Jewel Beetle Opsin Duplication and Divergence Is the Mechanism for Diverse Spectral Sensitivities

Abstract

The evolutionary history of visual genes in Coleoptera differs from other well-studied insect orders, such as Lepidoptera and Diptera, as beetles have lost the widely conserved short-wavelength (SW) insect opsin gene that typically underpins sensitivity to blue light (∼440 nm). Duplications of the ancestral ultraviolet (UV) and long-wavelength (LW) opsins have occurred in many beetle lineages and have been proposed as an evolutionary route for expanded spectral sensitivity. The jewel beetles (Buprestidae) are a highly ecologically diverse and colorful family of beetles that use color cues for mate and host detection. In addition, there is evidence that buprestids have complex spectral sensitivity with up to five photoreceptor classes. Previous work suggested that opsin duplication and subfunctionalization of the two ancestral buprestid opsins, UV and LW, has expanded sensitivity to different regions of the light spectrum, but this has not yet been tested. We show that both duplications are likely unique to Buprestidae or the wider superfamily of Buprestoidea. To directly test photopigment sensitivity, we expressed buprestid opsins from two Chrysochroa species in Drosophila melanogaster and functionally characterized each photopigment type as UV- (356-357 nm), blue- (431-442 nm), green- (507-509 nm), and orange-sensitive (572-584 nm). As these novel opsin duplicates result in significantly shifted spectral sensitivities from the ancestral copies, we explored spectral tuning at four candidate sites using site-directed mutagenesis. This is the first study to directly test opsin spectral tuning mechanisms in the diverse and specious beetles.

Keywords: Drosophila; Buprestidae; Coleoptera; Insect vision; Spectral tuning; Visual pigment.

Fig. 1.

Opsin evolution in Buprestidae. (A) Species topology of the beetle family Buprestidae based on Evans et al. (2015) and Cai et al. (2022) showing all subfamilies, with the exception of Galbellinae, which is nested within Chrysochroinae and Buprestinae. All genera whose opsins have been previously described are shown: Agrilus, Aphanisticus, Sphenoptera, Chrysochroa, Steraspis, Chrysobothris, and Acmaeodera, as well as Capnodis and Ptosima (this study). Coraebus is also included as spectral sensitivity has been characterized (Meglič et al. 2020). The superfamily Byrrhoidea has been proposed as sister to superfamily Buprestoidea (Buprestidae + Schizopodidae) (McKenna et al. 2019; Cai et al. 2022). The proposed timings for UV and LW buprestid opsin duplication events are indicated by an arrow and subfamilies with no available opsin or spectral sensitivity data are indicated with a question mark. Asterisks indicate genera where spectral sensitivity have been characterized previously (Coraebus (Meglič et al. 2020)) and in this study (Chrysochroa). ML phylogenetic relationship of buprestid UV1 and UV2 opsins (B) and LW1 and LW2 opsins (C) New sequences from this study are indicated in bold. Node values are UFboot supports based on 10,000 replicates. See supplementary figure S1, Supplementary Material online for the full topology.

Fig. 2.

Spectral sensitivities of cowpea weevil and monarch butterfly visual pigments ectopically expressed in Drosophila. (A) Spectral response of transgenic Drosophila (n = 6) expressing the cowpea weevil Callosobruchus maculatus UV (363 nm, pink line) or LW (511 nm, green line) opsin with fitted visual pigment templates (Stavenga et al. 1993) (dashed lines) and predicted photopigment λ_max. Spectral response of adult C. maculatus is shown in gray (n = 3). (B) Spectral response of transgenic Drosophila (n = 5) expressing monarch butterfly (Danaus plexippus) SW opsin with fitted visual pigment template (Stavenga et al. 1993) (dashed line) and predicted photopigment λ_max. Error shown is standard deviation. See supplementary dataset S1, Supplementary Material online for sensitivity data and fitted templates.

Fig. 3.

Spectral sensitivities of Chrysochroa visual pigments ectopically expressed in Drosophila. Photoreceptor response of transgenic Drosophila expressing Chrysochroa mniszechii (A) and Ch. rajah (B) UV1, UV2, LW1, or LW2 opsins. Mean response and standard deviation (n = 6) (left) and the mean responses fitted with a visual pigment template (right) (Stavenga et al. 1993). The estimated λ_max for each visual pigment is shown (nm). Prior to visual pigment template fitting, mean responses were renormalized between 0 and 1. Data for the LW2 opsin is shown for 315–550 nm and 450–700 nm and the curve was fit to the 315–550 nm testing range (see text). See supplementary dataset S1, Supplementary Material online for sensitivity data and fitted templates.

Fig. 4.

Spectral sensitivities of Chrysochroa rajah mutant photopigments. (A) 3D modeling of Ch. rajah UV2 and LW2 opsins highlighting all sites within the binding pocket surrounding the chromophore. Arrows indicate the sites mutated in this study and the chromophore in A. (B) Spectral sensitivities of Drosophila expressing wild-type UV2 opsin (n = 6), opsin with single mutations Q198A, F285Y and opsin with both mutations (n = 4). (C) Spectral sensitivities of Drosophila expressing wild-type LW2 opsin (n = 6) and opsin with single mutations C140T, V227C and opsin with both mutations (n = 4). Visual pigment templates (Stavenga et al. 1993) were fitted to wild-type UV2 and LW2 spectral responses in the equivalent wavelength testing range to mutant opsins (315–550 nm). The λ_max of fitted templates is also shown (nm). See supplementary dataset S1, Supplementary Material online for sensitivity data and fitted templates.