Protanopia (red color-blindness) in medaka: a simple system for producing color-blind fish and testing their spectral sensitivity

BACKGROUND: Color perception is important for fish to survive and reproduce in nature. Visual pigments in the retinal photoreceptor cells are responsible for receiving light stimuli, but the function of the pigments in vivo has not been directly investigated in many animals due to the lack of color-blind lines and appropriate color-perception tests. METHODS: In this study, we established a system for producing color-blind fish and testing their spectral sensitivity. First, we disrupted long-wavelength-sensitive (LWS) opsins of medaka (Oryzias latipes) using the CRISPR/Cas9 system to make red-color-blind lines. Single guide RNAs were designed using the consensus sequences between the paralogous LWSa and LWSb genes to simultaneously introduce double-frameshift mutations. Next, we developed a non-invasive and no-prior-learning test for spectral sensitivity by applying an optomotor response (OMR) test under an Okazaki Large Spectrograph (OLS), termed the O-O test. We constructed an electrical-rotary cylinder with black/white stripes, into which a glass aquarium containing one or more fish was placed under various monochromatic light conditions. The medaka were irradiated by the OLS every 10 nm, from wavelengths of 700 nm to 900 nm, and OMR was evaluated under each condition. RESULTS: We confirmed that the lws (−) medaka were indeed insensitive to red light (protanopia). While the control fish responded to wavelengths of up to 830 nm (λ = 830 nm), the lws (−) mutants responded up to λ = 740 nm; however, this difference was not observed after adaptation to dark: both the control and lws (−) fish could respond up to λ = 820 ~ 830 nm. CONCLUSIONS: These results suggest that the lws (−) mutants lost photopic red-cone vision, but retained scotopic rod vision. Considering that the peak absorption spectra (λ(max)) of medaka LWSs are about 560 nm, but the light-adapted control medaka could respond behaviorally to light at λ = 830 nm, red-cone vision could cover an unexpectedly wide range of wavelengths, and behavioral tests could be an effective way to measure spectral sensitivity. Using the CRISPR/Cas9 and O-O systems, the establishment of various other color-blind lines and assessment of their spectra sensitivity could be expected to proceed in the future. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s12863-017-0477-7) contains supplementary material, which is available to authorized users.