Interview with the authors: testing the role of ecological selection on color pattern variation

The color variation that exists among individuals has lent itself to the study of selection since Darwin. Recently, Zaman, Hubert, and Schoville (2019) investigated the effects of selection on the diversity of the wing color pattern in the butterfly Parnassius clodius across a large portion of its range. These researchers found evidence supporting the idea that coloration may serve as a warning signal to predators, providing some predator avoidance benefits to individuals. In addition, the variation of solar radiation and precipitation observed geographically across sites was negatively correlated with the amount of melanin observed at each site. This suggests that the occurrence of melanin may provide a selective advantage in the form of thermoregulatory function. For more information on how selection influences butterfly wing coloration, please see the full article and the interview with Dr. Schoville below. 

Parnassius clodius mating, male located below female. Photo by Sean Schoville.

What led to your interest in this topic / what was the motivation for this study? Khuram and I were both interested in butterfly color pattern variation, and in particular, cases where there might be competing selective pressures acting on wing pattern phenotypes. Most work on butterfly wing patterns focuses on predator-prey interactions and aposematic colors, but butterfly wings are essential to flight performance and important in thermal regulation. A number of recent papers have shown that butterfly color pattern appears to be responding to climate warming, and then there are well-known cases (such as alpine Colias, thanks to Ward Watt) where thermoregulation has been linked to basking behavior and the pigment on wings. Thus, in examining variation in Parnassius clodius (which occurs over a broad elevation and latitudinal range), we hoped that we could decouple environmental signals that might act on different wing color elements. 

What difficulties did you run into along the way? Sampling our butterflies across this large region was a major challenge, particularly as adult flight times are rather short (two to three weeks). And then we were surprised by the strong difference in adult phenology across sites (some adults are active in May, others in late July). This is evident regionally (Utah versus Washington), and across elevation within a region. In the end, it required three years of effort, with some very long road trips from Wisconsin.

What is the biggest or most surprising innovation highlighted in this study? After some initial efforts to link aposematic variation (red eyespots) indirectly to predator communities (through climate variables that might covary with predator abundance), we realized this was too tenuous. So, we were delighted to discover publicly available data on bird abundance. While this did not solve the problem (perhaps due to lack of spatial resolution in the bird data), I think using this data to analyze butterfly wing patterns was one of the more innovative aspects of our paper. We had a much easier time linking spatial climate data to melanization (dark pigmentation). As an aside, this raises the important point that some data, i.e. abiotic environmental data, is much easier to come by than biotic data. This is unfortunate, as we expect biotic selective forces to be equally or more important drivers of microevolution in some cases.

Lead author on this study, Khuram Zaman, after a day of collecting samples. Photo by Sean Schoville.

Moving forward, what are the next steps in this area of research? We’d like to extend this work in two directions. First, we’d like to connect color pattern traits to underlying genes and measures of heritability. Although the genes controlling butterfly color pattern are well studied, to date no representative of the snow Apollo subfamily Parnassinae have been included in these efforts. Members of the family are tremendously variable and quite stunning in their dramatic contrasts of color. Second, experiments are needed to link our inferences of ecological selection to fitness differences, as well as performance in the field. Physiological assays of melanic variants, coupled with mechanistic thermodynamic models, have been developed for Colias butterflies (see Joel Kingsolver and Lauren Buckley’s work). This type of modeling could provide important connections to conservation of Parnassius clodius populations under changing climates, and might perhaps extend to conservation work on other highly threatened Parnassius species.

What would your message be for students about to start developing or using novel techniques in Molecular Ecology? The development of novel approaches is a key part of advancing biological knowledge, but it can be a daunting endeavor given the breadth and scope of the scientific literature nowadays. Integrating multiple approaches, on the other hand, can equally help to advance our knowledge and provide opportunities to address long-standing questions. This is the direction we took in this paper. My personal view is that students should to try to master multiple techniques (assemble a toolkit, so to speak) and apply those techniques to fundamental problems. Hopefully, it’s a lot of fun in the process and leads to interesting collaborations!

What have you learned about methods and resources development over the course of this project? We are entering a golden age of data-rich resources, in terms of spatial environmental data and genomics data. These increasingly provide the power to test refined hypotheses about evolutionary and ecological processes, and are becoming more accessible to all researchers. One of my favorite accomplishments in the paper is using genetic covariance data among populations (relatedness data) as a covariate in fitting morphology ~ environment models. The use of such population contrasts is important in controlling for non-independence in the data due to ancestry. While we have known about the importance of genetic covariance in hypothesis testing for some time (thanks to Joseph Felsentstein’s work), it is only recently possible to use genome-wide data. This provides very precise measures that are highly informative, and enabled us to rule out the role of genetic drift as a driver of wing pattern variation. 

Describe the significance of this research for the general scientific community in one sentence. Our research demonstrates that butterfly wing color patterns evolve in response local climate conditions, as a way to regulate body temperature.

Describe the significance of this research for your scientific community in one sentence. Our work demonstrates that elements of butterfly wing pattern phenotypes respond independently to different sources of selection, with climate variation acting on thermoregulatory ability as an important driver of butterfly color pattern.

Parnassius clodius basking to thermoregulate. Photo by Sean Schoville.

Zaman K, Hubert MK, Schoville SD. 2019. Testing the role of ecological selection on color pattern variation in the butterfly Parnassius clodius. Molecular Ecology 28:50586-5102.

Interview with the authors: Environmental heterogeneity and not vicariant biogeographic barriers generate community‐wide population structure in desert‐adapted snakes

Phylogeographic studies have long focused on striking biogeographic barriers, and comparative phylogeography often looks for shared divergence across such barriers as evidence of shared responses to similar environments across taxa. However, in addition to such barriers, geographic distances and local adaptation to environmental heterogeneity may shape genetic divergence. In their recent Molecular Ecology paper, Myers and colleagues collect genomic data from 13 co-distributed species of snakes from Southwestern North America and evaluate the relative importance of biogeographic barriers, geographic distance, and environmental heterogeneity in structuring genetic divergence. Much of the previous phylogeographic work in this region has focused on divergence across a prominent biogeographic barrier: the Cochise Filter Barrier (CFB), which separates the Sonoran and Chihuahuan Deserts, and divergence across this barrier has been suggested to be an important factor driving divergence in snakes from the region. Though they expected to find a prominent role of this barrier, instead, Myers and colleagues find strong support for geographic distance and environmental heterogeneity as important factors structuring genetic divergence, but less support for biogeographic barriers. Further, they find that different variables contribute most to divergence across the 13 taxa studied, highlighting the importance of species-specific responses to environmental variation. Read the full article here, and read below for a behind-the-scenes interview with lead author Edward Myers.

What led to your interest in this topic / what was the motivation for this study? 
As a research team we have a general interest in what factors are promoting population genetic differentiation and whether codistributed species have similar evolutionary histories in response to shared environmental changes over time. Specifically, in this system where there is a well known biogeographic barrier (Cochise Filter Barrier; CFB), we were interested in whether entire assemblages of taxa show similar population structure. Initially the motivation for this study was to assess the degree of co-divergence across the CFB, however, as we analyzed these data it became clear that we needed to incorporate spatial and environmental data to understand population divergence. This study also allowed me spend a significant amount of time in the field collecting tissue samples from snakes!

What difficulties did you run into along the way? 
One of the biggest difficulties with this study was handling and analyzing all the generated data. We had almost 400 samples sequenced for radseq, so processing and analyzing these data took a significant amount of computational time. Also, one difficulty was the logistics of collecting fresh tissue samples for all of these species across the southwestern US and northern Mexico, but issues like this are easily over come by collaborating.

What is the biggest or most surprising finding from this study? 
The biggest surprise from this study is that patterns of isolation-by-distance and isolation-by-environment are more important in explaining population genetic differentiation than a commonly cited biogeographic barrier. This result really stresses the importance of incorporating spatial analyses when analyzing phylogeographic data because aspatial analyses may result in spurious results of population structure and mislead our ideas of what is driving population divergence and speciation.

Moving forward, what are the next steps for this research? 
Moving forward I plan to generate whole genome sequence data for species within this system to understand what loci may be under selection in response to environmental heterogeneity. Given the strong signature of IBE I expect to find patterns of strong selection along transects of temperature and precipitation across the Sonoran and Chihuahuan Deserts. Further, I am interested in how other regions globally that have been cited as important biogeographic barriers in phylogeographic studies might also be strongly influenced by patterns of IBD and IBE, and not vicariant barriers.

What would your message be for students about to start their first research projects in this topic? 
There is so much great work published in the field of landscape genetics and comparative phylogeography and I would suggest that students start by combing through that work first. But as general advice I would suggest that students really explore their data in a meaningful way and spend some time thinking about what factors could be responsible for similar patterns observed in a genomic data set (e.g., IBD vs vicariance or selection vs historical demography).

What have you learned about science over the course of this project? 
I have really learned that genomic data should be carefully analyzed as to not be influenced by preconceived ideas of the system that you might be working within. Also, I think that this is becoming more and more true, but you have to collaborate in order to do great science.

Describe the significance of this research for the general scientific community in one sentence.
This work demonstrates that codistributed species do not have shared evolutionary histories, and that they do not respond to the same landscape and shared environment in similar ways.

Describe the significance of this research for your scientific community in one sentence.
Our work shows that simple patterns of isolation-by-distance and isolation-by-environment have contributed to population genetic differentiation more so than commonly cited biogeographic barriers.

Full article: Myers EA, Xue AT, Gehara M, et al.Environmental heterogeneity and not vicariant biogeographic barriers generate community‐wide population structure in desert‐adapted snakes. Mol Ecol. 2019;28:4535–4548.