Nicholas M. Justyn

In a hybrid zone between two tropical lekking birds, yellow male plumage of one species has introgressed asymmetrically replacing white plumage of another via sexual selection. Here, we present a detailed analysis of the plumage trait to uncover its physical and genetic bases and trace its evolutionary history. We determine that the carotenoid lutein underlies the yellow phenotype and describe microstructural feather features likely to enhance color appearance. These same features reduce predicted water shedding capacity of feathers, a potential liability in the tropics. Through genome-scale DNA sequencing of hybrids and each species in the genus, we identify BCO2 as the major gene responsible for the color polymorphism. The BCO2 gene tree and genome-wide allele frequency patterns suggest that carotenoid-pigmented collars initially arose in a third species and reached the hybrid zone through historical gene flow. Complex interplay between sexual selection and hybridization has thus shaped phenotypes of these species, where conspicuous sexual traits are key to male reproductive success.

Carotenoid pigmentation produces the yellow and red coloration of birds and other vertebrates, but our understanding of the genetic architecture of carotenoid ornamentation is largely limited to studies of novel color variants observed in captively bred populations. The complexity of carotenoid-based color evolution in nature remains poorly characterized. Here, we examine the long-tailed finch Poephila acuticauda , an Australian songbird with two hybridizing subspecies that differ in bill coloration: yellow in western subspecies acuticauda and red in eastern subspecies hecki . We characterize the carotenoid composition of each subspecies and find that yellow bills can be explained by the loss of C(4)-oxidation, thus blocking yellow dietary pigments from being metabolized to red. Combining linked-read genomic sequencing and reflectance spectrophotometry measurements of bill color collected from wild-sampled finches and laboratory crosses, we identify four loci that together explain 53% of variance in this trait. The two loci of largest effect contain the genes CYP2J19 , an essential enzyme for the ketolation via C(4)-oxidation of dietary carotenoids, and TTC39B , an enhancer of ketocarotenoid production. Evolutionary genealogy reconstruction indicates that the red-billed phenotype is ancestral and yellow alleles at both CYP2J19 and TTC39B arose and fixed in acuticauda approximately 100 kya. Yellow alleles then introgressed into hecki less than 5 kya. Across all four loci, acuticauda derived variants show evidence of selective sweeps, implying that yellow bill coloration has been favored by natural selection. Our study suggests that the frequent adaptive evolutionary transitions between red and yellow ornamentation in nature can have a simple genetic basis. Significance: We studied variation in carotenoid ornamentation of an Australian songbird with two hybridizing subspecies that differ in bill color: one yellow and the other red. We identified a single metabolic process, C(4)-oxidation, underlying the distinct carotenoid composition of these two bill colors. Genetic association mapping revealed four major effect loci that explained most of the observed variation the trait, including the oxidative ketolation enzyme CYP2J19 and the carotenoid ketolation enhancer gene TTC39B . Evolutionary reconstruction indicates that yellow alleles are derived, ancient (~100 kya), and under positive selection. This has driven their recent (<5 kya) adaptive introgression across the hybrid zone. These findings have important implications for understanding the role of natural selection in phenotypic evolution in natural systems.

Male Painted Buntings (Passerina ciris) display at least 6 distinct plumage colors that encapsulate much of the visible light spectrum, yet the specific mechanisms responsible for generating this diversity of color have not been identified. Here, we show that metabolically derived carotenoids and nanostructures capable of producing structural color were ubiquitous across feather patches. We used digital photography, light microscopy, spectrophotometry, carotenoid extraction, and high-performance liquid chromatography to show that the resulting color of each feather patch depended on the concentration of carotenoids, melanins, and underlying feather nanostructures. For example, we found that the blue-violet head feathers contained low concentrations of ketolated carotenoids, which is not typical of blue-violet structurally colored feathers. Additionally, the red breast and orange belly feathers contained a green tuned structural color visible after carotenoid extraction, which is not typical of feathers that contain ketolated carotenoids. Although, none of these abnormal combinations of carotenoids and structural coloration appeared to significantly impact feather color. Conversely, we found the purple rump, dark green greater coverts, and bright yellow-green mantle feather coloration resulted from the combination of high concentrations of carotenoids and the presence of structural color. For the first time, we identify the combination of red ketolated carotenoids and blue structural color as a mechanism to produce purple feather coloration. Identifying the specific mechanisms that give rise to the diversity of colors within this species will facilitate the study of the—to date—unknown signaling functions of colors produced through the combination of carotenoids and nanostructures in Painted Buntings and other songbirds.

Unlike humans, some animals have evolved a physiological ability to deposit porphyrins, which are pigments produced during heme synthesis in cells, in the skin and associated integument such as hair. Given the inert nature and easiness of collection of hair, animals that present porphyrin-based pigmentation constitute unique models for porphyrin analysis in biological samples. Here we present the development of a simple, rapid, and efficient analytical method for four natural porphyrins (uroporphyrin I, coproporphyrin I, coproporphyrin III and protoporphyrin IX) in the Southern flying squirrel Glaucomys volans, a mammal with hair that fluoresces and that we suspected has porphyrin-based pigmentation. The method is based on capillary liquid chromatography-mass spectrometry (CLC-MS), after an extraction procedure with formic acid and acetonitrile. The resulting limits of detection (LOD) and quantification (LOQ) were 0.006–0.199 and 0.021–0.665 µg mL⁻¹, respectively. This approach enabled us to quantify porphyrins in flying squirrel hairs at concentrations of 3.6–353.2 µg g⁻¹ with 86.4–98.6% extraction yields. This method provides higher simplicity, precision, selectivity, and sensitivity than other methods used to date, presenting the potential to become the standard technique for porphyrin analysis.

While the specific mechanisms of colour production in biological systems are diverse, the mechanics of colour production are straightforward and universal. Colour is produced through the selective absorption of light by pigments, the scattering of light by nanostructures or a combination of both. WhenTigriopus californicuscopepods were fed a carotenoid-limited diet of yeast, their orange-red body coloration became faint, but their eyespots remained unexpectedly bright red. Raman spectroscopy indicated a clear signature of the red carotenoid pigment astaxanthin in eyespots; however, refractive index matching experiments showed that eyespot colour disappeared when placed in ethyl cinnamate, suggesting a structural origin for the red coloration. We used transmission electron microscopy to identify consecutive nanolayers of spherical air pockets that, when modelled as a single thin film layer, possess the correct periodicity to coherently scatter red light. We then performed microspectrophotometry to quantify eyespot coloration and confirmed a distinct colour difference between the eyespot and the body. The observed spectral reflectance from the eyespot matched the reflectance predicted from our models when considering the additional absorption by astaxanthin. Together, this evidence suggests the persistence of red eyespots in copepods is the result of a combination of structural and pigmentary coloration.

Most studies of animal coloration focus on spectral colours, which are colours evoked by single peaks within the wavelengths of visible light. It is poorly understood how non-spectral colours (those produced by a combination of reflectance peaks) are produced, despite their potential significance to both animal communication and biomimicry. Moreover, although both pigmentary and structural colour production mechanisms have been well characterized in feathers independently, their interactions have received considerably less attention, despite their potential to broaden the available colour spectrum. Here, we investigate the colour production mechanisms of the purple feathers of the purple-breasted cotinga (Cotinga cotinga). The purple feather colour results from both the coherent scattering of light by a sphere-type nanomatrix of β-keratin and air (spongy layer) in the barbs, which produces a blue–green colour, and the selective absorption of light in the centre of the bird-visible spectrum by the methoxy-carotenoid, cotingin. This unusual combination of carotenoid and nanostructure with a central air vacuole, in the absence of melanin, is a blueprint of a synergistic way to produce a non-spectral colour that would be difficult to achieve with only a single colour production mechanism.

When working with a citizen science database like eBird, there are many possible ways to filter or subsample observations. Here, we discuss the potential biases and assumptions that surround different subsampling approaches or filtering that can be done to the eBird database. Restricting observations to species that are known to frequently hybridize, a specific time of the year, or a specific location, has the potential to greatly inflate the calculated per individual rate of hybridization. Such filtering also assumes that researchers know a birds’ capacity to hybridize with all other species in its range, which we argue is an unfounded assumption. We ultimately conclude that a limited filtering approach is ideal when using a citizen science database to attempt to address a broad question such as: what is the per individual rate of hybridization across all of the bird species in the United States? This article is protected by copyright. All rights reserved

We consider four key challenges related to estimating per-individual rates of hybridization in wild birds: (1) what is the meaning of the term “hybrid”?, (2) the importance of distinguishing between shared DNA sequences and on-going hybridization between populations, (3) the perils of focusing exclusively on known hybrid zones, and (4) the implications of very low rates of per individual hybridization. Because our focus is on using phenotype to recognize hybrids, we define a hybrid as an individual with a phenotype that is intermediate between two parental species. We emphasize the value of quantifying the rate of between-species mating that is ongoing in current populations and distinguish this endeavor from estimates of gene flow between populations based on genomic analysis, which can reflect both current and ancient hybridization. We restate the importance of quantifying per individual rates of hybridization among all birds without prejudging which birds are presumed to engage in hybridization. And finally, we emphasize that evidence for strong prezygotic sorting is not necessarily evidence that mate choice is a driver of speciation.

The rate of hybridization among taxa is a central consideration in any discussion of speciation, but rates of hybridization are difficult to estimate in most wild populations of animals. We used a successful citizen science dataset, eBird, to estimate the rates of hybridization for wild birds in the United States. We calculated the frequency at which hybrid individuals belonging to different species, families, and orders of birds were observed. Between 1 January 2010 and 31 December 2018, a total of 334,770,194 species records were reported to eBird within the United States. Of this total, 212,875 or 0.064% were reported as hybrids. This estimate is higher than the rate of hybridization (0.00167%) reported by Mayr based on impressions from a career studying museum specimens. However, if the 10 most influential hybrid species are removed from the eBird dataset, the rate of hybridization decreases substantially to about 0.009%. We conclude that the rate of hybridization for individuals in most bird species is extremely low, even though the potential for birds to produce fertile offspring through hybrid crosses is high. These findings indicate that there is strong prezygotic selection working in most avian species.