Monarch and viceroy butterflies are the most famous example of defensive wing color mimicry, but little is known about the visual system of either species. Like the typical adult insect, monarchs and viceroys have color vision mediated by ultraviolet (UV), blue (B) and long (L) wavelength-sensitive visual pigments. Monarch eyes also have heterogeneously-expressed filtering pigments that may extend the range of color vision in the red-green part of the visible spectrum. By contrast, viceroy eyes lack these filtering pigments and are color-blind in the same part of the spectrum, which includes their own orange wing color. In addition, there is significant variation in the absorbance spectrum maximum of the L sensitive visual pigment among this genus of butterflies, with the viceroy L pigment being significantly blue-shifted. Behavioral tests will be used to examine comparative color vision, and wing reflectance data from viceroys and monarchs will be collected and used in the context of mating preference experiments. Behavioral experiments will then be performed to test the hypothesis that the orange-blind viceroys use the UV-reflecting white spots as a cue when selecting mates. Moreover, Drosophila transgenesis and site-directed mutagenesis will functionally evaluate the native visual pigments of these butterflies and use mutant visual pigments to define residues under positive selection. Comparative color vision and mating preference data will provide important biological insights into positive selection on the eyes and color vision system of butterflies for red-green color-blindness. This project will provide an excellent training environment for a postdoctoral associate in multi-level integrated approaches. Undergraduates at UCI from the NSF-funded Minority Science Program student pool will be recruited and trained in molecular biology. Lectures on butterfly vision and behavior will be developed for an upper-division course on insect physiology and made publicly available on the web.
Intellectual merit: Color vision is potentially of adaptive value to animals because color signals may be used to find food, avoid predators, and find mates. Color vision is physiologically based on the presence of spectrally distinct photoreceptors in the eye and appropriate neuronal wiring in the brain comparing the excitations of those photoreceptors. Consequently, to demonstrate that an animal has color vision requires behavioral experiments. Expanded color vision in the yellow to orange range is potentially of adaptive value to monarch butterflies (Danaus plexippus) because they have orange wings, because they need to recognize potential mates, and because their host plants (milkweeds) have yellow and orange flowers. In collaboration with the Weiss lab at Georgetown University, we produced theoretical calculations and conducted behavioral experiments using filtered lights which showed that monarch butterfly eyes contain a fourth undescribed class of photoreceptor, which permits expanded color vision in the yellow to orange range compared to related butterflies. These results, together with prior work in our lab examining the spatial expression of photoreceptor (opsin) mRNAs in the eyes of monarchs allowed us to conclude that this fourth class of photoreceptor is due to a specific physiological mechanism -- the heterogeneous expression of a deep orange filter pigment (i.e., non-opsin) in the eye. This study is the first to examine color vision in monarch butterflies, and adds to the handful of existing studies of butterfly species (Papilio, Pieris, and Heliconius) whose color vision systems have begun to be characterized using anatomical, physiological, molecular, and behavioral experiments. Butterfly color vision is mainly based on three classes of visual pigments with peak sensitivity in the ultraviolet (UV, 300-400 nm), blue (B, 400-500 nm) and long wavelength (LW, 500-600 nm) range of the light spectrum. Visual pigments are light-sensitive molecules composed of a chromophore and an opsin protein that is encoded by UV, B, or LW opsin gene. The compound eyes of most butterflies including monarch butterflies contain three opsin genes. However, some butterflies have four or more opsins as a result of gene duplication. In a second study, we found that Heliconius erato is one of those butterflies which has two UV opsin genes. To trace the evolutionary origin of this gene, we searched for eye opsin genes from different passion-vine butterflies, and found that the UV opsin gene duplication only occurs in the genus Heliconius but not in close butterfly relatives. Using statistical tests, we showed that this gene has evolved under positive selection following gene duplication. Consistently, our functional study of the eyes of H. erato revealed the presence of two UV-absorbing visual pigments in the adult compound eye. Along with the duplicated UV opsin gene, we have also discovered that Heliconius butterflies use a wing pigment, 3-hydroxykynurenine (3-OHK), that looks yellow to human eyes but can reflect both UV and long-wavelength light. In contrast, the yellow wing pigments from close relatives do not contain 3-OHK and only reflect long-wavelength light. We next modeled the color signals to examine theoretically how butterflies perceive wing color variation. Visual models of how butterflies perceive wing color variation indicate this has resulted in an expansion of the number of distinguishable yellow colors on Heliconius wings. In conclusion, our study suggests that UV-yellow wing pigments (color signal) evolved in a correlated manner with a UV opsin gene duplication (color vision). To follow up, we investigated the molecular and physiological context of this adaptive event (i.e. the UV gene duplication). We examined the evolutionary forces shaping the UV opsin genes in contrast to B and LW opsin genes across a diversity of passion-vine butterfly species. We also estimated the wavelength of peak sensitivity of visual pigments in some of the investigated species. From the statistical and physiological results, we not only confirmed our prior finding of positive selection of the duplicated UV opsin gene using a more conservative statistical test but also provided new evidence for purifying selection on the other three eye opsins of Heliconius. Broader impacts: Personnel trained by this project include three postdoctoral researchers, two PhD students, and three undergraduate researchers. One of the undergraduate researchers has now started a Ph.D. program in Computational Biology at Cornell University. Results of this research were featured in U.S. News and World Report and a variety of other news venues.
