Evolution of butterfly seasonal plasticity driven by climate change varies across life stages.

Photoperiod is a common cue for seasonal plasticity and phenology, but climate change can create cue-environment mismatches for organisms that rely on it. Evolution could potentially correct these mismatches, but phenology often depends on multiple plastic decisions made during different life stages and seasons that may evolve separately. For example, Pararge aegeria (Speckled wood butterfly) has photoperiod-cued seasonal life history plasticity in two different life stages: larval development time and pupal diapause. We tested for climate change-associated evolution of this plasticity by replicating common garden experiments conducted on two Swedish populations 30 years ago. We found evidence for evolutionary change in the contemporary larval reaction norm-although these changes differed between populations-but no evidence for evolution of the pupal reaction norm. This variation in evolution across life stages demonstrates the need to consider how climate change affects the whole life cycle to understand its impacts on phenology.

Extensive transcriptomic profiling of pupal diapause in a butterfly reveals a dynamic phenotype.

Diapause is a common adaptation for overwintering in insects that is characterized by arrested development and increased tolerance to stress and cold. While the expression of specific candidate genes during diapause have been investigated, there is no general understanding of the dynamics of the transcriptional landscape as a whole during the extended diapause phenotype. Such a detailed temporal insight is important as diapause is a vital aspect of life cycle timing. Here, we performed a time-course experiment using RNA-Seq on the head and abdomen in the butterfly Pieris napi. In both body parts, comparing diapausing and nondiapausing siblings, differentially expressed genes are detected from the first day of pupal development and onwards, varying dramatically across these formative stages. During diapause there are strong gene expression dynamics present, revealing a preprogrammed transcriptional landscape that is active during the winter. Different biological processes appear to be active in the two body parts. Finally, adults emerging from either the direct or diapause pathways do not show large transcriptomic differences, suggesting the adult phenotype is strongly canalized.

Local adaptation of life cycles in a butterfly is associated with variation in several circadian clock genes.

Many insects exhibit geographical variation in voltinism, the number of generations produced per year. This includes high-latitude species in previously glaciated areas, meaning that divergent selection on life cycle traits has taken place during or shortly after recent colonization. Here, we use a population genomics approach to compare a set of nine Scandinavian populations of the butterfly Pararge aegeria that differ in life cycle traits (diapause thresholds and voltinism) along both north-south and east-west clines. Using a de novo-assembled genome, we reconstruct colonization histories and demographic relationships. Based on the inferred population structure, we then scan the genome for candidate loci showing signs of divergent selection potentially associated with population differences in life cycle traits. The identified candidate genes include a number of components of the insect circadian clock (timeless, timeless2, period, cryptochrome and clockwork orange). Most notably, the gene timeless, which has previously been experimentally linked to life cycle regulation in P. aegeria, is here found to contain a novel 97-amino acid deletion unique to, and fixed in, a single population. These results add to a growing body of research framing circadian gene variation as a potential mechanism for generating local adaptation of life cycles.

Phylogenetic reconstruction of ancestral ecological networks through time for pierid butterflies and their host plants.

The study of herbivorous insects underpins much of the theory that concerns the evolution of species interactions. In particular, Pieridae butterflies and their host plants have served as a model system for studying evolutionary arms races. To learn more about the coevolution of these two clades, we reconstructed ancestral ecological networks using stochastic mappings that were generated by a phylogenetic model of host-repertoire evolution. We then measured if, when, and how two ecologically important structural features of the ancestral networks (modularity and nestedness) evolved over time. Our study shows that as pierids gained new hosts and formed new modules, a subset of them retained or recolonised the ancestral host(s), preserving connectivity to the original modules. Together, host-range expansions and recolonisations promoted a phase transition in network structure. Our results demonstrate the power of combining network analysis with Bayesian inference of host-repertoire evolution to understand changes in complex species interactions over time.

Bayesian Inference of Ancestral Host-Parasite Interactions under a Phylogenetic Model of Host Repertoire Evolution.

Intimate ecological interactions, such as those between parasites and their hosts, may persist over long time spans, coupling the evolutionary histories of the lineages involved. Most methods that reconstruct the coevolutionary history of such interactions make the simplifying assumption that parasites have a single host. Many methods also focus on congruence between host and parasite phylogenies, using cospeciation as the null model. However, there is an increasing body of evidence suggesting that the host ranges of parasites are more complex: that host ranges often include more than one host and evolve via gains and losses of hosts rather than through cospeciation alone. Here, we develop a Bayesian approach for inferring coevolutionary history based on a model accommodating these complexities. Specifically, a parasite is assumed to have a host repertoire, which includes both potential hosts and one or more actual hosts. Over time, potential hosts can be added or lost, and potential hosts can develop into actual hosts or vice versa. Thus, host colonization is modeled as a two-step process that may potentially be influenced by host relatedness. We first explore the statistical behavior of our model by simulating evolution of host-parasite interactions under a range of parameter values. We then use our approach, implemented in the program RevBayes, to infer the coevolutionary history between 34 Nymphalini butterfly species and 25 angiosperm families. Our analysis suggests that host relatedness among angiosperm families influences how easily Nymphalini lineages gain new hosts. [Ancestral hosts; coevolution; herbivorous insects; probabilistic modeling.].

Local adaptation of photoperiodic plasticity maintains life cycle variation within latitudes in a butterfly.

The seasonal cycle varies geographically and organisms are under selection to express life cycles that optimally exploit their spatiotemporal habitats. In insects, this often means producing an annual number of generations (voltinism) appropriate to the local season length. Variation in voltinism may arise from variation in environmental factors (e.g., temperature or photoperiod) acting on a single reaction norm shared across populations, but it may also result from local adaptation of reaction norms. However, such local adaptation is poorly explored at short geographic distances, especially within latitudes. Using a combination of common-garden rearing and life cycle modeling, we have investigated the causal factors behind voltinism variation in Swedish populations of the butterfly Pararge aegeria, focusing on a set of populations that lie within a single degree of latitude but nonetheless differ in season length and voltinism. Despite considerable differences in ambient temperature between populations, modeling suggested that the key determinant of local voltinism was in fact interpopulation differences in photoperiodic response. These include differences in the induction thresholds for winter diapause, as well as differences in photoperiodic regulation of larval development, a widespread but poorly studied phenomenon. Our results demonstrate previously neglected ways that photoperiodism may mediate insect phenological responses to temperature, and emphasize the importance of local adaptation in shaping phenological patterns in general, as well as for predicting the responses of populations to changes in climate.

Genetic variation underlying local adaptation of diapause induction along a cline in a butterfly.

Diapause is a life history strategy allowing individuals to arrest development until favourable conditions return, and it is commonly induced by shortened day length that is latitude specific for local populations. Although understanding the evolutionary dynamics of a threshold trait like diapause induction provides insights into the adaptive process and adaptive potential of populations, the genetic mechanism of variation in photoperiodic induction of diapause is not well understood. Here, we investigate genetic variation underlying latitudinal variation in diapause induction and the selection dynamics acting upon it. Using a genomewide scan for divergent regions between two populations of the butterfly Pararge aegeria that differ strongly in their induction thresholds, we identified and investigated the patterns of variation in those regions. We then tested the association of these regions with diapause induction using between-population crosses, finding significant SNP associations in four genes present in two chromosomal regions, one with the gene period, and the other with the genes kinesin, carnitine O-acetyltransferase and timeless. Patterns of allele frequencies in these two regions in population samples along a latitudinal cline suggest strong selection against heterozygotes at two genes within these loci (period, timeless). Evidence for additional loci modifying the diapause decision was found in patterns of allelic change in relation to induction thresholds over the cline, as well as in backcross analyses. Taken together, population-specific adaptations of diapause induction appear to be due to a combination of alleles of larger and smaller effect size, consistent with an exponential distribution of effect sizes involved in local adaption.

Microevolutionary selection dynamics acting on immune genes of the green-veined white butterfly, Pieris napi.

Insects rely on their innate immune system to successfully mediate complex interactions with their microbiota, as well as the microbes present in the environment. Previous work has shown that components of the canonical immune gene repertoire evolve rapidly and have evolutionary characteristics originating from interactions with fast-evolving microorganisms. Although these interactions are likely to vary among populations, there is a poor understanding of the microevolutionary dynamics of immune genes, especially in non-Dipteran insects. Here, we use the full set of canonical insect immune genes to investigate microevolutionary dynamics acting on these genes between and among populations by comparing three allopatric populations of the green-veined white butterfly, Pieris napi (Linné; Lepidoptera, Pieridae). Immune genes showed increased genetic diversity compared to genes from the rest of the genome and various functional categories exhibited different types of signatures of selection, at different evolutionary scales, presenting a complex pattern of selection dynamics. Signatures of balancing selection were identified in 10 genes, and 17 genes appear to be under positive selection. Genes involved with the cellular arm of the immune response as well as the Toll pathway appear to be enriched among our outlier loci, regardless of functional category. This suggests that the targets of selection might focus upon an entire pathway, rather than functional subsets across pathways. Our microevolutionary results are similar to previously observed macroevolutionary patterns from diverse taxa, suggesting that either the immune system is robust to dramatic differences in life history and microbial communities, or that diverse microbes exert similar selection pressures.

Metabolome dynamics of diapause in the butterfly Pieris napi: distinguishing maintenance, termination and post-diapause phases.

Diapause is a deep resting stage facilitating temporal avoidance of unfavourable environmental conditions, and is used by many insects to adapt their life cycle to seasonal variation. Although considerable work has been invested in trying to understand each of the major diapause stages (induction, maintenance and termination), we know very little about the transitions between stages, especially diapause termination. Understanding diapause termination is crucial for modelling and predicting spring emergence and winter physiology of insects, including many pest insects. In order to gain these insights, we investigated metabolome dynamics across diapause development in pupae of the butterfly Pieris napi, which exhibits adaptive latitudinal variation in the length of endogenous diapause that is uniquely well characterized. By employing a time-series experiment, we show that the whole-body metabolome is highly dynamic throughout diapause and differs between pupae kept at a diapause-terminating (low) temperature and those kept at a diapause-maintaining (high) temperature. We show major physiological transitions through diapause, separate temperature-dependent from temperature-independent processes and identify significant patterns of metabolite accumulation and degradation. Together, the data show that although the general diapause phenotype (suppressed metabolism, increased cold tolerance) is established in a temperature-independent fashion, diapause termination is temperature dependent and requires a cold signal. This revealed several metabolites that are only accumulated under diapause-terminating conditions and degraded in a temperature-unrelated fashion during diapause termination. In conclusion, our findings indicate that some metabolites, in addition to functioning as cryoprotectants, for example, are candidates for having regulatory roles as metabolic clocks or time-keepers during diapause.

Idiosyncratic development of sensory structures in brains of diapausing butterfly pupae: implications for information processing.

Diapause is an important escape mechanism from seasonal stress in many insects. A certain minimum amount of time in diapause is generally needed in order for it to terminate. The mechanisms of time-keeping in diapause are poorly understood, but it can be hypothesized that a well-developed neural system is required. However, because neural tissue is metabolically costly to maintain, there might exist conflicting selective pressures on overall brain development during diapause, on the one hand to save energy and on the other hand to provide reliable information processing during diapause. We performed the first ever investigation of neural development during diapause and non-diapause (direct) development in pupae of the butterfly Pieris napi from a population whose diapause duration is known. The brain grew in size similarly in pupae of both pathways up to 3 days after pupation, when development in the diapause brain was arrested. While development in the brain of direct pupae continued steadily after this point, no further development occurred during diapause until temperatures increased far after diapause termination. Interestingly, sensory structures related to vision were remarkably well developed in pupae from both pathways, in contrast with neuropils related to olfaction, which only developed in direct pupae. The results suggest that a well-developed visual system might be important for normal diapause development.

Why stay in a bad relationship? The effect of local host phenology on a generalist butterfly feeding on a low-ranked host.

BACKGROUND: In plant-feeding insects, the evolutionary retention of polyphagy remains puzzling. A better understanding of the relationship between these organisms and changes in the metabolome of their host plants is likely to suggest functional links between them, and may provide insights into how polyphagy is maintained. RESULTS: We investigated the phenological change of Cynoglossum officinale, and how a generalist butterfly species, Vanessa cardui, responded to this change. We used untargeted metabolite profiling to map plant seasonal changes in both primary and secondary metabolites. We compared these data to differences in larval performance on vegetative plants early and late in the season. We also performed two oviposition preference experiments to test females' ability to choose between plant developmental stages (vegetative and reproductive) early and late in the season. We found clear seasonal changes in plant primary and secondary metabolites that correlated with larval performance. The seasonal change in plant metabolome reflected changes in both nutrition and toxicity and resulted in zero survival in the late period. However, large differences among families in larval ability to feed on C. officinale suggest that there is genetic variation for performance on this host. Moreover, females accepted all plants for oviposition, and were not able to discriminate between plant developmental stages, in spite of the observed overall differences in metabolite profile potentially associated with differences in suitability as larval food. CONCLUSIONS: In V. cardui, migratory behavior, and thus larval feeding times, are not synchronized with plant phenology at the reproductive site. This lack of synchronization, coupled with the observed lack of discriminatory oviposition, obviously has potential fitness costs. However, this "opportunistic" behavior may as well function as a source of potential host plant evolution, promoting for example the acceptance of new plants.

Evolutionary history of host use, rather than plant phylogeny, determines gene expression in a generalist butterfly.

BACKGROUND: Although most insect species are specialized on one or few groups of plants, there are phytophagous insects that seem to use virtually any kind of plant as food. Understanding the nature of this ability to feed on a wide repertoire of plants is crucial for the control of pest species and for the elucidation of the macroevolutionary mechanisms of speciation and diversification of insect herbivores. Here we studied Vanessa cardui, the species with the widest diet breadth among butterflies and a potential insect pest, by comparing tissue-specific transcriptomes from caterpillars that were reared on different host plants. We tested whether the similarities of gene-expression response reflect the evolutionary history of adaptation to these plants in the Vanessa and related genera, against the null hypothesis of transcriptional profiles reflecting plant phylogenetic relatedness. RESULT: Using both unsupervised and supervised methods of data analysis, we found that the tissue-specific patterns of caterpillar gene expression are better explained by the evolutionary history of adaptation of the insects to the plants than by plant phylogeny. CONCLUSION: Our findings suggest that V. cardui may use two sets of expressed genes to achieve polyphagy, one associated with the ancestral capability to consume Rosids and Asterids, and another allowing the caterpillar to incorporate a wide range of novel host-plants.

Slowed aging during reproductive dormancy is reflected in genome-wide transcriptome changes in Drosophila melanogaster.

BACKGROUND: In models extensively used in studies of aging and extended lifespan, such as C. elegans and Drosophila, adult senescence is regulated by gene networks that are likely to be similar to ones that underlie lifespan extension during dormancy. These include the evolutionarily conserved insulin/IGF, TOR and germ line-signaling pathways. Dormancy, also known as dauer stage in the larval worm or adult diapause in the fly, is triggered by adverse environmental conditions, and results in drastically extended lifespan with negligible senescence. It is furthermore characterized by increased stress resistance and somatic maintenance, developmental arrest and reallocated energy resources. In the fly Drosophila melanogaster adult reproductive diapause is additionally manifested in arrested ovary development, improved immune defense and altered metabolism. However, the molecular mechanisms behind this adaptive lifespan extension are not well understood. RESULTS: A genome wide analysis of transcript changes in diapausing D. melanogaster revealed a differential regulation of more than 4600 genes. Gene ontology (GO) and KEGG pathway analysis reveal that many of these genes are part of signaling pathways that regulate metabolism, stress responses, detoxification, immunity, protein synthesis and processes during aging. More specifically, gene readouts and detailed mapping of the pathways indicate downregulation of insulin-IGF (IIS), target of rapamycin (TOR) and MAP kinase signaling, whereas Toll-dependent immune signaling, Jun-N-terminal kinase (JNK) and Janus kinase/signal transducer and activator of transcription (JAK/STAT) pathways are upregulated during diapause. Furthermore, we detected transcriptional regulation of a large number of genes specifically associated with aging and longevity. CONCLUSIONS: We find that many affected genes and signal pathways are shared between dormancy, aging and lifespan extension, including IIS, TOR, JAK/STAT and JNK. A substantial fraction of the genes affected by diapause have also been found to alter their expression in response to starvation and cold exposure in D. melanogaster, and the pathways overlap those reported in GO analysis of other invertebrates in dormancy or even hibernating mammals. Our study, thus, shows that D. melanogaster is a genetically tractable model for dormancy in other organisms and effects of dormancy on aging and lifespan.

Energy and lipid metabolism during direct and diapause development in a pierid butterfly.

Diapause is a fundamental component of the life cycle in the majority of insects living in environments characterized by strong seasonality. The present study addresses poorly understood associations and trade-offs between endogenous diapause duration, thermal sensitivity of development, energetic cost of development and cold tolerance. Diapause intensity, metabolic rate trajectories and lipid profiles of directly developing and diapausing animals were studied using pupae and adults of Pieris napi butterflies from a population in which endogenous diapause has been well studied. Endogenous diapause was terminated after 3 months and termination required chilling. Metabolic and post-diapause development rates increased with diapause duration, while the metabolic cost of post-diapause development decreased, indicating that once diapause is terminated, development proceeds at a low rate even at low temperature. Diapausing pupae had larger lipid stores than the directly developing pupae, and lipids constituted the primary energy source during diapause. However, during diapause, lipid stores did not decrease. Thus, despite lipid catabolism meeting the low energy costs of the diapausing pupae, primary lipid store utilization did not occur until the onset of growth and metamorphosis in spring. In line with this finding, diapausing pupae contained low amounts of mitochondria-derived cardiolipins, which suggests a low capacity for fatty acid ß-oxidation. While ontogenic development had a large effect on lipid and fatty acid profiles, only small changes in these were seen during diapause. The data therefore indicate that the diapause lipidomic phenotype is developed early, when pupae are still at high temperature, and retained until post-diapause development.

Specialist and generalist oviposition strategies in butterflies: maternal care or precocious young?

Herbivorous insects specialized on a narrow set of plants are believed to be better adapted to their specific hosts. This hypothesis is supported by observations of herbivorous insect species with a broader diet breadth which seemingly pay a cost through decreased oviposition accuracy. Despite many studies investigating female oviposition behavior, there is a lack of knowledge on how larvae cope behaviorally with their mothers' egg-laying strategies. We have examined a unique system of five nymphalid butterfly species with different host plant ranges that all feed on the same host plant. The study of this system allowed us to compare at the species level how oviposition preference is related to neonate larval responses in several disadvantageous situations. We found a general co-adaptation between female and larval abilities, where species with more discriminating females had larvae that were less able to deal with a suboptimal initial feeding site. Conversely, relatively indiscriminate females had more precocious larvae with better abilities to cope with suboptimal sites. Despite similarities between the tested species with similar host ranges, there were also striking differences. Generalist and specialist species can be found side by side in many clades, with each clade having a specific evolutionary history. Such clade-specific, phylogenetically determined preconditions apparently have affected how precisely a broad or narrow diet breadth can be realized.

Molecular phylogeny of Lymantriinae (Lepidoptera, Noctuoidea, Erebidae) inferred from eight gene regions.

To understand the evolutionary history of Lymantriinae and test the present higher-level classification, we performed the first broad-scale molecular phylogenetic analysis of the subfamily, based on 154 exemplars representing all recognized tribes and drawn from all major biogeographical regions. We used two mitochondrial genes (cytochrome c oxidase subunit I and 16S ribosomal RNA) and six nuclear genes (elongation factor-1α, carbamoylphosphate synthase domain protein, ribosomal protein S5, cytosolic malate dehydrogenase, glyceraldehyde-3-phosphate dehydrogenase and wingless). Data matrices (in total 5424 bp) were analysed by parsimony and model-based evolutionary methods (maximum likelihood and Bayesian inference). Based on the results of the analyses, we present a new phylogenetic classification for Lymantriinae composed of seven well-supported tribes, two of which are proposed here as new: Arctornithini, Leucomini, Lymantriini, Orgyiini, Nygmiini, Daplasini trib. nov. and Locharnini trib. nov. We discuss the internal structure of each of these tribes and address some of the more complex problems with the genus-level classification, particularly within Orgyiini and Nygmiini.

Mechanisms of macroevolution: polyphagous plasticity in butterfly larvae revealed by RNA-Seq.

Transcriptome studies of insect herbivory are still rare, yet studies in model systems have uncovered patterns of transcript regulation that appear to provide insights into how insect herbivores attain polyphagy, such as a general increase in expression breadth and regulation of ribosomal, digestion- and detoxification-related genes. We investigated the potential generality of these emerging patterns, in the Swedish comma, Polygonia c-album, which is a polyphagous, widely-distributed butterfly. Urtica dioica and Ribes uva-crispa are hosts of P. c-album, but Ribes represents a recent evolutionary shift onto a very divergent host. Utilizing the assembled transcriptome for read mapping, we assessed gene expression finding that caterpillar life-history (i.e. 2nd vs. 4th-instar regulation) had a limited influence on gene expression plasticity. In contrast, differential expression in response to host-plant identified genes encoding serine-type endopeptidases, membrane-associated proteins and transporters. Differential regulation of genes involved in nucleic acid binding was also observed suggesting that polyphagy involves large scale transcriptional changes. Additionally, transcripts coding for structural constituents of the cuticle were differentially expressed in caterpillars in response to their diet indicating that the insect cuticle may be a target for plant defence. Our results state that emerging patterns of transcript regulation from model species appear relevant in species when placed in an evolutionary context.

Larval transcriptomes reflect the evolutionary history of plant-insect associations.

In this study, we investigated whether patterns of gene expression in larvae feeding on different plants can explain important aspects of the evolution of insect-plant associations, such as phylogenetic conservatism of host use and re-colonization of ancestral hosts that have been lost from the host repertoire. To this end, we performed a phylogenetically informed study comparing the transcriptomes of 4 nymphalid butterfly species in Polygonia and the closely related genus Nymphalis. Larvae were reared on Urtica dioica, Salix spp., and Ribes spp. Plant-specific gene expression was found to be similar across butterfly species, even in the case of host plants that are no longer used by two of the butterfly species. These results suggest that plant-specific transcriptomes can be robust over evolutionary time. We propose that adaptations to particular larval food plants can profitably be understood as an evolved set of modules of co-expressed genes, promoting conservatism in host use and facilitating re-colonization. Moreover, we speculate that the degree of overlap between plant-specific transcriptomes may correlate with the strength of trade-offs between plants as resources and hence to the probability of colonizing hosts and complete host shifts.

Nonvolatile chemical cues affect host-plant ranking by gravid Polygonia c-album females.

In a multiple-choice test, the preference of egg-laying Polygonia c-album (comma butterfly) females was studied for oviposition on plants bearing surrogate leaves treated with crude methanol extracts obtained from leaves of seven host-plant species: Humulus lupulus, Urtica dioica, Ulmus glabra, Salix caprea, Ribes nigrum, Corylus avellana, and Betula pubescens. The ranking order of surrogate leaves treated with host-plant extracts corresponded well to that reported on natural foliage, except R. nigrum. Thus, host-plant choice in P. c-album seems to be highly dependent on chemical cues. Moreover, after two subsequent fractionations using reversed-phase chromatography the nonvolatile chemical cues residing in the most polar water-soluble fractions evidently provided sufficient information for egg-laying females to discriminate and rank between the samples of more and less preferred plants, since the ranking in these assays was similar to that for natural foliage or whole methanol extracts, while the physical traits of the surrogate leaves remained uniform.

A region of the sex chromosome associated with population differences in diapause induction contains highly divergent alleles at clock genes.

Developmental plasticity describes the capacity of individuals with the same genotype to induce permanent change in a phenotype depending on a specific external input. One well-studied example of adaptive developmental plasticity is the induction of facultative diapause in insects. Studies investigating the inheritance of diapause induction have suggested diverse genetic origins. However, only few studies have performed genome-wide scans to identify genes affecting the induction decision. Here we compare two populations of the butterfly Pieris napi that differ in the propensity to enter diapause, and despite showing a low genome-wide divergence, we identify a few genomic regions that show high divergence between populations. We then identified a single genomic region associated with diapause induction by genotyping diapausing and directly developing siblings from backcrosses of these populations. This region is located on the Z chromosome and contained three circadian clock genes, cycle, clock, and period. Additionally, period harbored the largest number of SNPs showing complete fixation between populations. We conclude that the heritable basis of between-population variation in the plasticity that determines diapause induction resides on the Z chromosome, with the period gene being the prime candidate for the genetic basis of adaptive plasticity.