In a paper published in PNAS in 1999, Douglas Schemske and Toby Bradshaw showed, through field crosses of a bee-pollinated and a hummingbird-pollinated monkeyflower (Mimulus) plant, that genes of large effect on pollinator preference have had a role in floral evolution and premating reproductive isolation of these plants. Seventeen years after the paper was published, I spoke to Douglas Schemske about his motivation to carry out this study, memories of field work and what we have learnt since about evolution of floral traits in monkeyflowers.
Citatin: Schemske, D. W., & Bradshaw, H. D. (1999). Pollinator preference and the evolution of floral traits in monkeyflowers (Mimulus). Proceedings of the National Academy of Sciences, 96(21), 11910-11915.
Date of interview: 21 September 2016 (via Skype)
Hari Sridhar: What was your motivation to do this particular piece of work? From looking at your publication profile, I realized that, by this time, you’d been working already on plant-pollinator interactions for more than 20 years. And in 1995, you published your first paper on this particular system. Could you give us a sense of the motivation for doing the work presented in this particular paper in relation to what you had done on this system earlier?
Douglas Schemske: Sure. The main motivation for the entire project was to understand the traits and the genes that contribute to speciation. So, in collaboration with Toby Bradshaw, who was at the University of Washington, we started working on this system that was first identified by members of the Carnegie Plant Biology group at Stanford. You may know they published a number of really outstanding, groundbreaking papers on the ecology and genetics of adaptation and speciation in California plants. So, they had identified this system as a good one because the two species are closely related, they are intercrossable…they’d done a lot of work on reciprocal transplants, but nothing on, really, the mechanisms of speciation, and nothing at all on pollination, other than just to describe it. Toby Bradshaw knew a lot about techniques and emerging molecular technologies specifically QTL (Quantitative Trait Locus) mapping. So, we decided that we could cross these species, produce a mapping population, and begin investigating the genetics of floral traits related pollination and then carry that out to the field. So, we first published a paper on QTLs and then wanted to know what the pollinators are. So, we took the plants down to Yosemite and we were able to use the pollinators, basically as assays, to determine which traits are most important and how they might contribute to reproductive isolation. So, in a nutshell, that’s what the background is for that.
HS: Stepping back a bit, how did you get interested in plant-pollinator interactions? I’m guessing that’s what you also did for your PhD, many years before this?
DS: Yeah. My primary interests early on were actually in fish. And, I also realized that I wanted to understand, both, the ecology and the evolution of fish, and look at components of fitness. And that was going to be hard. So, I switched to plants, basically, because I’m lazy. I like plants okay, but you can do so many things with them that you can’t do with vertebrates. And then I became fascinated with pollination, having read papers by Darwin and many others, that it was just a case where I could take my interest in animals and also study interactions between plants and animals. So, that’s kind of how it got started. And also, at about that time, I went to the tropics, in an OTS course, and that just completely opened my eyes to the importance of biotic interactions and, specifically, pollination biology.
HS: Would you remember the year in which you went for the OTS course?
DS: Yeah. 1973.
HS: You say, “A theory suggesting that mutations of large effect can often be beneficial during the early stages of adaptation.” From reading what you wrote, it seems like “large effects” and “major genes” were an important motivation for this piece of work. Can you say a little more about the role of this paper, by Orr in Evolution, in motivating your work?
DS: It was a general motivation for the entire project, but not specifically Orr. There was a long standing controversy about what the genetic basis of adaptation and speciation might be. Fisher, first, proposed this infinitesimal model where there are many mutations of small effect. And Orr revisited that and said, “Hey, you know, ecologically, that doesn’t really make a lot of sense. I can imagine a case where populations are far from their optimum, and maybe initially a big step would be advantageous.” So, we thought this was a good system to test those alternate hypotheses.
HS: Can you tell us a little bit more about how the collaboration between you and Professor Bradshaw started?
DS: So, that is sort of an interesting story. I was at the University of Washington, as was he. He was a research scientist in the forestry department, and I was on the faculty in plant biology. He had previously mainly done work on molecular genetics and, at that time, was working on poplars, and was really one of the pioneers on the whole poplar genomics research area. And, at that time, I was teaching a graduate course on plant evolution and he sat in on that. And I would often say, well, here are the controversies, and then, we don’t really have a genetic model or an approach to investigating these questions. He came up once and said, “Well, you know, I know how to do that. If you want to get together, we might be able to do something.” That’s exactly how it started. So it was me bringing this sort of ecological background and him bringing the molecular techniques. It’s been a spectacular collaboration ever since.
HS: Did he do all the molecular work and did your group do all the experimental work in the field?
DS: Well, he did the molecular work, and then he and I and a cast of thousands did all the fieldwork. I mean, he was always there in the field and he enjoyed it and participated just like everybody else. That’s something that I knew how to do, but he has such good intuition and interest in biology in general that it was a natural fit for him as well.
HS: Did you say a cast of thousands?
DS: We had quite a few. Not quite thousands.
HS: Were these students?
DS: Not so much. I don’t usually have students do the work on my projects. I want them to do their own thing. Now, having said that, I will hire them now and then to do things, but it’s not like they are working directly on the projects. So, for the paper that we’re talking about, specifically, I had an undergraduate in the lab help, I had another researcher help, one of Toby’s people came down, and then me and him. That was basically it.
HS: How did you decide to do the experiments in Yosemite?
DS: So, the original paper published by the Carnegie group in the monograph that came out of this is by Hiesey, Knobs and Björkman (1971; Carnegie Inst. Washington Publ. 628, 1–213); they’re at the Carnegie institution. They did all of their work in Yosemite or in the vicinity and they identified places where these two species are sympatric. So, rather than reinvent the wheel, we just went back exactly to those places.
HS: You mention another site where you made observations in 1998. Was that also in Yosemite?
DS: No, but it was just outside Yosemite. The bee-pollinated species was primarily found at high-elevation, ant the hummingbird species was at low elevation, but they overlapped at about 4200 feet in Sierras. So, any place you can find a riparian habitat at that elevation, you’d often find both of them there.
HS: Do you continue to work in these sites today. Is the greenhouse still used for experiments?
DS: Well, no to both. We got another grant after this work. And Toby pursued a number of molecular genetic studies in the floral traits. I moved to Michigan State. We continue to collaborate on a few things, but we’re not doing it anymore, for a variety of reasons. One is just the regulations in trying to introduce engineered lines of plants is pretty challenging. It’s also challenging to make them. So, we’re no longer doing that, but that collaboration lasted until just a few years ago.
HS: When was the last time you visited either of these sites?
DS: Oh, I would say, just guessing, maybe three, four years ago. Something like that.
HS: Have these sites changed a lot from the time you work there for these experiments?
DS: Not so much. I mean, as I said before, these species live along fairly large rivers. And, if there is a large snowpack and a rapid melting, then the banks are scoured out and it actually produces a lot of good habitat for these seeds to establish and germinate. So there is a constant flux in how high the river is, how much bank scouring there has been, but, more or less, it’s quite similar to the way it’s been. I mean, that’s another advantage of working in Yosemite – it’s a protected area. So there’s not going to be a shopping mall that goes in; it’s protected.
HS: In the paper you talk about two experiments – one experiment that was earlier published as two papers (1,2) and then a second experiment that this paper reports on. Was the study planned this way from the beginning?
DS: Yeah, pretty much. So, the very first paper was extremely preliminary even though it was successful. We used that basically to motivate the next studies to get some money to do it in a more comprehensive fashion. But from the very start, this was the plan, to do the crosses through reciprocal transplants, grow the plants in the field, look at pollinators, identify the QTL that underlie the traits and then make various lines that are engineered to have particular traits that we expose to pollinators. So, one of the papers we did shortly after I arrived here was to take a single trait, carotenoid concentration, and make so called near-isogenic lines, where we moved the QTL from the hummingbird species into the bee-pollinated species and vice versa. And we put those out in Yosemite as well, so then we can look at the effect of a single trait and a single QTL that underlies that trait.
HS: Do you remember approximately how long the writing of this paper took and where and when you did most of the writing?
DS: I probably don’t remember any of that. The way we typically work on these things is the lead author would do a draft – so that would be me – I would send it to Toby, and he would almost instantly get back to me, we’d have a back and forth and then off it would go. I think in this case, it went pretty fast because the data were straight forward, and we were shooting for, you know, so to speak, a high impact journal, which had a page limit. So it was pretty fast.
HS: Did you and Toby meet often to discuss this project, or were the discussions about writing mostly done over email?
DS: We would talk almost daily. And even when coming to MSU, we would have lots of communication, more on phone than email. Email doesn’t really lend itself to long rambling conversations. He and I work best when we’re in the same room.
HS: You say, “Despite striking morphological differences, these two monkey flowers are very closely related. A phylogeny based on DNA sequence from the internal transcribed spacer of nuclear ribosomal RNA places M. cardinalis and the Sierra Nevada form of M. lewisii together and distinct from Rocky Mountain and Cascade Range populations of M. lewisii and other members of the section Erythranthe (A. Yen, R. G. Olmstead, H.D.B. and D.W.S., unpublished work). Was this published later as a paper?
DS: Yes, it was. Dick Olmstead was a plant systematist at Washington, and he had a student named Paul Beardsley, who took up this project and published two very nice papers looking at the phylogeny of Mimulus over the entire group, and also our Erythranthe section.
HS: Did this paper have a relatively smooth ride through peer review? And was PNAS the first place you submitted this to?
DS: Now, that’s a good question. So, back in the day, it was important to get papers in the high visibility journals. I guess it still is, but I sort of ignore all that. We submitted it to Science and they chose not to review it and send it back almost instantly. And then I revised and sent it to PNAS. The reviews were reasonable and had some good points. So, we modified it and they accepted it. One funny thing is that shortly after the PNAS paper came out, Science contacted me and said they were doing a general review of speciation and could we send them some of the images of our plants? And I said, No, I mean, you weren’t interested in even reviewing it, so I don’t see why you would like to see the photographs. It wasn’t a good career move, but they kind of pissed me off in the first place.
HS: Who took the lovely photographs in the paper?
DS: I think I did. But I’m not good at it. You know, it took a lot of time and screwing around, but I’m definitely not good at it.
HS: How was the paper received in academia and the popular press? Did it attract a lot of attention?
DS: Oh, I guess yes to both. So, I remember when it came out, I was fishing in Montana. And one of the writers for the New York Times contacted me and did a story on it. She was quite interested in our finding that there was large effects that might contribute to adaptation. So there was that. And then in, you know, academic circles, I think everybody thought it was a good contribution. There was some controversy about, just in general, whether there are so called pollination syndromes. And whether red, in particular, is attractive to hummingbirds and de-attractive to bees; you know, sort of a minor thing. But I think the paper’s held up pretty well.
HS: I would like to go over the list of people you acknowledge to get a sense of a little more about how these people helped. Can we do that?
DS: Yeah.
HS: B. Best.
DS: B. Best was one of my former wives. She was also a biologist, and she was with me in Seattle.
HS: J. Coyne.
DS: J. Coyne is Jerry Coyne, who, with Allen Orr, wrote the most recent book on speciation and is a world leader in studies of evolution and speciation. A very good friend of mine and the most critical person I know; and I say that in a positive way.
HS: D. Ewing.
DS: He was the greenhouse manager. You know, we grew hundreds and hundreds, thousands of plants, and without his expertise it just wouldn’t have happened.
HS: B. Frewen
DS: Yeah, that’s Barb Frewen, who was a student with Toby. She did some of the pigment extractions and contributed to some of the molecular work and also to some of the field work.
HS: J. McKay.
DS: John McKay. I think he hadn’t yet started grad school. So, he was just looking for opportunities to do research in plant evolution and he contributed to some of the molecular work. He’s now a very successful evolutionary plant biologist, and he and I collaborate on the Arabidopsis work that we’re doing now.
HS: K. Otto.
DS: He was a research tech. in Toby’s lab who did much of the molecular work.
HS: Y. Sam.
DS: Okay, this is a long one, but I’ll explain. It’s pretty humorous. So that refers to the cartoon character Yosemite Sam. Do you know anything about that?
HS: No.
DS: Okay, you have to check it out online. I’ll make this fast. So, when we grew up all these hybrids, we used a series of photographic filters to see the images as they might appear to pollinators. And bumblebees have a different color vision than vertebrates, so they can’t really see red as a color. We took all these filtered images, and one of the F2 genotypes, through this filtered image, looked exactly like the cartoon character Yosemite Sam. So, in all of our papers, we acknowledge Y. Sam. We don’t say Yosemite, we say Y. Sam. We also submitted this image to Nature, because we published our first paper on this system in Nature and said, would you guys be interested in publishing this? And they said, no. But the reason was that if they did so, they would have to publish many photographs of potatoes that look like the President Richard Nixon. It’s the best rejection letter I’ve ever had. So that’s where Y. Sam comes from.
HS: K. Ward.
DS: She was a student in my lab, who helped in growing the plants and doing some molecular work.
HS: You also mention that, typically three to five observers made all the observations for the experiments. Were these people also from the university?
DS: Yes, yeah. Some of the people that I already mentioned were involved in that. Because we had 200 some plants out there, they were large, right, and in the course of a week, I think, we saw 12,000 pollinator visits. So it took a number of people running around and following these things.
HS: E. Sugden
DS: I think he was an instructor in zoology. I’m not 100% sure of that. But he was an expert on bee biology and he’d worked in that region and was familiar with the bees. So, he identified all the bees for us.
HS: J. van Wagtendonk and P. Moore.
DS: Yeah, so both of those people were administrators at Yosemite National Park, and were essential in getting us permission and access to the field sites.
HS: What kind of impact did this paper have on your career and the future trajectory of your research?
DS: I think it did have a direct impact because it was a story that’s accessible to almost anyone. Everybody knows flowers, everybody knows pollinators, and although it’s not as if we found something totally novel and striking – I mean, we’re studying adaptation – it’s just that, I think, flowers made it so much more accessible. So, you know, this has been reprinted in a few textbooks, and people know the images, just because it resonated with so many people. So, in that sense, it gave my career a little bit of a boost, I suppose. It also made me much more interested in understanding the underlying genetics of adaptation. And because I was in the golden years of my career, I actually decided to switch from Mimulus, which is beautiful and elegant but doesn’t provide all the genetic tools that I needed before I would retire. So I moved on to Arabidopsis, but basically taking exactly the same fundamental approach as we did with Mimulus, but just that there’s no significant pollination story to tell. But the conceptual perspective is identical.
HS: Today, 17 years after this paper was published, would you say that the main conclusions from this paper still hold true, more or less?
DS: I would say so. Yeah, I mean, there’s probably debate about that. You know, the basic question is what is the genetic basis of traits that contribute to adaptation and speciation? And what we found is that there are a few of large effect and many of small, and I think that is, more or less, a general rule, but there’s still so few studies that I wouldn’t say that we have the complete answer. But for this kind of scenario, I think I think it’s held up pretty well.
HS: If you were to redo these experiments today, would you do anything differently given the advances in genetic tools, theory and statistical techniques?
DS: Um, I wouldn’t change the field experiments at all. I mean, they worked far better than I ever thought. So, I don’t think I could improve on that. But certainly the genetic techniques have improved dramatically, so, instead of using the genetic markers we used, we would use single nucleotide polymorphisms, we’d have more of them, we’d probably do several more rounds of crossing to get the genome to recombine more. But other than that, that’s the way people would still do things just using a different set of markers.
HS: Did you build upon work that was done in this paper?
DS: Oh, yeah, absolutely. The paper here demonstrated that there are basically two traits that have a very large effect on visitation – nectar and color, specifically these carotenoid pigments. So with that information, then we pursued both of those, and color had a more successful outcome than nectar. But then, as I mentioned earlier, we produce these near-isogenic lines, put those lines in the field and showed that that single substitution in the background of the parents had a large effect on both bees and hummingbirds. So it was all sort of a continuous idea that we would get down to the mechanisms finer and finer and finer. And that’s certainly what we did.
HS: You say, “We cannot exclude the possibility that other, unmeasured traits may contribute to pollinator visitation, and that these may be linked to the traits included in our study, or have pleiotropic effects on those traits”. Subsequent to this study, were any other traits discovered as being important?
DS: Yes. So, Toby and his group worked a little bit on scent. And at the time, we didn’t think that scent was that important, and they found out that it did make some contribution. And that was work that was done exclusively in his lab. Beyond that, nothing else has emerged.
HS: In the very last sentence of the paper, you say, “Further studies are needed to determine whether our results can be generalized to other plant taxa where closely related species differ in their major pollinators”. Could you reflect on that and say to what extent that’s happened?
DS: Ours, I think, was really the first to look at the case where closely related species had very different pollinators, attempting to look at the traits and genes involved. Since then, there have been similar studies in Petunias, and there too they find very large effects, particularly for flower color. And in their case, it’s a bee versus hawk moth pollination. They also found differences in scent. We’ve since worked on a tropical group in the genus Costus, where, in the new world tropics, there’s about 55 species and half of them are bee-pollinated half, half is hummingbird-pollinated. Hummingbird pollination is derived seven different times. And we’ve done some genetic work on that and it’s a similar story – we find large effect QTL for flowering. The nectar production is the same for both of those, so, in that case, there are actually fewer traits that have to evolve. There’s been similar studies in Iris. So, I think it appears to be a general case that there’s one or two large effect genes that contribute to pollinator isolation in closely related groups.
HS: This paper has been cited over 500 times. Do you have a sense of what it mostly gets cited for?
DS: I think it’s probably cited for two things. One is, how pollinators contribute to speciation. and that’s probably the main one, because it’s a case where there’s a biotic interaction where adaptation directly leads to reproductive isolation. And then, second, I think, probably, the idea that some large effect genes and QTL can contribute to adaptation and speciation.
HS: Have you ever read this paper after it was published?
DS: Yes, because I would sometimes have my class read it and discuss it and criticize it. So, I had to make sure that I remembered what I had done. In that context, I have read it a few times.
HS: If you compare this paper to the papers you write today, do you notice anything strikingly different?
DS: That’s a good question. Well, I don’t think so, but I really dislike writing. I love doing the work, I love finding the answer, and I don’t really like writing so much. So, I try to write all of my papers in the same way – to have a big conceptual question, clear hypotheses and clear results And that gets hard when you don’t have the complete story. In this case, at least, the story was pretty complete. I would say now there’s more statistical issues, particularly when one is doing genetic mapping approaches. You have to be careful to recognize that a QTL is not necessarily a gene. I’d say more caution, perhaps, enters my writing when I’m trying to link molecular approaches to evolutionary questions.
HS: Would you count this as one of your favorites among all the papers you have written?
DS: Yeah, it was. It was a spectacular collaboration and Toby Bradshaw remains one of my very best friends. He’s a brilliant guy. And a lot of the success I have is because of my conversations with him. I love doing fieldwork. That worked beautifully. It was a fabulous sight. The pollinators were exquisite. So, virtually every part of it was great. And I think it’s a good story.
HS: What would you say to a student who is about to read this paper today? What should he or she take away from this paper published 17 years ago? Would you add any caveats?
DS: Well, first, before they read this, I would ask them to read other things that pre-dated it, to think broadly about how populations adapt to new environments, how that adaptation can lead, eventually, to the formation of species through reproductive isolating barriers. This is just one of many examples of how that might happen. They should judge the importance of combining field work with some molecular evidence and how new techniques might improve it. But I will also emphasize that the basic questions are still in place; the new technology has not changed that, in my mind, at all. That’s kind of the warning I would give.
0 Comments