Association of feather colour with constitutively active melanocortin 1 receptors in chicken Maria K. Ling 1 , Malin C. Lagerstro¨m 1 , Robert Fredriksson 1 , Ronald Okimoto 2 , Nicholas I. Mundy 3 , Sakae Takeuchi 4 and Helgi B. Schio¨th 1 1 Department of Neuroscience, Uppsala University, Uppsala, Sweden; 2 Department of Poultry Science, University of Arkansas, Fayetteville, Arkansas, USA; 3 Department of Biological Anthropology, University of Oxford, Oxford, UK; 4 Department of Biology, Faculty of Science, Okayama University, Okayama, Japan Seven alleles of the chicken melanocortin (MC) 1 receptor were cloned into expression vectors, expressed in mam- malian cells and pharmacologically characterized. Four of the clones e +R ,e +B&D ,e wh /e y ,E Rfayoumi gave receptors to which melanocortin stimulating hormone (a-MSH) and NDP-MSH bound with similar IC 50 values and responded to a-MSH by increasing intracellular cAMP levels in a dose- dependent manner. Three of the cMC1 receptors; e b ,Eand E R , did not show any specific binding to the radioligand, but were found to be constitutively active in the cAMP assay. The E and E R alleles are associated with black feather col- our in chicken while the e b allele gives rise to brownish pigmentation. The three constitutively active receptors share a mutation of Glu to Lys in position 92. This mutation was previously found in darkly pigmented sombre mice, but constitutively active MC receptors have not previously been shown in any nonmammalian species. We also inserted the Glu to Lys mutation in the human MC1 and MC4 recep- tors. In contrast with the chicken clones, the hMC1-E94K receptor bound to the ligand, but was still constitutively active independently of ligand concentration. The hMC4- E100K receptor did not bind to the MSH ligand and was not constitutively active. The results indicate that the structural requirements that allow the receptor to adapt an active conformation without binding to a ligand, as a con- sequence of this E/K mutation, are not conserved within the MC receptors. The results are discussed in relationship to feather colour in chicken, molecular receptor structures and evolution. We suggest that properties for the ÔE92K switchÕ mechanism may have evolved in an ancestor common to chicken and mammals and were maintained over long time periods through evolutionary pressure, probably on closely linked structural features. Keywords: G-protein coupled; MSH; melanocortin receptor; polymorphism. Spontaneous or constitutive G-protein coupled receptor (GPCR) activity was first convincingly described for the d-opioid receptor [1], and was further established by the demonstration of constitutive activity in chimeras of the a 1B and a 2 receptors (summarized in [2]). Later it was shown that mutations in the human rhodopsin gene can constitutively activate transducin in the absence of retinal and light [3]. It is now known that naturally occurring constitutively active GPCRs are found to be responsible for a diverse array of inherited as well as somatic genetic disorders [4,5]. The melanocortins (a-MSH/ACTH), secreted from a frog pituitary, were in 1912 found to cause pigmentation. In higher vertebrates including aves and mammals, these peptides are expressed throughout the body, and are involved in a variety of physiological regulatory functions [6–8]. In the skin, the melanocortins are synthesized locally (for birds [9], for mammals [10]), and act through a GPCR named melanocortin (MC) 1 receptor to regulate melano- genesis. Keratinocytes probably serve as the main physio- logical source of melanocortins acting on melanocytes in the epidermis and hair follicle. The MC1 receptor couples through G proteins to adenylate cyclase to stimulate tyrosinase, the rate-limiting enzyme in the synthesis of both classes of melanin pigments, eumelanin and phaeomelanin. A low level of tyrosinase expression leads to increased phaeomelanin synthesis, while elevated levels of tyrosinase, that can result from a-MSH stimulation of melanocytes, divert the intermediates primarily along the eumelanin synthetic pathway (for reviews see [8,10–11]). The extension (E) locus, together with agouti (A) locus, regulates the relative amount of black pigment (eumelanin) and red/yellow pigment (phaeomelanin) in mammals [12]. The E locus encodes the MC1 receptor and the A locus encodes the agouti peptide, an antagonist of the MC1 receptor. Dominant alleles at extension result in dark brown or black coat colour, while animals homozygous for recessive alleles have yellow or red coats. The opposite is true for the agouti allele. A dominant mutation at the A Correspondence to H. B. Schio ¨ th, Department of Neuroscience, Biomedical Center, Box 593, 75 124 Uppsala, Sweden. Fax:+4618511540, E-mail: helgis@bmc.uu.se Abbreviations: GPCR, G-protein coupled receptor; IC 50 , 50% inhibitory concentration; EC 50 , 50% effective concentration; IL, intracellular loops; MC, melanocortin; MSH, melanocortin stimulating hormone; NDP-MSH, [Nle4, D -Phe7]a-MSH; TM2, transmembrane region 2. (Received 3 December 2002, revised 26 January 2003, accepted 6 February 2003) Eur. J. Biochem. 270, 1441–1449 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03506.x locus (A y /A) in mice causes uniform yellow coat and obesity. A similar phenotype, but with normal body weight, is found in recessive yellow mice (e/e), where a loss of function mutation in the MC1 receptor results in animals producing phaeomelanin [13]. The dominant extension locus alleles, sombre (E SO and E SO)3J )andtobacco (E tob ), which all have dominant melanizing effects, result from point mutations that produce constitutively active receptors. The sombre alleles produce a fairly uniform black coat, while tobacco darkening only involves the dorsal portion of the animal. The tobacco alleles produce a receptor that remains hormone responsive but produces a greater activation of adenylyl cyclase than does the wild-type allele. Both sombre alleles were found to produce constitutively active receptors, defined as being able to significantly elevate adenylyl cyclase activity in the absence of ligand, and thereby enhance the eumelanin production resulting in a dark pigmentation in mice [13]. Null-mutations of POMC, the precursor for the melanocortins, causes yellow coat, obesity and adrenal insufficiency. These mice are somewhat darker, Ôdirty blondÕ suggesting that the basal MC1 receptor activity in the absence of ligands may be higher than a nonfunctioning receptor [14]. Mutations in MC1 receptors, related to hair or skin colour, have been found in several other mammalian species, although pharmacological characterization of these changes has usually not been carried out. Two species, whose receptors have been pharmacologically investigated, are fox and sheep. Va ˚ ge cloned and characterized the fox MC1 receptor and found a mutation that caused constitu- tive activity in the dark coated Alaska Silver fox [15]. This dominant mutation was however, also found in foxes with significant red coat colouration, and it was suggested that the fox agouti protein counteracted the signalling activity of a constitutively active fox MC1 receptor. In sheep, two mutations in the MC1 receptor showed complete cosegre- gation with dominant black coat colour in a family lineage. These mutations were transferred into the corresponding mouse receptor in which they produced constitutive activity [16]. Loss of function mutations are common in the human MC1 receptor and these are over-represented in Northern European redheads and in individuals with pale skin [17,18]. These variants are a risk factor, possibly independent of skin type, for melanoma [19]. Variants of the MC1 receptor gene were found to be associated with the extension (E) locus in chickens and it was proposed that they might be linked to feather pigmentation [20]. A dominant mutation, identical to one of the mutations in the sombre mouse, was found in chicken with black feathers [20]. The very same mutation, Glu to Lys in transmembrane region 2 (TM2), has also been shown to be present in melanic individuals of the bananaquit, a neotropical passerine bird, and absent in all yellow individuals [21]. The molecular pharmacology of these mutations has however, not yet been investigated. In this study, we performed the first expression studies on avian MC receptor genes. We expressed and pharmacologi- cally characterized seven polymorphic variants of the chicken MC1 receptors derived from different E locus alleles. We also introduced the Glu to Lys mutation in to the human MC1 and MC4 receptors to investigate the pharma- cological impact of these mutations. Experimental Receptor clones Oligonucleotide primers were designed to amplify the entire coding region of each cMC1 receptor variant. To facilitate cloning and establishment of orientation of the PCR- amplified DNA fragments, recognition sites for HindIII and BamHI were introduced into 5¢ and 3¢ primers, respectively, by altering original nucleotide sequences. The primer sequences were 5¢-GGAAGCTTTGTAGGTGCTGCA GTT-3¢ for the 5¢ primer and 5¢-CATGGATCCTCCTC CTGTCTGTGCCGC-3¢ for the 3¢ primer, corresponding to positions )55 to )78 and 1049–1072, respectively, of the cMC1 receptor gene, where the A of the translation initiation codon ATG was defined as +1. The amplified DNA fragmentswere cloned into pGEM3Zf(+), sequenced, and subsequently subcloned into pCEP4 Turbo expression vector [22] and resequenced. The human MC1 [23] and human MC4 receptors [24] were used as templates for the mutagenesis. Site-directed mutagenesis The E94K mutation in hMC1 and E100K mutation in hMC4, were introduced into the receptor coding sequence by PCR. Two complementary oligonucleotides were designed to contain the required mutation. The hMC1- E94K primers were (5¢ primer) 5¢-CAGGAGGATGA CGGCCGTCTTCAGCACGTTGCTCCC-3¢ and (3¢ pri- mer) 5¢-GGGAGCAACGTGCTGAAGACGGCCGTCA TCCTCCTG-3¢. The hMC4-E100K primers were (5¢ primer) 5¢-TAGGGTGATGATAATGGTTTTTGATCC ATTTGAAAC-3¢ and (3¢ primer) 5¢-GTTTCAAATG GATCAAAAACCATTATCATCACCCTA-3¢. The pri- mers were hybridized to opposite strands of the receptor gene and the complete MC receptor coding sequence was amplified. The end-primer was complementary to 3¢ or 5¢ depending on which mutagenesis primer was used (forward or reverse). The two products were used as templates and linked together in a second PCR, in which only the end- primers were used. Expression DNA for transfection was prepared using Qiagen Plasmid Maxi Kit (Merck). HEK 293 EBNA cells 50–70% confluent on 10-cm plates, were transfected with 15 lgofthe construct using FuGENE TM Transfection Reagent (Boeh- ringer Mannheim), diluted in Optimem medium (Gibco BRL). After transfection, cells were grown in Dulbecco’s MEM/Nut Mix F-12 (Gibco BRL) containing 10% foetal bovine serum (Biotech Line), 2.4 m ML -glutamine, 2.5 mgÆmL )1 G418, 2.5 lgÆmL )1 amphotericin, and 100 lgÆmL )1 kanamycin solution (all from Gibco BRL) until harvesting, after 48 h. Binding assays Intact transfected cells were re-suspended in 25 m M Hepes buffer (pH 7.4) containing 2.5 m M CaCl 2 ,1m M MgCl 2 and 2 gÆL )1 bacitracin. Competition experiments were 1442 M. K. Ling et al. (Eur. J. Biochem. 270) Ó FEBS 2003 performed in a final volume of 100 lL. The cells were incubatedin96-wellplatesfor3hat37°Cwithconstant concentration of [ 125 I]NDP-MSH and appropriate concen- trations of competing unlabelled ligands, [Nle4, D -Phe7] a-MSH (NDP-MSH) or a-MSH. The incubations were terminated by filtration through Glass Fibre Filters, Filter- mat A (Wallac Oy, Turku, Finland), which had been presoaked in 0.3% poly(ethylenimine), using a TOMTEC Mach III cell harvester (Orange, CT, USA). The filters were washed with 5.0 mL 50 m M Tris pH 7.4 at 4 °C and dried at 60 °C. The dried filters were then treated with MeltiLex A (Perkin Elmer) melt-on scintillator sheets and counted in a Wallac 1450 (Wizard automatic Microbeta counter). The results were analysed with a software package suitable for radioligand binding data analysis ( PRISM 3.0, Graphpad, San Diego, CA, USA). The binding assays were performed in duplicate and repeated three times. Nontransfected HEK293-EBNA cells did not show any specific binding to [ 125 I]NDP-MSH. NDP-MSH was radio-iodinated by the chloramine T method and purified by HPLC. NDP-MSH and a-MSH were purchased from Neosystem (France). cAMP assay Semi-stable cells, expressing the receptors of interest, were harvested in growth media containing 3-isobutyl-1-methyl- xanthine from Sigma. Twohundred microlitres of the cells were added to tubes containing appropriate concentra- tions of a-MSH and incubated for 30 min at 37 °C. After stimulation the cells were lysed and cAMP extracted using 4.4 M perchloric acid. The cAMP extract was then neutralized with 5 M KOH and centrifuged. The intracel- lular cAMP produced was measured in 50 lLofthe supernatant after addition of [ 3 H]cAMP and bovine adrenal binding protein and incubation for 2 h at 4 °C. Standards containing nonlabeled cAMP were assayed in the same manner. The incubates were then harvested by adding activated carbon. After centrifugation, the super- natant was removed into scintillation tubes and counted in a Tri-carb Liquid scintillation analyser after addition of 3 mL scint solution (Ready Safe TM ; Beckman Coulter). The cAMP assay was performed in duplicate and repeated three times. The results were analysed using the PRISM 3.0 software package (Graphpad, San Diego, CA, USA). The protein concentrations were measured using Bio-Rad Protein Assay (Bio-Rad, Solna, Sweden) with BSA as standard. Results The amino acid sequences for the seven polymorphic cMC1 receptors derived from different E locus alleles are shown in Table 1. Four clones; e +R ,e +B&D ,e wh /e y ,E Rfayoumi resulted in competition curves for a-MSH and NDP-MSH using iodinated [ 125 I]NDP-MSH as radioligand. The binding curves are shown in Fig. 1. The IC 50 for the MSH ligands are shown in Table 2. Three clones, e b , E and E R , did not induce any specific binding on the transfected cells. We also measured the intracellular cAMP in response to varying concentrations of a-MSH. The cells transfected with the four cMC1 receptors, that showed competition curves in the binding assays, also responded to a-MSH by accumulation of intracellular cAMP in a dose-dependent manner. The cAMP curves are shown in Fig. 2, and the EC 50 values are showninTable2. The three cMC1 receptors; e b ,Eand E R , did not show any specific binding to the radioligand or respond to a-MSH by accumulation of intracellular cAMP. The results are shown in Fig. 3. The basal levels were, however, significantly increased for the cells transfected with the three clones in all experimental points, as compared with nontransfected HEK-293 EBNA cells, that were used as a control. The cells transfected with the receptors (e +R , e +B&D , e wh /e y , E Rfayoumi ) also served as controls, as at the initial level and at the lowest concentrations of a-MSH, the cAMP levels were lower than that observed for the cells transfected with e b ,Eand E R . All experiments were performed with a similar number of cells; the amount of protein per well was determined and the cAMP values were normalized accordingly. The cAMP levels for these three clones were significantly higher for all experimental points, also when corrected for the cell number. The experiments were repeated three times and qualitatively the results were the same each time. The activation level did, however, not reach the maximum level of the other clones. e b ,Eand E R cMC1 receptors were partially activated to  20–60% of the maximal activation of a ÔnormallyÕ functioning cMC1 receptor (e wh /e y ). The cAMP levels did, however, vary between the repeats and the internal order between the three clones was not always the same. Therefore, we do not draw the conclusion that the cAMP levels for the e b ,Eand E R cMC1 receptors transfected cells differed from each other. The Glu92 residue in TM2 (see Table 1) is conserved within the family of MC receptor subtypes. The e b ,Eand E R cMC1 receptors have Lys in this position. We inserted a Table 1. Amino acid positions in the seven different cMC1 receptors. Dominance: E >E R >e wh >e + >e b >e y . Allele 71 92 133 143 213 215 Line e + Met Glu Leu Thr Arg His Richardson’s RJF (e +R ) e + – – – – Cys – B & D RJF (e +B&D ) E Thr Lys – – Cys – Black Australorp (E) e b Thr Lys – – Cys Pro Smyth Brown line (e b ) E R – Lys – – – – ADOL line 0 (E R ) E R – – Gln – – – Fayoumi (E Rfayoumi ) e wh /e y – – – Ala – – NHR, RIR, Buff Min(e wh ) – Indicates that the amino acid in this position is the same as the one as described in the allele at the top. Note that e wh and e y are identical in amino acid sequence. Ó FEBS 2003 Constitutively active MC1 receptors in chicken (Eur. J. Biochem. 270) 1443 Lys in the corresponding position in the human MC1 and MC4 receptors by site-directed mutagenesis and generated clones termed hMC1-E94K and hMC4-E100K. The clones were expressed and assayed pharmacologically in the same manner as the clones mentioned above. The binding results are shown in Fig. 4. The cells transfected with hMC1-E94K, in contrast with the chicken receptors with Lys in position 92, did bind NPD-MSH. The IC 50 value was about 18-fold lower than that of the wild-type human MC1 receptor. The cells transfected with hMC4-E100K did not, however, show any specific binding. The cAMP results are shown in Fig. 5. The cells transfected with hMC1-E94K showed significantly higher cAMP values at all experimental points, independent of concentration of a-MSH, as compared with the non- transfected cells. The cells transfected with hMC4-E100K, in contrast with the chicken receptors with Lys in position 92, showed the same low levels as nontransfected cells. Discussion Like the extension locus in mammals, the extended black (E) locus of the chicken controls the relative amount of eumelanin and phaeomelanin in melanocytes. The locus was originally localized on chromosome 1, but recent genetic and FISH analyses revealed that it is located on a microchromosome [25,26]. Several alleles exhibiting differ- ent pigmentation have been described and there is an intricate hierarchy among them. But in general, alleles that make more eumelanin are dominant over those that make less eumelanin. Unlike the mammalian extension locus, the phenotype of each allele is expressed mainly in chicks and adult females. In fact, the phenotypes of adult males with different nonmelanic alleles, including e + , e wh , e y ,ande b ,are similar, having black-breasted red feather pattern. The only phenotypic difference observed among them is the under- colour, the fluff of the feathers next to the skin; it is white or cream for the e wh and e y malesandgreyforthee + and e b .In melanic alleles (E, E R ,andE Rfayoumi ), adult males are black in all areas and the flight feathers are also black for E, while the other melanic alleles (E R and E Rfayoumi )givehalf-red half-black feathers. Fig. 6 shows the adult E locus colour patterns on a wild-type background for all other feather colour genes. Our results indicate that the polymorphic e b , E and E R chicken MC1 receptors are constitutively active. These three receptors all share a mutation of Glu to Lys in position 92 (see Table 1). Previously it has been shown that constitutive activation of the MC1 receptor in darkly pigmented sombre mice results from the very same mutation in position 92 [13]. L98P mutation in the MC1 receptor in mouse, D119N in Fig. 1. Competition curves of [Nle4, D -Phe7]a-MSH (u) and a-MSH (m) obtained with transfected HEK-293 (EBNA) cells using a fixed concen- tration of 0.2 n M [ 125 I][Nle 4 , D -Phe 7 ]a-MSH for four cMC1 receptors, e +R ,e +B&D ,e wh /e y ,E Rfayoumi . Data points represent means of duplicates and error bars indicate standard error of the mean (SEM). Table 2. Pharmacological characterization of chicken MC1 receptors after expression in HEK cells. The IC 50 values were obtained from competition using [ 125 I] [Nle4, D-Phe7] a-MSH as radioligand and a-MSH and [Nle4, D-Phe7] a-MSH as competitors. The EC 50 values are obtained in intracellular cAMP assay using a-MSH as stimulator. Receptor a-MSH (IC 50 ) (nmolÆL )1 ) NDP-MSH (IC 50 ) (nmolÆL )1 ) a-MSH (EC 50 ) (nmolÆL )1 ) e +(R) 1350 ± 680 9.76 ± 2.51 449 ± 69 e wh /e y 827 ± 167 13.7 ± 8.8 20.4 ± 3.5 e +(B & D) 601 ± 117 6.00 ± 0.40 21.9 ± 5.3 E Rfayoumi 1080 ± 137 4.79 ± 0.20 36.2 ± 0.5 1444 M. K. Ling et al. (Eur. J. Biochem. 270) Ó FEBS 2003 sheep and C125R in Alaska silver fox that cause dark pigmentation have also been shown to be constitutively active [15,27]. The pharmacology of constitutively active MC receptors has not been previously shown in any nonmammalian species, and these results show that the function of this mutation seems to be similar in chicken and mice. The results add further support to the hypothesis that the same point mutation (E92K) in the bananaquit, Coereba flaveola associated with the melanic plumage morph [21] is constitutively active. The clear association between the mainly black feather colour of chickens possessing E and E R and the presence of constitutively active MC1 receptor is in line with the previous observation of the E92K mutation in other species. Males possessing the e b allele are described above. Among the females, the allele gives rise to brownish pigmentation, although according to the results from the cAMP assay for this receptor, perhaps a darker pheno- type would have been expected, as for the other two constitutively active cMC1 receptors. Likewise, E Rfayoumi and e wh /e y alleles exhibiting phenotypes similar to E R allele and yellow-red pigmentation, respectively, were found to encode normally functioning cMC1 receptors, which bind agonist and couple to G-protein in a ‘normal’ manner. Thus, our pharmacological results cannot com- pletely explain the association of cMC1 variants with the corresponding phenotypes. It is possible that the expres- sion levels of cMC1 receptor vary with different alleles, which results in phenotypic difference in alleles encoding cMC1 receptors with similar pharmacological character- istics. Alternatively, specific amino acid substitutions observed in each allele alter the interaction of cMC1 receptor with unidentified factors expressed specifically in melanocytes. Further analyses are thus required to clarify molecular mechanisms for all the different feather pig- mentation patterns in chicken. Considering the other polymorphic residues in the chicken clones, the results indicate that the changes of Fig. 2. Generation of cAMP in HEK-293 (EBNA) cells transfected with four cMC1 receptors: e +R ,e +B&D ,e wh /e y ,E Rfayoumi in response to a-MSH. The cAMP levels were normalized to the amount of protein. Data points represent means of duplicates and error bars indicate standard error of the mean (SEM). Fig. 3. Cells transfected with the three cMC1 receptors e b (n), E (s), E R (m) show elevated cAMP levels and independence of the concentration of a-MSH in comparison to basal cAMP levels in HEK-293 (EBNA) cells (j). The cAMP levels were normalized to the amount of protein. e wh /e y (d) (from Fig. 2) is shown for comparison. Data points repre- sent means of duplicates and error bars indicate standard error of the mean (SEM). Ó FEBS 2003 Constitutively active MC1 receptors in chicken (Eur. J. Biochem. 270) 1445 Met71 to Thr, Leu133 to Gln, Thr143 to Ala, Arg213 to Cys, or His215 to Pro do not influence the pharmacological function of the receptors. Met71, Leu133 and His215 are conserved through all the MC receptors cloned so far. These residues are believed to be in the first, second and third intracellular loops (IL), respectively. Thr143 alternates between Thr and Ser, and is found in IL2, while Arg213 alternates between Arg and Cys, and is found in IL3, within the entire MC receptor family. It is interesting that mutation of Cys215 to Gly in the human MC1 receptor (correspond- ing to Arg213) resulted in failure to generate cAMP signal in response to the agonists a-MSH [28]. It seems therefore that the function of the MC receptor requires a polar residue in this position, while a hydrophobic nonpolar residue causes disconnection of the signal transduction. As the key residues within the MC receptor family are highly conserved, our new results provide additional information for generating molecular models of the binding and activation process of the MC receptors that is an important part of rational drug discovery [8,29]. In order to shed further light on the structural require- ments needed for generating a constitutively active MC receptor, we introduced the corresponding E92K mutation into the human MC1 and MC4 receptors. It was surprising that the mutant human MC1 receptor showed differences in pharmacology as compared with the mutant chicken MC1 receptor. It is intriguing that in contrast with the chicken receptor with Lys, hMC1-E94K bound the ligand but was still constitutively active, independently of the ligand. This pharmacology is remarkable, as this receptor seems to have the ability to bind the ÔagonistÕ peptide ligand with high (albeit slightly lower) affinity, and undergo the conforma- tional changes that this binding interaction is proposed to have, without influencing the interaction with the G-protein. Even though inactivating mutations associated with red hair are common in the hMC1 receptors, no constitutive activating MC1 receptor has yet been found in humans. It seems clear that the structural properties of the hMC1 receptor are different from those of chicken, mouse and fox, where the constitutive E/K mutation leads to loss of binding. It is possible that the evolutionary pressure on the hMC1 receptor is altered due to changes in the physiological importance of body hair colour and therefore the structural pharmacology is evolving differently as compared to MC1 receptors of species in which colour has a more defined function. It would be interesting to see if this property is unique to humans or if it has evolved earlier in primates. In order to find out if these pharmacological properties are shared beyond the MC1 receptors, we also introduced the E92K mutation into the hMC4 receptor. This receptor is mainly expressed in the central regions of the brain, including the hypothalamus where it is an important regulator of food intake. The MC4 receptor shares 60% of the amino acids with the MC1 receptor, but has a completely different physiological role with no known overlap in function. The MC1 and MC4 receptors do, however, share the unique property that they both have natural antagonists, the agouti and the agouti-related peptide, respectively, in addition to their natural agonist, a-MSH. The hMC4-E100K, in contrast with the hMC1 receptor mutation, did not bind the MSH ligand and was not constitutively active. No naturally occurring activation mutants have been found for MC receptors other than MC1, but several inactivation mutants of the MC4 recep- tors are related to obesity [30,31]. The results indicate that the structural ability to form constitutively active receptors with a single amino acid mutation of the conserved Glu in TM2 has either not been maintained, or was never present in the MC4 receptors. Unlike constitutively activating mutations in many GPCRs that give increased agonist efficacy or affinity, these MC1 receptor mutations have the opposite effect. The molecular mechanism of constitutive activation of the MC1 receptor has been studied by inserting the mutations into the mouse MC1 receptor [16]. These authors proposed a ligand-mimetic model which explains Fig. 4. Competition curve of [Nle4, D -Phe7]a-MSH obtained with transfected HEK-293 (EBNA) cells using a fixed concentration of 0.2 n M [ 125 I][Nle 4 , D -Phe 7 ]a-MSH for the human wild-type MC1 (u), human MC1 (E94K) (m), and wild-type MC4 (j), human MC4 (E100K) (r) receptors. Data points represent means of duplicates and error bars indicate standard error of the mean (SEM). Fig. 5. cAMP levels of normal cAMP in HEK-293 (EBNA) cells (j) and cells transfected with human MC1 (E94K) (d), human MC4 (E100K) (u) receptors, showing cAMP levels independent of the con- centration a-MSH. The cAMP levels were adjusted to the amount of protein. e wh /e y (.) (from Fig. 2) is shown for comparison. Data points represent means of duplications and error bars indicate standard deviations. 1446 M. K. Ling et al. (Eur. J. Biochem. 270) Ó FEBS 2003 the lower (read nonexistent) ligand affinity and efficacy. It was proposed that the mutations transformed the receptor into its active form, not by disrupting the internal constraint as proposed by the early rhodopsin studies and the ternary allosteric model, but by indirectly mimicking ligand-bind- ing. The amino acid residues that were mutated in the mouse MC1 receptor are all conserved within the human MC1 and further studies are thus needed to explain the pharmacological differences between the mouse and human MC1 receptors. The reptilian ancestors of chickens diverged about 300– 350 million years ago from the lineage leading to mammals [32]. It is remarkable that the same single amino acid mutation forming a constitutive active receptor is found in so evolutionary distant species. Recent studies show remarkable evolutionary conservation within the primary sequence and pharmacology of MC receptors (including the E92) for at least 450 million years [33]. Our mutations of the hMC1 and hMC4 receptors indicate that the structural requirements that allows the receptors to fall into a constitutively active state without binding to a ligand are not automatically conserved within the MC receptors or within the MC1 receptors. Therefore, we find it unlikely that it is simply the ability to bind MSH peptides, the main pharmacological property shared by the MC1 and MC4 receptors, that is the structural feature that automatically allows the ÔE92K switchÕ mechanism to work. We find it also unlikely that the complex structural requirements for single amino acid activation switch were independently developed in chicken and mammals. It is thus tempting to speculate that some structural properties evolved once before the divergence of the avian and mammalian lineages. The intriguing question is by which mechanism this property could have been conserved over such long time period. One theoretical possibility is that the specific mutation was ancestral and that an E/K92 heterozygote had a selective advantage over the two homozygotes, thereby preserving both the alleles and also the structural constraints. This is, however, very unlikely due to the long evolutionary times and the divergent lineages involved. Fig. 6. Cartoon representation of the adult E locus colour patterns on a wild-type background for the feather colour genes. E (dominant extended black) birdsareblackinallareasinbothsexes.Malesthataree + ,e b , e wh or e y (wild-type, partridge or brown, dominant wheaten or recessive wheaten, respectively) all have the black-breasted red feather pattern. The only difference is that the wheaten males have a white or cream feather under- colour and the e + and e b males have a grey under-colour. E R (birchin) males are different in the wings. The flight feathers are all black instead of the half-red half-black feathers of the recessive alleles. Birchin females have black bodies and gold hackle feathers. Both male and female birchins have black melanin in the epidermis of their shanks. The brown (e b ) females have brown stippled backs and wings and brown stippled breasts, where the wild-type (e + ) females have brown stippled backs and wings and salmon breasts. Both females have a grey under-colour. Dominant wheaten (e wh ) and recessive wheaten (e y ) females have the same phenotype, having the salmon colour of the wild-type breast extended into the plumage of the back and wings. They have a white or cream under-colour and black is diluted in the female plumage. Ó FEBS 2003 Constitutively active MC1 receptors in chicken (Eur. J. Biochem. 270) 1447 There are only a few examples of the maintenance of specific alleles by heterozygote advantage (e.g. MHC class II alleles, haemoglobin S) and in these cases the maintenance has been for only thousands to at most a few million years [34]. Moreover, it is notable that the E92K mutation in bananaquit (mentioned above) is believed to have occurred recently in that lineage [21]. We believe therefore that it is most likely that the structural properties for the ÔE92K switchÕ mechanism were maintained through evolutionary pressure on closely linked structural features. These pro- perties were subsequently retained in certain lineages, like mouse and birds, and lost in others, like humans. Further structural characterization of the protein structure and studies on the MC receptors in more ancient tetrapod groups, such as reptiles, may shed further light into the mechanism on how structural properties for a single amino acid switch mechanism may have survived such a long evolutionary distance. Acknowledgements We thank Prof D. Larhammar, Uppsala University for valuable criticism. The studies were supported by the Swedish Medical Research council (MRC), the Swedish Society for Medical Research (SSMF), Svenska La ¨ karesa ¨ llskapet, A ˚ ke Wibergs Stiftelse and Melacure Therapeutics AB, Uppsala, Sweden to H.S., the Japanese Society for Promotion of Science (Grant-in Aid for Scientific Research) to S.T. References 1. Costa, T. & Herz, A. (1989) Antagonists with negative intrinsic activity at delta opioid receptors coupled to GTP-binding proteins. Proc.NatlAcad.Sci.USA86, 7321–7325. 2.Lefkowitz,R.,Cotecchia,S.,Samama,P.&Costa,T.(1993) Constitutive activity of receptors coupled to guanine nucleotide regulatory proteins. Trends Pharmacol. Sci. 14, 303–307. 3. Robinson,P.,Cohen,G.,Zhukovsky,E.&Oprian,D.(1992) Constitutive active mutants of rhodopsin. Neuron 9, 719–725. 4. Leurs, R., Smit, M., Alewijnse, A. & Timmerman, H. (1998) Agonist-independent regulation of constitutively active G-protein- couple receptors. Trends Biochem. Sci. 23, 418–422. 5. Sadee,W.,Hoeg,E.,Lucas,J.&Wang,D.(2001)Geneticvari- ations in human G protein-coupled receptors: implications for drug therapy. AAPS Pharmsci. 3,E22. 6. Takeuchi, S. & Takahashi, S. (1998) Melanocortin receptor genes in the chicken – tissue distributions. Gen. Comp. Endocrinol. 112, 220–231. 7. Takeuchi, S., Teshigawara, K. & Takahashi, S. (2000) Widespread expression of Agouti-related protein (AGRP) in the chicken: a possible involvement of AGRP in regulating peripheral melano- cortin systems in the chicken. Biochim. Biophys. Acta 1496, 261– 269. 8. Schio ¨ th, H.B. (2001) The physiological role of melanocortin receptors. Vitamin Hormones 63, 195–232. 9. Takeuchi, S., Teshigawara, K. & Takahashi, S. (1999) Molecular cloning and characterization of the chicken pro-opiomelanocortin (POMC) gene. Biochim. Biophys. Acta 1450, 452–459. 10. Slominski, A., Wortsman, J., Luger, T., Paus, R. & Solomon, S. (2000) Corticotropin releasing hormone and proopiomelanocortin involvement in the cutaneous response to stress. Physiol. Rev. 80, 979–1020. 11. Jackson, I. (1997) Homologous pigmentation mutations in human, mouse and other model organisms. Hum. Mol. Genet. 6, 1613–1624. 12.Bultman,S.,Michaud,E.&Woychik,R.(1992)Molecular characterization of the mouse agouti locus. Cell 71, 1195–1204. 13. Robbins, L., Nadeau, J., Johnson, K., Kelly, M., Roselli-Rehfuss, L.,Baack,E.,Mountjoy,K.&Cone,R.(1993)Pigmentation phenotypes of variant extension locus alleles results from point mutations that alter MSH function. Cell 72, 827–834. 14.Yaswen,L.,Diehl,N.,Brennan,M.&Hochgeschwender,U. (1999) Obesity in the mouse model of pro-opiomelanocortin deficiency responds to peripheral melanocortin. Nature Med. 5, 1066–1070. 15. Va ˚ ge, D., Lu, D., Klungland, H., Lien, S., Adalsteinsson, S. & Cone, R. (1997) A non-epistatic interaction of the agouti and extensioninthefox,Vulpes vulpes. Nature Genet. 15, 311–315. 16. Lu, D., Va ˚ ge, D. & Roger, D. (1998) A ligand-mimetic model for constitutive activation of the melanocortin-1 receptor. Mol. Endocrinol. 12, 592–604. 17. Valverde, P., Healy, E., Jackson, I., Rees, J.L. & Thody, A.J. (1995) Variants of the melanocyte-stimulating hormone receptor geneareassociatedwithredhairandfairskininhumans.Nature Genet. 11, 328–330. 18. Schio ¨ th, H., Philips, S., Rudzish, R., Birch-Machin, M., Wikberg, J. & Rees, J. (1999) Loss of function mutations of the human melanocortin 1 receptor are common and are associated with red hair. Biochem. Biophys. Res. Commun. 260, 488–491. 19. Rees, J.L., Birch-Machin, M., Flanagan, N., Healy, E., Phillips, S. & Todd, C. (1999) Genetic studies of the human melanocortin-1 receptor. Ann. NY Acad. Sci. 885, 134–142. 20. Takeuchi, S., Suzuki, H., Yabuuchi, M. & Takahashi, S. (1996) A possible involvement of melanocortin 1-receptor in regulating feather color pigmentation in the chicken. Biochim. Biophy Acta 1308, 164–168. 21. Theron, E., Hawkins, K., Bermingham, E., Ricklefs, R. & Mundy, N. (2001) The molecular basis of an avian plumage polymorphism in the wild: a melanocortin-1-receptor point mutation is perfectly associated with the melanic plumage morph of the bananaquit, Coerebra flaveola. Curr. Biol. 11, 1–20. 22. Marklund, U., Bystrom, M., Gedda, K., Larefalk, A., Juneblad, K., Nystrom, S. & Ekstrand, A.J. (2002) Intron-mediated expression of the human neuropeptide YY (1) receptor. Mol. Cell Endocrinol. 188, 85–97. 23. Chhajlani, V. & Wikberg, J. (1992) Molecular cloning and expression of the human melanocyte stimulating hormone receptor cDNA. FEBS Lett. 309, 417–420. 24.Gantz,I.,Konda,Y.,Tashiro,T.,Shimoto,Y.,Miwa,H., Munzert, G., De Watson, S.J.I., Valle, J. & Yamada, T. (1993) Molecular cloning of a novel melanocortin receptor. J. Biol. Chem. 268, 8246–8250. 25. Sazanov, A., Masabanda, J., Eward, D., Takeuchi, S., Tixier- Boichard, M., Buitkamp, J. & Fries, R. (1998) Evolutionarily conseved telomeric localization of BBC1 and MC1R on a micro- chromosome questions the identity of MC1R and a pigmentation locus on chromosome 1 in chicken. Chromosome Res. 6, 651–654. 26. Okimoto, R., Ellet, A.E. & Takeuchi, S. (2000) Melanocortin 1-receptor (MC1-R.) gene polymorphisms associated with chicken E locus alleles. Poultry Sci. 79,9. 27. Va ˚ ge, D., Klungland, H., Lu, D. & Cone, R. (1999) Molecular and pharmacological characterization of dominant black coat color in sheep. Mamm. Genome 10, 39–43. 28. Frandberg, P.A., Doufexis, M., Kapas, S. & Chhajlani, V. (2001) Cysteine residues are involved in structure and function of mela- nocortin 1 receptor: Substitution of a cysteine residue in trans- membrane segment two converts an agonist to antagonist. Biochem. Biophys. Res. Commun. 281, 851–857. 29. Schio ¨ th, H., Yook, P., Muceniec, R., Wikberg, J. & Szardenings, M. (1998) Chimeric melanocortin MC1 and MC3 receptors: 1448 M. K. Ling et al. (Eur. J. Biochem. 270) Ó FEBS 2003 Identification of domains participating in binding of melanocyte- stimulating hormone peptides. Mol. Pharmacol. 54, 154–161. 30. Vaisse, C., Clement, K., Durand, E., Hercberg, S., Guy-Grand, B. & Froguel, P. (2000) Melanocortin-4 receptor mutations are a frequent and heterogeneous cause of morbid obesity. J. Clin. Invest. 106, 253–262. 31. Hinney, A., Schmidt, A., Nottebom, K., Heibult, O., Becker, I., Ziegler,A.,Gerber,G.,Sina,M.,Gorg,T.,Mayer,H.,Siegfried, W., Fichter, M., Remschmidt, H. & Hebebrand, J. (1999) Several mutations in the melanocortin-4 receptor gene including a non- sense and a frameshift mutation associated with dominantly inherited obesity in humans. J. Clin. Endocrinol. Metab. 84, 1483– 1486. 32. Benton, M.J. (1993) The Fossil Record 2, 2nd edn. Chapman & Hall, London, UK. 33. Ringholm, A., Fredriksson, R., Poliakova, N., Yan, Y.L., Postlethwait, J.H., Larhammar, D. & Schioth, H.B. (2002) One melanocortin 4 and two melanocortin 5 receptors from zebrafish show remarkable conservation in structure and pharmacology. J. Neurochem. 82, 6–18. 34. Hartl, D.L. & Clark, A.G. (1997) Principles of Population Gene- tics, 3rd edn. Sinauer, Sunderland, MA, USA. Ó FEBS 2003 Constitutively active MC1 receptors in chicken (Eur. J. Biochem. 270) 1449 . Association of feather colour with constitutively active melanocortin 1 receptors in chicken Maria K. Ling 1 , Malin C. Lagerstro¨m 1 , Robert. FEBS 2003 Constitutively active MC1 receptors in chicken (Eur. J. Biochem. 270) 14 43 Lys in the corresponding position in the human MC1 and MC4 receptors