This Title All WIREs
How to cite this WIREs title:
WIREs Dev Biol
Impact Factor: 3.754

Mimicry in butterflies: co‐option and a bag of magnificent developmental genetic tricks

Full article on Wiley Online Library:   HTML PDF

Can't access this content? Tell your librarian.

Butterfly wing patterns are key adaptations that are controlled by remarkable developmental and genetic mechanisms that facilitate rapid evolutionary change. With swift advancements in the fields of genomics and genetic manipulations, identifying the regulators of wing development and mimetic wing patterns has become feasible even in nonmodel organisms such as butterflies. Recent mapping and gene expression studies have identified single switch loci of major effects such as transcription factors and supergenes as the main drivers of adaptive evolution of mimetic and polymorphic butterfly wing patterns. We highlight several of these examples, with emphasis on doublesex, optix, WntA and other dynamic, yet essential, master regulators that control critical color variation and sex‐specific traits. Co‐option emerges as a predominant theme, where typically embryonic and other early‐stage developmental genes and networks have been rewired to regulate polymorphic and sex‐limited mimetic wing patterns in iconic butterfly adaptations. Drawing comparisons from our knowledge of wing development in Drosophila, we illustrate the functional space of genes that have been recruited to regulate butterfly wing patterns. We also propose a developmental pathway that potentially results in dorsoventral mismatch in butterfly wing patterns. Such dorsoventrally mismatched color patterns modulate signal components of butterfly wings that are used in intra‐ and inter‐specific communication. Recent advances—fuelled by RNAi‐mediated knockdowns and CRISPR/Cas9‐based genomic edits—in the developmental genetics of butterfly wing patterns, and the underlying biological diversity and complexity of wing coloration, are pushing butterflies as an emerging model system in ecological genetics and evolutionary developmental biology. WIREs Dev Biol 2018, 7:e291. doi: 10.1002/wdev.291 This article is categorized under: Gene Expression and Transcriptional Hierarchies > Regulatory Mechanisms Comparative Development and Evolution > Regulation of Organ Diversity Comparative Development and Evolution > Evolutionary Novelties
Sex‐limited mimicry and polymorphism in butterflies. The Danaus‐Hypolimnas mimicry ring illustrates a rare example where both the Batesian model and the mimic are polymorphic, female forms of the Batesian mimic H. missipus mimicking D. chrysippus in a form‐specific manner. On the other hand, in P. dardanus and P. polytes, multiple female forms mimic distinct species of models. In the last two species, the male and male‐like female forms are nonmimetic, which in P. polytes also represents the ancestral phenotype. These species represent some of the best examples of the degree to which natural selection—through predation—may drive nearly perfect and polymorphic wing pattern resemblance between Batesian models and mimics. Mimicry is controlled by co‐option of major developmental genes in both the Papilio, whereas the molecular genetic basis in Hypolimnas is still unknown. See Figure in Box for the explanation of arrows.
[ Normal View | Magnified View ]
Dorsoventral mismatch of wing color patterns is widespread in butterflies, irrespective of mimicry. (a) Dorsoventral mismatch is illustrated by distinct dorsal (on the left) and ventral (on the right) wing coloration in A. lyncida, whereas the two wing surfaces have matching patterns in A. nephele. Such dorsoventral mismatch has a special significance in mimetic butterflies where selection for sex‐limited mimicry and efficacy of the conflicting signal components of predator avoidance and sexual attraction, have led to sex‐ and surface‐specific wing patterns. For example, while the nonmimetic Elymnias singhala has somewhat similar wing color patterns on both wing surfaces, in E. caudata the ventral wing surface has remained inconspicuous to aid in crypsis, whereas the dorsal surface has diverged into novel male and female patterns, the female producing a superb mimetic resemblance to its toxic model D. genutia. (b) Determination of the dorsoventral boundary in insect wings. The gene network is known in Drosophila wing imaginal discs, based on which we present a developmental genetic hypothesis for butterflies. Asterisks denote genes that are expressed at the corresponding sites in both Drosophila and butterfly wing discs (see Table S1).
[ Normal View | Magnified View ]
Mimetic wing patterns in butterflies are controlled by co‐option of major developmental genes. Co‐opted genes and their distinct functions in embryonic development, metamorphosis, wing development, and color patterning in Drosophila and butterflies are listed. (a) Gene sets 1–5 have shared functions between Drosophila and butterflies, whereas 6–22 have known and possible co‐options in butterflies. (b) Wing patterning genes in butterflies may serve multiple functions in response to selection for mimicry (natural selection) and mate choice (sexual selection). The assigned functions are based on published literature, although it is possible that additional functions for these genes will be discovered in the future. (c) Functional space of genes involved in wing development and color patterning in Drosophila and butterflies also illustrates co‐option of genes at distinct developmental stages. Refs are cited in Table S1.
[ Normal View | Magnified View ]
Polymorphism gone wild in mimetic butterfly species. H. bolina and P. memnon exhibit multiple female forms, many of which are neither male‐like and ancestral nor mimetic, and their wing pattern elements appear to recombine frequently to produce a wide array of color pattern forms. In case of H. bolina, the wing pattern diversity may have been produced by relaxed predation pressures on islands, that is, under neutral processes—as many island populations are prominently variable with novel, nonmimetic wing patterns. Males, on the other hand, show limited diversity of wing patterns within and across populations. In P. memnon, wing pattern elements and tails appear to switch occasionally between female forms, creating almost all possible permutations. The resultant morphological diversity is probably either selectively neutral or mildly deleterious. Shown here is only a selection of form diversity in these species. Genetics and development of these wing patterns are largely unknown, and likely differ from the genetic architectures so far known in other polymorphic, mimetic butterflies such as P. polytes and Heliconius.
[ Normal View | Magnified View ]
The striking wing patterns of the Neotropical Heliconius butterflies are products of their aposematism, taxonomic diversity and loose reproductive isolation between species. Wing patterning alleles must have initially evolved under genetic drift and selection for aposematism as well as for Müllerian mimicry, and have subsequently been widely introgressed across species. H. numata has a supergene architecture that controls polymorphic mimicry. H. erato and H. melpomene show parallel diversification in wing patterning for Müllerian mimicry (top three forms), but also show hybrid and intermediate forms. Gene expression of optix, which controls red wing areas in Heliconius is shown on the right (antibody staining: image courtesy of Arnaud Martin).
[ Normal View | Magnified View ]
The CDS regions of doublesex (dsx) show hundreds of SNPs in lepidopteran genomes, some of which are nonsynonymous, indicating that dsx may be a hotspot of adaptive molecular evolution in some regions but highly conserved in others. (a) Exon usage in the sex‐specific isoforms and other variants are shown along with the SNPs (vertical gray bars) unique to each sequenced genome. It is yet unknown whether nonuniversal exon usage and the species‐specific SNPs across different exons are adaptive, and how they might relate to sex‐limited mimicry and other adaptive traits in butterflies. (b) Species pairwise comparison of the total number of SNPs of dsx, which shows some correlation between phylogenetic relatedness and the number of genetic differences. (c–d) dsx splice variants (a and c) and their tissue‐ (c) and form‐specific (d) expression that controls female‐limited mimetic polymorphism in P. polytes. Data are from GenBank and LepBase (panels a–b), and Kunte et al. (panels c–d). The mimetic female form is produced by upregulation of dsx (d).
[ Normal View | Magnified View ]

Browse by Topic

Gene Expression and Transcriptional Hierarchies > Regulatory Mechanisms
Comparative Development and Evolution > Regulation of Organ Diversity
Comparative Development and Evolution > Evolutionary Novelties