Violet-Blue Light Photobiological Effect on Cultured Corneal and Pigment Retinal Cells

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The study's concept is interesting. The idea of ​​controlling light sources is sound, as not all of them are safe, especially those emitting in the blue spectrum.

 

The idea of ​​testing the potential safety of 405 nm light, which is used for disinfection, for the human eye of an operator present in the room during disinfection is very important. The data presented in the study show that exposure to 405 nm light on ARPE-19 cell culture and cornea results in an average 10% reduction in cell viability.

 

I would also like to emphasize the importance of the developed irradiation setup. This is a common problem when irradiating cell culture plates, ensuring uniform illumination of each well. The authors have proposed an excellent solution to this problem. This is novel and convenient: a 3D-printed prototype with several LEDs centered on the 405 nm wavelength was used to ensure uniform light distribution during exposure.

 

 

Notes.

However, there are a couple of important caveats regarding the biological component of the study.

First, there are studies on the effects of light, including 405 nm, on ARPE-19 cell cultures, showing that oxygen stress does occur, at least from the mitochondria (doi: 10.1111/aos.13812).

 

The second point is much more significant. ARPE-19 cells do not contain lipofuscin (LG) and melanin granules (DOI: 10.1016/j.exer.2014.01.016), which are present in this tissue directly in the eye (Kennedy, C.J.; Rakoczy, P.E.; Constable, I.J. Lipofuscin of the retinal pigment epithelium: A review. Eye 1995, 9, 763–771). This is important because There are many studies showing that the presence of LG in RPE cells is a risk factor for the development of pathologies (Holz, F.G.; Pauleikhoff, D.; Klein, R.; Bird, A.C. Pathogenesis of lesions in late age-related macular disease. Am. J. Ophthalmol. 2004, 137, 504–510.; Katz, M.L. Potential role of retinal pigment epithelial lipofuscin accumulation in age-related macular degeneration. Arch. Gerontol. Geriatrics2002, 34, 359–370.) at least due to its phototoxicity and the generation of reactive oxygen species under the influence of light (Boulton M, Dontsov A, Jarvis-Evans J, et al. Lipofuscin is a photoinducible free radical generator. J PhotochemPhotobiol B Biol 1993; 19: 201–204). Exposure of ARPE-19 cells filled with LG to blue light (wavelength 390–470 nm) significantly reduces their viability and induces apoptosis (https://doi.org/10.1016/S0891-5849(96)00555-2, DOI: 10.1016/s0891-5849(01)00582-2, Olchawa, M.M.; Furso, J.A.; Szewczyk, G.M.; Sarna, T.J. Lipofuscin-mediated photic stress inhibits phagocytic activity of ARPE-19 cells; effect of donors’ age and antioxidants. Free Radic. Res. 2017, 51, 799–811, https://doi.org/10.3390/ijms232012234).

 

Thus, taking into account the comments, I propose shifting the focus of the discussion of the results slightly from the fact that the study data showed that 405 nm light is safe for selected ocular structures. To the fact that a highly effective model lighting setup was developed and used in a preliminary model experiment on ARPE-19 cells and the cornea. That these studies were conducted within the framework of the idea of ​​testing the safety of disinfecting light in medical practice. And that the next step will be testing this on ARPE-19 cells loaded with LG to assess the physiological risks to the human RPE. Perhaps the discussion should also include discussion of the proposed protective measures for the operator during such light disinfection procedures, which include wearing glasses with yellow filters that block light up to 450 nm, perhaps with references to studies that mention such light-correcting glasses.

 

Minor comments:

 

1) ARPE-19 is retinal pigment epithelial cell cultures, and the term "retinal cell cultures" is not very correct.

It would be more accurate to write about retinal pigment epithelial cell cultures, not retinal cell cultures.

 

2) Lines 202-203. Finally, BCE C/D-1b cells treated for 9 hours showed a reduction in cell 202 viability of approximately 4.2% (reduction of 0.018 log10, CI 0.01-0.03) (Figure 3).

In Figure 3B, I see changes for ARPE-19, but for BCE cells, the control and irradiated cells are similar. This is inconsistent with the text.

 

3) Line 222 - From the collected data, we can conclude that light radiation does not exert a cytotoxic effect on ocular tissue cells after both short and long exposures.

You just wrote that light reduces the viability of ARPE-19 cells by 12%, and then you say there's no effect. How is that possible?

I agree that the effect on corneal epithelium is less significant, given the data provided.

Author Response

REVISOR 1

The study's concept is interesting. The idea of controlling light sources is sound, as not all of them are safe, especially those emitting in the blue spectrum.

The idea of testing the potential safety of 405 nm light, which is used for disinfection, for the human eye of an operator present in the room during disinfection is very important. The data presented in the study show that exposure to 405 nm light on ARPE-19 cell culture and cornea results in an average 10% reduction in cell viability.

I would also like to emphasize the importance of the developed irradiation setup. This is a common problem when irradiating cell culture plates, ensuring uniform illumination of each well. The authors have proposed an excellent solution to this problem. This is novel and convenient: a 3D-printed prototype with several LEDs centered on the 405 nm wavelength was used to ensure uniform light distribution during exposure.

We sincerely thank the reviewer for the positive assessment of our study and for recognizing the clinical relevance of evaluating 405 nm light safety for operators during disinfection procedures. We are particularly pleased that the reviewer highlighted the novelty and effectiveness of our 3D-printed irradiation setup.

Note 1:

However, there are a couple of important caveats regarding the biological component of the study.

First, there are studies on the effects of light, including 405 nm, on ARPE-19 cell cultures, showing that oxygen stress does occur, at least from the mitochondria (doi: 10.1111/aos.13812).

The second point is much more significant. ARPE-19 cells do not contain lipofuscin (LG) and melanin granules (DOI: 10.1016/j.exer.2014.01.016), which are present in this tissue directly in the eye (Kennedy, C.J.; Rakoczy, P.E.; Constable, I.J. Lipofuscin of the retinal pigment epithelium: A review. Eye 1995, 9, 763–771). This is important because There are many studies showing that the presence of LG in RPE cells is a risk factor for the development of pathologies (Holz, F.G.; Pauleikhoff, D.; Klein, R.; Bird, A.C. Pathogenesis of lesions in late age-related macular disease. Am. J. Ophthalmol. 2004, 137, 504–510.; Katz, M.L. Potential role of retinal pigment epithelial lipofuscin accumulation in age-related macular degeneration. Arch. Gerontol. Geriatrics2002, 34, 359–370.) at least due to its phototoxicity and the generation of reactive oxygen species under the influence of light (Boulton M, Dontsov A, Jarvis-Evans J, et al. Lipofuscin is a photoinducible free radical generator. J PhotochemPhotobiol B Biol 1993; 19: 201–204). Exposure of ARPE-19 cells filled with LG to blue light (wavelength 390–470 nm) significantly reduces their viability and induces apoptosis (https://doi.org/10.1016/S0891-5849(96)00555-2, DOI: 10.1016/s0891-5849(01)00582-2, Olchawa, M.M.; Furso, J.A.; Szewczyk, G.M.; Sarna, T.J. Lipofuscin-mediated photic stress inhibits phagocytic activity of ARPE-19 cells; effect of donors’ age and antioxidants. Free Radic. Res. 2017, 51, 799–811, https://doi.org/10.3390/ijms232012234).

Thus, taking into account the comments, I propose shifting the focus of the discussion of the results slightly from the fact that the study data showed that 405 nm light is safe for selected ocular structures. To the fact that a highly effective model lighting setup was developed and used in a preliminary model experiment on ARPE-19 cells and the cornea. That these studies were conducted within the framework of the idea of testing the safety of disinfecting light in medical practice. And that the next step will be testing this on ARPE-19 cells loaded with LG to assess the physiological risks to the human RPE. Perhaps the discussion should also include discussion of the proposed protective measures for the operator during such light disinfection procedures, which include wearing glasses with yellow filters that block light up to 450 nm, perhaps with references to studies that mention such light-correcting glasses.

We are deeply grateful to the reviewer for the insightful and constructive feedback. We fully agree with the suggestion to refine the focus of our manuscript. In line with this, we have significantly expanded the "Limitations of the Study" section within the Discussion to address the biological caveats you raised. As suggested, we have integrated a detailed discussion regarding the absence of lipofuscin and melanin in the ARPE-19 model, acknowledging that their presence in vivo significantly increases the susceptibility of the RPE to phototoxicity and ROS generation (citing the references provided, such as Olchawa et al., 2017 and Katz, 2002). By explicitly listing these factors in the Limitations section, we now clarify that our results represent a baseline cytocompatibility profile and that testing LG-loaded cells is the necessary next step to accurately assess physiological risks. Furthermore, we have shifted the emphasis of the manuscript to highlight the innovation of our 3D-printed irradiation setup as a standardized tool for preliminary screenings. We have also enriched the Discussion by proposing practical protective measures for operators, such as the use of yellow-filter glasses (blue-blockers), to mitigate potential risks during disinfection procedures. These revisions ensure a more nuanced biological interpretation while maintaining the technological value of the study.

Minor comments:

Note 2:

ARPE-19 is retinal pigment epithelial cell cultures, and the term "retinal cell cultures" is not very correct.

It would be more accurate to write about retinal pigment epithelial cell cultures, not retinal cell cultures.

We thank the reviewer for this important terminological clarification. We agree that the term "retinal cell cultures" is imprecise as it could imply a broader range of retinal cells, including neurons and photoreceptors. Following your suggestion, we have carefully revised the manuscript to ensure that the term "retinal pigment epithelial cell cultures" (or RPE) is used throughout the text when referring to the ARPE-19 model. This change ensures greater scientific accuracy and correctly identifies the specific cellular layer being investigated.

Note 3:

Lines 202-203. Finally, BCE C/D-1b cells treated for 9 hours showed a reduction in cell 202 viability of approximately 4.2% (reduction of 0.018 log10, CI 0.01-0.03) (Figure 3).

In Figure 3B, I see changes for ARPE-19, but for BCE cells, the control and irradiated cells are similar. This is inconsistent with the text.

Thank you very much for pointing out the inconsistency. The result reported in the text was not consistent with that shown in the figure. We have corrected the typo and updated the data in the 'Results' paragraph.

Note 4:

Line 222 - From the collected data, we can conclude that light radiation does not exert a cytotoxic effect on ocular tissue cells after both short and long exposures.

You just wrote that light reduces the viability of ARPE-19 cells by 12%, and then you say there's no effect. How is that possible? I agree that the effect on corneal epithelium is less significant, given the data provided.

We appreciate the reviewer pointing out this inconsistency, a concern also raised by the other reviewer. We take your point regarding the interpretation of the viability data. While a 12% reduction in ARPE-19 viability may often fall within the range of experimental variation, especially when considering the standard deviation, we agree that dismissing it entirely as having "no cytotoxic effect" is imprecise and contradicts the reported data. In response to this comment, we have revised the manuscript to describe this result more accurately. Instead of stating there was "no effect," we now characterize it as a "weak decrease in cell viability." This ensures a more nuanced and cautious interpretation of the biological data. We have updated the corresponding sections in the Results and Conclusions to reflect this change, ensuring that our claims are fully aligned with the observed data.

Reviewer 2 Report

Comments and Suggestions for Authors

This study is a reasonable and, seemingly carefully executed culture study.  In general, however, the authors over interpret their findings (they cannot go from this cell line to the safety of the whole eye).  There are a number of both major and minor issues that are detailed below.

1. More information is necessary about the LEDs that were used.  I suggest providing a graph showing the spectrum.  LEDs are not a single wavelength, their waveband can be quite wide.
2. The authors repeatedly extrapolate from isolated cell lines (ARPE-19, bovine corneal endothelial cells) to whole-eye safety, including implications for photobiological standards and human exposure. The retina’s blue-light hazard is not mediated by RPE viability alone. Photoreceptors, Müller glia, retinal circuitry, and optical filtering (lens, macular pigment) are critical.
3. ARPE-19 cells lack normal outer segment phagocytosis, phototransduction coupling, and realistic chromophore loading. Corneal endothelium is not the most light-vulnerable corneal layer (epithelium and subbasal nerve plexus are).  This is not really an ocular safety study.
4. The authors are pretty glib at dismissing 12-13% loss in the MTT signal (ARPE-19)- “no cytotoxic effect” I am actually pretty sure that is biologically meaningful.
5. These exposure paradigms are not particularly realistic. Cells are irradiated from below with no ocular optics.  Constant irradiance for 3-9 hours does not in any way “mimic indoor workers”  (real life exposure is intermittent).
6. The ROS assessment is too narrow. ROS conclusions rely almost entirely on DCFH-DA fluorescence at one time point and catalase expression.  There was no assessment of lipid peroxidation, DNA oXidation, mitochondrial stress, but then you state that there was “no oxidative stress”
7. Apoptosis is under-sampled.  
8. You note that the results could inform updates to IEV photobiological safety standards. This is over-reach to the extreme.  Two cell-lines, one wavelength, one irradiance range, no invivo or optical modeling, etc
9. Using odds ratios for continuous absorbance data is unconventional and unexplaines.
10. No correction for multiple comparsions.
11. ROS figure reports a single experiment (n = 48 wells not biological replicates)
12. Your cell model choice was not the best.  ARPE-19 cells are already known to be remarkably resistant to oxidative stress. Bovine corneal cells are not neurons and metaboloically robust.  No photoreceptors, no other retinal cells…

 

Author Response

REVISOR 2

This study is a reasonable and, seemingly carefully executed culture study. In general, however, the authors over interpret their findings (they cannot go from this cell line to the safety of the whole eye). There are a number of both major and minor issues that are detailed below.

We sincerely thank the reviewer for this insightful observation. In response to this concern, we have carefully revised the manuscript to avoid overinterpretation.

Note 5:

More information is necessary about the LEDs that were used.  I suggest providing a graph showing the spectrum.  LEDs are not a single wavelength, their waveband can be quite wide.

We thank the reviewer for this valuable feedback. We have expanded the description of the LED source in the Methods section. The device (SST-10-UV-A130-F405-00, Luminus Devices) has a nominal peak wavelength of 405 nm, a wavelength bin of 405–410 nm and an FWHM spectral bandwidth of approximately 10 nm at 500 mA, in line with the manufacturer's specifications. We would like to clarify that the reported wavelength refers to the nominal peak emission and not to a monochromatic source.

Note 6:

The authors repeatedly extrapolate from isolated cell lines (ARPE-19, bovine corneal endothelial cells) to whole-eye safety, including implications for photobiological standards and human exposure. The retina’s blue-light hazard is not mediated by RPE viability alone. Photoreceptors, Müller glia, retinal circuitry, and optical filtering (lens, macular pigment) are critical.

We fully agree with the reviewer’s assessment. This study is intended to be exploratory and does not claim to describe the complex photobiological effects on the entire eye, which involve multiple cell types and physiological filters not captured in our model. However, we selected the ARPE-19 cell line as it is a well-validated tool for investigating RPE-specific cellular responses. These cells display many properties of native RPE, including characteristic cobblestone morphology, the formation of polarized structures on porous supports, and the expression of RPE-selective markers such as CRALBP. Due to these RPE-like features, these cells are frequently used in retinal research to identify fundamental molecular mechanisms (doi:10.3389/fnins.2022.938089).

In response to this feedback, we have revised the manuscript to ensure that our conclusions are strictly confined to the specific in vitro responses observed. We have also explicitly stated throughout the text that these results cannot be used to define whole-eye safety.

Note 7:

ARPE-19 cells lack normal outer segment phagocytosis, phototransduction coupling, and realistic chromophore loading. Corneal endothelium is not the most light-vulnerable corneal layer (epithelium and subbasal nerve plexus are).  This is not really an ocular safety study.

We appreciate the reviewer’s perspective and have adjusted the focus of our manuscript accordingly. We have shifted the emphasis from "whole-eye safety" to the specific data showing that 405 nm light is safe for the selected ocular cell structures under the conditions tested. It is important to clarify that these studies were conducted within the framework of evaluating the safety of disinfecting light for medical practice; our goal was to establish a preliminary study of biocompatibiliy for individual cell layers. To address the reviewer’s concerns, we have added a dedicated "Limitations of the Study" paragraph in the Discussion section. In this section, we explicitly discuss the physiological limitations of using isolated cell lines, such as the lack of outer segment phagocytosis and phototransduction. We have also stated that the next necessary step will be testing ARPE-19 cells loaded with lipofuscin granules (LG) to more accurately assess the physiological risks to the human RPE. We have revised the entire text to ensure our conclusions are strictly proportionate to these in vitro findings.

Note 8:

The authors are pretty glib at dismissing 12-13% loss in the MTT signal (ARPE-19)- “no cytotoxic effect” I am actually pretty sure that is biologically meaningful.

We take the reviewer’s point regarding the interpretation of the MTT signal. While a 12–13% reduction in viability may often fall within the range of experimental variation, especially when considering the standard deviation, we agree that dismissing it entirely as having "no cytotoxic effect" may be imprecise. In response to this comment, we have revised the manuscript to describe this result more accurately. Instead of stating there was "no effect," we now characterize it as a  "weak decrease in cell viability".  This ensures a more nuanced and cautious interpretation of the biological data. We have updated the corresponding sections in the Results and Discussion to reflect this change, ensuring that the description is more aligned with the observed data.

Note 9:

These exposure paradigms are not particularly realistic. Cells are irradiated from below with no ocular optics.  Constant irradiance for 3-9 hours does not in any way “mimic indoor workers”  (real life exposure is intermittent).

We agree that the exposure paradigm does not replicate real-world ocular exposure conditions. The model was designed to provide controlled, homogeneous irradiance at the cellular level, not to reproduce anatomical optical filtering or intermittent exposure patterns. By adopting this exposure paradigm (3–9 hours of continuous irradiation), we aimed to test the cells under the most demanding conditions possible. We believe that if the cell layers demonstrate safety under such rigorous, continuous exposure, it provides a robust baseline for the lower, intermittent levels typical of real-world indoor environments. We have clarified this rationale in the Discussion, specifying that this was a stress test designed to identify the upper limits of cellular tolerance.

Note 10:

The ROS assessment is too narrow. ROS conclusions rely almost entirely on DCFH-DA fluorescence at one time point and catalase expression.  There was no assessment of lipid peroxidation, DNA oXidation, mitochondrial stress, but then you state that there was “no oxidative stress”

We agree that the ROS assessment could be more comprehensive. To strengthen our findings, we have included an analysis of SOD1 expression, which, together with Catalase, provides a clearer picture of the cellular antioxidant defense system. Our data show that these enzymes are not downregulated following exposure; the fact that their expression levels remain stable, alongside the DCFH-DA results, indicates that the redox balance within the exposed cells is maintained. We have revised the manuscript to clarify that this specific antioxidant profile suggests a lack of acute oxidative stress under our experimental conditions. Furthermore, we have updated the "Limitations" section to acknowledge that while these markers are stable, other specific pathways such as lipid peroxidation or DNA oxidation were not explored in this study.

Note 11:

Apoptosis is under-sampled.

We appreciate the reviewer’s comment regarding the depth of the apoptosis assessment. To ensure a rigorous evaluation of programmed cell death, our study specifically focused on the expression of key molecular markers in both ARPE-19 and BCE C/D-1b cell lines.

As widely recognized in the literature, Caspase-3 acts as the primary executioner enzyme in the apoptotic cascade, and its cleavage is a definitive hallmark of apoptosis activation (doi:10.1074/jbc.273.16.9357; doi:10.1080/01926230701320337). Our results demonstrate no cleavage of Caspase-3, alongside stable expression of the total protein, following light exposure in both cell models.

To further strengthen these findings, in the revised version of the manuscript, we have also included a Western Blot analysis of Cytochrome c. The release of Cytochrome c from the mitochondria to the cytosol is a pivotal upstream event in the intrinsic apoptotic pathway. Our data show no increase in Cytochrome c expression following treatment, consistent with the lack of Caspase-3 activation.

We believe that the absence of both Cytochrome c involvement and Caspase-3 cleavage provides robust molecular evidence that the major apoptotic pathways are not triggered under our experimental conditions.

Note 12:

You note that the results could inform updates to IEV photobiological safety standards. This is over-reach to the extreme.  Two cell-lines, one wavelength, one irradiance range, no invivo or optical modeling, etc

We agree that our findings alone are insufficient to inform direct revisions to photobiological safety standards. We have therefore modified the relevant statements, presenting our results as preliminary mechanistic evidence that could contribute to the wider debate.

Note 13:

Using odds ratios for continuous absorbance data is unconventional and unexplaines

We agree. Odds ratios apply to binary outcomes and were inadvertently mentioned in the Methods section. Our analyses were performed on continuous fluorescence/absorbance values by comparing exposed samples to the corresponding controls using log-transformed ratios and reporting mean effects with 95% confidence intervals (back-transformed to percent change for interpretation). We have removed the odds ratio wording and revised the Methods accordingly.

Note 14:

No correction for multiple comparsions.

We acknowledge this point. Our analyses focus on estimation (effect sizes with 95% confidence intervals) for pre-specified comparisons versus controls, rather than formal hypothesis testing across a large family of post‑hoc comparisons; therefore, a multiple-testing adjustment was not applied. We have clarified this in the Statistical Analysis section.

Note 15:

ROS figure reports a single experiment (n = 48 wells not biological replicates)

We acknowledge the reviewer’s comment regarding the number of biological replicates. To ensure the robustness and reproducibility of our findings, we have performed two additional independent experiments for the ROS assessment. The updated Figure now reflects the results from three independent biological replicates (n=3), rather than multiple technical replicates from a single session. We have updated the figure, the corresponding results, and the figure legend to accurately describe this expanded dataset.

Note 16:

Your cell model choice was not the best.  ARPE-19 cells are already known to be remarkably resistant to oxidative stress. Bovine corneal cells are not neurons and metaboloically robust.  No photoreceptors, no other retinal cells…

We acknowledge the reviewer’s perspective regarding the metabolic resilience of ARPE-19 and bovine corneal cells. We agree that these models do not capture the full physiological vulnerability of the retina, particularly the sensitivity of neurons and photoreceptors. As specified in the revised "Limitations of the Study" section, our goal was to establish a preliminary cytocompatibility screening using standardized models. In that section, we now explicitly discuss the limitations of using isolated ARPE-19 cells, such as the absence of outer segment phagocytosis and, crucially, the lack of lipofuscin granules (LG), which are known to increase RPE sensitivity to blue light. We have clarified that these findings represent a baseline safety profile for individual cell layers and that further studies involving LG-loaded cells and more complex models will be essential to accurately assess the risks to the human RPE and neuroretina.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors took all my comments into account and carefully revised the article. I recommend the manuscript for publication.

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have adequately responded to the suggested revisions.