Discussion
Our analysis of more than 20 million SNPs shows that eastern and western North American monarchs have extremely low genome-wide genetic differentiation. We did not detect any fixed nucleotide differences between eastern and western monarchs, and even the smallest window (100bp) size analyses indicated low maximum FSTvalues of 0.06, indicating a lack of genomic islands of differentiation. The windows with maximum genetic differentiation between eastern and western monarchs were low compared to the genome-wide average genetic differentiation between subpopulations in other butterflye species (Nadeau et al. , 2013; Talla et al. , 2019; Martin et al. , 2020). We found an almost perfect genome-wide correlation between nucleotide diversity in eastern and western monarchs, and genome-wide phylogenetic analyses indicated no clustering of eastern and western monarchs. Both ∂a∂i and Tajima’s D results suggest that eastern and western monarchs have a similar effective population size. Importantly, both methods contrast with population census data, which show much smaller numbers of western than eastern monarchs (Schultz, Brown, Pelton & Crone, 2017; Malcolm, 2018; Pelton, Schultz, Jepsen, Black & Crone, 2019), and support the notion of frequent genetic exchange between these monarchs.
Our results are in line with previous studies based on more limited genetic markers, including allozymes (Shephard et al. , 2002) and microsatellites (Lyons et al. , 2012). These findings also support observational, geographical, and tagging studies that have suggested regular interchange between eastern and western monarchs (Brower & Pyle, 2004; Dingle, Zalucki, Rochester & Armijo-Prewitt, 2005; Morris, Kline & Morris, 2015). Because the Rocky Mountains form a dispersal barrier for monarchs, the high levels of interchange between eastern and western monarchs indicated by our genomic analyses most likely occur during both the spring migration – when north-flying monarchs from Mexico could end up west of the Rocky Mountains (Brower & Pyle, 2004) – and autumn migration, when monarchs from western North America can end up migrating south to Mexico (Morris et al. , 2015; Billings, 2019). Indeed, population genetic analyses of microsatellites are consistent with a radial dispersal of monarchs from Mexico, including north-ward dispersal to western North America (Pierce, Altizer, Chamberlain, Kronforst & de Roode, 2015). Further tagging studies will be necessary to map the migration routes of western monarchs (Dingle et al. , 2005; James et al. , 2018) and to determine where actual genetic exchanges between eastern and western monarchs are occurring. It is interesting to note that in our study, three samples from Big Sur, California, appeared to cluster in both the ADMIXTURE and SAGUARO analyses. This suggests potential genetic sub-structuring in western North America, consistent with the non-random migratory pathways of western overwintering monarchs inferred by tagging and stable isotope studies (Naganoet al. , 1993; Yang et al. , 2016).
As with many migratory species, monarch migration has a genetic basis, and genome comparisons between migratory and non-migratory populations have revealed strong evidence for the existence of migration-related genes (Zhan et al. , 2014). While migration per se is genetically determined and associated with large-effect alleles, the lack of genomic divergence observed here suggests that differences in migration routes, distances and destinations for migratory monarchs are not. Our flight trials clearly demonstrated differences in flight performance between eastern and western monarchs, and these phenotypic differences may be driven by either a large number of small-effect alleles or by differential gene expression (Liedvogel et al. , 2011) induced by environmental triggers in eastern and western North America. To conclusively discern between these two alternatives, one would ideally carry out genetic crosses between eastern and western monarchs, and then release both parental genotypes and cross-progeny offspring on both sides of the Rocky Mountains to study migratory behavior, for example through the use of radio-tra­cking tagged monarchs . However, current United States Department of Agriculture regulations prohibit the transfer and release of monarchs across the Rocky Mountains – partly based on the assumption that eastern and western monarchs are genetically distinct populations – preventing such a definitive study design.
However, some progress could be made by comparing the transcriptomes of eastern and western monarchs throughout the year. Our preliminary gene expression studies showed a trend for differential gene expression, and one gene related to non-muscular motor activity was significantly differentially expressed. While these results show that eastern and western monarchs may differentially express migration-related genes during active flight, it is likely that many other genes are differentially expressed both during active flight, and in the developmental stages leading up to migration. Previous studies have shown different transcriptome profiles between breeding and migratory monarchs in eastern North America, including differences in expression of genes related to juvenile hormone production (Zhu, Gegear, Casselman, Kanginakudru & Reppert, 2009). Other studies have shown that non-migratory monarchs in Australia, which evolved from migratory monarchs in North America (Zhan et al. , 2014), have retained the ability to enter reproductive diapause (Freedman et al. , 2017), and that exposing eastern North American monarchs to artificial light and temperature conditions disrupts migration orientation behavior (Tenger-Trolander, Lu, Noyes & Kronforst, 2019). Our study further suggests that environmental variation on the east and west of the Rocky Mountains triggers monarchs to follow different pathways to develop into eastern and western migrants. Such factors may include the different species of milkweeds that monarchs use in eastern and western North America (Woodson, 1954; Dilts et al. , 2019), as recent work shows that milkweeds can significantly affect wing morphology (Davis & de Roode, 2018; Freedman & Dingle, 2018; Decker, Soule, de Roode & Hunter, 2019). A transcriptomics study examining differences during development and flight between eastern and western monarchs would be important in uncovering gene regulatory networks involved in migration ability, and further shine light on how these highly similar genomes can give rise to divergent migratory behavior.
Ultimately, determining the genetic or epigenetic basis of differential migration in eastern and western monarchs will not only advance our understanding of migration genetics, but may also have relevance for conservation biology. The population size of eastern migrating monarchs has dwindled over the last three decades (Vidal & Rendón-Salinas, 2014; Malcolm, 2018; Boyle, Dalgleish & Puzey, 2019), with some estimates indicating a decline over 80% from a high in 1996 (Semmens et al. , 2016). While studies disagree on the primary cause, an emerging picture is that monarch population decline is due to a combination of illegal logging at the Mexican overwintering sites, climate change, agriculture-induced loss of milk­weed host plants in North America, and reduced availability of nectar sources along the fall migration flyways (Pleasants & Oberhauser, 2013; Vidal, López-García & Rendón-Salinas, 2014; Inamine, Ellner, Springer & Agrawal, 2016; Thogmartin et al. , 2017; Boyle et al. , 2019; Saunders et al. , 2019; Wilcox, Flockhart, Newman & Norris, 2019). Western monarch population size has also declined (Espeset et al. , 2016; Schultz et al. , 2017), reaching critically low levels in the 2018-2019 migrating season (Pelton, 2018; Pelton et al. , 2019). Monarch migration has been coined an endangered phenomenon (Brower et al. , 2012), and the population decline has led a group of organizations and scientists to petition the US Fish and Wildlife Service to protect monarchs under the Endangered Species Act (Center for Biological Diversity, Center for Food Safety, Xerces Society & Brower, 2014). Following recent advances in merging evolutionary biology with conservation biology (Hendry et al. , 2011; Lankau, Jørgensen, Harris & Sih, 2011; Sgro, Lowe & Hoffmann, 2011; Smith, Kinnison, Strauss, Fuller & Carroll, 2014), a crucial aspect of this process is to determine the adaptive capacity of monarch butterflies. This includes asking how much adaptive genetic variation monarch populations harbor, and which populations must be preserved to allow the species to adapt to changing conditions and to preserve the processes that allow evolution to occur. If future studies reveal that differential eastern and western migration is driven by gene expression rather than by genetic differentiation, then this would suggest that preservation of eastern monarchs could potentially rescue western migration and vice versa . Future studies, ideally involving reciprocal translocation experiments, will be needed to address this important question.