Altered Visual Function in Short-Wave-Sensitive 1 (sws1) Gene Knockout Japanese Medaka (Oryzias latipes) Larvae

Visual perception plays a crucial role in foraging, avoiding predators, mate selection, and communication. The regulation of color vision is largely dependent on opsin, which is the first step in the formation of the visual transduction cascade in photoreceptor cells. Short-wave-sensitive 1 (sws1) is a visual pigment that mediates short-wavelength light transduction in vertebrates. The depletion of sws1 resulted in increased M-opsin in mice. However, there is still no report on the visual function of sws1 in teleost fish. Here, we constructed the sws1 knockout medaka using CRISPR/Cas9 technology. The 6 dph (days post-hatching) medaka sws1−/− larvae exhibited significantly decreased food intake and total length at the first feeding stage, and the mRNA levels of orexigenic genes (npy and agrp) were significantly upregulated after feeding. The swimming speed was significantly reduced during the period of dark-light transition stimulation in the sws1-mutant larvae. Histological analysis showed that the thickness of the lens was reduced, whereas the thickness of the ganglion cell layer (GCL) was significantly increased in sws1−/− medaka larvae. Additionally, the deletion of sws1 decreased the mRNA levels of genes involved in phototransduction (gnb3b, grk7a, grk7b, and pde6c). We also observed increased retinal cell apoptosis and oxidative stress in sws1 knockout medaka larvae. Collectively, these results suggest that sws1 deficiency in medaka larvae may impair visual function and cause retinal cell apoptosis, which is associated with the downregulation of photoconduction expression and oxidative stress.


Introduction
The animal visual systems play vital roles in foraging, avoiding predators, mate selection, and communication [1]. In vertebrates, the retina is a multilayered tissue whose photoreceptor cells express opsin genes that are the core of vision at the molecular level [2]. Compared with other vertebrates, teleost fish have a unique set of visual opsins [3][4][5]. In fish, UV (short-wavelength sensitivity 1: sws1), blue (short-wavelength sensitivity 2: sws2), green (medium-wavelength sensitivity: rh2), and red (long-wavelength sensitivity: lws) opsins are expressed in cones and mediate color perception under strong light, and rod opsin (rh1) is expressed in rod cells and plays a role in dim light [6,7]. Some species have only a subset of these photopigment types, while others have multiple copies. Fish typically have fewer copies of short-wavelength-sensitive opsin (sws1 and sws2) than long-wavelength-sensitive opsin genes, and sws1 does not appear to have been ancestrally duplicated in teleost fish [8].
sws1 belongs to the G-protein-coupled receptor family, and is highly expressed in planktonic fish owing to its ability to enhance ultraviolet light sensitivity [9,10]. To date, most fish studied have the SWS1 pigment, although many fish lose SWS1 (UV-sensitive) with maturity. This heterochronous expression may cause changes in spectral sensitivity, thereby adjusting the visual system of fish to adapt to environmental spectra [11,12]. For instance, African cichlid fishes that feed on zooplankton, algae, and phytoplankton display higher levels of sws1 expression than those that feed on fish or benthic invertebrates [13]. In mice, the disruption of sws1 resulted in the increased expression of M-opsin but did not cause retinal degeneration [14]. Recently, it was demonstrated that sws1 deletion not only led to head and eye malformations but also retinal developmental defects in rainbow trout [15]. However, little is known about the relationship between sws1 and visual dysfunction in fish larvae.
The teleost fish, medaka (Oryzias latipes) is an excellent model organism for the investigation of visual development and function [16,17]. Medaka has outstanding vision and still retains all four of the ancient pyramidal subtypes of vertebrates [18,19]. In the present study, we constructed sws1 knockout medaka using CRISPR/Cas9 technology. We identified a sws1 mutant zebrafish line with a 214 bp deletion, which resulted in a frameshift and premature stop codon. We further explored the visual function of sws1 in larvae.

Generation of sws1 Knockout Medaka
We detected the expression pattern of sws1 in medaka. Sws1 was mainly expressed in the eye among adult tissues, and further RT-qPCR analysis showed that sws1 expression was significantly higher in adult tissues than in larval tissues ( Figure S1). Subsequently, we used the CRISPR/Cas9 technique to generate sws1 mutant lines and better understand the function of sws1 during medaka larval development. The sws1 −/− larvae line with 214 bp deletion (sws1 −/− ) in the 7-tm_1 domain resulted in frameshift and premature stop codon ( Figure 1A,B). Compared with the WT, the sws1 −/− larvae showed a normal survival rate and morphology at 6 dph ( Figure S2). In situ hybridization analysis revealed that sws1 expression was significantly diminished in the sws1 −/− larvae ( Figure 1C). The mRNA levels of sws1 (p = 0.003), sws2a (p = 0.041), sws2b (p = 0.016), and rh2-b (p = 0.044) were down-regulated in the sws1 −/− mutant medaka at 6 dph ( Figure 1D). Cells 2023, 12, x FOR PEER REVIEW 2 of 12 most fish studied have the SWS1 pigment, although many fish lose SWS1 (UV-sensitive) with maturity. This heterochronous expression may cause changes in spectral sensitivity, thereby adjusting the visual system of fish to adapt to environmental spectra [11,12]. For instance, African cichlid fishes that feed on zooplankton, algae, and phytoplankton display higher levels of sws1 expression than those that feed on fish or benthic invertebrates [13]. In mice, the disruption of sws1 resulted in the increased expression of M-opsin but did not cause retinal degeneration [14]. Recently, it was demonstrated that sws1 deletion not only led to head and eye malformations but also retinal developmental defects in rainbow trout [15]. However, little is known about the relationship between sws1 and visual dysfunction in fish larvae. The teleost fish, medaka (Oryzias latipes) is an excellent model organism for the investigation of visual development and function [16,17]. Medaka has outstanding vision and still retains all four of the ancient pyramidal subtypes of vertebrates [18,19]. In the present study, we constructed sws1 knockout medaka using CRISPR/Cas9 technology. We identified a sws1 mutant zebrafish line with a 214 bp deletion, which resulted in a frameshift and premature stop codon. We further explored the visual function of sws1 in larvae.

Generation of sws1 Knockout Medaka
We detected the expression pattern of sws1 in medaka. Sws1 was mainly expressed in the eye among adult tissues, and further RT-qPCR analysis showed that sws1 expression was significantly higher in adult tissues than in larval tissues ( Figure S1). Subsequently, we used the CRISPR/Cas9 technique to generate sws1 mutant lines and better understand the function of sws1 during medaka larval development. The sws1 −/− larvae line with 214 bp deletion (sws1 −/− ) in the 7-tm_1 domain resulted in frameshift and premature stop codon ( Figure 1A,B). Compared with the WT, the sws1 −/− larvae showed a normal survival rate and morphology at 6 dph ( Figure S2). In situ hybridization analysis revealed that sws1 expression was significantly diminished in the sws1 −/− larvae ( Figure 1C). The mRNA levels of sws1 (p = 0.003), sws2a (p = 0.041), sws2b (p = 0.016), and rh2-b (p = 0.044) were down-regulated in the sws1 −/− mutant medaka at 6 dph ( Figure 1D).

Decreased Food Intake in the Larvae of sws1 Knockout Medaka
To assess feeding capacity in sws1 −/− medaka, we tested the food intake at 6 dph of the sws1 mutant and WT. The results showed that there was significantly decreased food intake in the sws1 −/− medaka compared with the WT medaka at 6 dph (p < 0.001) (Figure 2A,B). Subsequently, we continued to analyze growth performance at the firstfeeding stage. There was reduced growth compared with the WT within seven days of first feeding (p < 0.05) ( Figure 2C). The relative expression levels of orexigenic genes (npy (p = 0.034) and agrp (p = 0.012)) in sws1 −/− mutant larvae were significantly increased after feeding ( Figure 2D,E). No difference in the mRNA levels of the anorexigenic gene (pomc) was observed between the WT and sws1 −/− medaka (p > 0.05) ( Figure 2F).
Cells 2023, 12, x FOR PEER REVIEW 3 of 12 medaka in the larval eyes at 6 dph. All data are expressed as means ± S.E.M (n = 6). * p < 0.05 per Student's t-test.

Decreased Food Intake in the Larvae of sws1 Knockout Medaka
To assess feeding capacity in sws1 −/− medaka, we tested the food intake at 6 dph of the sws1 mutant and WT. The results showed that there was significantly decreased food intake in the sws1 −/− medaka compared with the WT medaka at 6 dph (p < 0.001) ( Figure  2A,B). Subsequently, we continued to analyze growth performance at the first-feeding stage. There was reduced growth compared with the WT within seven days of first feeding (p < 0.05) ( Figure 2C). The relative expression levels of orexigenic genes (npy (p = 0.034) and agrp (p = 0.012)) in sws1 −/− mutant larvae were significantly increased after feeding ( Figure 2D,E). No difference in the mRNA levels of the anorexigenic gene (pomc) was observed between the WT and sws1 −/− medaka (p > 0.05) ( Figure 2F). Quantification of the food intake between WT and sws1 −/− medaka larvae (n = 6/group). (C) Total length measurements of medaka within seven days of first feeding (n = 20/group); d, day. The transcriptional levels of orexigenic genes (npy (D) and agrp (E)) and anorexigenic gene (pomc (F)) in WT and sws1 −/− medaka larvae (n = 10/group, with 6 replicates). Results are shown as mean ± SEM. * p < 0.05. ns, not significant per Student's t-test.

Disruption of sws1-Impaired Medaka Larvae Swimming Behavior
Next, we performed a photoperiod stimulation test at 6 dph for WT and sws1 −/− mutant larvae. Compared with the WT, the sws1 −/− larvae displayed a decrease in swimming speed during periods of dark-light and light-dark transition stimulation (p < 0.05) ( Figure 5).

Disruption of sws1-Impaired Medaka Larvae Swimming Behavior
Next, we performed a photoperiod stimulation test at 6 dph for WT and sws1 −/− mutant larvae. Compared with the WT, the sws1 −/− larvae displayed a decrease in swimming speed during periods of dark-light and light-dark transition stimulation (p < 0.05) ( Figure 5).

Disruption of sws1-Impaired Medaka Larvae Swimming Behavior
Next, we performed a photoperiod stimulation test at 6 dph for WT and sws1 −/− mutant larvae. Compared with the WT, the sws1 −/− larvae displayed a decrease in swimming speed during periods of dark-light and light-dark transition stimulation (p < 0.05) ( Figure 5).

Disruption of sws1-Impaired Medaka Larvae Swimming Behavior
Next, we performed a photoperiod stimulation test at 6 dph for WT and sws1 −/− mutant larvae. Compared with the WT, the sws1 −/− larvae displayed a decrease in swimming speed during periods of dark-light and light-dark transition stimulation (p < 0.05) ( Figure 5).

Induced Oxidative Stress in sws1-Deficient Medaka Larvae
To evaluate the oxidative stress activities in medaka larvae, the activities of total superoxide dismutase (T-SOD), catalase (CAT), glutathione peroxidase (GSH-Px), and malondialdehyde (MDA) were detected. Compared with the WT larvae, CAT activity and GSH-Px content were not significantly changed, but the T-SOD activity was significantly reduced (p = 0.007) ( Figure 6A-C). Meanwhile, we observed an MDA content in the sws1 −/− mutant larvae that was dramatically increased (p < 0.001) ( Figure 6D).

Induced Oxidative Stress in sws1-Deficient Medaka Larvae
To evaluate the oxidative stress activities in medaka larvae, the activities of total superoxide dismutase (T-SOD), catalase (CAT), glutathione peroxidase (GSH-Px), and malondialdehyde (MDA) were detected. Compared with the WT larvae, CAT activity and GSH-Px content were not significantly changed, but the T-SOD activity was significantly reduced (p = 0.007) ( Figure 6A-C). Meanwhile, we observed an MDA content in the sws1 −/− mutant larvae that was dramatically increased (p < 0.001) ( Figure 6D).

Depletion of sws1 Led to Retinal Cell Apoptosis in Medaka Larvae
To investigate whether apoptosis activation mediates the effects of visual function in medaka larvae, apoptosis was assessed in the WT and sws1 −/− larvae at 6 dph. There were few TUNEL-positive cells in the WT retinas. In comparison, the sws1 −/− mutant larvae had significantly more apoptotic cells in the eye (p = 0.045) ( Figure 7A,B). Caspase-9 is a relatively upstream caspase in apoptotic signaling [20]. At 6 dph, the activity of caspase-9 in the sws1 −/− medaka larvae was found to be higher than that in the WT larvae (p = 0.005 ( Figure 7C).

Depletion of sws1 Led to Retinal Cell Apoptosis in Medaka Larvae
To investigate whether apoptosis activation mediates the effects of visual function in medaka larvae, apoptosis was assessed in the WT and sws1 −/− larvae at 6 dph. There were few TUNEL-positive cells in the WT retinas. In comparison, the sws1 −/− mutant larvae had significantly more apoptotic cells in the eye (p = 0.045) ( Figure 7A,B). Caspase-9 is a relatively upstream caspase in apoptotic signaling [20]. At 6 dph, the activity of caspase-9 in the sws1 −/− medaka larvae was found to be higher than that in the WT larvae (p = 0.005 ( Figure 7C).

Discussion
In this study, we created a medaka mutant line with a 214 bp deletion via CRISPR/Cas9 gene editing technology. The deletion caused the loss of seven transmembrane domain 1 (7-tm_1). We observed that the knockout of sws1 caused impaired feeding ability and hypoactivity in the first feeding stage. Histological analysis and subsequent experiments indicated that impaired feeding ability and hypoactivity might be the causes of the affected retinal structures and downregulation of phototransduction related genes. We further found that sws1 −/− individuals had reduced resistance to oxidative stress and increased apoptosis in the larval eye.
Visual perception is the main sensory way for fish larvae to detect and capture prey [21]. Fish with photopic vision utilize more cones, whereas those with scotopic vision, or dark-adapted eyes (twilight vision), use rods [22]. Ultraviolet vision plays a crucial role in identifying zooplankton because UV light scatters in water bodies, while near-transparent zooplankton absorb UV light in contrast [23,24]. The UV-sensitive opsin (sws1) expressed by UV cones is sensitive to UV light, and its expression level determines the sensitivity of UV cones [25]. Flamarique [9] treated young rainbow trout with thyroid hormone, resulting in the reduction in single cone cells expressing sws1 mRNA in the retina and the reduction in sensitivity to zooplankton. Similarly, the tbx2b zebrafish mutant with reduced UV cones also found zooplankton with a smaller mean positional distance and angle than wild zebrafish [26]. Yoshimatsu et al. [27] further revealed that the UV cone is the main input of the visual predator-prey circuit of zebrafish larvae through two-photon imaging in vivo, transcriptome, and computational models. In sws1 knockout medaka, food intake was shown to be significantly reduced, accompanied by slow growth during the first feeding phase (Figure 2A-C). Correspondingly, the appetite-orexigenic genes (npy and agrp) were significantly upregulated after feeding in the sws1 mutants ( Figure 2D,E). Our explanation is that the sws1 −/− mutant larvae have an impaired feeding ability and still do not reach

Discussion
In this study, we created a medaka mutant line with a 214 bp deletion via CRISPR/Cas9 gene editing technology. The deletion caused the loss of seven transmembrane domain 1 (7-tm_1). We observed that the knockout of sws1 caused impaired feeding ability and hypoactivity in the first feeding stage. Histological analysis and subsequent experiments indicated that impaired feeding ability and hypoactivity might be the causes of the affected retinal structures and downregulation of phototransduction related genes. We further found that sws1 −/− individuals had reduced resistance to oxidative stress and increased apoptosis in the larval eye.
Visual perception is the main sensory way for fish larvae to detect and capture prey [21]. Fish with photopic vision utilize more cones, whereas those with scotopic vision, or darkadapted eyes (twilight vision), use rods [22]. Ultraviolet vision plays a crucial role in identifying zooplankton because UV light scatters in water bodies, while near-transparent zooplankton absorb UV light in contrast [23,24]. The UV-sensitive opsin (sws1) expressed by UV cones is sensitive to UV light, and its expression level determines the sensitivity of UV cones [25]. Flamarique [9] treated young rainbow trout with thyroid hormone, resulting in the reduction in single cone cells expressing sws1 mRNA in the retina and the reduction in sensitivity to zooplankton. Similarly, the tbx2b zebrafish mutant with reduced UV cones also found zooplankton with a smaller mean positional distance and angle than wild zebrafish [26]. Yoshimatsu et al. [27] further revealed that the UV cone is the main input of the visual predator-prey circuit of zebrafish larvae through two-photon imaging in vivo, transcriptome, and computational models. In sws1 knockout medaka, food intake was shown to be significantly reduced, accompanied by slow growth during the first feeding phase (Figure 2A-C). Correspondingly, the appetite-orexigenic genes (npy and agrp) were significantly upregulated after feeding in the sws1 mutants ( Figure 2D,E). Our explanation is that the sws1 −/− mutant larvae have an impaired feeding ability and still do not reach satiation after half an hour of feeding compared with the WT, so their appetitive genes are still upregulated. Numerous studies have implicated npy and agrp in the regulation of appetite and food intake in teleosts [28][29][30]. In mice, the NPY + and AgRP + signals in the hypothalamic arcuate nucleus (ARC) in the fasting group were significantly stronger than those in the feeding group [31]. Our result is consistent with the expression of npy and agrp during fasting. The mediating and integrating role of sws1 in feeding and appetite needs to be investigated in further studies.
The retina is a highly complex tissue structurally composed of three layers of nerve cell bodies (ONL, INL, and GCL) and two layers of synapses (OPL and IPL) [32]. In each layer, there are specific cells with varied physiological functions. An important retinal neuron is the GCL, which connects the retina to the optic tectum in the brain [33]. Histological results showed that the thickness of the GCL was significantly increased, whereas the lens became thinner in the sws1 −/− mutant larvae (Figure 3), which may have led to impaired visual function [34]. The spectral specificity of visual pigments is determined via the opsin synthesized by cones and rods in the ONL [35]. Here, there was no obvious histological damage in the ONL, indicating that it is likely a molecular effect. Our quantification showed that the transcription levels of sws1, sws2a, sws2b, and rh2-b were significantly decreased in sws1 −/− mutant larvae ( Figure 1D). Opsin accumulation in photoreceptors may be reduced as transcription levels of opsin genes decline [36]. In addition, many previous studies have suggested that photoreceptor defects may lead to the decreased sensitivity of fish larvae to light stimulation [37][38][39].
Subsequently, we tested the sensitivity of the fish larvae to light stimulation, and we discovered that the sws1 −/− larvae showed a reduction in swimming speed compared with the WT larvae in the light/dark behavior test ( Figure 5). The light/dark cycle is considered to be the main factor that entrains the biological rhythm of teleosts through its influence on feeding physiology and gene expression [40][41][42]. Studies have shown that both clock genes and feeding behavior exhibit a circadian pattern [43]. Rods, cones, and melanopsin are the only photoreceptors that contribute to circadian light entrainment [44]. Therefore, sws1 loss may affect circadian photoentrainment. In addition, intrinsic photosensitive retinal ganglion cells (ipRGCs) are also involved in regulating circadian photoentrainment, whereas ipRGCs reside in the GCL [45,46]. In this study, changes in the thickness of ganglion cells may also have affected circadian photoentrainment. Furthermore, the genes gnb3b, grk7a, grk7b, and pde6c, which are associated with phototransduction, were downregulated in sws1 −/− mutant medaka at 6 dph ( Figure 4). In phototransduction, opsin converts optical signals into electrical signals by generating electrochemical signals from captured photons via pigments and intracellular mechanisms [2,47]. Gnb3b, grk7a, grk7b, and pde6c form the downstream phototransduction pathway of cone opsin [48,49]. In zebrafish, gnb3b disruption exhibited aberrant cone photoreceptor formation [50]. It was also found that grk7a-knockdown larvae had slower photoresponse recovery and that their temporal contrast sensitivity was reduced [51]. Also, similar results were observed in pde6c zebrafish mutant larvae [52]. In addition, it has been reported that the thinning of the GCL prevents electrical signals from being transmitted to the central nervous system [53,54]. Therefore, the decreased activity of the sws1 −/− mutant larvae may have mainly been due to impaired retinal light signal reception and transmission.
Our study further found that the absence of sws1 reduced the resistance of larvae to oxidative stress. The levels of CAT and GSH-Px in the WT and sws1 −/− mutant larvae were not statistically significantly different ( Figure 6B,C). These results suggested that CAT and GSH-Px may not be involved in antioxidation to sws1 in medaka larvae. The T-SOD activity and MDA content in the sws1 −/− mutant larvae were significantly altered ( Figure 6A,D), demonstrating that sws1 disruption caused oxidative damage. Excess free radicals lead to the overproduction of MDA, a lipid peroxidation product that indicates the severity of oxidative stress [55]. On the other hand, oxidative stress is one potential factor in the cellular apoptosis process. Oxidative stress releases mitochondrial cytochrome c, which activates apoptogenic proteins [56,57]. The loss of sws1 led to retinal cell apoptosis in medaka larvae ( Figure 7A,B). Additionally, caspase (cysteine-requiring aspartate protein) is a protease family that plays an important role in the process of cell apoptosis, and caspase-9 is the initiator of apoptosis signal transduction [58][59][60]. In our study, the activity of caspase-9 in the sws1 −/− medaka larvae was significantly increased compared with the wild-type medaka ( Figure 7C), suggesting that the deletion of sws1 resulted in the activation of caspase-9. Our characterization of retinal cell apoptosis in sws1 genes is similar to that of Blanch et al. [61]. Moreover, previous studies have shown that eye development defects in zebrafish larvae are caused by the apoptosis of neuronal progenitor cells in the eye [62]. Therefore, we speculated that the abnormal development of the eye in the sws1 −/− mutant larvae is also related to the increased level of apoptosis.
In summary, we revealed an essential role of sws1 in vision-guided behavioral alteration and visual development. Our results implicate sws1 signaling as a regulator of reactions to prey capture and light-dark transitions in larval medaka, with sws1 −/− medaka larvae showing decreased food intake and swimming speed during the period of dark-light and light-dark transition stimulation. As another new discovery, we observed microphthalmia and increased retinal cell apoptosis in sws1 knockout medaka larvae, which may have been related to oxidative stress. Our results provide insights into understanding the mechanisms of sws1 mutation-mediated vision-guided behavioral and visual developmental alteration in medaka.

Medaka Lines and Maintenance
The wild-type medaka is an orange strain, which was kept at 26-28 • C and a 14 h light/10 h dark cycle. Medaka embryos were raised at 28 • C in medaka embryo medium (MEM) [63]. Larvae at 6 days post-hatching (dph) were given live Artemia twice daily after the yolk sac had almost been consumed. All medaka protocols were approved by the Institutional Animal Care and Use Ethics Committee of Huazhong Agricultural University (approval reference number HZAUFI-2020-0024).

Generating sws1 −/− Mutants Using CRISPR/Cas9 Technology
The single-guide RNAs (sgRNAs) of medaka sws1 (ENSORLG00000019293) gene were designed using CCTOP web (https://cctop.cos.uni-heidelberg.de/). The sequences of single-guide RNAs (sgRNAs) and PCR primers are shown in Supplementary Table S1. sgRNAs were cloned into pMD-18T vector and were synthesized using TranscriptAid T7 High Yield Transcription kit (Thermo, Scientific, San Diego, CA, USA). The compounds of sgRNAs (50 ng/µL) and Cas9 protein (NEB, Ipswich, MA, USA) were coinjected into one-or two-cell stage wild-type embryos. The F0 medaka were outcrossed with the wild type to generate F1 medaka. The target region was amplified using ordinary PCR with sws1 test-F and sws1 test-R. The reaction conditions were 95 • C for 30 s, 60 • C for 30 s, and 72 • C for 30 s for 40 cycles. We detected that the sws1 mutation type is a large deletion, and we could preliminarily judge whether the F1 generation was a heterozygous mutant using PCR. Then, the mutation type was determined via sequencing. The F1 heterozygous with the same mutation in-crossed to obtain F2 homozygous, and all experiments were conducted with F3 homozygous.

Histological Assessment, In Situ Hybridization (ISH), and TUNEL Staining
Whole medaka at 6 dph were preserved in 4% paraformaldehyde solution (PFA) for 24 h, dehydrated in 70-100% ethanol, embedded in paraffin, and sectioned in thicknesses of 4 µm (Leica, Wetzlar, Germany). Then, the sections were stained with (H&E) according to standard protocols. The slides were imaged via slice digital scanning (Pannoramic250, Pannoramic250 MIDI, 3D HISTECH). The thickness of each retinal layer in right eye was measured using ImageJ 1 software according to the Chen et al. method [37].
For in situ hybridization (ISH), the 6 dph sections hybridized with the DIG-labeled RNA probes (Roche, Basel, Switzerland) at 65 • C for 12-16 h and the hybridization signals were visualized with nitroblue etrazolium chloride (NBT)/5-bromo-4-chloro-3-indolyl phosphate (BCIP) (NBT/BCIP) (Sigma, Cibolo, TX, USA) staining, as described in our previous study [64]. The signals from the right eye of the larvae were observed under an optical microscope (Olympus, Tokyo, Japan). The probes primers are listed in Supplementary Table S1. A TUNEL staining kit (catalog no. G1501) (Servicebio, Wuhan, China) was used to identify apoptotic cells in the medaka 6 dph sections, following the manufacturer's instructions. The TUNEL-positive cells on the slides were observed using a fluorescence microscope (NIKON ECLIPSE C1, Tokyo, Japan), and the quantification of TUNEL-positive cells was performed by manually counting the number of positively labeled cones in the retina.

Light/Dark Behavior Analysis
Light/dark behavioral tests were conducted between 15:00 and 17:00 using a Danio-Vision Observation Chamber (Noldus Information Technology, Wageningen, The Netherlands) linked to Etho Vision XT13 software. The 6 dph larvae were plated into a 24-well plate (diameter 1.56 cm wells) with 1 mL of MEM (individual larvae per 24-well plate). The larvae were acclimated for 10 min at 28 • C, and the larval locomotor activity was tested in response to dark-light conversion (3 min light/3 min dark/3 min light/3 min dark) based on the protocol by Huang et al. [65] with some modifications. The average swimming speed (cm/s) for each individual larva was collected every 60 s, and each experiment was repeated 3 times. Further analysis was performed using custom Open Office Org 2.4 software.

Larvae Feeding Assays
For larvae food intake, 6 dph larvae were fed with Artemia in wells of a 6-well plate with 8 mL of MEM (diameter 3.48 cm wells) for 30 min (6 larvae per 6-well plate). Then, larvae were anesthetized with MS-222 and some larvae were fixed with 4% PFA overnight for food intake analysis, and others were frozen with liquid nitrogen for quantitative analysis of appetite genes. The orange area of Artemia in the digestive tract was photographed using a stereomicroscope and measured with Image J 1 software. The amount of food ingested by medaka larvae was determined following the procedure described previously [66].

Growth Performance and Survival Rate
For the growth performance and survival rate, 6 dph larvae were fed with abundant Artemia twice daily for 7 days. Twenty WT and sws1 −/− medaka larvae were randomly selected, anaesthetized, and fixed with 4% PFA for total-length measurement using Image J1. The experiment was repeated 3 times.

RNA Isolation and Quantitative RT-PCR
All fish were sampled in the light phase of the light/dark cycle. The adult fish tissues (n = 3 individuals) and 6 dph larval eyes (n = 6 individuals) were collected and frozen in liquid nitrogen. Total RNA was extracted from each sample according to the RNAiso instructions and the cDNA reverse-transcribed using a HiScript ® III 1st Strand cDNA Synthesis Kit (Vazyme, Nanjing, China). Quantitative RT-PCR was performed with ChamQ universal SYBR qRT-PCR master mix (Vazyme, Nanjing, China) on a CFX 96 Touch System (Bio-Rad, Hercules, CA, USA). The data were collected using the 2 −∆∆Ct value method and normalized to β-actin [67]. The primers of opsin genes and phototransduction-related genes are set in Table S2, and the primers of appetite genes were carried out as described previously [68].

Statistical Analysis
All results are presented as means ± standard error of the mean (S.E.M), and the normality of the data was first tested using the Shapiro-Wilk test. The differences were analyzed using Student's t-test, and differences were considered significant at p < 0.05. All statistical analyses were carried out using IBM SPSS Statistics 25 software.
Author Contributions: K.L. and X.-F.L. designed and performed the research, analyzed experimental data, and wrote the original draft. J.W. performed the experiment and acquired the data, and S.T. analyzed the data. Y.W., L.Z. and F.C. raised the fish and screened the mutants. All authors have read and agreed to the published version of the manuscript. Institutional Review Board Statement: All experimental protocols were approved by the Institutional Animal Care and Use Ethics Committee of Huazhong Agricultural University (an approval reference number HZAUFI-2020-0024).

Informed Consent Statement: Not applicable.
Data Availability Statement: All data are available from the corresponding author by request.