The Role of Sphingosine-1-Phosphate Receptor 2 in Mouse Retina Light Responses

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Shrestha, AP, Stiles, M, Grambergs, RC, et al have investigated the functional role of S1P receptor subtype 2 (S1PR2) in the mouse retina. The authors conducted tests on wild-type and S1PR2 knockout mice, revealing that S1PR2 knockout mice displayed increased responses to light stimuli, thicker retinal layers, and altered synaptic markers, implying a distinct role for S1PR2 in retinal structure and synaptic organization, possibly affecting the retina's light-related functions.

The authors quantified the changes in the retinal functionality and structure through electroretinography (ERG) and optical coherence tomography (OCT) and examined synaptic gene and protein expression with immunohistochemistry (IHC) and quantitative reverse transcription polymerase chain reaction (RT-qPCR). The authors tried to map the function of S1PR2, in retinal structure and synaptic organization, potentially affecting light-mediated retinal function.

The authors provide evidence of S1PR2's involvement in light response, retinal structure, and synaptic transmission. The IHC and RT-PCR evidence are highly appreciated which helps to understand that S1PR2 also influences retinal adaptation to light and dark conditions, affecting key synaptic genes and proteins. Moreover, the authors also suggest that S1PR2 may play a crucial role in balancing excitatory-inhibitory transmission, possibly by inhibiting GABAergic synapses via Nogo-A signaling but may need more evidence to support this statement. The novelty of this study is that S1PR2 is a potential target in retinal diseases, connecting neural retina function, and organization.

The experiments designed for this study are justified. Overall, the results are significant. The introduction and the discussion were written clearly. The mentioned animal ethics is also much appreciated.

Nonetheless, the article seemed to possess good value toward the function of S1PR2 in the neural retina, especially a significant impact on retinal layer thickness, the expression of synaptic genes and proteins, as well as light responses at both functional and cellular levels, underscoring its pivotal role in shaping synaptic structure and retinal function.

Overall, the clarity of the text needs a few readjustments. The manuscript has minor typographical and grammatical errors. The results and the figures were consistent based on the written legends and results. The quantitative analyses are much appreciated. In general, the manuscript can accomplish the caliber of quality for consideration for publication in the Journal of “Biomolecules” with minor changes. The authors are advised to consider the comments below:

Comments:

 

1.      Page 6 / Line 237 / Requesting to use either “animals” or “mice”.

2.      Page 13 / Line 436 / It seems that the verb were does not agree with the subject. Consider changing the verb form “were” to “was”.

3.      I would request the authors to be consistent with their writing throughout the manuscript, either using PKCa or PKCa. Check page 13 – “S1PR2 KO mice exhibit illumination-dependent changes in rod bipolar cell morphology and synaptic protein expression.

4.      Page 16/ Line 484/ The word which doesn’t seem to fit this context. Consider replacing it and using “with”.

5.      Page 19 / Line 543 / It seems that the verb is does not agree with the subject. Consider changing the verb form “is” to “are”.

6.      Figure 2/ It is understandable that it is difficult to get a high-resolution image from an OCT (Figure 2, C). It would be helpful or would be easier to understand if the authors provided a histology image comparing WT vs S1PR2 KO in the supplementary.

7.      Page 7 / Role of S1PR2 in the expression of synaptic proteins / Line 277 – “Table 1” only mentioned the list of antibodies, not raw quantification data, which seems irrelevant here to mention.

8.      I would request the authors to be consistent with their marking throughout the manuscript like Figure 3 – Graph marking G, H, and I (mark them from the corner following other images). Figure 4 – The marking of A, B, C, D, E, and F is on the opposite side compared to other images.

9.      Figures 3, 4, 5, and 6 / Please mark all the retina layers in the SIPR2 KO for proper representation (3D, 4A & D, 5, 6C).

10.   Please check the image size and reorganize the images (Figure 4), as well as either use only a scale bar or a scale bar with written value.

11.   I would request the authors to provide a quantification (graph) of fluorescence intensity (multiple ROI) of Rhodopsin, MAGUK, and SV2 marker comparing WT vs S1PR2 KO in the supplementary. The % Area labeled by IHC would be variable due to tissue conditions or sectioning or due to different locations (distance from the optic nerve). Moreover, In Figures 5 and 6 the expression of MAGUK and SV2 is higher (visually appearing in the representative image).

12.   Figures 8 and 9 / Please mark the retinal layers in the representative images.

Comments on the Quality of English Language

The manuscript has minor typographical and grammatical errors. Overall, the clarity of the text is good and needs very few readjustments.

Author Response

Dear Reviewer, Thank you to this reviewer for their great interest and constructive feedback on our manuscript. We have incorporated our comments to the best of our knowledge.

 

 

1. Page 6 / Line 237 / Requesting to use either “animals” or “mice”.

Response: Thank you, we have removed “animals”.

 

1. Page 13 / Line 436 / It seems that the verb were does not agree with the subject. Consider changing the verb form “were” to “was”.

Response: Thank you, we have updated.

 

1. I would request the authors to be consistent with their writing throughout the manuscript, either using PKCa or PKCa. Check page 13 – “S1PR2 KO mice exhibit illumination-dependent changes in rod bipolar cell morphology and synaptic protein expression.

Response: Thank you, we have updated.

 

 

1. Page 16/ Line 484/ The word which doesn’t seem to fit this context. Consider replacing it and using “with”.

Response: Thank you, we have updated.

 

 

1. Page 19 / Line 543 / It seems that the verb is does not agree with the subject. Consider changing the verb form “is” to “are”.

Response: Thank you, however, the verb “is” agrees with the subject “expression,”.

 

 

1. Figure 2/ It is understandable that it is difficult to get a high-resolution image from an OCT (Figure 2, C). It would be helpful or would be easier to understand if the authors provided a histology image comparing WT vs S1PR2 KO in the supplementary.Authors have mentioned in the Fig 2 that they have observed increase length of SR (synaptic ribbon) in OA model, do they think this is compensatory mechanism for the loss of ribbon/s. Also, how will they explain this phenomenon with Knockout models of SR.  

Response: Thank you for understanding and for the suggestion. We have provided histological images for WT vs S1PR2 KO retina in the supplement, Supplement Figure S2.

 

1. Page 7 / Role of S1PR2 in the expression of synaptic proteins / Line 277 – “Table 1” only mentioned the list of antibodies, not raw quantification data, which seems irrelevant here to mention.

Response: Thank you, we have updated.

 

 

1. I would request the authors to be consistent with their marking throughout the manuscript like Figure 3 – Graph marking G, H, and I (mark them from the corner following other images). Figure 4 – The marking of A, B, C, D, E, and F is on the opposite side compared to other images.

Response: Thank you, we have updated.

 

1. Figures 3, 4, 5, and 6 / Please mark all the retina layers in the SIPR2 KO for proper representation (3D, 4A & D, 5, 6C).

Response: Thank you, we have updated.

 

10. Please check the image size and reorganize the images (Figure 4), as well as either use only a scale bar or a scale bar with written value.

Response: Thank you, we have updated.

 

1. I would request the authors to provide a quantification (graph) of fluorescence intensity (multiple ROI) of Rhodopsin, MAGUK, and SV2 marker comparing WT vs S1PR2 KO in the supplementary. The % Area labeled by IHC would be variable due to tissue conditions or sectioning or due to different locations (distance from the optic nerve). Moreover, In Figures 5 and 6 the expression of MAGUK and SV2 is higher (visually appearing in the representative image).Is it possible to show loss of SR in Fig 2 and how does the distribution of Synaptic vesicle look at Electron microscopy level? 

Response: Thank you, we have updated. Please see the new figures, supplementary figures 3-5.

 

1. Figures 8 and 9 / Please mark the retinal layers in the representative images.

Response: Thank you, we have updated.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Sphingosine-1-phosphate receptor 2 (S1PR2) is an intriguing topic. It has been previously established to play a significant role in central nervous system (CNS) stress responses, making it relevant in the highly CNS-like environment of the retina. This study provides valuable insights and offers a methodological approach to studying visual signaling in the retina. However, there are still some key questions that need to be addressed.  I have found almost the same article in PubMed (bioRxiv. 2023 Sep 5:2023.09.01.555709. doi: 10.1101/2023.09.01.555709). Hence, I must reject this manuscript in its current status of this manuscript.

 

1.         The scotopic ERG results in Figure 1B show that S1PR2 KO mice exhibit higher A-wave amplitudes at flash intensities > 4.0 cd∙s/m². However, the results in Figure 1A do not quantitatively support this finding. Therefore, the results at higher flash intensities may be necessary. The same issue arises in Figure 1D-F.

 

2.         In addition, there may be some controversy regarding the OCT results in Figure 2.

 

(1)      Firstly, the results in Figure 2A and Figure 2B are too blurry to rule out the possibility of width misjudgment due to differences in focus depth.

(2)      Secondly, there is a noticeable visual difference in the length and width of the optic nerve (the black cavity in the center of the retina) in Figure 2A and Figure 2B. To ensure consistency in the magnification factor, it may be necessary to provide scale bars for both the vertical and horizontal axes.

(3)      The significant differences between Figure 2D and 2E may require further clarification. For instance, in Figure 2D, the results show that the RNFL and OPL thickness in S1PR2 KO mice are significantly greater than in WT mice. However, the thickness difference between the two groups is even larger in the ONL, yet this doesn't appear to be significantly different on the graph. This phenomenon is further magnified in Figure 2E.

 

    In summary, providing clear H&E stained retinal sections may better support the OCT and statistical results. Furthermore, confirming the effects of S1PR2 KO on various retinal layers, especially ONL, IS, and OS, might be necessary.

 

3.         Continuing from the above, the results and reports related to the ONL may need more careful scrutiny and conclusions.

(1)      Figures 2D and 2E reveal that the ONL thickness in S1PR2 KO mice increases with age (3 months vs. 6 months), accompanied by a decrease in photoreceptor outer segment (OS) thickness when compared to WT mice.

 

(2)      The cells density and tissue thickness of DAPI-stained cells in S1PR2 KO mice (Figures 3A and 3D) appears visually greater than that in WT mice. However, the results in Figure 3G suggest a reduction in the number of ONL cell nuclei in S1PR2 KO mice

The above results suggest that S1PR2 KO may contribute to photoreceptor performance. This could be a reason for the elevated A-wave in ERG, and it likely involves the generation of downstream B-wave signals. Therefore, it should not be overlooked, even though there are still concerns about the ERG results in Figure 1.

 

4.         It seems that there are some missing labels in the figures. Are the scale bars in Figure 4 and Figure 5 50 µm? Furthermore, the descriptions in Figures 5, 8, and 9 are too brief to confirm the relative locations or even orientations of the areas displayed in the retina. It is suggested that at least two areas should be labeled, similar to Figure 6, for assisting readers in identification.

 

5.         Above issues also occurred in Figure 7 (and possibly in other figures). It is recommended to review and provide basic labeling to help readers at least determine the orientation and relative positions, or to provide low-magnification retinal images with marked imaging areas. In addition, we may have overlooked some information, making it difficult to identify the dendritic areas labeled with white arrows (Figures 7A-D) and axon areas (Figures 7E-F). Therefore, we cannot provide recommendations regarding this result.

 

In summary, this article suggests that S1PR2 KO enhances pre-synaptic and post-synaptic signals in bipolar cells, resulting in stronger visual signals. It emphasizes that this effect is achieved through morphological changes in dendrites and axons rather than an increase in photoreceptor cells. To support the arguments presented in this article, it might be necessary to determine the impact of photoreceptor cells on visual signals. Therefore, some of the data in this study is subject to debate and requires further validation, particularly regarding the main conclusions.

Author Response

Dear Reviewer: We appreciate the time and energy the reviewer committed and the value of their comments. We kindly inform the reviewer that bioRxiv. 2023 Sep 5:2023.09.01.555709. doi: 10.1101/2023.09.01.555709 is a preprint only.

 

 

1)   The scotopic ERG results in Figure 1B show that S1PR2 KO mice exhibit higher A-wave amplitudes at flash intensities > 4.0 cd∙s/m². However, the results in Figure 1A do not quantitatively support this finding. Therefore, the results at higher flash intensities may be necessary. The same issue arises in Figure 1D-F.

Response: Figures 1A and 1D are examples of Waveforms of Scotopic and Photopic ERGs, respectively. These are from a single mouse with single flash and these are representative of the data. These are not quantitative and not for comparing with the data in Figures 1B, 1C, and 1F.

 

 

2)    In addition, there may be some controversy regarding the OCT results in Figure 2.

 

• Firstly, the results in Figure 2A and Figure 2B are too blurry to rule out the possibility of width misjudgment due to differences in focus depth.

Response: These are representative images to show the layers. We provided representative Histological images in Supplement Figure S2 in the revised manuscript. As described in the methods, OCT images were used to measure the thickness of different retinal layers by the ‘automated image segmentation software provided by the manufacturer was used for the segmentation of the images, and each segmentation was manually verified. Thickness data was collected from an equidistant region between 0.6-0.8 mm away from the center of the optic nerve on both sides of the image. Therefore, four readings were collected from the two images (horizontal and vertical) per eye’.

 

• Secondly, there is a noticeable visual difference in the length and width of the optic nerve (the black cavity in the center of the retina) in Figure 2A and Figure 2B. To ensure consistency in the magnification factor, it may be necessary to provide scale bars for both the vertical and horizontal axes.

Response: As described in the methods, each OCT image was captured for the 2 mm central retina at the best focus, so there is no chance of any magnification difference. Again, the retinal layers were measured at the central retina 0.6-0.8 mm away from the center of the optic nerve on both sides of the image. Therefore, no scaling is required for these representative images, and the optic nerve does not play any role in these measurements.

 

• The significant differences between Figure 2D and 2E may require further clarification. For instance, in Figure 2D, the results show that the RNFL and OPL thickness in S1PR2 KO mice are significantly greater than in WT mice. However, the thickness difference between the two groups is even larger in the ONL, yet this doesn't appear to be significantly different on the graph. This phenomenon is further magnified in Figure 2E.

Response: A significant difference was determined by statistics from the data of N= 64 from WT and 64 from S1P2 KO readings. It is likely that the ONL thickness was not significantly different because of greater variability.

 

In summary, providing clear H&E stained retinal sections may better support the OCT and statistical results. Furthermore, confirming the effects of S1PR2 KO on various retinal layers, especially ONL, IS, and OS, might be necessary.

Response: We provided H&E stained images for retinal sections in Supplement Figure S2. Analysis of layer thickness at the central retina of 8 WT and 9 S1P2 KO mice did not show a significant difference in the ONL, though the thickness of the entire central retina appears to be higher in the KO mice.

 

3)    Continuing from the above, the results and reports related to the ONL may need more careful scrutiny and conclusions.

(1)      Figures 2D and 2E reveal that the ONL thickness in S1PR2 KO mice increases with age (3 months vs. 6 months), accompanied by a decrease in photoreceptor outer segment (OS) thickness when compared to WT mice.

Response: Thank you for this observation and the comments. However, statistical analysis does not support a significant difference between WT and KO mice for ONL and OS thickness.

 

(2)      The cells density and tissue thickness of DAPI-stained cells in S1PR2 KO mice (Figures 3A and 3D) appears visually greater than that in WT mice. However, the results in Figure 3G suggest a reduction in the number of ONL cell nuclei in S1PR2 KO mice

Response: Thank you for this observation and the comments. However, statistical analysis does not support a significant difference between WT and KO mice for ONL and OS DAPI-stained cells.

 

4)    It seems that there are some missing labels in the figures. Are the scale bars in Figure 4 and Figure 5 50 µm? Furthermore, the descriptions in Figures 5, 8, and 9 are too brief to confirm the relative locations or even orientations of the areas displayed in the retina. It is suggested that at least two areas should be labeled, similar to Figure 6, for assisting readers in identification.

Response: Thank you, we have updated.

 

In summary, this article suggests that S1PR2 KO enhances pre-synaptic and post-synaptic signals in bipolar cells, resulting in stronger visual signals. It emphasizes that this effect is achieved through morphological changes in dendrites and axons rather than an increase in photoreceptor cells. To support the arguments presented in this article, it might be necessary to determine the impact of photoreceptor cells on visual signals. Therefore, some of the data in this study is subject to debate and requires further validation, particularly regarding the main conclusions.

Response: Thank you for the acknowledgment.

 

 

Author Response File: Author Response.pdf