being used. Some of the sites were ones that were near ones that Chris worked on and some were the exact same as hers. So, some of the sites from 1990 are still being used now, but we’ve expanded the research program a lot because there’s many other stick insect species across California, and we’re trying to understand the whole radiation across different temporal and spatial scales. So, a lot of new sites have been added, but some of those core areas in the Santa Ynez mountains near Santa Barbara are still being used. In fact, in 2018 we published a paper in Science that presents 25 years of data from these core sites near Santa Barbara. The results show that evolution, at least on the scale of decades and tens of generations, can be predictable and repeat itself.
HS: What kind of data are you collating?
PN: It depends. We always collect basic observational data on the frequencies of different color morphs, because this is the traits that’s the easiest measure and has been consistently collected year after year (and is key for the aforementioned paper in Science). So, if you went to a locality and you caught 100 stick insects, almost every single time you would record what proportion of those were the striped morph phenotype or the unstriped morph, and there’s a few other morphs that aren’t talked about in the 2002 paper that we also designate. So this is very basic phenotypic information we have for very many years. Other types of information are more scattered, depending on the project or the experiment or what the different student was doing. We have other types of data on mating isolation or on the chemicals on the skin of the insects. But the really consistent data has been about the color morphs.
HS: Over these 15 years, do you think that these sites have changed in any way?
PN: Yes, it’s gotten drier. There was a drought in California around the year 2013, and so you can notice, I mean, anecdotally, just your eyes; you notice that plants are drier, and the densities of the Timema are changing a bit. And then, of course there’s urbanization. So, there are some sites where new houses have been built or the land has been sold or the land has been, you know, cut down. So, some of the older sites don’t get used anymore because now they’re in kind of urban or semi-urban areas. And this is happening all over the world. Still, many of the sites up higher in the mountains have not changed, not obviously, as they remain lush and undeveloped.
HS: Would you know what has happened to the lab you used, which you said was John Endler’s lab?
PN: John moved to Australia. There was a new a new biologist who moved to UCSB named Todd Oakley and he’s kind my contact; helping me when I go there. Now, what we often do, because I have larger grants, is we often rent a house where we can raise our insects in the house and do our experiments inside of this house. So, you don’t tend to use the campus anymore. We have kind of this field station, if you will, that’s in between the field sites and the town. One of the difficult things to get is ethanol, actually, because you cannot fly with ethanol, obviously, on a plane. And getting very high grade ethanol to preserve your samples is difficult, so Todd in Santa Barbara always provides us with some ethanol, so we can preserve our samples.
HS: When you did the experiments in 2001, were you staying in the town?
PN: Yeah, I mean, it’s an ecological reserve, it’s right in the borders of town, but it’s a protected area technically. It’s called Coal Oil Point Reserve. Cris is the manager there, alongwith her husband. They have guest facilities, not just for me, not just for people working on Timema, but for researchers that want to come and do biology in the area. It’s very near that the University of Santa Barbara campus.
HS: Did you have people to help you with fieldwork?
PN: I did a lot on my own, but I have had field assistants over the years, and now of course I have students and post-docs. In the beginning it was mostly me and, and for that paper, it was primarily me, yes.
HS: Is that also true of the mating experiments and the lab work?
PN: Yeah, most of that I did on my own, yes. Particularly for that paper. Later in my PhD, I started doing some more labor-intensive measurements where I would have undergraduate helpers and assistance and this kind of thing, but for that paper it was primarily on my own.
HS: Can we go over the names of people you’ve acknowledged, to get a sense of how you knew them and how they helped?
PN: Yes. Cris Sandoval pioneered the system, really, and introduced me to it. Bernie Crespi was my PhD advisor that introduced me to Cris, and her and Bernie had discussed working on speciation in Timema. And, I came at this moment, so they introduced me to the system, and I really kind of started to work on the project through Bernie and Cris. John Endler gave me lab space in Santa Barbara, particularly for that paper. And these days, I get a bit of help from Todd Oakley in Santa Barbara,
HS: T. Reimchen.
PN: That was my undergraduate supervisor Tom Reimchen.
HS: H. Rundle.
PN: He was another scientist working on the same topic really, ecological speciation, in stickleback fish, rather than insects. He was really someone I discussed a lot…in fact, you may know there’s a paper in Science in 2000 that’s really conceptually similar to our Nature paper, actually, about parallel speciation in stickleback, and he’s the first author of that paper. This 2002 Nature paper was similar in spirit. It was obviously different because it was in insects and plants, not fish, and there was a component of predation rather than competition. So, there were aspects that set it aside, but Howard was really working on similar topics for his PhD. So, he and I just discussed a lot about speciation, And we wrote a review paper about ecological speciation that was published in 2005 that has been cited quite a lot. And so we ended up working together a little bit after that.
HS: A. Mooers.
PN: He was also at Simon Fraser University and on my PhD committee. Bernie was my main supervisor, but Arne was on my committee, and he was very involved. He was not one of these lazy committee members. He was always giving very constructive, useful feedback and was interested in the project. So that was his role.
HS: F. Breden.
PN: Same as Arne. He was the other member of my PhD committee. Felix Breden. He’s also at Simon Fraser University. Felix and Arne and Bernie, they formed a group of evolutionary biologists. And they have an acronym for their lab – FAB Lab. So this FAB lab was really the kind of environment that I was in for my PhD, and that’s why the three of them were all acknowledged; well, Bernie’s an author, of course on the paper. But Arne and Felix are now good friends. I still see all of them. Now that, you know, you get older, and you mature from the student-advisor relationship into becoming colleagues and friends. So it’s always nice to see them when I visit Vancouver still.
HS: R. Vos.
PN: Rutger was Arne’s PhD student. He was a good friend during my PhD years. We discussed science a lot. There was no formal kind of role. But him and I were talking about science everyday in the lab.
HS: C. Parent.
PN: That was another PhD student. Same thing as Rutger Vos, really a PhD student of Bernie’s, again, during this time, the people that I discussed with the most, or that read versions of the manuscript before I submitted it; this kind of thing.
HS: J. Joy.
PN: Same thing. A PhD student of Bernie’s.
HS: S. Springer.
PN: Same thing. A student of Bernie’s. But here there’s an interesting story, actually. I don’t know if you put these kinds of anecdotes in but what happened with this manuscript, the first time we submitted it, we had the mitochondrial DNA sequence data, and the reviewers wanted us to add data from a nuclear gene – to substantiate the phylogenetic patterns. Now, this seems crazy in today’s day and age – because we’re sequencing whole genomes – but back then it was still the time where you wanted to have more than one gene to build your phylogeny. And so, what I tried to do in the lab was to get a nuclear gene working, basically, with PCR. But I’m not very good at lab work, so my reactions were failing. I had good DNA, I had the mitochondrial sequence, and I just needed to get some nuclear sequence data to revise the manuscript; I couldn’t do it. And Steve was very good at lab work and molecular biology. Actually, I didn’t know about this. He just ran a PCR with my primers and did the reaction for me, one night, and when I came in the morning, there was a gel with all the bands showing that the reaction had worked. I had to do more samples after that, but in the beginning, we weren’t sure if the problem was me, or if the problem was the reagents or the primers or something else. But what Steve did was confirmed that really the problem was me! So, Steve was also a student, but he played this extra kind of little role in helping with the lab work.
HS: D. Schluter.
PN: Yeah, so Dolph was the person that inspired me to work on ecological speciation. Some of these key papers in the late 90s that I mentioned were by Dolph. He was also on my PhD Advisory Committee. He was at the other university at the other side of town, in Vancouver, at UBC, and he was also on my committee. So, he formally contributed to the thesis this way, but it was also because he was, literally, like the world expert on ecological speciation. So, he was helping with the project.
HS: M. Fulton.
PN: Oh, yeah, that is an ex-girlfriend that did help me a little bit in the field, actually. She helped me collect some insects and so she can actually be acknowledged for a bit of field work.
HS: D McLaren.
PN: Same thing. He helped a bit with the field work.
HS: Were these people who helped students at the University in Santa Barbara?
PN: So, M. Fulton was an ex-girlfriend. Dale McLaren was just a friend of mine. Ah, let me see. I’m pulling up the paper now to see whose names are in there. But I think, for this paper, they were usually just people I knew. Later on, for future papers, I’ve often had students helping, but this was the very beginning where I didn’t even really know students. I’m trying to find the Acknowledgements. Yeah, same for B. Mickelson; that was a friend of mine that helped with the field work. M. Vankoeveringe was a technician in Bernie’s lab. He just helped with organizing the lab. T. Luchin was also a friend that helped with some of the analysis, actually.
HS: E. Rolán-Alvarez.
PN: Yeah, he was someone I did not know. He’s another scientist in Spain working on speciation. He provided software, which, I think, was already published in papers or something, but which helped with some of the analyses.
HS: For your phylogenetic analysis individuals from a different mountain, Ojala, was used as an out-group. Was this mountain also a part of the Santa Ynez mountains?
PN: Yes, it’s connected. In the valleys between mountains, you tend to not have the Timema because the host plants don’t grow or the climate’s not right. And so even if you have ridges of connected mountains, they’re kind of semi-isolated, and so you end up with different mountain tops being really kind of their own monophyletic group that can act as an outgroup. But the range of this entire species is really in that Santa Barbara area. I mean, it’s not only where I worked, but it’s not a very, very widespread species.
HS: Do you remember how long you took to write up this manuscript and where and when you did most of the writing?
PN: I would have done most of the writing in the fall of 2001. And initially, actually, I wrote it in a longer format. I’d written it to submit to a journal like Evolution, or, you know, one of the top journals in the field of evolutionary biology. And then, at some point, I think, actually it was one night when I was reading Howard’s Rundle’s 2000 paper in Science when I realized, well, really, nobody else has these kinds of results. And so, I wrote Bernie and I said, should we may be write it for Nature, which is always crazy because your chances of publication are extremely low. Bernie said, Yeah, let’s go for it, and so, at that point, I had to change it from the format of a regular manuscript into the Nature format. So, that took more work, but it wasn’t really that challenging. The papers that were publishing now as a community have so much more data in them; they’re much more challenging. This one was challenging, let’s say it took a few months, but it was not overly challenging.
HS: Did you write it when you were back in Canada?
PN: Yes.
HS: Do you remember if it had a relatively easy ride through peer review?
PN: It went through two or three rounds of review. It wasn’t easy, but wasn’t that hard either. It was, let’s say, warmly received even in the first round. There were a number of criticisms and it was clear they were not going to accept it right away, but they didn’t ‘trash’ the paper. One of the main criticisms in the first round was that we only had mitochondrial DNA for the phylogenetics. So that’s what led to the story about trying to get a nuclear marker working. And then there were a few other conceptual criticisms about the framing and the analysis. But the really big one in the first round of review was: you need to really add another marker to the phylogeny. So we did that and rewrote it. And then, in the second round, I think they were much happier with it because we had clarified all of the kind of conceptual and analytical and writing issues and added this marker. But there were a few issues that persisted, about really whether it was color pattern that differed between populations that affected mate choice or some other trait. It turns out that’s it another trait; the chemicals on the insects cuticle. So we had a second round of revision that was really to clarify that. And then it got in.
HS: Did it attract a lot of attention when it was published? Were the results considered controversial?
PN: One of the big issues, and this is still an issue in evolutionary biology, is how to interpret phylogenetic trees when you have gene flow between populations. This is because gene flow can blur or even erase signals of evolutionary history. For example, you can have geographic areas where you have very old lineages, and so they will be quite genetically divergent. However, if these lineages at some point occur together and start mating, then they start becoming genetically similar, even though they were genetically divergent before. In this manner, gene flow can obscure evolutionary history. So this was a little bit of a controversy, trying to interpret really what the phylogenetic patterns represented, what the word ‘parallel’ meant, in terms of, did these insect ecotypes really originate over and over and over and over again, you know, in each hillside, or did they originate once and then spread and come into contact and interbreed. And this is still – even 15-20 years later – something that the concept of parallel evolution with gene flow needs to consider. That said, a lot of progress has been made in data and analyses, and there is increasing acceptance that evolution can repeat itself even from a common pool of existing genetic variation; it need not always rely on completely independent new mutations.
HS: Did it also attract attention in the popular press?
PN: A little bit. I was quite junior at this point so I didn’t pursue much of this. There was an article in the Smithsonian magazine. It was more focused on Cris than myself because she had pioneered this system. So I didn’t really play a big role in that. There were a few things but not as much. And I didn’t really pursue anything. Nowadays if I have a higher profile paper I would be a little bit more aware to, maybe, contact the media or contact people in the media that had covered our work before and tried to see if we could get some press. But back then I was a PhD student and it didn’t even really cross my mind, to tell you the truth.
HS: Do you have a sense of what this paper gets mostly cited for?
PN: I would say for the general idea of ecological speciation; when populations adapt to different environments they change in ways that create new species. I don’t think it’s cited much because of the fact that it’s insects and host plants. It really cited mostly in the general context of ecological speciation.
HS: Do you think that this paper had any kind of direct impact on your career itself?
PN: Yes. Probably more than it should have. I mean, obviously getting a high profile paper early in one’s career helps. I stayed productive, obviously, through my scientific career, but it’s very hard to disentangle cause and effect when you have something important like this happen early in your career. It couldn’t have hurt, because, especially at that early career stage, I think it’s hard for more junior researchers to separate themselves from one another. And so having a high profile paper, I’m sure must have helped, because every application you can say, I have this paper in Nature. It’s hard to say how much, but I think it must have helped.
HS: Did it also have a strong influence on your subsequent research trajectory?
PN: Yes. The big question that was left unanswered by the paper, and by my thesis in general, was how far can this process of ecological speciation go? Can it really push populations to the point that they’re split into very distinct species? In this paper, these insects have some unwillingness to mate with one another when they’re from different host plants, but they’ll still do it; they still mate with one another if they have no other choice. And there was other data that we’d collected since that show that the hybrid offering don’t do particularly poorly; they do a little bit poorly, but not very poorly, in terms of survival. And so the insects on different plants are not really distinct species by any kind of definition, even if you argue about species definitions. So they’ve started the process of speciation as a consequence of adapting to different plants, but will they ever finish it? And this is one of the big questions in speciation research still, trying to understand these different steps or stages in this continuous process of speciation. And that’s really what’s driven a lot of work in this system and speciation research generally in the last five or 10 years, trying to figure out what these different steps are in speciation and what’s the role of natural selection in those different steps?
HS: Would you say the main conclusions from this paper still hold true, more or less?
PN: The core conclusion, yes, in terms of, host-plant adaptation promotes the evolution of reproductive isolation, at least to some extent. And this gets to the answer to the previous question – having a little bit of reproductive isolation isn’t really going to make you a distinct species; you need a lot of reproductive isolation. And so there could be many other factors. And in fact, this was why the title to the paper was carefully chosen. Even back in my PhD years, Bernie and I were careful to not put the word ‘speciation’ in the title. Because it was not clear if this kind of beginning or onset of speciation would really lead to distinct species. And so, we left ‘speciation’ out of the title and used ‘reproductive isolation; instead. So I think, at that level, it still holds, that adaptation promotes reproductive isolation. But really how important that is for speciation is a bit unclear, and in fact, we even have a paper that’s been in review for a while now that argues that there may be really a lot of other factors that are involved to really get you distinct species. There may be genetic or geographic factors that are really important. So, the core conclusions hold, but there’s likely a lot more to speciation than just adapting gradually to different host plants.
HS: If you were to do redo this experiment today, would you do anything differently given, the advances in technology, techniques in the lab, and in the theory underlying all of this?
PN: I would certainly do a few things differently, given both conceptual and technical advances. One thing is that, in this particular paper, all the individuals that were used were wild-caught individuals. So we did not control for their prior life experience or any life experiences their mothers had, etc. So there are these potential effects of plasticity. To some extent we addressed this issue in a future paper in 2003 in the Proceedings of the Royal Society of London. There, we actually reared individuals on different host plants to see if there was a strong effect rearing environment, and we found there was not, at least not within a generation. But now that genetic tools are better and we can breed these insects with more success it would be useful to better understand the relative roles of genes versus environmental experience in mate choice and speciation. The other obvious change would be to use modern genome sequencing to improve the phylogenetic results, well beyond just using two genes.
HS: You say that adaptation to a host plant is probably linked to prediction. Do we know more about that aspect?
PN: Yeah. It’s clear that these different morphs or these different color patterns that the insects bear confer camouflage and crypsis on different host plants. There was a bit of an argument for that in the 2002 paper. We actually had a paper, also from my thesis, in 2006 in PNAS that was about the predation side of the story. But these color patterns don’t contribute to the mating isolation. So, you have these two different axes of differentiation. The mating preference is diverging for some kind of yet unknown reason (that we now think is to do with the chemicals on their skin), and then you have the color patterns and the camouflage traits diverging because of the predators.
HS: In the abstract of the paper, you say that this is probably only the second example of parallel evolution of reproductive isolation. Subsequent to this study, have more such cases documented?
PN: Yes, many examples of parallel speciation specifically have since been documented, and even more cases of ecological speciation generally. And there are different kinds of evidence. There’s evidence for parallel speciation, this repeated evolution of reproductive isolation has been documented in other systems of insects and fish and snails. And then all kinds of other evidence have built up for a role for natural selection in speciation. So now, is reasonably accepted that natural selection will play some role in speciation, and will promote speciation often. But there’s a remaining big question of whether it can really complete the process, or whether it’s enough on its own, or whether you need some other forms of selection or some geographic isolation or some particular types of genetic changes or mutations to arise.
HS: Have there been other examples documented that involve host plant specialization?
PN: Yes, certainly. Even before our paper there was a very nice example published in leaf beetles, it’s citation number three in our Nature paper, published in the journal Evolution by Daniel Funk. His design was very similar to ours, albeit with a more limited number of comparisons. Specifically, like in our paper, Dan compared pairs of populations on different host plants to those on the same host, and used molecular data to control for population divergence time. His paper was a real inspiration to me, and the main advance in ours was having more population pairs such that we could statistically compare different-host pairs to same-host pairs. Since then, other examples have emerged in other insects. Jeff Feder’s work on host races of Rhagoletis flies also really inspired me.
HS: Did you continue to work with Bernard Crespi and Cristina Sandoval after your PhD?
PN: Yes. We followed up some of Cris’s earlier work on predation and stuff we looked at other forms of reproductive isolation, such as host plant preferences. We also studied the chemicals used in mate choice, and have used modern genome sequencing methods to elucidate the genetic changes driving adaptation and speciation.
HS: Have you ever read the paper after it was published?
PN: Yes, but rarely, because, in all honesty, I think the paper is simplistic, and well-engrained in my thinking. Now that I have a larger collaborative network and it is possible to collect so much genetic data, often we’ll have much more complex results that are nuanced. So I do read old papers again of mine quite often, but usually more complex ones where I need to remind even myself of particular details.
HS: The few times when you’ve gone back to reading it, has it, mostly, been to check details?
PN: Yes, usually to check some detail, some number, like, was it 80% or 70%? Usually, it’s just to toggle my memory because I’m writing a new paper and I want to get the context right or the nuance right. Usually, it’s for that.
HS: If you compare this paper with papers you write today, do you think there are any striking differences in the way you write?
PN: I would say this paper is written in a straightforward, linear way. As alluded to above, the complex and nuanced nature of more current papers makes its sometimes difficult to write in this way, although in honesty I am striving for clarity and simplicity wherever possible. However, it’s not always possible to tell a linear story where A leads to B leads to C. That’s a consequence of the data, I would say. We have so much data now that it almost forces a slightly different style of analysis and writing and thinking. Another factor is that I now write papers with a larger number of co-authors with diverse expertise, so one has to balance the views and writing style of a greater number of scientists.
HS: Would you count this as one of your favorites?
PN: If I am to be honest, I would say no, it is not one of my favorites. As I alluded to previously, although the results were important, I think, at the time, they were fairly straightforward and not entirely surprising. In other words, the results largely matched the expectations at the time. I do like the paper because it addressed a timely topic and was published in a high-profile journal – it really helped me career-wise – but I didn’t actually struggle very much with this paper and the results didn’t fundamentally surprise or confuse me. Many of my favorite papers tend to be the ones I struggle with, because I don’t understand the results or they’re confusing or we needed several more experiments to understand them, or because everybody’s arguing about what the results mean.
HS: If I asked you to pick one or two of your favorites, which would they be?
PN: I really like a paper we published a few years ago in Current Biology, that concerns how rapid evolution within species can affect entire ecological communities. The experiments in that paper showed that rapid evolution of departures from perfect camouflage within species of Timema, caused by gene flow, affects population and community dynamics (Figure 7). Specifically, conspicuousness of maladapted individuals attracts bird predators that prey upon T. cristinae but also on co-occurring arthropod species, reducing species richness of communities as ‘collateral damage’. In turn, this reduced abundance and richness of arthropods cascades to reduce plant herbivory. This work was instrumental in demonstrating how the evolutionary process of gene flow acting within a single species can have manifold ecological consequences for many species.
A topic related to eco-evolutionary dynamics is our ability to predict rapid evolution; my favorite paper to date is probably the aforementioned one we published in Science in 2018. Although prediction is a core component of all the sciences, evolutionary biology is often portrayed as a descriptive or historical science, rather than a predictive one. Nonetheless, the predictability of evolution can be quantified, for example by testing how well existing time-series predict future evolutionary change. Classic debates concerning whether evolution is predictable have gained new urgency as environmental changes facilitate the spread of infectious diseases and force species to adapt or risk extinction. This paper in Science highlighted how predictive ability can be constrained by limited data that cause poor understanding of deterministic natural selection, and how data limits can be reduced with feasible empirical effort. In this context, rapid evolution can be predictable, particularly if the mechanisms of evolution are understood.
HS: What would you say to a student who’s about to read this paper today? What should he take away from this paper? Would you add any caveats that they should keep in mind when reading it?
PN: I think the core message goes back to something we’ve already discussed. That’s this idea that adaptation or natural selection often promotes speciation. So, I think the core of this paper still holds, but how strongly so and whether it can really do so on its own, to the point that we get truly distinct species, is unknown. So, I think the paper is correct in terms of natural selection is part of the puzzle for understanding speciation, but the caveat would be that, I think it’s still unknown, in this system and in most systems actually, how far natural selection, on its own, can really push things. That would be the caveat. And that’s why, I guess, 15 years later, I’m happy I didn’t put ‘speciation’ in the title and that I used ‘reproductive isolation’ instead.
0 Comments