Transcriptomics analysis reveals molecular alterations underpinning spaceflight dermatology

Background Spaceflight poses a unique set of challenges to humans and the hostile spaceflight environment can induce a wide range of increased health risks, including dermatological issues. The biology driving the frequency of skin issues in astronauts is currently not well understood. Methods To address this issue, we used a systems biology approach utilizing NASA’s Open Science Data Repository (OSDR) on space flown murine transcriptomic datasets focused on the skin, biochemical profiles of 50 NASA astronauts and human transcriptomic datasets generated from blood and hair samples of JAXA astronauts, as well as blood samples obtained from the NASA Twins Study, and skin and blood samples from the first civilian commercial mission, Inspiration4. Results Key biological changes related to skin health, DNA damage & repair, and mitochondrial dysregulation are identified as potential drivers for skin health risks during spaceflight. Additionally, a machine learning model is utilized to determine gene pairings associated with spaceflight response in the skin. While we identified spaceflight-induced dysregulation, such as alterations in genes associated with skin barrier function and collagen formation, our results also highlight the remarkable ability for organisms to re-adapt back to Earth via post-flight re-tuning of gene expression. Conclusion Our findings can guide future research on developing countermeasures for mitigating spaceflight-associated skin damage.

The authors performed comprehensive analyses to investigate the effects of spaceflight on skin using public RNA-seq data.They identified 102 key genes involved in spaceflight response and showed that cell cycle-and lipogenesis-associated genes were changed in murine skin.In the analyses of RNA-seq data, they compared the mice to the astronaut.In this study, different tissues (skin, blood, and hair follicles) obtained from different experiments using different mouse strains were compared.This made the interpretation of the results quite complicated and weakened the conclusion validity.As the experimental design, the comparisons are not suitable, while the reviewer understands the limitation of data from samples obtained in the space.This reviewer thinks that correlation of data between murine skin and human skin has not been fully verified.In addition, this study did not reveal whether the identified genes or pathways actually contribute to skin pathogenesis in the space.Also, it should be specified in the Methods section whether the skin samples used in this study contain both epidermis and dermis, or only epidermis.Some interesting genes were picked up and described in the text,.However, it is hard to find those genes in the comprehensive figures.The authors should reconsider figure layouts.
In Figure 3B, 5A and 6B, many genes have reversed expression patterns depending on the mouse strains (BALB,C3H and C57BL).This indicates that mouse strains in addition to spaceflight influence the gene expressions.To elucidate whether the gene expression changes in each mouse strain are correlated with those in humans, the authors should add figures that show the comparison of the same genes examined in murine skin with those in human skin data from the Inspiration4 studies.Murine skin data should be compared with human skin data.
In Figure 4A and Figure 5B, blood data from JAXA CFE and NASA Twins and hair follicle data from JAXA should not be presented together.Human blood data should be presented relative to mouse blood data.The validity of comparing gene expression in murine skin and human blood should be well discussed.
In Line 362, the authors described that Vitamin D supplementation is already used on the ISS.The authors should clearly mention whether the astronauts who supplied skin samples took Vitamin D supplementation in the space.
Minor comments: Line 196: Figs.3C and 3B should be Figs.3B and 3C.In line 297-316, citation of "Figure 6" is lacked.Line 270: FLG should be spelled out at the first appearance in Line 249. Figure 9: the font size is different in some drug names.
Reviewer #2 (Remarks to the Author): Review of Cope et al., More than a Feeling: Dermatological Changes Impacted by Spaceflight The authors describe an analysis of unpublished mouse RNAseq datasets and their comparison with previously published human datasets.All datasets are RNAseq datasets related to dermatological analyses under space conditions.The heterogeneity of the datasets requires a complex analysis strategy performed by the authors using classical and AI-based analyses.The analysis strategy is plausible, and the presentation of results is largely appropriate.However, the presentation of the results lacks precision, which is absolutely necessary due to the complex analysis situation.1.The missing description of important abbreviations makes it difficult to follow the article without bioinformatics or immunological knowledge.Abbreviations are not explained in the text: Gene Ontology Biological Process (GOBP), differential expressed genes (DEG), Peripheral blood mononuclear cells (PBMC), log fold change (LFC; are those log2 or log10?),Molecular Signatures Database (MSigDB) etc. 2. Abbreviations that are explained are not explained at their first occurrence (e.g.FDR).3. In addition, facts are presented in an incomprehensibly abbreviated manner: "union of intra-MHU-2 combinations", in figure 2 4. Different FDR thresholds are applied.Here, the use of FDRs above the 10% threshold is questionable and requires at least a detailed explanation: Figure 5, FDR ≤ 0.25 Lines 126-130: Please indicate the cluster method and genes (genes in text or in a supplement).Lines 138-139 : Please give the GLYCAM1 P-value.Line 166 : Please report the results for the 102 key genes in a supplement.Lines 175-176 "For example, in the case of the gene with the lowest p-value, LAMA1 (Laminin Subunit Alpha 1)": Please indicate the significance level or P-value and the compared conditions.Line 166: Please mention string and the confidence level (0.7).Lines 204-205 "It is interesting to note the significantly lower upregulation of these metabolic pathway genes for female mice.":Please indicate the significance level or P-value Lines 231-294: The data interpretation is unclear.The total number of regulated genes in the studies is needed for the interpretation of the overlap.Lines 330-331: "One of these models demonstrated how the upregulation of D2HGDH" the bioinformatical context is needed: number and size of models.Lines 356-357: "Calcitriol and L-asparaginase exhibited significant activation scores for the majority of the datasets": please give the scores at least in a supplement.Lines 644: 0.05 is a normal threshold for FDR and not "highly significant".Lines 665-670: Please indicate if the analysis is done by your team or is done in the previous study.Figures 1 B and 3 D: text size and colour selection is suboptimal.

Dear Editor and Reviewers,
We thank both the editor and reviewers for the comments.We believe based on these final revisions we have made that this manuscript is now stronger and much improved.We have addressed the comments by the reviewers and our responses appear in red font below the original reviewer comment.We look forward to the next steps.

Reviewers' comments:
Reviewer #1 (Remarks to the Author): The authors performed comprehensive analyses to investigate the effects of spaceflight on skin using public RNA-seq data.They identified 102 key genes involved in spaceflight response and showed that cell cycle-and lipogenesis-associated genes were changed in murine skin.In the analyses of RNA-seq data, they compared the mice to the astronaut.In this study, different tissues (skin, blood, and hair follicles) obtained from different experiments using different mouse strains were compared.This made the interpretation of the results quite complicated and weakened the conclusion validity.As the experimental design, the comparisons are not suitable, while the reviewer understands the limitation of data from samples obtained in the space.This reviewer thinks that correlation of data between murine skin and human skin has not been fully verified.
Thank you for the reviewer's comment, we have modified language to frame the contributions of the investigation more thoroughly, including in the abstract.The potential links to dermatological manifestations in astronauts have been made clearer, especially in the discussion.
1. Also, it should be specified in the Methods section whether the skin samples used in this study contain both epidermis and dermis, or only epidermis.
We thank the reviewer, we made use of public data, and unfortunately we do not have detailed accounts of the dissection procedures beyond the metadata in the database from which the data was obtained (which we point the reader to in the methods section).A note has also been added to the study limitations.
2. Some interesting genes were picked up and described in the text,.However, it is hard to find those genes in the comprehensive figures.The authors should reconsider figure layouts.
At the request of the reviewer we have revised figure layouts and visibility is now improved.For example, Figure 3A now shows only cross-mission genes mentioned throughout the text, and some heat maps have been moved to supplemental so that larger text can be shown for gene names in the main figures.
3. In Figure 3B, 5A and 6B, many genes have reversed expression patterns depending on the mouse strains (BALB, C3H and C57BL).This indicates that mouse strains in addition to spaceflight influence the gene expressions.To elucidate whether the gene expression changes in each mouse strain are correlated with those in humans, the authors should add figures that show the comparison of the same genes examined in murine skin with those in human skin data from the Inspiration4 studies.Murine skin data should be compared with human skin data.
Thanks to the excellent reviewer suggestion in Figure 5 we now show the skin health genes in human data from the Inspiration4 mission alongside the murine skin data.
4. In Figure 4A and Figure 5B, blood data from JAXA CFE and NASA Twins and hair follicle data from JAXA should not be presented together.Human blood data should be presented relative to mouse blood data.The validity of comparing gene expression in murine skin and human blood should be well discussed.
Based on the comments of the reviewer and studies finding differences between gene expression in skin and blood samples (e.g.PMIDs: 29228364, 26091259, 22299064), we have prioritized demonstrating key genes involved in space flight response in murine skin and subsequently investigated the modifications in expression of these genes both in murine and human skin and hair follicles; but given the availability of blood samples from astronauts, we have also included the latter to our analysis.Moreover, time points and experimental conditions/study design for analysis of spaceflight response of human skin and murine skin differed.Due to this limitation, although we did try to correlate the results between murine skin and human skin, and seek common alterations in gene expression, a one on one comparison was not the main focus of this study.We have rather prioritized on outlining the results obtained under different experimental conditions, side by side.And, in cases where we have seen a global signature in skin tissue, such as a specific gene down regulation in both human and murine skin, we did try to explain some of the alterations in murine skin with the results acquired from human blood, assuming that a similar change would be observed in mice blood.Nevertheless, despite the differences in study design, we did still seek global signatures in all datasets; human blood & skin and mice skin.In addition to refocusing the figures on skin and hair follicle data by moving blood sample data into supplemental figures, we have also rephrased the language so that we are not implying direct comparison between blood and skin.
5. In Line 362, the authors described that Vitamin D supplementation is already used on the ISS.The authors should clearly mention whether the astronauts who supplied skin samples took Vitamin D supplementation in the space.
We have added the following text to the that section to provide more clarity for vitamin D supplementation: In the blood markers section of the results we write "Notably, this decrease in vitamin D is observed despite consistent programmatic vitamin D supplementation since the collection of blood samples 62 ."and in the i4 section of the methods we write "Records of vitamin D supplement consumption were not available for the i4 crew."We have addressed these minor comments and thank the reviewer for their thoroughness.
Reviewer #2 (Remarks to the Author): Review of Cope et al., More than a Feeling: Dermatological Changes Impacted by Spaceflight The authors describe an analysis of unpublished mouse RNAseq datasets and their comparison with previously published human datasets.All datasets are RNAseq datasets related to dermatological analyses under space conditions.The heterogeneity of the datasets requires a complex analysis strategy performed by the authors using classical and AI-based analyses.The analysis strategy is plausible, and the presentation of results is largely appropriate.However, the presentation of the results lacks precision, which is absolutely necessary due to the complex analysis situation.
We thank the reviewer for their insightful comments.We have substantially revised the manuscript to improve the presentation to bring more precision with the results.We have also refined our analysis to further make the presentation and the results more clear and precise.
1.The missing description of important abbreviations makes it difficult to follow the article without bioinformatics or immunological knowledge.Abbreviations are not explained in the text: Gene Ontology Biological Process (GOBP), differential expressed genes (DEG), Peripheral blood mononuclear cells (PBMC), log fold change (LFC; are those log2 or log10?),Molecular Signatures Database (MSigDB) etc.
2. Abbreviations that are explained are not explained at their first occurrence (e.g.FDR).
In addition to introducing additional acronyms, we have edited to explain them during their first occurrences.
3. In addition, facts are presented in an incomprehensibly abbreviated manner: "union of intra-MHU-2 combinations", in figure 2 4. Different FDR thresholds are applied.Here, the use of FDRs above the 10% threshold is questionable and requires at least a detailed explanation: Figure 5, FDR ?0.25 To accommodate the reviewer's comment, we have updated the plots to show FDR < 0.05 and have provided all the Gene Set Enrichment Analysis (GSEA) values as supplemental data for readers to reference.Throughout the paper we use FDR < 0.1 for defining significant genes, and FDR < 0.05 for defining significant pathways.Additionally, we removed the term "union of intra-MHU-2 combinations" as it was indeed confusing; we have tried to simplify processes throughout, including the derivation of "key genes", which are now referred to as "cross-mission genes", which we believe to be easier to interpret.
While we have decided on a more stringent cutoff of FDR < 0.05 for pathways, we note that the scientists that created the GSEA algorithm do consider an appropriate cutoff for statistics to be FDR < 0.25.
4. Lines 126-130: Please indicate the cluster method and genes (genes in text or in a supplement).

6.
Minor comments:  a. Line 196: Figs.3C and 3B should be Figs.3B and 3C  b.In line 297-316, citation of "Figure6" is lacked c.Line 270: FLG should be spelled out at the first appearance in Line 249 d.Figure9: the font size is different in some drug names.