| dcterms.abstract | Color vision in primates is a subject that has long been of interest to evolutionary biologists and anthropologists, largely owing to the unique capacity for trichromatic color vision in humans and many other primates (e.g., Jacobs 1981, 1993). Trichromatic color vision affords individuals the ability to make additional, more salient chromatic distinctions between red and green hues that most other placental mammals, which are primarily dichromatic (i.e., red-green colorblind), are unable to make (e.g., Jacobs 1981, 1993). This ability has been tied to two genetic mechanisms, making primate color vision an illustrative case study in molecular evolution. Humans and other catarrhines, as well as New World howling monkeys (Alouatta), have one autosomal, short-wavelength (S) opsin gene, and two opsin genes on the X chromosome, one resulting in sensitivity to medium wavelengths of light (M) and the other to long wavelengths of light (L), allowing for routine trichromacy (Kainz et al. 1998; Dulai et al. 1999; Jacobs and Deegan 1999; Nathans 1999). Many platyrrhines and lemurs have one autosomal, S opsin gene and one opsin gene on the X chromosome, for which there are two alleles (M and L), allowing for dichromatic color vision (e.g., Surridge et al. 2003). In some taxa, the X-linked opsin gene is polymorphic, providing heterozygous females the potential for trichromacy, while males and homozygous females are red-green color blind (Jacobs 1998; Tan and Li 1999; Jacobs and Deegan 2003). Trichromatic color vision appears to characterize all diurnal catarrhines and platyrrhines through one mechanism or another (reviewed in Surridge et al. 2003; Jacobs 2007, 2008, 2009; Kawamura et al. 2012), but the same cannot be said of lemuriforms, in which polymorphic trichromacy has been identified in some diurnal and cathemeral species, while many others are strictly dichromatic (e.g., Tan and Li 1999; Leonhardt et al. 2009; Veilleux and Bolnick 2009; Bradley et al. 2009). Among diurnal haplorhines, trichromatic color vison has long been thought to result from positive selection favoring trichromacy over dichromacy (e.g., see Surridge et al. 2003 for review). Multiple hypotheses have been proposed suggesting various fitness-related tasks for which trichromatic color vision might be advantageous (e.g., Osorio and Vorobyev 1996; Lucas et al. 1998, 2003; Coss and Ramakrishnan 2000; Pessoa et al. 2014); the most long-standing hypothesis suggests advantages are conferred during foraging, primarily when foraging on red food items (e.g., fruit or young leaves; e.g., Allen 1879; Mollon 1989; Lucas et al. 1998, 2003). This hypothesis has received some support from studies that have modeled color perception of dietary items for different color vision phenotypes (e.g., Osorio and Vorobyev 1996; Lucas et al. 1998, 2003; Sumner and Mollon 2000a, b; Dominy and Lucas 2001; Regan et al. 2001; Osorio et al. 2004). In addition, trichromatic foraging advantages have been observed in captive settings (e.g., Caine and Mundy 2000; Smith et al. 2003), but there is limited evidence for such advantages in wild populations (e.g., Vogel et al. 2007; Hiramatsu et al. 2008; Melin et al. 2008, 2009). Although the selective pressure(s) favoring trichromatic color vision is unknown, the apparent near ubiquity of trichromacy in haplorhines suggests there is positive selection (e.g., Surridge et al. 2003), and molecular studies have identified signatures of balancing selection to maintain color vision variation in platyrrhines (Hiwatashi et al. 2010; Kawamura et al. 2012). Color vision in lemurs has been comparatively understudied, but similar adaptive explanations related to foraging behavior have been proposed to account for color vision polymorphisms in some species (Leonhardt et al. 2009). That said, there is a large amount of variation in color vision capacities across this lineage, and it is unclear why some lemur species have polymorphic trichromatic color vision, while other, closely related species are dichromatic (e.g., Tan and Li 1999; Tan et al. 2005; Bradley et al. 2009; Leonhardt et al. 2009; Veilleux and Bolnick 2009). Thus, the overall goal of this dissertation was to identify potential evolutionary mechanisms that might account for observed color vision variation in lemurs. In so doing, this dissertation had three objectives: 1) to characterize the color vision capacity of a population of red-bellied lemurs (Eulemur rubriventer) in Ranomafana National Park (RNP) in southeastern Madagascar, 2) to examine color vision evolution in the genus Eulemur using phylogenetic methods to estimate ancestral color vision state, and 3) to explore potential nonadaptive and adaptive explanations for the type of color vision observed in this population of E. rubriventer. To address the first objective, I sequenced exon 5 of the X-linked M/L opsin gene for 87 individual red-bellied lemurs (NX chromosomes = 134). I found that this population is strictly dichromatic and has a single M/L opsin variant with peak spectral sensitivity at 558 nm (i.e., L opsin is fixed; chapter 2). When placed in a comparative context, this result identifies E. rubriventer as unique among other species of Eulemur, for which data are available; some species/populations of Eulemur are dichromatic, but the peak spectral sensitivity of the M/L opsin is 543 nm (i.e., M opsin is fixed), while other species/populations are polymorphic, with both M and L opsins present within a population (Tan and Li 1999; Bradley et al. 2009; Leonhardt et al. 2009; Veilleux and Bolnick 2009; Bradley et al. in prep). To better understand the evolutionary history of color vision in this lineage, I compiled data on color vision phenotypes for strepsirrhines and mapped these data onto two time-calibrated phylogenetic trees (chapter 2). I then used a maximum likelihood approach to infer the ancestral state of Eulemur. Overall, results suggest that an M/L opsin polymorphism was likely the ancestral Eulemur condition. Therefore, this result suggests that the population of E. rubriventer in RNP likely lost polymorphic trichromatic color vision. Given that trichromatic color vision in other primates is thought to be adaptive, this begs the question of why a potentially advantageous trait would be lost from this population. To address this question (objective 3), I explored the potential for a recent genetic bottleneck in the population of red-bellied lemurs in RNP (chapter 3). Madagascar has suffered from recent and large-scale forest loss (e.g., Harper et al. 2007), which, combined with other threats such as hunting (e.g., Schwitzer et al. 2014), has resulted in population declines, and genetic bottlenecks in a number of lemur species (e.g., Fredsted et al. 2007; Olivieri et al. 2008; Craul et al. 2009; Brenneman et al. 2012; Parga et al. 2012; Holmes et al. 2013). Genetic bottlenecks provide a potential nonadaptive mechanism through which genetic variation can be lost, because the impact of genetic drift increases in small populations, such that the strength of drift can be greater than selection and even result in loss of advantageous alleles (Futuyma 1998). Using genotypes for 7 variable microsatellite loci from 55 adult red-bellied lemurs, I found that the E. rubriventer population in RNP exhibited significant heterozygosity excess compared to mutation-drift equilibrium, which suggests this population may have experienced a recent population bottleneck. However, I also found that this population did not exhibit significantly low M ratios (i.e., ratio of number of alleles to range in allele size) compared to mutation-drift equilibrium, which is not indicative of a genetic bottleneck. Taken together, these results provide mixed evidence that there was a recent genetic bottleneck in this population, and, therefore, this hypothesis cannot be rejected. Thus, polymorphic trichromatic color vision may have been lost through nonadaptive mechanisms in the population of E. rubriventer in RNP, and the L opsin may be fixed in this population as a result of genetic drift. Although nonadaptive mechanisms might play a role in the evolution of color vision in this population, alternatively/additionally, there may be adaptive explanations (objective 3). In chapter 4, I addressed foraging hypotheses that might result in relaxed or disruptive selection (i.e., selection against trichromacy) on polymorphic color vision. I also addressed the potential for directional selection favoring the L opsin in red-bellied lemurs. Using color modeling methods, I compiled reflectance data from 40 species including 72 plant parts consumed by red-bellied lemurs in RNP and modeled how these food items would be perceived by a trichromatic and dichromatic Eulemur. I used this approach to first address the hypothesis that food items consumed by E. rubriventer are primarily “dull†in coloration (e.g., green and brown; Dew and Wright 1998; Birkinshaw 2001), and therefore largely inconspicuous to a trichromat, which could result in relaxed selection to maintain polymorphic trichromatic color vision. I found that red-green chromaticities (only available to a trichromat) of many food items, particularly ripe fruit, were greater than chromaticities of background foliage, suggesting that trichromatic color vision, theoretically, offers a potential advantage in detecting many food items. I then modeled the chromatic and luminance contrasts of each food item from its leaf background as perceived by the two Eulemur dichromatic phenotypes (L opsin vs. M opsin) to determine if chromatic contrasts were significantly greater for the L opsin, which could result in directional selection favoring the L opsin over the M opsin. I found that chromatic contrasts were significantly greater for dichromats with the M opsin, but luminance contrasts were significantly greater for dichromats with the L opsin. This result suggests that E. rubriventer may rely on luminance cues during foraging, which could lead to relaxed selection to maintain polymorphic trichromacy or even selection against trichromatic color vision, as chromatic information may interfere with luminance vision (Osorio et al. 1998). Either of the two mechanisms could result in loss of polymorphic color vision. At the same time, fixation of the L opsin may be adaptive for maximizing luminance contrast and may have been driven to fixation through directional or purifying selection, suggesting a potential adaptive explanation for color vision in the E. rubriventer population in RNP. In sum, the results of this dissertation suggest that color vision evolution in lemur populations is likely complex and may be influenced by both adaptive and nonadaptive mechanisms. Therefore, caution is warranted in assuming that observed color vision variation results from adaptation alone. Ultimately, evolutionary processes likely vary across lineages, species, and populations, and identifying how they vary will be important for understanding how evolution shaped the diversity in lemur color vision that we see today. | |
