Radiation of the polymorphic Little Devil poison frog (Oophaga sylvatica) in Ecuador

Expression patterns of melanin-related genes are linked to crypsis and conspicuousness in a pumpkin toadlet

Colour signals play pivotal roles in different communication systems, and the evolution of these characters has been associated with behavioural ecology, integumentary production processes and perceptual mechanisms of the species involved. Here, we present the first insight into the molecular and histological basis of skin colour polymorphism within a miniaturized species of pumpkin toadlet, potentially representing the lowest size threshold for colour polytypism in tetrapods. Brachycephalus actaeus exhibits a coloration ranging from cryptic green to conspicuous orange skin, and our findings suggest that colour morphs differ in their capability to be detected by potential predators. We also found that the distribution and abundance of chromatophores are variable in the different colour morphs. The expression pattern of coloration related genes was predominantly associated with melanin synthesis (including dct, edn1, mlana, oca2, pmel, slc24a5, tyrp1 and wnt9a). Up-regulation of melanin genes in grey, green and brown skin was associated with higher melanophore abundance than in orange skin, where xanthophores predominate. Our findings provide a significant foundation for comparing and understanding the diverse pathways that contribute to the evolution of pigment production in the skin of amphibians.

Deep divergences among inconspicuously colored clades of Epipedobates poison frogs

Poison frogs (Dendrobatidae) are famous for their aposematic species, having a combination of diverse color patterns and defensive skin toxins, yet most species in this family are inconspicuously colored and considered non-aposematic. Epipedobates is among the youngest genus-level clades of Dendrobatidae that includes both aposematic and inconspicuous species. Using Sanger-sequenced mitochondrial and nuclear markers, we demonstrate deep genetic divergences among inconspicuous species of Epipedobates but relatively shallow genetic divergences among conspicuous species. Our phylogenetic analysis includes broad geographic sampling of the inconspicuous lineages typically identified as E. boulengeri and E. espinosai, which reveals two putative new species, one in west-central Colombia (E. sp. 1) and the other in north-central Ecuador (E. aff. espinosai). We conclude that E. darwinwallacei is a junior subjective synonym of E. espinosai. We also clarify the geographic distributions of inconspicuous Epipedobates species including the widespread E. boulengeri. We provide a qualitative assessment of the phenotypic diversity in each nominal species, with a focus on the color and pattern of inconspicuous species. We conclude that Epipedobates contains eight known valid species, six of which are inconspicuous. A relaxed molecular clock analysis suggests that the most recent common ancestor of Epipedobates is ∼11.1 million years old, which nearly doubles previous estimates. Last, genetic information points to a center of species diversity in the Chocó at the southwestern border of Colombia with Ecuador. A Spanish translation of this text is available in the supplementary materials.

Continuous Variation in an Aposematic Pattern Affects Background Contrast, but Is Not Associated With Differences in Microhabitat Use

Variation in aposematic signals was once predicted to be rare, yet in recent years it has become increasingly well documented. Despite increases in the frequency with which polytypism and polymorphism have been suggested to occur, population-wide variance is rarely quantified. We comprehensively sampled a subpopulation of the poison frog Oophaga sylvatica, a species which is polytypic across its distribution and also shows considerable within-population polymorphism. On one hand, color pattern polymorphism could be the result of multifarious selection acting to balance different signaling functions and leading to the evolution of discrete sub-morphs which occupy different fitness peaks. Alternatively, variance could simply be due to relaxed selection, where variation would be predicted to be continuous. We used visual modeling of conspecific and heterospecific observers to quantify the extent of within population phenotypic variation and assess whether this variation produced distinct signals. We found that, despite considerable color pattern variation, variance could not be partitioned into distinct groups, but rather all viewers would be likely to perceive variation as continuous. Similarly, we found no evidence that frog color pattern contrast was either enhanced or diminished in the frogs’ chosen microhabitats compared to alternative patches in which conspecifics were observed. Within population phenotypic variance therefore does not seem to be indicative of strong selection toward multiple signaling strategies, but rather pattern divergence has likely arisen due to weak purifying selection, or neutral processes, on a signal that is highly salient to both conspecifics and predators.

Genetic isolation by distance underlies colour pattern divergence in red-eyed treefrogs (Agalychnis callidryas)

Investigating the spatial distribution of genetic and phenotypic variation can provide insights into the evolutionary processes that shape diversity in natural systems. We characterized patterns of genetic and phenotypic diversity to learn about drivers of colour-pattern diversification in red-eyed treefrogs (Agalychnis callidryas) in Costa Rica. Along the Pacific coast, red-eyed treefrogs have conspicuous leg colour patterning that transitions from orange in the north to purple in the south. We measured phenotypic variation of frogs, with increased sampling at sites where the orange-to-purple transition occurs. At the transition zone, we discovered the co-occurrence of multiple colour-pattern morphs. To explore possible causes of this variation, we generated a single nucleotide polymorphism data set to analyse population genetic structure, measure genetic diversity and infer the processes that mediate genotype-phenotype dynamics. We investigated how patterns of genetic relatedness correspond to individual measures of colour pattern along the coast, including testing for the role of hybridization in geographic regions where orange and purple phenotypic groups co-occur. We found no evidence that colour-pattern polymorphism in the transition zone arose through recent hybridization. Instead, a strong pattern of genetic isolation by distance indicates that colour-pattern variation was either retained through other processes such as ancestral colour polymorphisms or ancient secondary contact, or else it was generated by novel mutations. We found that phenotype changes along the Pacific coast more than would be expected based on genetic divergence and geographic distance alone. Combined, our results suggest the possibility of selective pressures acting on colour pattern at a small geographic scale.

Contrasting environmental drivers of genetic and phenotypic divergence in an Andean poison frog (Epipedobates anthonyi)

Phenotypic and genetic divergence are shaped by the homogenizing effects of gene flow and the differentiating processes of genetic drift and local adaptation. Herein, we examined the mechanisms that underlie phenotypic (size and color) and genetic divergence in 35 populations (535 individuals) of the poison frog Epipedobates anthonyi along four elevational gradients (0-1800 m asl) in the Ecuadorian Andes. We found phenotypic divergence in size and color despite relatively low genetic divergence at neutral microsatellite loci. Genetic and phenotypic divergence were both explained by landscape resistance between sites (isolation-by-resistance, IBR), likely due to a cold and dry mountain ridge between the northern and southern elevational transects that limits dispersal and separates two color morphs. Moreover, environmental differences among sites also explained genetic and phenotypic divergence, suggesting isolation-by-environment (IBE). When northern and southern transects were analyzed separately, genetic divergence was predicted either by distance (isolation-by-distance, IBD; northern) or environmental resistance between sites (IBR; southern). In contrast, phenotypic divergence was primarily explained by environmental differences among sites, supporting the IBE hypothesis. These results indicate that although distance and geographic barriers are important drivers of population divergence, environmental variation has a two-fold effect on population divergence. On the one hand, landscape resistance between sites reduces gene flow (IBR), while on the other hand, environmental differences among sites exert divergent selective pressures on phenotypic traits (IBE). Our work highlights the importance of studying both genetic and phenotypic divergence to better understand the processes of population divergence and speciation along ecological gradients.

Hybridization of Two Species of Japanese Toads, Bufo torrenticola and Bufo japonicus formosus, in the Central Part of Japan

Interspecific hybridization between two species of Japanese toads, Bufo torrenticola (Bt) and Bufo japonicus formosus (Bjf), was not thought to have occurred due to the differences in their spawning habitats, although their hybridization was reported by previous research in a limited area. Such hybridization could have pivotal effects, such as genetic admixture, on Bt, which has a limited distribution and habitat. In this study, to clarify the details of hybridization, a total of 192 individuals were collected from multiple parapatric regions and an only Bjf allopatric region and analyzed based on morphology, mtDNA sequences, and nuclear DNA microsatellite loci. Each individual was identified based on morphological type and mtDNA lineage, and two mismatched combinations, Bt or an intermediate morphological type and the Bjf mtDNA lineage, were detected. In particular, the combination of Bt type and Bjf lineage was dominant in parapatric regions, and it was considered that asymmetric hybridization in which Bjf became a matriline had occurred. In addition, a Bayesian clustering method based on the microsatellite loci was conducted to detect a genetic admixture of the two toad species. Individuals representing a combination of the Bt type and Bjf lineage in the parapatric regions were largely assigned to the Bt cluster or equally to both clusters. These results indicate that backcross with Bt has been repeated, and therefore the traits of Bt might be maintained in this study.

Balanced polymorphisms and their divergence in a Heliconius butterfly

The evolution of mimicry in similarly defended prey is well described by the Müllerian mimicry theory, which predicts the convergence of warning patterns in order to gain the most protection from predators. However, despite this prediction, we can find great diversity of color patterns among Müllerian mimics such as Heliconius butterflies in the neotropics. Furthermore, some species have evolved the ability to maintain multiple distinct warning patterns in single populations, a phenomenon known as polymorphic mimicry. The adaptive benefit of these polymorphisms is questionable since variation from the most common warning patterns is expected to be disadvantageous as novel signals are punished by predators naive to them. In this study, we use artificial butterfly models throughout Central and South America to characterize the selective pressures maintaining polymorphic mimicry in Heliconius doris. Our results highlight the complexity of positive frequency-dependent selection, the principal selective pressure driving convergence among Müllerian mimics, and its impacts on interspecific variation of mimetic warning coloration. We further show how this selection regime can both limit and facilitate the diversification of mimetic traits.

Rapid toxin sequestration modifies poison frog physiology

Poison frogs sequester chemical defenses from their diet of leaf litter arthropods for defense against predation. Little is known about the physiological adaptations that confer this unusual bioaccumulation ability. We conducted an alkaloid-feeding experiment with the Diablito poison frog (Oophaga sylvatica) to determine how quickly alkaloids are accumulated and how toxins modify frog physiology using quantitative proteomics. Diablito frogs rapidly accumulated the alkaloid decahydroquinoline within 4 days, and dietary alkaloid exposure altered protein abundance in the intestines, liver and skin. Many proteins that increased in abundance with decahydroquinoline accumulation are plasma glycoproteins, including the complement system and the toxin-binding protein saxiphilin. Other protein classes that change in abundance with decahydroquinoline accumulation are membrane proteins involved in small molecule transport and metabolism. Overall, this work shows that poison frogs can rapidly accumulate alkaloids, which alter carrier protein abundance, initiate an immune response, and alter small molecule transport and metabolism dynamics across tissues.

Complex patterns of differentiation and gene flow underly the divergence of aposematic phenotypes in Oophaga poison frogs

Hybridization and introgression can have complex consequences for both species evolution and conservation. Here, we investigated the origin and characteristics of a putative hybrid zone between two South American poison dart frog species, Oophaga anchicayensis and the critically endangered Oophaga lehmanni, which are heavily sought after on the illegal pet market. Using a combination of phenotypic (49 traits) and genomic (ddRADseq) data, we found that the putative hybrids are morphologically distinct from their parental species and confirmed genomic signatures of admixture in these populations. Several lines of evidence (hybrid indices, interspecific hybrid heterozygosity, genomic clines, comparisons with simulated hybrids and demographic modelling) support the conclusion that these populations are not comprised of early-generation hybrids and thus, they probably did not arise as a result of illegal translocations associated with wildlife trafficking. Instead, they probably represent an independent lineage which has persisted through isolation and has only relatively recently re-established gene flow with both parental species. Furthermore, we detected signals of differential introgression from parental species into these hybrid populations which suggest relaxed stabilizing selection on these aposematic colour morphs, potentially via context-dependent female choice. These populations thus provide a fascinating window into the role of hybridization, isolation and female choice in the diversification of South American poison dart frogs. In addition, our results underline the importance of landscape conservation measures to protect, not only known localities of nominal species, but also the phenotypic and genomic variation harbored by admixed lineages which represent crucial repositories for the impressive diversity in this system.

Weak warning signals can persist in the absence of gene flow

Aposematic organisms couple conspicuous warning signals with a secondary defense to deter predators from attacking. Novel signals of aposematic prey are expected to be selected against due to positive frequency-dependent selection. How, then, can novel phenotypes persist after they arise, and why do so many aposematic species exhibit intrapopulation signal variability? Using a polytypic poison frog (Dendrobates tinctorius), we explored the forces of selection on variable aposematic signals using 2 phenotypically distinct (white, yellow) populations. Contrary to expectations, local phenotype was not always better protected compared to novel phenotypes in either population; in the white population, the novel phenotype evoked greater avoidance in natural predators. Despite having a lower quantity of alkaloids, the skin extracts from yellow frogs provoked higher aversive reactions by birds than white frogs in the laboratory, although both populations differed from controls. Similarly, predators learned to avoid the yellow signal faster than the white signal, and generalized their learned avoidance of yellow but not white. We propose that signals that are easily learned and broadly generalized can protect rare, novel signals, and weak warning signals (i.e., signals with poor efficacy and/or poor defense) can persist when gene flow among populations, as in this case, is limited. This provides a mechanism for the persistence of intrapopulation aposematic variation, a likely precursor to polytypism and driver of speciation.

Phenotypic and genetic diversity in aposematic Malagasy poison frogs (genus Mantella)

Intraspecific color variation has long fascinated evolutionary biologists. In species with bright warning coloration, phenotypic diversity is particularly compelling because many factors, including natural and sexual selection, contribute to intraspecific variation. To better understand the causes of dramatic phenotypic variation in Malagasy poison frogs, we quantified genetic structure and color and pattern variation across three closely related species, Mantella aurantiaca, Mantella crocea, and Mantella milotympanum. Although our restriction site-associated DNA (RAD) sequencing approach identified clear genetic clusters, they do not align with current species designations, which has important conservation implications for these imperiled frogs. Moreover, our results suggest that levels of intraspecific color variation within this group have been overestimated, while species diversity has been underestimated. Within major genetic clusters, we observed distinct patterns of variation including: populations that are phenotypically similar yet genetically distinct, populations where phenotypic and genetic breaks coincide, and populations that are genetically similar but have high levels of within-population phenotypic variation. We also detected admixture between two of the major genetic clusters. Our study suggests that several mechanisms-including hybridization, selection, and drift-are contributing to phenotypic diversity. Ultimately, our work underscores the need for a reevaluation of how polymorphic and polytypic populations and species are classified, especially in aposematic organisms.

Molecular physiology of chemical defenses in a poison frog

Poison frogs sequester small molecule lipophilic alkaloids from their diet of leaf litter arthropods for use as chemical defenses against predation. Although the dietary acquisition of chemical defenses in poison frogs is well-documented, the physiological mechanisms of alkaloid sequestration has not been investigated. Here, we used RNA sequencing and proteomics to determine how alkaloids impact mRNA or protein abundance in the Little Devil Frog (Oophaga sylvatica) and compared wild caught chemically defended frogs to laboratory frogs raised on an alkaloid-free diet. To understand how poison frogs move alkaloids from their diet to their skin granular glands, we focused on measuring gene expression in the intestines, skin, and liver. Across these tissues, we found many differentially expressed transcripts involved in small molecule transport and metabolism, as well as sodium channels and other ion pumps. We then used proteomic approaches to quantify plasma proteins, where we found several protein abundance differences between wild and laboratory frogs, including the amphibian neurotoxin binding protein saxiphilin. Finally, because many blood proteins are synthesized in the liver, we used thermal proteome profiling as an untargeted screen for soluble proteins that bind the alkaloid decahydroquinoline. Using this approach, we identified several candidate proteins that interact with this alkaloid, including saxiphilin. These transcript and protein abundance patterns suggest the presence of alkaloids influences frog physiology and that small molecule transport proteins may be involved in toxin bioaccumulation in dendrobatid poison frogs.

Radiation of the polymorphic Little Devil poison frog (Oophaga sylvatica) in Ecuador

Some South American poison frogs (Dendrobatidae) are chemically defended and use bright aposematic colors to warn potential predators of their unpalatability. Aposematic signals are often frequency-dependent where individuals deviating from a local model are at a higher risk of predation. However, extreme diversity in the aposematic signal has been documented in poison frogs, especially in Oophaga. Here, we explore the phylogeographic pattern among color-divergent populations of the Little Devil poison frog Oophaga sylvatica by analyzing population structure and genetic differentiation to evaluate which processes could account for color diversity within and among populations. With a combination of PCR amplicons (three mitochondrial and three nuclear markers) and genome-wide markers from a double-digested RAD (ddRAD) approach, we characterized the phylogenetic and genetic structure of 199 individuals from 13 populations (12 monomorphic and 1 polymorphic) across the O. sylvatica distribution. Individuals segregated into two main lineages by their northern or southern latitudinal distribution. A high level of genetic and phenotypic polymorphism within the northern lineage suggests ongoing gene flow. In contrast, low levels of genetic differentiation were detected among the southern lineage populations and support recent range expansions from populations in the northern lineage. We propose that a combination of climatic gradients and structured landscapes might be promoting gene flow and phylogenetic diversification. Alternatively, we cannot rule out that the observed phenotypic and genomic variations are the result of genetic drift on near or neutral alleles in a small number of genes.