American Journal of Primatology 67:463–470 (2005) RESEARCH ARTICLE Color Vision Pigment Frequencies in Wild Tamarins (Saguinus spp.) A.K. SURRIDGE1, S.S. SUÁREZ2, H.M. BUCHANAN-SMITH3, A.C. SMITH3,4, AND N.I. MUNDY5 1 School of Biological Sciences, University of East Anglia, Norwich, England 2 Department of Anthropology, New York University and the New York Consortium in Evolutionary Primatology (NYCEP), New York, New York 3 Scottish Primate Research Group, Department of Psychology, University of Stirling, Stirling, Scotland 4 Department of Life Sciences, Anglia Polytechnic University, Cambridge, England 5 Department of Zoology, University of Cambridge, Cambridge, England The adaptive importance of polymorphic color vision found in many New World and some prosimian primates has been discussed for many years. Polymorphism is probably maintained in part through a heterozygote advantage for trichromatic females, as such individuals are observed to have greater foraging success when selecting ripe fruits against a background of forest leaves. However, recent work also suggests there are some situations in which dichromatic individuals may have an advantage, and that variation in color vision among individuals possessing different alleles may also be significant. Alleles that confer a selective advantage to individuals are expected to occur at a higher frequency in populations than those that do not. Therefore, analyzing the frequencies of color vision alleles in wild populations can add to our understanding of the selective advantages of some color vision phenotypes over others. With this aim, we used molecular techniques to determine the frequencies of color vision alleles in 12 wild tamarin groups representing three species of the genus Saguinus. Our results show that allele frequencies are not equal, possibly reflecting different selective regimes operating on different color vision phenotypes. Am. J. Primatol. 67:463–470, 2005. c 2005 Wiley-Liss, Inc. Key words: balancing selection; color vision; photopigment; polymorphism; tamarin Contract grant sponsor: BBSRC; Contract grant number: 98/S11498; Contract grant sponsor: J. William Fulbright Foreign Scholarship Board; Contract grant number: 1999/2000; Contract grant sponsor: Wenner-Gren Foundation; Contract grant number: Gr. 6560; Contract grant sponsor: L.S.B. Leakey Foundation; Contract grant sponsor: American Society of Primatologists. Correspondence to: A.K. Surridge, School of Biological Sciences, University of East Anglia, Norwich NR4 7TJ, England. E-mail: email@example.com Received 27 October 2004; revised 18 February 2005; revision accepted 10 March 2005 DOI 10.1002/ajp.20200 Published online in Wiley InterScience (www.interscience.wiley.com). r 2005 Wiley-Liss, Inc. 464 / Surridge et al. INTRODUCTION Eutherian mammals have a poor color sense compared to many birds and reptiles. This is probably due to a loss of photopigments during the nocturnal phase of their evolution. Primates, however, have uniquely re-evolved a form of color vision. We have a good understanding of how this trichromatic color vision has evolved in primates, as many of the molecular and physiological adaptations accompanying this change are known (for a recent review see Jacobs and Rowe ). What is less well understood are the selective processes that led to the evolution of color vision, and the reasons why primates alone evolved this sense. The advantages offered by red-green color vision may involve a superior foraging ability, as it appears that primate photopigments are optimized for detecting both ripe fruit and young leaves against a dappled background of mature leaves [Dominy & Lucas, 2001; Lucas et al., 1998; Osorio & Vorobyev, 1996; Regan et al., 2001; Sumner & Mollon, 2000a,b]. Color vision in different primate species is remarkably variable. Primates with three different opsin proteins (one short-wavelength sensitive (SWS) and two middle- to long-wavelength sensitive (M- to LWS)) expressed in separate classes of cone photoreceptors in the retina have trichromatic vision. Those with two different opsin proteins (one SWS and one M- to LWS) have dichromatic vision. Trichromacy has been achieved in two ways. In Old World primates (OWP) and howler monkeys (Alouatta spp.), a gene duplication of the X-linked M- to LWS locus has occurred. In most New World primates (NWP) and some prosimians, a single X-linked M- to LWS locus has two or more alleles, giving heterozygous females trichromacy [Jacobs, 1984; Mollon et al., 1984]. However, in these species all males and homozygous females have dichromatic vision. This polymorphism within species allows us to test hypotheses regarding the selective advantages of color vision in primates by comparing the foraging abilities of individuals with different vision phenotypes. Experiments on captive marmosets have revealed a foraging advantage for trichromatic individuals based on food color [Caine & Mundy, 2000; Caine et al., 2003]. The same result was observed in a study of captive tamarins that used naturalized color stimuli based on those encountered in the wild [Smith et al., 2003a]. These studies reiterate the advantages of trichromacy over dichromacy for finding food, and provide support for hypotheses evoking a role for heterozygote advantage, as a form of balancing selection, in maintaining the polymorphic color vision system in platyrrhines [Mollon et al., 1984]. However, this polymorphic color vision system found in some primates still represents a puzzle [Surridge et al., 2003]. Opsin protein alleles have remained unchanged, in both function and number, over long periods of evolutionary time in the face of both selection and genetic drift [Surridge & Mundy, 2002]. An inevitable conclusion from this is that photopigments, finely tuned by selection, represent some form of adaptive peak that maintains a population containing both dichromats and trichromats. This notion is difficult to reconcile with the hypothesis that trichromacy is generally better than dichromacy. An alternative (but not mutually exclusive) explanation for the stability of the polymorphism is that frequency-dependent selection operates on different color vision phenotypes in NWP [Mollon et al., 1984; Osorio et al., 2004]. Under this model, allele frequencies could change over space and time, depending on the availability of different food resources. Alleles that confer a selective advantage to individuals are expected to occur at a higher frequency than those that do not. Am. J. Primatol. DOI 10.1002/ajp Opsin Allele Frequencies in Wild Tamarins / 465 There is some preliminary evidence that allele frequencies are not equal in callitrichid primates (marmosets and tamarins). In these species, three opsin alleles are observed, with the intermediate allele (with lmax 5 556 nm) being found at a lower frequency than the two extreme alleles (543 and 563 nm) in previous studies [Rowe & Jacobs, 2004; Surridge & Mundy, 2002]. However, many of the individuals used in those studies were housed in captive populations, and small initial population sizes and inbreeding may result in allele frequencies that do not resemble those found in wild populations. In the present study we determined opsin allele frequencies in wild callitrichids (taxonomy following Rylands et al. ) to obtain deeper insights into the fascinating color vision polymorphism of NWP. MATERIALS AND METHODS Sampling Two mixed-species groups of saddleback (Saguinus fuscicollis) and moustached tamarins (S. mystax) were observed for 164 full days (1,612 hr) from January 2000 until December 2000 at the Estación Biológica Quebrada Blanco II in northeastern Peru (41210 S, 731090 W). During this time fecal samples were collected from individually identifiable animals for DNA extraction (for further details see Smith et al. [2003b]). The samples were stored in 100% ethanol at ambient temperature. Eight neighboring groups of red-bellied moustached tamarins (S. labiatus) were live-trapped and sedated for genetic sample collection between the months of April and June 2000 as part of a larger, 18-month behavioral field study at the Tahuamanu Biological Station located in the Department of Pando, northwestern Bolivia (111240 S, 69110 W). Hair samples were tweezed from each animal and stored at ambient temperature until DNA was extracted. DNA Analysis and Determination of Color-Vision Genotype DNA was extracted from fecal samples with the use of a QIAamp DNA Stool Kit (Qiagen, Crawley, UK), and from hair samples with a QIAamp DNA Minikit (Qiagen, Crawley, UK) following the manufacturer’s recommended protocol. In callitrichids, differences in spectral tuning among the three functional alleles (543, 556, and 563 nm) have been attributed to substitutions at four amino acid positions: position 180 in exon 3, positions 229 and 233 in exon 4, and position 285 in exon 5 [Shyue et al., 1998]. Although a recent report showed that positions in exon 4 do not contribute significantly to functional changes in spectral alleles [Hiramatsu et al., 2004], this exon was included in our typing process for thoroughness. The combinations of amino acids at each of these respective sites for each allele are as follows: 543 nm 5 Ala Ile Ser Ala; 556 nm 5 Ala Phe Ser Thr; and 563 nm 5 Ser Phe Gly Thr. Color-vision status was determined for each individual by polymerase chain reaction (PCR) and sequencing of exons 3, 4, and 5 of the X-linked opsin gene. The nucleotide substitutions that code for each amino acid as described above were determined directly from DNA sequence traces. Because the material used for DNA extraction (feces and hair) yield low concentrations of genomic DNA, each of the three exons were amplified individually, with the combination of nucleotides of all three used to determine the genotype. All three exons were amplified in order to obtain three independent amplifications (and therefore three replicates) per individual to eliminate spurious results due to genotyping errors. The entire procedure (extraction, Am. J. Primatol. DOI 10.1002/ajp 466 / Surridge et al. amplification, and sequencing) was performed twice for fecal samples (on two separately collected samples) to provide six replicates per individual and to eliminate the chance of individuals being incorrectly assigned as homozygotes due to allelic dropout (for further details see Surridge et al. ). PCR and sequencing were performed as described previously [Surridge & Mundy, 2002; Surridge et al., 2002]. The observed allele frequencies were compared with those expected (assuming that the frequency of each allele was equal in each population) by means of Chi-square goodness-of-fit tests. RESULTS Color vision genotypes were obtained for 71 individuals (49 males and 22 females), yielding a total of 93 X chromosomes examined. The frequency of each allele for each species is shown in Fig. 1. For S. mystax (15 X chromosomes in 12 individuals) the 543 nm allele is the most common in our sample, with the 556 nm allele being absent. The numbers of each allele detected differ significantly from that expected if the allele frequencies had been equal (w2 5 6.23, Po0.05). For S. fuscicollis (17 X chromosomes in 12 individuals) the 556 nm allele is the most common allele; however, the numbers of each allele detected do not differ from the values expected with equal allele frequencies. For S. labiatus (61 X chromosomes in 47 individuals) the 543 nm allele is again the most common allele, with the 556 nm allele being the least common, and allele counts show a significant deviation from equal-frequency expectations (w2 5 9.40, Po0.01). We found no evidence for recombinant alleles in our sample of homozygous individuals, in agreement with previous data for callitrichids [Surridge & Mundy, 2002]. However, because exons are amplified individually in the typing process, haplotypes cannot be determined in heterozygous individuals. Putative recombinant alleles have been detected in capuchins [Boissinot et al., 1998] and squirrel monkeys [Cropp et al., 2002]. S. mystax (n=15) Saguinus labiatus (n=61) 1 1 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 543nm 556nm 563nm Allele 543nm 556nm 563nm Allele S. fuscicollis (n=17) 1 0.8 0.6 0.4 0.2 0 543nm 556nm 563nm Allele Fig. 1. Frequencies of the three opsin alleles (with lmax at 543, 556, and 563 nm) in three species of wild tamarins (Saguinus spp.). N 5 number of X-chromosomes examined. Am. J. Primatol. DOI 10.1002/ajp Opsin Allele Frequencies in Wild Tamarins / 467 Saguinus spp. (n=137) Callithrix spp. (n=25) 1 1 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 543nm 556nm 543nm 563nm 556nm 563nm Allele Allele Leontopithecus spp. (n=32) Callimico goeldii (n=39) 1 1 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 543nm 556nm 563nm 543nm 556nm 563nm Allele Allele Saimiri spp. (n=261) 1 0.8 0.6 0.4 0.2 0 535nm 550nm 561nm Allele Fig. 2. Frequencies of opsin alleles determined in previous studies. Data for Saguinus, Callithrix, Callimico, and Leontopithecus are taken from Surridge and Mundy . Data for Saimiri are from Cropp et al. . N 5 number of X-chromosomes examined. For comparison, the allele frequencies for each genus of callitrichid sampled in our previous study [Surridge & Mundy, 2002], together with those from a study of squirrel monkeys (Saimiri spp. [Cropp et al., 2002]), are given in Fig. 2. For Saguinus, Leontopithecus, and Callimico the observed number of each allele differs significantly from that expected for equal frequency alleles, with the 556 nm allele being the rarest allele (the 556 nm allele was completely absent from our sample of 39 Callimico X chromosomes) in all cases. In contrast, the 556 nm allele was the most common allele in Callithrix; however, the number of observed alleles did not differ from that expected for equal allele frequencies for either Callithrix or Saimiri. DISCUSSION Allele Frequencies in Wild Tamarins Are Not Equal Our sample of 93 wild tamarin X-chromosomes indicates that the frequencies of opsin alleles are not equal. Some variation in allele frequencies across the three species was observed, and notably the rarest allele (556 nm) in S. labiatus and Am. J. Primatol. DOI 10.1002/ajp 468 / Surridge et al. S. mystax was the most common allele in S. fuscicollis. S. fuscicollis and S. mystax form mixed-species troops, with each species foraging preferentially in different strata in the tropical rain-forests (S. fuscicollis is found at lower heights than its congeners [Heymann & Buchanan-Smith, 2000]). This may account, in part, for the differences in allele frequencies between the two species (illustrated in Fig. 1). However, because of the relatively small number of chromosomes examined in S. mystax and S. fuscicollis, together with the limited range of geographical sampling, these results should be interpreted with some caution. Nevertheless, when all three species are combined, our data from wild tamarins are in agreement with that of our previous study of captive tamarins in revealing the 556 nm allele to be the rarest allele. In our sample of wild tamarins, the frequencies of each allele were as follows: 543 nm 0.53, 556 nm 0.17, and 563 nm 0.30. The allele frequencies of the tamarin species in our previous study were 0.30, 0.15, and 0.55, respectively [Surridge & Mundy, 2002]. Over all callitrichid primates analyzed to date, the average allele frequencies are 0.39, 0.17, and 0.44 (for 326 X-chromosomes). Do Some Photopigments Have a Selective Advantage Over Others? Our finding that the intermediate 556 nm allele is the least common allele in wild tamarins provides some evidence that it may also be the allele that is least favored by selection. This is in agreement with a study showing that the 543/ 563 nm trichromat is expected to outperform either of the intermediate trichromats (543/556 nm or 556/563 nm) at fruit detection, irrespective of the effects of variable illumination [Osorio et al., 2004]. This effect was demonstrated in a study of human observers in which visual stimuli were chosen to simulate cone excitations as observed in the cones of callitrichids viewing fruit [Rowe & Jacobs, 2004]. Similarly, in spider monkeys (in which alleles are of slightly different lmax), modeling of fruit detection by different color-vision phenotypes has shown that the 535/562 nm trichromat performs better than the intermediate 535/550 nm and 550/562 nm trichromats [Riba-Hernández et al., 2004]. Despite this apparent selective disadvantage for the intermediate opsin allele, it has been maintained for several millions of years in callitrichids (with the possible exception of Callimico, for which more data are needed to confirm the presence or absence of this allele in the wild [Surridge & Mundy, 2002]). There are two possible explanations for this. First, if trichromats always have an advantage over dichromats, this allele would be maintained through a heterozygote advantage for trichromats carrying the allele (even though it may not be favored among trichromats). Second, there may be an advantage for longerwavelength alleles in dichromats [Osorio et al., 2004]. Therefore, dichromats carrying the 556 nm allele could be favored over those carrying the 543 nm allele. Hence, this system with three alleles is balanced by selection for widely spaced photopigments in trichromats and longer wavelength pigments in dichromats. Furthermore, the presence of three alleles may represent a compromise between a selective advantage of polymorphism whereby more alleles result in more heterozygous individuals that are fitter, and a functional constraint on the opsin protein. Frequency-Dependent Selection and the Importance of Trichromacy to NWP If different color vision phenotypes have different levels of foraging fitness depending on subtle differences in color and light [Osorio et al., 2004; Regan Am. J. Primatol. DOI 10.1002/ajp Opsin Allele Frequencies in Wild Tamarins / 469 et al., 2001], then individuals may forage within visually distinct niches [Tovée et al., 1992], reducing competition from conspecifics. In this situation, allelic polymorphism would be maintained by frequency-dependent selection, such that alleles are most beneficial when they are rare. It is interesting to note that such selection may operate above the species level. When allele frequencies of species that form polyspecific associations (in this case S. fuscicollis and S. mystax) are combined, they do not differ significantly from equal-frequency expectations (w2 5 3.84; Pr0.20). Hence, alleles that might be favored in one species are balanced by the favoring of different alleles in another species. Further support for different roles for different phenotypes comes from the observation that trichromatic marmosets perform poorly when faced with the task of detecting camouflaged food items [Caine et al., 2003]. In the wild, cryptic food items (e.g., camouflaged insects) that are overlooked by trichromats may be detected by conspecific dichromats [Mollon et al., 1984]. The role of cooperative behavior within this system is obscure. To date there is no evidence to suggest that other members of a group benefit from another individual’s trichromacy through social foraging. In a study of foraging behavior of tamarins, trichromatic females did not consistently lead the group to fruiting trees [Smith et al., 2003b]. In summary, while frequency-dependent selection would act to maintain opsin polymorphism through long periods of time, differing opsin allele frequencies support lines of evidence that suggest that some color vision alleles have a selective advantage over others at this present point in evolutionary time [Osorio et al., 2004]. Subtle differences in a species’ ecology might shift these selective advantages, with a corresponding shift in allele frequencies. Further studies on larger spatial and temporal scales are required to test this. Notably, no correlates between opsin allele frequencies and biogeography or behavior were detected in a study of squirrel monkeys [Cropp et al., 2002]. In addition, genetic drift, migration, and social group composition (including mate choice and inbreeding avoidance) may have important influences on allele frequencies. This may be deduced from fine-scale analyses of wild primate groups. A comparative study of foraging ecology across different platyrrhine genera (such as tamarins and squirrel monkeys) with opsin alleles of different spectral sensitivities could provide further insight into this fascinating color vision system. ACKNOWLEDGMENTS We thank two anonymous reviewers for their comments on the manuscript. This work was funded by the BBSRC (98/S11498 to H.M.B.-S. and N.I.M.). Funding for fieldwork on Saguinus labiatus was provided by the J. William Fulbright Foreign Scholarship Board (1999/2000), the Wenner-Gren Foundation (grant no. Gr. 6560, 1999/2000), the L.S.B. Leakey Foundation (1999/2000), and the American Society of Primatologists (Conservation Small Grant, 1999/2000). Special thanks go to field assistants Stephanie Dammermann, Leeann Haggerty, Rina Aviram, Laura Miller, Edilio Nacimiento, and Rafael Suárez. We are grateful to Dr. E. Montoya of the Proyecto Peruano de Primatologia and Biologo R. Pezo of the Universidad Nacional de la Amazonia Peruana (UNAP), Iquitos, Peru, for help and support with logistical matters. Particular thanks are due to Ney Shahuano, who provided unflagging assistance in the field. REFERENCES Boissinot S, Tan Y, Shyue S-K, Schneider H, Sampaio I, Neiswanger K, Hewett-Emmett D, Li W-H. 1998. Origins and antiquity of X-linked triallelic color vision systems in Am. J. Primatol. DOI 10.1002/ajp 470 / Surridge et al. New World monkeys. Proc Natl Acad Sci USA 95:13749–13754. Caine NG, Mundy NI. 2000. Demonstration of a foraging advantage for trichromatic marmosets (Callithrix geoffroyi) dependent on food colour. Proc R Soc Lond B Biol Sci 267: 439–444. Caine NG, Surridge AK, Mundy NI. 2003. Dichromatic and trichromatic marmosets (Callithrix geoffroyi) differ in relative foraging ability for red-green colour-camouflaged and non-camouflaged food. Int J Primatol 24:1163–1175. Cropp S, Boinski S, Li W-H. 2002. Allelic variation in the squirrel monkey X-linked colour vision gene: biogeographical and behavioural correlates. J Mol Evol 54: 734–745. Dominy NJ, Lucas PW. 2001. Ecological importance of trichromatic vision to primates. Nature 410:363–366. Heymann E, Buchanan-Smith HM. 2000. The behavioural ecology of mixed-species troops of callitrichine primates. Biol Rev 75: 169–190. Hiramatsu C, Radlwimmer FB, Yokoyama S, Kawamura S. 2004 Mutagenesis and reconstitution of middle-to-long-wave-sensitive visual pigments of New World monkeys for testing the tuning effect of residues at sites 229 and 233. Vision Res 44:2225–2231. Jacobs GH. 1984. Within-species variations in visual capacity among squirrel monkeys (Saimiri sciureus): colour vision. Vision Res 24:1267–1277. Jacobs GH, Rowe MP. 2004. Evolution of vertebrate color vision. Clin Exp Optom 87: 206–216. Lucas PW, Darvell BW, Lee PKD, Yuen TDB, Choong MF. 1998. Colour cues for leaf food selection by long-tailed macaques (Macaca fascicularis) with a new suggestion for the evolution of trichromatic colour vision. Folia Primatol 69:139–152. Mollon JD, Bowmaker JK, Jacobs GH. 1984. Variations of colour vision in a New World primate can be explained by polymorphism of retinal photopigments. Proc R Soc Lond B Biol Sci 222:373–399. Osorio D, Vorobyev M. 1996. Colour vision as an adaptation to frugivory in primates. Proc R Soc Lond B Biol Sci 263:593–599. Osorio D, Smith AC, Vorobyev M, BuchananSmith HM. 2004. Detection of fruit and the selection of primate visual pigments for colour vision. Am Nat 164:696–708. Regan BC, Julliot C, Simmen B, Vienot F, Charles-Dominique PC, Mollon JD. 2001. Fruits, foliage and the evolution of primate Am. J. Primatol. DOI 10.1002/ajp colour vision. Proc R Soc Lond B Biol Sci 356:229–283. Riba-Hernández P, Stoner KE, Osorio D. 2004. Effect of polymorphic colour vision for fruit detection in the spider monkey Ateles geoffroyi, and its implications for the maintenance of polymorphic colour vision in platyrrhine monkeys. J Exp Biol 207: 2465–2470. Rowe MP, Jacobs GH. 2004. Cone pigment polymorphism in New World monkeys: are all pigments created equal? Vis Neurosci 21: 217–222. Rylands AB, Schneider H, Langguth A, Mittermeier RA, Groves CP, Rodriguez-Luna E. 2000. An assessment of the diversity of New World primates. Neotrop Primates 8:61–93. Shyue S-K, Boissinot S, Schneider H, Sampaio I, Schneider MP, Abee CR, Williams L, et al. 1998. Molecular genetics of spectral tuning in New World monkey colour vision. J Mol Evol 46:697–702. Smith AC, Buchanan-Smith HM, Surridge AK, Osorio D, Mundy NI. 2003a. The effect of colour vision status on the detection and selection of fruits by tamarins (Saguinus spp.). J Exp Biol 206:3159–3165. Smith AC, Buchanan-Smith HM, Surridge AK, Mundy NI. 2003b. Leaders of progressions in wild mixed-species troops of saddleback (Saguinus fuscicollis) and mustached tamarins (S. mystax), with emphasis on color vision and sex. Am J Primatol 61: 145–157. Sumner P, Mollon JD. 2000a. Catarrhine photopigments are optimised for detecting targets against a foliage background. J Exp Biol 203:1963–1986. Sumner P, Mollon JD. 2000b. Chromaticity as a signal of ripeness in fruits taken by primates. J Exp Biol 203:1987–2000. Surridge AK, Mundy NI. 2002. Trans-specific evolution of opsin alleles and the maintenance of trichromatic colour vision in Callitrichine primates. Mol Ecol 11: 2157–2170. Surridge AK, Smith AC, Buchanan-Smith HA, Mundy NI. 2002. Single-copy nuclear DNA sequences obtained from non-invasively collected primate feces. Am J Primatol 56: 185–190. Surridge AK, Osorio D, Mundy NI. 2003. Evolution and selection of colour vision in primates. Trends Ecol Evol 18:198–205. Tovée MJ, Bowmaker JK, Mollon JD. 1992. The relationship between cone pigments and behavioural sensitivity in a New World monkey (Callithrix jacchus jacchus). Vision Res 32:867–878.