Amino acid residue at position 188 determines the UV-sensitive bistable property of vertebrate non-visual opsin Opn5

Opsins are G protein-coupled receptors specialized for photoreception in animals. Opn5 is categorized in an independent opsin group and functions for various non-visual photoreceptions. Among vertebrate Opn5 subgroups (Opn5m, Opn5L1 and Opn5L2), Opn5m and Opn5L2 bind 11-cis retinal to form a UV-sensitive resting state, which is inter-convertible with the all-trans retinal bound active state by photoreception. Thus, these opsins are characterized as bistable opsins. To assess the molecular basis of the UV-sensitive bistable property, we introduced comprehensive mutations at Thr188, which is well conserved among these opsins. The mutations in Opn5m drastically hampered 11-cis retinal incorporation and the bistable photoreaction. Moreover, T188C mutant Opn5m exclusively bound all-trans retinal and thermally self-regenerated to the original form after photoreception, which is similar to the photocyclic property of Opn5L1 bearing Cys188. Therefore, the residue at position 188 underlies the UV-sensitive bistable property of Opn5m and contributes to the diversification of vertebrate Opn5 subgroups.

One of the major claims in the manuscript is that the T188C mutation of Opn5m converted the photoreaction profile and the binding selectivity of retinal isomers to mimic those of Opn5L1. In Opn5L1, it is reported that (1) the light illumination photo-isomerizes all-trans-retinal to 11-cis, and (2) the retinal transiently forms covalent interaction with C188 (Sato et al., 2018). However, in the T188C mutant of Opn5m, the retinal is isomerized to 13-cis (Fig. 2d,h,l), and the authors do not provide any direct evidence of the adduct formation between T188C and retinal. The mutation surely affects the bistable property of Opn5m, but it is too much to claim that the mutant mimics the property of Opn5L1.
Minor points 1. Page 10, lines 3-6: "The crystal structure of squid rhodopsin, which is a prototypical bistable opsin, shows that the residue at position 188 is accommodated in a β-sheet structure of the second extracellular loop and is located within 5 Å of the retinal polyene chain (Fig. S5a)." How close are the sequences of squid rhodopsin and Opn5m? The authors should at least show the sequence alignment between squid rhodopsin and Opn5m, and they can also show the homology model of Opn5m based on the crystal structure of squid rhodopsin.
2. The authors show the results of "Binding selectivity of retinal isomers" separately in Fig. 1 and 2, but this is very difficult to see. The authors can merge Fig. 1 and 2, keep action spectra to the main figure, and move the bar graphs to supplementary.

Reviewer #2 (Remarks to the Author):
This is an interesting study that confirms the role of Thr188 in the functionality of Opn5m. The account however is very difficult to follow, partly through grammatical errors, partly through being too condensed and partly because the Figures 1 and 2 which are at the centre of the investigation, are difficult to interpret. It is imperative that the experiments are much more extensively explained, and the rational involved, in the Results section.
Here are my major comments Firstly, why no page or line numbering? Results page 5: The reader needs to know what is meant by "all possible 19 mutations". Coding/non-coding changes? 1. Change to read "wild type and mutant recombinant proteins" 2. What does this sentence mean? "Among these mutants, we detected the photopigments of seven mutants….." Only these mutant proteins produced photopigments? 3. Say what form each retinal produces. 4. Change to read "wild-type and mutant Opn5m". the grammatical structure used here and elsewhere in the text is very poor. 5. Defien what is meant by "purified" Page 6 6. Need to state more clearly the illumination used in the experiments. 7. Is it a "speculation" or a "conclusion"? 8. Avoid using "mutants" -it is mutant protein of mutant Opn5m 9. The various experiments presented in the Figs need to be described in the Results -it is not sufficient to expect the reader to extract this from Fig legends and M&M. 10. "Larger amount" -meaning what? Give a range if single figure not possible. 11. "More preferentially" -bad grammar -either preferentially or not! 12. "is responsible" -to say "is required". Other sites may also be involve so to say that this is all down to one site may be going too far. Page 7 13. Use of "mutants" again. Page 8 14. Significance ot temperature not explained. Page 9 15. "all possible 19 mutations" -define 16. Move at least some of the description of the effect of the different mutations into the Results section.

Responses to Reviewer #1
Thank you very much for your critical reading of our manuscript and for your valuable comments. We revised the manuscript based on your comments.
The following are our responses to your comments.
Comment 1: The title of the paper is "Identification of the amino acid residue responsible for UV-sensitive bistable property of vertebrate non-visual opsin Opn5", but the authors only analyzed the single amino acid (T188) of Opn5 from the single species (Xenopus tropicalis). The authors have to perform more comprehensive analysis, including the analysis of other residues and other Opn5 homologs, otherwise, the scope of the manuscript is too specific for the broad audience of Communications Biology.
Response: According to your comment, we performed additional experiments. However, in the T188C mutant of Opn5m, the retinal is isomerized to 13-cis (Fig.   2d,h,l), and the authors do not provide any direct evidence of the adduct formation between T188C and retinal. The mutation surely affects the bistable property of Opn5m, but it is too much to claim that the mutant mimics the property of Opn5L1.
Response: Thank you for your important comment. According to your suggestion, we weakened the claim about T188C mutant Opn5m throughout the manuscript. And we revised the sentences in the "Discussion" section as follows.
Page 14: "Therefore, it is suggested that the structural differences in the retinal binding pocket between Opn5L1 and T188C mutant Opn5m can alter the rate of the photocycle and the selectivity of the retinal isomerization. Additional mutations in Opn5m would be necessary to completely mimic the molecular properties of Opn5L1 by recapitulating the molecular evolutionary process from bistable opsin to photocyclic opsin." Comment 3: Page 10, lines 3-6: "The crystal structure of squid rhodopsin, which is a prototypical bistable opsin, shows that the residue at position 188 is accommodated in a β-sheet structure of the second extracellular loop and is located within 5 Å of the retinal polyene chain (Fig. S5a)." How close are the sequences of squid rhodopsin and Opn5m? The authors should at least show the sequence alignment between squid rhodopsin and Opn5m, and they can also show the homology model of Opn5m based on the crystal structure of squid rhodopsin.
Response: According to your comment, we added the sequence comparison among bovine rhodopsin, squid rhodopsin, Opn5m and Opn5L1 in Fig. S1 and described the sequence similarity (28 %) between squid rhodopsin and Xenopus Opn5m in the legend of Fig. S1. We also showed the locations of three residues (positions 167, 188 and 212) in the crystal structure of squid rhodopsin (Fig. 1a).
Comment 4: The authors show the results of "Binding selectivity of retinal isomers" separately in Fig. 1 and 2, but this is very difficult to see. The authors can merge Fig. 1 and 2 Response: According to your comment, we revised the phrase throughout the manuscript.
Comment 3: What does this sentence mean? "Among these mutants, we detected the photopigments of seven mutants……" Only these mutant proteins produced photopigments?
Response: We tried to obtain the recombinant Opn5m proteins whose residue at position 188 was replaced by the other 19 natural amino acid residues. We expressed these mutant recombinant proteins in cultured cells and reconstituted the photo-pigments by adding 11-cis or all-trans retinal to the collected cell membranes. Among these mutant proteins, we could produce the photo-pigments of T188C mutant protein (Fig. 1) and six other mutant proteins (T188S, T188G, T188A, T188N, T188M and T188Q) (Fig. 2), and could not detect the formation of the photo-pigments of the other 12 mutant proteins by the absorption spectrum analysis. To clarify this point, we revised the sentences in the "Results" section as follows.
Page 8: "To reveal the roles of the well-conserved threonine residue at position 188 in the Opn5m subgroup, we replaced Thr188 with the other 18 natural amino acid residues in addition to the cysteine of the T188C mutant protein.
We expressed the mutant recombinant proteins in cultured cells and reconstituted the photo-pigments by adding 11-cis or all-trans retinal to the collected cell membranes. Among these mutant Opn5m, we could produce the photo-pigments of six mutant proteins, T188S, T188G, T188A, T188N, T188M, and T188Q (Fig. 2), but could not detect the formation of the photo-pigments of the other 12 mutant proteins by absorption spectrum analysis." Comment 4: Say what form each retinal produces.
Response: In this experiment, we added 11-cis or all-trans retinal to the cell membranes which contained the mutant recombinant proteins. To clarify this point, we revised the sentences in the "Results" section as follows.
Page 5: "We reconstituted their photo-pigments by adding 11-cis or all-trans retinal to the collected cell membranes and purified them by affinity column chromatography using Rho1D4-conjugated agarose (Fig. 1b-1e)." Page 8: "We expressed the mutant recombinant proteins in cultured cells and reconstituted the photo-pigments by adding 11-cis or all-trans retinal to the collected cell membranes." Comment 5: Change to read "wild-type and mutant Opn5m". the grammatical structure used here and elsewhere in the text is very poor.
Response: According to your comment, we revised the phrase throughout the manuscript.
Comment 6: Define what is meant by "purified".
Response: We prepared the recombinant Opn5m proteins which were tagged with the epitope sequence of the anti-bovine rhodopsin monoclonal antibody Rho1D4 at the C-terminus and purified them by affinity column chromatography using Rho1D4-conjugated agarose. We already described the procedure in the "Materials and Methods" section. However, to clarify this point, we revised the sentences in the "Results" section as follows.
Page 5: "We reconstituted their photo-pigments by adding 11-cis or all-trans retinal to the collected cell membranes and purified them by affinity column chromatography using Rho1D4-conjugated agarose (Fig. 1b-1e)." Comment 7: Need to state more clearly the illumination used in the experiments.
Response: According to your comment, we added the illumination wavelength in the legends of all the figures and described the illumination lamps and the optical filters used for the measurement of the absorption spectra in the "Materials and Methods" section as follows.
Page 16: "The sample was irradiated with light which was generated by a 1-kW tungsten halogen lamp (Master HILUX-HR, Rikagaku seiki) and passed through optical filters (Y-52 or UV-D36C, AGC Techno Glass). The transient absorption spectra of Opn5m T188C mutant protein were recorded using a Shimadzu UV2450 spectrophotometer at 10 ± 0.1 C. The spectra were measured in the dark and after 2 min irradiation with light which was generated by a 1-kW tungsten halogen lamp and passed through an optical filter (Y-52, AGC Techno Glass). The transient absorption spectra of Opn5m T188C mutant protein were also recorded using a high-speed CCD camera spectrophotometer (C10000 system, Hamamatsu Photonics) kept at 37 ± 0.1 C by a temperature controller (pqod, QUANTUM Northwest) to accelerate the photocyclic reaction, according to the methods used in our previous study for Opn5L1. The flash light for irradiation was generated by a short-arc power flash (SA-200, Nissin Electronic; pulse duration of ~170 s and flash lamp input of 200 J/F) and passed through an optical filter (Y-48, AGC Techno Glass)." Comment 8: Is it a "speculation" or a "conclusion"?
Response: In this study, we showed that wild-type Opn5m protein purified after the addition of 11-cis retinal exclusively contained 11-cis retinal. By contrast, wild-type Opn5m protein purified after the addition of all-trans retinal contained a substantial amount of 11-cis retinal in addition to all-trans retinal. Thus, we concluded that wild-type Opn5m protein preferentially binds 11-cis retinal. To clarify this point, we revised the sentence in the "Results" section as follows.
Page 6: "Thus, we concluded that wild-type Opn5m protein preferentially binds 11-cis retinal." Comment 9: Avoid using "mutants"it is mutant protein of mutant Opn5m Response: According to your comment, we revised the phrase throughout the manuscript. Response: To clarify the relationship between the explanations in the main text and the figures, we inserted the figure numbers in several sentences of the main text and revised the sentences in the "Results" section as follows.
Page 8: "All the mutant proteins purified after the addition of all-trans retinal also had the absorbance in the visible region (black curve of Figs. 2g-2l). These