Major Breeding Plumage Color Differences of Male Ruffs (Philomachus pugnax) Are Not Associated With Coding Sequence Variation in the MC1R Gene

Sequence variation in the melanocortin-1 receptor (MC1R) gene explains color morph variation in several species of birds and mammals. Ruffs (Philomachus pugnax) exhibit major dark/light color differences in melanin-based male breeding plumage which is closely associated with alternative reproductive behavior. A previous study identified a microsatellite marker (Ppu020) near the MC1R locus associated with the presence/absence of ornamental plumage. We investigated whether coding sequence variation in the MC1R gene explains major dark/light plumage color variation and/or the presence/absence of ornamental plumage in ruffs. Among 821bp of the MC1R coding region from 44 male ruffs we found 3 single nucleotide polymorphisms, representing 1 nonsynonymous and 2 synonymous amino acid substitutions. None were associated with major dark/light color differences or the presence/absence of ornamental plumage. At all amino acid sites known to be functionally important in other avian species with dark/light plumage color variation, ruffs were either monomorphic or the shared polymorphism did not coincide with color morph. Neither ornamental plumage color differences nor the presence/absence of ornamental plumage in ruffs are likely to be caused entirely by amino acid variation within the coding regions of the MC1R locus. Regulatory elements and structural variation at other loci may be involved in melanin expression and contribute to the extreme plumage polymorphism observed in this species.

Birds display a wide range of variation in plumage coloration and pattern that has long fascinated biologists because of its importance in sexual selection, speciation, and adaptation (Roulin 2004). The ruff (Philomachus pugnax) is a lekking sandpiper which exhibits major dark/light color differences in melanin-based male breeding plumage that is closely associated with a genetic polymorphism for alternative male mating behavior (Hogan-Warburg 1966;Van Rhijn 1973;Lank et al. 1995). Three genetic male morphs persist in ruff populations: 1) dark-plumed territorial "Independents"; 2) lightplumed non-territorial "Satellites"; and 3) small female-like males called "Faeders" that lack ornamental plumage (Jukema and Piersma 2006;Lank et al. 2013). The extensive individual variation in melanin-based coloration of ornamental neck ruffs and head tufts of male ruffs has been well described (Hogan-Warburg 1966;Höglund and Lundberg 1989;Ekblom et al. 2012;Van Rhijn et al. 2014) and a wide range of plumage colors and patterns exists within both independent and satellite morph types. Independents range from black, dark rust, light rust to ivory, with occasional patches of white, but always contain predominantly dark rust or black feathers in ruff, head tufts, or both. In contrast, satellites are predominantly white in ornamental plumage color, but range from white, ivory to straw yellow, with secondary ruff colors of light to medium rust, but lack solid black feathers in the ruff or head tufts (Figure 1). This hypervariability has been attributed to diversifying selection for individual identity signalling Dale et al. 2001). Finally, the faeder males grow breeding plumage typical of females, lacking ornamental feather growth and conspicuous plumage colors (Jukema and Piersma 2006) (Figure 1).
Sequence variation in the melanocortin-1 receptor (MC1R) gene explains major dark/light plumage color variation in several avian species as well as coat color variation in many mammals (Hubbard et al. 2010;Roulin and Ducrest 2013). Mutations that increase the activation of MC1R result in increased synthesis of eumelanin, producing shades of black or brown, whereas mutations reducing MC1R activation lead to increased synthesis of phaeomelanin producing shades of rust (Mundy 2005). For example, single base-pair mutations in the coding sequence of the MC1R gene perfectly associate with dark and light plumage types in bananaquit (Coereba flaveola), lesser snow goose (Anser c. caerulescens), Arctic skua (Stercorarius parasiticus) and chestnut-bellied monarch (Monarcha castaneiventris) (Uy et al. 2009;Theron et al. 2001;Mundy et al. 2004). In ruffs, we previously found that the presence/absence of male breeding plumage (the putative Faeder locus) (Lank et al. 2013) was associated with a microsatellite marker predicted to be in the close vicinity of MC1R (Farrell et al. 2013b), in addition to evidence of linkage disequilibrium between white ruff coloration and the Satellite locus (Farrell 2013). We therefore investigated whether coding sequence variation in the MC1R gene explains major dark/light breeding plumage color variation and/or the presence/absence of ornamental plumage in male ruffs.

Sample Population
Forty-four male ruffs with a diversity of breeding plumage phenotypes previously used in linkage mapping (Farrell et al. 2013a) were selected from a captive-bred population maintained by DBL at Simon Fraser University in Burnaby, British Columbia. The sample included 12 dark-plumed males, 12 light-plumed males, and 20 unornamented female-like faeder males (Table 1). The dark-plumed males are assumed to have higher levels of eumelanin within their dark feathers, whereas the light-plumed males are assumed to have phaeomelanin or lack pigmentation Figure 1. The wide range of melanic plumage coloration and patterns exhibited by male ruffs in their ornamental breeding plumage. Independents (top row) are predominantly dark in color, ranging from black, dark rust, light rust to ivory, with occasional patches of white. Satellites (middle row) are predominantly white in color and range from white, ivory to straw yellow, with secondary colors of light to medium rust, but lack solid black in the ruff or head tufts. Faeder males (bottom row) lack ornamental plumage and closely resemble the breeding plumage of ruff females (female far right) (Photos by D.B.L., C.K., and L.L.F.).
(pure white) (Van Rhijn et al. 2014). Unornamented individuals were included because of the apparent close linkage of the Faeder locus with a microsatellite marker located near the predicted location of MC1R (Farrell et al. 2013b).

MC1R Genotyping and Analysis
Genomic DNA was extracted from blood samples stored in absolute ethanol (50 μL of blood in 1.5 mL of absolute ethanol) using an ammonium acetate precipitation method (Nicholls et al. 2000). A segment of the MC1R gene that encompassed sites previously associated with plumage polymorphism in birds was amplified using the conserved primers MSHR72 and MSHR9, with internal sequencing primers MSHR73 and MSHR74 (Mundy et al. 2004). Each 10-μL Polymerase Chain Reaction (PCR) contained approximately 10 ng of genomic DNA, 1 μL of each primer (5 μM), 3 µL ultrapure H 2 0 and 4 μL Qiagen Multiplex PCR Mix (Qiagen  (Sambrook et al. 1989). Purified template DNA was directly sequenced using Big Dye.v3.1 chemistry (PE Biosystems), according to the manufacturer's protocol using an ABI3730 Genetic Analyzer (Applied Biosystems). Amplified fragments were sequenced in the forward and reverse complementary directions and a consensus sequence was created using a modified version of the Phred  and Phrap/Cross_match/ Swat (Green 1996) software (PERL scripts provided by the NERC Biomolecular Facilty, UK). Sequences that did not form a consensus between their forward or reverse strands, but that were of good quality, were included in the study by alignment with those that did produce a consensus. Manual base calling and comparative analyses were performed in CODONCODE ALIGNER v 4.0 (http://www.codoncode. com/). Sequences were aligned in MEGA.v5.0 (Tamura et al. 2011) using ClustalW and sequences deposited in Genbank (Accession numbers LM993813-LM993852).

Results
A total of 821 bp of the MC1R gene corresponding to positions 79-899 of the aligned chicken MC1R gene sequence (Kerje et al. 2003) was sequenced from 44 male ruff individuals. Three polymorphic SNPs were found: 2 synonymous substitutions (A/G: Thr93, A/T: Val105) and 1 nonsynonymous substitution (A/G: His207Arg); the latter polymorphism is shared with dark/light plumage differences in the red-footed booby (Sula sula) (Baião et al. 2007). At all the other functional sites known to be associated with dark and light melanic plumage variation in other species, all ruff MC1R sequences were monomorphic ( Figure 2). None of the SNPs at the polymorphic ruff sites were associated with major dark/light plumage morphs or the presence/absence of breeding plumage (Table 1).

Conclusion
More than 150 genes are known to affect animal coloration and pattern (Barsh 1996;Hoekstra 2006) and ruffs have the greatest naturally evolved intraspecific plumage diversity among birds . Given that coding sequence variation in MC1R does not explain the major plumage color differences of male ruffs, we conclude that the control of plumage variation in ruffs is more complex than in other bird species with simple dark and light morphs. It is unlikely that the presence of dark or light coloration or the presence/absence of ornamental plumage in male ruffs is solely determined by amino acid variation within the MC1R locus. Because only a partial amino acid sequence of the coding region of the MC1R gene was sequenced, we cannot rule out the possibility that functional non-synonymous substitutions affecting plumage color may be present in the regions (approximately 9% based on other bird species) not sequenced. However, it seems more likely that regulatory polymorphisms account for the observed variation. The extensive individual variation suggests that ornamental plumage coloration is a polygenic trait, which might nonetheless involve MC1R along with other genes that affect the deposition of melanin Roulin and Ducrest 2013;Van Rhijn et al. 2014). The potential functional significance, implied by the apparent proximity of the Faeder locus and MC1R (Farrell et al. 2013b), remains to be determined through more detailed mapping and analysis. Promoter regions of MC1R remain as candidates that might influence ornamental plumage type in ruffs. For example, local expression differences in MC1R and/ or other pigmentation genes in the feather follicles of the neck ruff and head tufts could produce the observed hypervariability of the breeding plumage in this species.