Skin-derived G-CSF activates pathological granulopoiesis upon psoriasis

  1. Tomson Kosasih
  2. Tatsuya Morishima
  3. Sohyeon Lee
  4. Jungyeon Yoon
  5. Kanako Wakahashi
  6. Pilhan Kim
  7. Aiko Sada
  8. Hitoshi Takizawa

Reviewed by

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

Log in to save this article

Abstract

Psoriasis is an inflammatory skin disease initiated by environmental triggers and driven by disruption of T cell cytokine network in the cutaneous milieu. The fact that complete resolution of disease by targeting key inflammatory cytokines remains challenging indicates a contribution of other immune cells to the pathogenesis. Here, we study the role of neutrophils in psoriasis, the first-line innate immune defender that is short-lived but mobile and infiltrate into various tissues. We found that upon psoriasis induction, skin-resident endothelial cells are activated to produce G-CSF which activates emergency granulopoiesis in bone marrow and induces cutaneous infiltration and accumulation of neutrophil that are functionally overactive. Depletion of neutrophils or blockage of psoriasis-driven granulopoiesis by respective neutralizing antibodies results in reducing cutaneous neutrophil burden and mitigating psoriasis pathogenesis. This mechanism might be conserved in human psoriasis as confirmed by public RNA-seq database. Our findings uncovered and detailed the pathological crosstalk between skin and BM in psoriatic inflammation, proposing a potential therapeutic approach targeting cross-organ communication.

Article activity feed

  1. Review Commons

    Note: This response was posted by the corresponding author to Review Commons. The content has not been altered except for formatting.

    Learn more at Review Commons

    Reply to the reviewers

    Reviewer #1

    First, the authors have not convincingly shown that skin cells, or more specifically skin ECs, are a major source of circulating G-CSF in the psoriasis model as stated in the title and abstract. The data in Figure 4 show selective upregulation of Csf3 gene in skin ECs and their ability to secrete G-CSF upon IMQ treatment in vitro. However, the provided data do not address to what degree the skin EC-derived G-CSF contributes to the elevated level of circulating G-CSF. Additional experiments to selectively deplete G-CSF in skin ECs, or at least in skin cells of the affected site, are warranted to support the authors' claim. Does intradermal injection of G-CSF neutralizing antibody into the psoriatic skin reduce circulating levels of G-CSF?

    Author's response:

    Thank you for reviewer's comment. We agree with the Reviewer#1 that it is important to directly block G-CSF to the skin via intradermal injection and measure the G-CSF level in the serum afterwards. Therefore, we will perform intradermal injection of IgG-isotype or anti-G-CSF antibody into the IMQ-induced psoriatic mice.

    Another concern is insufficient demonstration of G-CSF-mediated emergency granulopoiesis in the psoriasis model. All data in Figure 5 were obtained from experiments with only n=3, and adding more replicates, in particular to those in Figure 5B, which show quite some variation in MPP numbers, is recommended. The relatively small reduction of BM granulocyte numbers (Figure 5C) compared to greater depletion of circulating granulocytes (Figure S5A) raises the possibility that it is the mobilization effect rather than granulopoiesis-stimulating effect that skin-derived G-CSF exerts to promote supply of circulating neutrophils that eventually infiltrate into the affected skin. This could also explain the negligible effect of IL-1blockade (Figure S4), which selectively shut off myelopoiesis-stimulating effect of IL-1 (Pietras et al. Nat Cell Biol 2016, PMID: 27111842). Are the HSPCs in the psoriasis model more cycling? Do they show myeloid-skewed differentiation when cultured ex vivo or upon transplantation?

    Author's response: Thank you for these critical comments. We agree to do the following experiments to address them:

    1. HSPCs quantification in Figure 5 especially the MPPs will be added with more replicates.

    2. We will assess cycling status of HSPCs by flow cytometric analysis of Ki67and Propidium Iodide to characterize G0, G1 and G2/M cell cycle phase.

    3. To test myeloid-skewed differentiation, Lin- c-Kit+ Sca-1+ cells containing HSPCs will be isolated from bone marrow of Vas/IMQ-treated mice and transplanted into lethally irradiated syngeneic mice.

    The authors' claim that skin-derived G-CSF "induces" neutrophil infiltration warrants further clarification. Alternative explanation is that the upregulated neutrophil-attracting chemokines (Figure S1D) could induce infiltration, whereas G-CSF increase the number of neutrophils to circulate in the vessels near the psoriatic skin. This notion seems supported elsewhere (Moos et al. J Invest Dermatol. 2019, PMID: 30684554). Can the infiltration be inhibited by systemically injecting neutralizing antibody of their receptor, CXCR2?

    Author's response: The manuscript focuses on the skin-derived G-CSF function as a long-distance signal for emergency granulopoiesis in the bone marrow upon psoriasis, not the chemoattractant property of it. The sentence of interest is "We found that upon psoriasis induction, skin-resident endothelial cells are activated to produce G-CSF which activates emergency granulopoiesis in bone marrow and induces cutaneous infiltration and accumulation of neutrophil that are functionally inflammatory." in line 28-30. In agreement with point #2 from Reviewer#2, the fact that neutrophil recruitment factors (CXCL1, CXCL2, and CXCL5) were upregulated in psoriatic skin (Figure S1D), suggesting a CXCL-mediated neutrophil recruitment. The sentence of concern need to be changed to "We found that upon psoriasis induction, skin-resident endothelial cells are activated to produce G-CSF which activates emergency granulopoiesis in bone marrow, leading to cutaneous accumulation of neutrophil that are functionally inflammatory.". This revised sentence has omitted the proposal that G-CSF directly dictates neutrophils mobilization to the skin, which is not the key message of the study. Therefore, we found that the CXCR2 (CXCLs receptor) blockade experiment may be of the benefit of future studies.

    It remains unclear how skin-derived G-CSF accumulates pathogenic neutrophils. The authors state "pathogenic granulopoiesis," but are the circulating neutrophils in the psoriatic mice already "pathogenic" or do they acquire pathogenic phenotype after cutaneous infiltration? Additional RNA-seq to compare circulating and infiltrated neutrophils would answer this question.

    Author's response: We appreciate this valuable comment. We will perform RNA-seq with the peripheral blood-circulating neutrophils (CD45+ CD11b+ Ly6G+ Ly6Cmid) versus skin-infiltrating neutrophils from both Vas/IMQ mice.

    In addition, how the accumulated pathogenic neutrophils exacerbate the psoriatic changes remains obscure. Although the authors have attempted to correlate Il17a gene expression in infiltrated neutrophils with psoriatic skin changes, the data do not address to what degree it contributes to cutaneous IL-17A protein levels. The data that cutaneous neutrophil depletion leads to subtle decrease in skin IL-17A expression (Figure 2H) rather supports alternative possibilities. For instance, as indicated elsewhere, IL-17A cutaneous tone could be enhanced by neutrophil-mediated augmentation of Th17 or gamma/delta T cell function (Lambert et al. J Invest Dermatol. 2019, PMID: 30528823). Does neutrophil depletion or G-CSF neutralization alter cell numbers or function of cutaneous Th17 and gamma/delta T cells?

    Author's response: Thank you for this insightful comment. To better understand the relative contribution of neutrophils to the cutaneous IL-17A tone in the psoriatic skin, we will perform flowcytometric analysis of Th17 and gamma/delta T cells which are widely known as the major source of IL-17 in psoriatic skin of IMQ-induced mice following injection of isotype-matched or anti-Ly6G antibody.

    Finally, as the above conclusions rely solely on the IMQ-induced acute psoriasis model, it would be informative if they could be derived from another psoriasis model. IMQ is known to induce unintended systemic inflammation due to grooming-associated ingestion (Gangwar et al. J Invest Dermatol. 2022, PMID: 34953514), and "pathological crosstalk between skin and BM in psoriatic inflammation" could be strengthened by an intradermal injection model.

    Author's response: We appreciate the reviewer for bringing this important point. Regarding the systemic inflammation upon psoriasis, the above-cited study reported increased IFN-B expression in the intestines of IMQ-ingested animal (Grine L et al. Sci Rep. 2016, PMID: 26818707 in Gangwar et al. J Invest Dermatol. 2022, PMID: 34953514). We examined several pro-inflammatory cytokines including IFN-b, IFN-g, and IL-6 and in contrast, found no systemic increase in all these cytokines, except for IFN-g downregulation (Explanation Figure 1), which suggests no evidence of grooming-associated ingestion.

    We also examined the *Csf3 *expression across several distinctively located tissues which showed a selective upregulation in the skin (Figure 4C), suggesting a skin-restricted perturbation. In addition, one study showed that IMQ-ingestion didn't alter number of gut injury-associated CXCR3+ macrophages nor did it aggravate skin inflammation (Pinget et al. Cell Reports. 2022, PMID: 35977500). Together, these findings support that IMQ-induced psoriasis by topical cutaneous application used in our study elicit a local inflammation but not systemic inflammation.

    The authors, however, realize that testing alternative psoriasis model such as intradermal injection of IL-23 (Chan et al. J Exp Med. 2006, PMID: 17074928) will strengthen the skin-local insults within the psoriasis model employed, and should be tested in the future.

    Minor comments

    Figure 1E shows multiple elongated Ly6G+ structures in d0-2 control and d0 IMQ skins that do not appear to be neutrophils.

    Author's response: We appreciate the Reviewer#1 pointing this issue. As mentioned by the Reviewer#1, the elongated structures detected in the intravital microscopy are not neutrophils, but autofluorescence from the skin bulge regions (Wun et al. J Invest Dermatol. 2005, PMID: 15816847). We have eliminated these unspecific signals from the transformation and quantification (Figure 1F, S1G, and S1H). We will also add an explanatory sentence in Materials and Methods section "Of note, the fluorescent signal with elongated structures resembling hair bulge were autofluorescence and thus removed from further analysis." to be more precise about our methods.

    In Figure 2C, the bottom GSEA seems to be showing type II IFN response, not type I IFN, according to the text.

    Author's response: Thank you for the comment, we will correct this misspelling.

    Author's response: We appreciate that Reviewer#1 bring up this point. We examined the kinetics of the bone marrow cellularity and GMPs across 4 days of psoriasis induction in mice. The bone marrow cell number was lowered along that span with lowermost count at 2 days. Consistent to the BM-cellularity, the GMP number was also lowered about one-third in the first 2 days of psoriasis. This kinetic is consistent with the previous report showing a rapid reduction of GMPs in the bone marrow within 2 days following systemic G-CSF administration driven emergency granulopoiesis (Hirai et al. Nat. Immunol. 2006, PMID: 16751774). From 2 days to 4 days, the GMP number rapidly increased to slightly above basal number (Explanation Figure 2). This timely coordinated expansion suggests a significant supply of GMPs from the differentiating upstream myeloid progenitors (Figure 3B).

    When the psoriatic mice with elevated G-CSF is injected with anti-G-CSF or IgG-isotype antibody, the bone marrow cellularity and GMP numbers at 4 days were (Explanation Figure 3). Firstly, as psoriasis reduced bone marrow cellularity (Explanation Figure 2), the unchanged number after anti-G-CSF injection indicates that administration of 10µg/day for 4 days does not significantly affect mobilization of psoriatic bone marrow cells. Secondly, the similar GMP numbers at 4 days psoriasis is plausibly due to snapshot analysis when it has already in the numerical recovery period (Explanation Figure 2). Importantly, the notion that anti-G-CSF injection to psoriatic mice reduced granulocytes in the bone marrow, peripheral blood, and skin suggesting G-CSF as a key mediator in psoriatic driven emergency granulopoiesis on top of unlikely case of ineffective anti-G-CSF treatment.

    Taken together, these data suggest a G-CSF mediated emergency granulopoiesis occurrence in the IMQ-induced psoriasis. We will put these data into a revised Figure.

    In Figures 6B, in which cluster of human skin cells IL-17A expression would be enriched?

    Author's response: Thank you for this important point. The IL-17A expression is found in the T-cell cluster (Explanation Figure 4). We also expected to see IL-17A contribution from other cell subset(s), in particular neutrophil. However, due to the fragile nature of neutrophils and thereby, technical difficulty to get their sequencing reads, this dataset (GSE173706) doesn't contain neutrophils, but rather monocytes, macrophages, and dendritic cells among the myeloid subset (Explanation Figure 5). With this, it leaves open the question on what potential contribution of IL-17A produced by neutrophils is in human psoriasis (Reich et al. Exp. Dermatol. 2015, PMID: 25828362).

    Figure 1E shows multiple elongated Ly6G+ structures in d0-2 control and d0 IMQ skins that do not appear to be neutrophils.

    Author's response: We appreciate the Reviewer#1 pointing this issue. As mentioned by the Reviewer#1, the elongated structures detected in the intravital microscopy are not neutrophils, but autofluorescence from the skin bulge regions (Wun et al. J Invest Dermatol. 2005, PMID: 15816847). We have eliminated these unspecific signals from the transformation and quantification (Figure 1F, S1G, and S1H). We will also add an explanatory sentence in Materials and Methods section "Of note, the fluorescent signal with elongated structures resembling hair bulge were autofluorescence and thus removed from further analysis." to be more precise about our methods.

    In Figure 2C, the bottom GSEA seems to be showing type II IFN response, not type I IFN, according to the text.

    Author's response: Thank you for the comment, we will correct this misspelling.

    Reviewer#2

    Interpretation of neutrophil transcriptomic changes (Figure 2)

    The RNA-seq analysis reveals substantial downregulation of several canonical pro inflammatory pathways in neutrophils from psoriatic skin, including IL-6, IL-1, and type II interferon signaling. The authors should discuss the functional relevance of this unexpected transcriptional repression. For example, does this indicate a shift toward specialized effector functions rather than classical cytokine responsiveness? More importantly, the most striking transcriptional change is the upregulation of NADPH oxidase-related genes (e.g., Nox1, Nox3, Nox4, Enox2). This suggests an oxidative stress-driven pathogenic mechanism, potentially more relevant than IL-17A production. Yet this aspect is not explored in the manuscript. Assessing ROS levels or oxidative neutrophil effector functions in this model would considerably strengthen the mechanistic link. Conversely, although IL-17A is upregulated in neutrophils, neutrophil depletion reduces total Il17a expression in skin only partially. This indicates that neutrophils are unlikely to be the dominant IL-17A source in the lesion. The authors' focus on neutrophil-derived IL 17A therefore seems overstated. A more rigorous assessment-e.g., conditional deletion of Il17a specifically in neutrophils-would be required to establish its true contribution. Taken together, the data suggest that oxidative programs, rather than IL-17A production, may represent the principal pathogenic axis downstream of neutrophils, and this deserves deeper discussion.

    Author's response: Thank you for raising this valuable views. We have agreed to address these critical points by the following approaches:

    1. To address the changes in NADPH oxidase-related gene signature, we will measure ROS production in the neutrophils from skin and peripheral blood with DHR123.

    2. Responding to the IL17A contribution by neutrophils, we will flow cytometrically assess the Th17 and gamma/delta T cell population in the skin of psoriatic mice treated with anti-Ly6G or isotype-matched antibody as was suggested by Reviewer#1.

    3. We will discuss downregulation of the canonical pro inflammatory and IL-17 pathways in the psoriatic neutrophils in the discussion.

    Human data reanalysis (Figure 6):

    The re-analysis of bulk and single-cell RNA-seq datasets is valuable but incomplete. Several mechanistically relevant questions could be addressed with the available data:

    2.1. GM-CSF (CSF2) is also strongly upregulated in psoriatic lesions (bulk RNA-seq). It would be informative to determine whether endothelial cells also express CSF2 in the scRNA-seq dataset, as this would suggest coordinated regulation of myeloid-supporting cytokines.

    2.2. Myeloid cell subsets should be examined more closely. A comparison of human myeloid transcriptomes with the mouse neutrophil RNA-seq would clarify whether similar IL-17A-related or NADPH oxidase-related signatures occur in human disease. In particular, which cell types express IL17A in human lesions?

    2.3. Chemokine production should be attributed to specific cell types. Bulk RNA-seq confirms strong induction of CXCL1, CXCL2, CXCL5, but the scRNA-seq dataset allows determining whether these chemokines originate from endothelial cells or other stromal/immune populations. This information is important for defining whether endothelial cells coordinate both neutrophil recruitment and granulopoiesis.

    Addressing these points would make the human-mouse comparison substantially stronger.

    Author's response: Thank you for pointing these important issues. By reanalyzing the dataset, we found several points regarding the comments, as follows:

    2.1) *CSF2 *is expressed by T-cell cluster in the human skin dataset (Explanation Figure 4), in agreement with previous murine study (Hartwig et al. Cell Reports. 2018, PMID: 30590032). We will add this data in the revised manuscript.

    2.2) In line with point#10 from Reviewer#1, the dataset clearly shows T-cell cluster as the main IL17A source (Explanation Figure 4 above). The dataset, however, doesn't contain phenotypic neutrophils (CEACAM (CD66b) and PGLYRP1) but monocytes, macrophages, and dendritic cells (Explanation Figure 5 above). This loss was probably due to a technical limitation given the difficulty in capturing sequencing reads from fragile neutrophils. Therefore, it is no longer possible to reanalyze IL-17 expression in the absence of neutrophils in the datapool.

    2.3) Reanalysis of CXCLs in the human scRNAseq dataset (GSE173706) clarified their secretion dynamics and cellular sources under normal and psoriatic condition. In normal skin, all examined cell subsets show only low CXCLs expression. In contrast, psoriatic skin exhibits significant CXCLs upregulation with distinct cell subsets clearly showing dramatic upregulation, potentially being the major CXCLs source. *CXCL1 *is markedly upregulated in fibroblasts, myeloid cells, and melanocyte and nerve cells. *CXCL2 *is strikingly upregulated to myeloid cells, while *CXCL5 *is hugely increased in fibroblasts, myeloid cells, and mast cells (Explanation Figure 7). Taken together, these results suggest that CXCLs upregulation in the psoriatic skin is coordinatively executed by both stromal and immune compartments. Of note, the endothelial cells show minimal changes in CXCLs expression, even downregulate *CXCL2 *in psoriasis, indicating that they are unlikely to be the major contributor to CXCL-mediated neutrophil recruitment.

    **Referees cross-commenting**

    I agree with Reviewer 1 that the contribution of EC-derived G-CSF to circulating G-CSF levels and to emergency myelopoiesis requires additional genetic or neutralization experiments to be fully established.

    Author's response: We appreciate that Reviewer#2 raised this key point. In addition to examining the serum G-CSF upon intradermal anti-G-CSF administration in point#1 from Reviewer#1 above, we will also examine the emergency myelopoiesis signs in vivo.

    Minor points

    Line 319: the text likely refers to Figure S4, not S3.

    Author's response: Thank you, we will correct the nomenclature.

    Line 338: "psoriatic" is misspelled.

    Author's response: Thank you, we will change this to "psoriatic".

    Reviewer #3

    • Place the work in the context of the existing literature (provide references, where appropriate).

    Psoriasis is extensively studied, a good recent reference- https://doi.org/10.1016/j.mam.2024.101306

    Author's response: Thank you for Reviewer#3's suggestion. The referenced study highlights the current paradigm that largely focus on skin-restricted mechanism and overlook potential cross-organ interaction in the psoriasis inflammation. Our findings provide a new insight into the skin-bone marrow crosstalk in the disease context. In addition, the suggested reference underscores the key roles of diverse innate immune cells including neutrophils, eosinophils, dendritic cells, etc. which is fundamental for our study and might also guide future exploration of additional innate cell subsets beyond neutrophils. We will therefore include the mentioned reference to our revised manuscript.

    • Do you have suggestions that would help the authors improve the presentation of their data and conclusions?

    It is all good. May add graphical-abstract.

    Author's response: Thank you for the reviewer's input, we agree that a graphical-abstract will help the readers more clearly grasp the key messages of our manuscript. We will include it in the revised manuscript.

    Major comments:

    • Should the authors qualify some of their claims as preliminary or speculative, or remove them altogether?

    No. It is very solid.

    Author's response: We appreciate the reviewer's view that the claims in our paper are solid.

    • Would additional experiments be essential to support the claims of the paper? Request additional experiments only where necessary for the paper as it is, and do not ask authors to open new lines of experimentation.

    Such a discovery clearly opens many options, and it is fascinating to suggest additional experiments for future studies. It is a complete study, best to publish as-is and let many to read and proceed with this new concept.

    Author's response: We thank the reviewer for noting that the current experimental evidence is complete that no additional experiments are necessary at this stage. We agree that the discovery opens prospective directions for future studies.

    • Are the suggested experiments realistic in terms of time and resources? It would help if you could add an estimated cost and time investment for substantial experiments.

    N/A - I suggest no additional experiments at this point. Get it published and see how many will follow this new direction!

    Author's response: We thank the reviewer for recognizing that the experimental data has been sufficient to be a foundation for the future research.

    • Are the data and the methods presented in such a way that they can be reproduced?

    Yes.

    Author's response: We thank the reviewer for recognizing that our methods are reproducible.

    • Are the experiments adequately replicated, and is the statistical analysis adequate?

    Yes. The data are of very high quality.

    Author's response: We are grateful that the reviewer view our replication strategy and statistical analysis are of a high quality.

    Minor comments:

    • Specific experimental issues that are easily addressable.

    None. It is good as-is. One may always suggest minor things- but this one is better published so many laboratories may rush for this new direction. I think it will be interesting studying some long-term impacts, and changes not only of neutrophils but also of other innate cells, such as DCs, Macrophages, and Eosinophils - so it is best to let laboratories that focus on these cells know of the discovery and pursue independent studies.

    Author's response: We appreciate the reviewer's assessment that our paper is already well set for the community to explore the newly proposed direction.

    • Are the text and figures clear and accurate?

    Yes.

    Author's response: We thank the reviewer's evaluation. We have ensured that the text and figures in our manuscript are clear and accurate. Once again, we thank the reviewer for the encouraging and constructive appraisal. We are pleased that the reviewer find the manuscript has already been strong and suitable for publication.

  2. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #3

    Evidence, reproducibility and clarity

    Summary:

    Provide a short summary of the findings and key conclusions (including methodology and model system(s) where appropriate).

    Study titled: "Skin-derived G-CSF activates pathological granulopoiesis upon psoriasis" by Kosasih and Takizawa. Paper show establishment of psoriasis model in C57BL/6 mice. They focus on neutrophils infiltration following the Imiquimod cream induction. Importantly, authors show that the induction of psoriasis in the skin cause a robust enhancement of granulopoiesis in the bone marrow. Mechanistically, G-CSF is produced in the skin, especially by endothelial cells. Blocking of G-CSF gained clear inhibition of psoriatic pathology. They further add human data showing that patient with psoriasis have more neutrophils and more G-CSF in their skin endothelial cells.

    Parts of the study are simply in line with previous knowledge (e.g.- neutrophils infiltration into psoriatic skin, IL17a). authors show some data that largely confirm the model used. Major discovery: skin endothelial cells are secreting G-CSF that induce granulopoiesis in the bone-marrow. This is a conceptual advancement of this study: psoriatic skin not only recruit neutrophils from the blood, but also enhance the generation of new neutrophils in the bone-marrow. That a major- psoriasis at the level of the model used must not be considered as a confined-pathology. It affect systematically, and might also benefit new systemic treatments. There are plenty of follow-up experiments to pursue now, so it is critical to publish this finding and let many laboratories to know of this new direction. I expect this study to attract high interest and many citations.

    Major comments:

    • Are the key conclusions convincing?

    Yes. The study has excellent data, with good quantification, and very solid support for the discovery and interpretations.

    • Should the authors qualify some of their claims as preliminary or speculative, or remove them altogether?

    No. It is very solid.

    • Would additional experiments be essential to support the claims of the paper? Request additional experiments only where necessary for the paper as it is, and do not ask authors to open new lines of experimentation.

    Such a discovery clearly opens many options, and it is fascinating to suggest additional experiments for future studies. It is a complete study, best to publish as-is and let many to read and proceed with this new concept.

    • Are the suggested experiments realistic in terms of time and resources? It would help if you could add an estimated cost and time investment for substantial experiments.

    N/A - I suggest no additional experiments at this point. Get it published and see how many will follow this new direction!

    • Are the data and the methods presented in such a way that they can be reproduced?

    Yes.

    • Are the experiments adequately replicated, and is the statistical analysis adequate?

    Yes. The data are of very high quality.

    Minor comments:

    • Specific experimental issues that are easily addressable.

    None. It is good as-is. One may always suggest minor things- but this one is better published so many laboratories may rush for this new direction. I think it will be interesting studying some long-term impacts, and changes not only of neutrophils but also of other innate cells, such as DCs, Macrophages, and Eosinophils - so it is best to let laboratories that focus on these cells know of the discovery and pursue independent studies.

    • Are prior studies referenced appropriately?

    Yes. I may suggest adding a recent review by Park and Jung, 2024, https://doi.org/10.1016/j.mam.2024.101306 to cover current concepts of innate immunity in psoriasis.

    • Are the text and figures clear and accurate?

    Yes.

    • Do you have suggestions that would help the authors improve the presentation of their data and conclusions?

    It is all good. May add graphical-abstract.

    Significance

    • Describe the nature and significance of the advance (e.g., conceptual, technical, clinical) for the field.

    Conceptual advancement - discovery of a major impact of psoriasis on bone-marrow granulopoiesis. Explicit finding of endothelial-cells G-CSF as a major communication moiety.

    Neutrophil recruitment and IL17A are well established. G-CSF of endothelial cells brings the conceptual advancement- psoriasis at the level induced by IMQ develops local pathology, but is tightly linked to systemic changes. The impact on bone-marrow granulopoiesis may have many implications. So far, it was largely considered that chronic inflammation may affect hematopoiesis, but this study reveals an acute and specific communication between skin and bone marrow. The neutrophils are not only recruited from blood- they are made anew, so the disease is enhanced significantly! This discovery led to a novel basic understanding and suggests novel therapeutic options.

    • State what audience might be interested in and influenced by the reported findings.

    Dermatologist, immunologist, haematologist - this one goes for a broad audience.

    • Define your field of expertise with a few keywords to help the authors contextualize your point of view. Indicate if there are any parts of the paper that you do not have sufficient expertise to evaluate.

    Immunology and hematology. I am not an expert of dermatology.

  3. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #2

    Evidence, reproducibility and clarity

    General assessment

    This is a well-written and carefully executed study that identifies skin-derived G-CSF as a key driver of pathological emergency granulopoiesis in an imiquimod-induced mouse model of psoriasis. The authors convincingly show that endothelial cells are the dominant source of G-CSF in inflamed skin, and that this cytokine mediates systemic hematopoietic skewing and neutrophil accumulation, ultimately aggravating psoriatic pathology. The eanalysis of human transcriptomic datasets strengthens the translational relevance of the findings. Overall, the conclusions are well supported by the data. However, several mechanistically relevant aspects remain underexplored, particularly regarding the functional state of psoriatic neutrophils and the human data integration. Addressing these points would substantially enhance the impact of the study.

    Major points

    1. Interpretation of neutrophil transcriptomic changes (Figure 2)

    The RNA-seq analysis reveals substantial downregulation of several canonical proinflammatory pathways in neutrophils from psoriatic skin, including IL-6, IL-1, and type II interferon signaling. The authors should discuss the functional relevance of this unexpected transcriptional repression. For example, does this indicate a shift toward specialized effector functions rather than classical cytokine responsiveness? More importantly, the most striking transcriptional change is the upregulation of NADPH oxidase-related genes (e.g., Nox1, Nox3, Nox4, Enox2). This suggests an oxidativestress-driven pathogenic mechanism, potentially more relevant than IL-17A production. Yet this aspect is not explored in the manuscript. Assessing ROS levels or oxidative neutrophil effector functions in this model would considerably strengthen the mechanistic link.

    Conversely, although IL-17A is upregulated in neutrophils, neutrophil depletion reduces total Il17a expression in skin only partially. This indicates that neutrophils are unlikely to be the dominant IL-17A source in the lesion. The authors' focus on neutrophil-derived IL17A therefore seems overstated. A more rigorous assessment-e.g., conditional deletion of Il17a specifically in neutrophils-would be required to establish its true contribution. Taken together, the data suggest that oxidative programs, rather than IL-17A production, may represent the principal pathogenic axis downstream of neutrophils, and this deserves deeper discussion.

    1. Human data reanalysis (Figure 6):

    The re-analysis of bulk and single-cell RNA-seq datasets is valuable but incomplete.

    Several mechanistically relevant questions could be addressed with the available data:

    2.1. GM-CSF (CSF2) is also strongly upregulated in psoriatic lesions (bulk RNA-seq). It would be informative to determine whether endothelial cells also express CSF2 in the scRNA-seq dataset, as this would suggest coordinated regulation of myeloid-supporting cytokines.

    2.2. Myeloid cell subsets should be examined more closely. A comparison of human myeloid transcriptomes with the mouse neutrophil RNA-seq would clarify whether similar IL-17A-related or NADPH oxidase-related signatures occur in human disease. In particular, which cell types express IL17A in human lesions?

    2.3. Chemokine production should be attributed to specific cell types. Bulk RNA-seq confirms strong induction of CXCL1, CXCL2, CXCL5, but the scRNA-seq dataset allows determining whether these chemokines originate from endothelial cells or other stromal/immune populations. This information is important for defining whether endothelial cells coordinate both neutrophil recruitment and granulopoiesis. Addressing these points would make the human-mouse comparison substantially stronger.

    Minor points

    1. Line 319: the text likely refers to Figure S4, not S3.
    2. Line 338: "psoriatic" is misspelled.

    Referees cross-commenting

    I agree with Reviewer 1 that the contribution of EC-derived G-CSF to circulating G-CSF levels and to emergency myelopoiesis requires additional genetic or neutralization experiments to be fully established.

    Significance

    The study is solid and potentially impactful, particularly for audiences working in inflammation and hematopoiesis, as it uncovers a cross-organ mechanism linking skinderived G-CSF to emergency granulopoiesis in psoriasis. My expertise lies in inflammation and hematopoiesis, and from this perspective several essential mechanistic issues remain insufficiently addressed. In particular, the neutrophil transcriptomic data highlight strong induction of NADPH oxidase-related pathways, which appears more biologically meaningful than the modest Il17a upregulation emphasized by the authors. Likewise, the human RNA-seq reanalyses leave open key questions regarding CSF2 expression, myeloid heterogeneity, and chemokine cellular sources. These issues affect the strength and interpretation of the central claims. For these reasons, I recommend major revision before the manuscript can be considered further.

  4. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #1

    Evidence, reproducibility and clarity

    Summary:

    A role of neutrophils in psoriasis pathogenesis has been highlighted by several past studies; however, how the neutrophils are recruited to the affected skin has not been fully understood. The work by Kosasih et al. tackles a relevant question and has investigated the effect of psoriatic skin inflammation on BM myelopoiesis. Using an IMQ-induced acute psoriasis mouse model, the authors derive 3 major conclusions: (1) skin ECs secrete G-CSF into circulation in response to psoriatic stress, (2) skin EC-derived G-CSF stimulates emergency granulopoiesis, and (3) skin EC-derived G-CSF induces infiltration and accumulation of pathogenic neutrophils in the affected skin. The authors provide many pieces of interesting data, but most of them remain correlative and insufficient to support the conclusions. Many of the experiments were performed in a small number of samples or mice (mostly with n=3), leaving the story still preliminary.

    Major comments:

    1. First, the authors have not convincingly shown that skin cells, or more specifically skin ECs, are a major source of circulating G-CSF in the psoriasis model as stated in the title and abstract. The data in Figure 4 show selective upregulation of Csf3 gene in skin ECs and their ability to secrete G-CSF upon IMQ treatment in vitro. However, the provided data do not address to what degree the skin EC-derived G-CSF contributes to the elevated level of circulating G-CSF. Additional experiments to selectively deplete G-CSF in skin ECs, or at least in skin cells of the affected site, are warranted to support the authors' claim. Does intradermal injection of G-CSF neutralizing antibody into the psoriatic skin reduce circulating levels of G-CSF?
    2. Another concern is insufficient demonstration of G-CSF-mediated emergency granulopoiesis in the psoriasis model. All data in Figure 5 were obtained from experiments with only n=3, and adding more replicates, in particular to those in Figure 5B, which show quite some variation in MPP numbers, is recommended. The relatively small reduction of BM granulocyte numbers (Figure 5C) compared to greater depletion of circulating granulocytes (Figure S5A) raises the possibility that it is the mobilization effect rather than granulopoiesis-stimulating effect that skin-derived G-CSF exerts to promote supply of circulating neutrophils that eventually infiltrate into the affected skin. This could also explain the negligible effect of IL-1blockade (Figure S4), which selectively shut off myelopoiesis-stimulating effect of IL-1 (Pietras et al. Nat Cell Biol 2016, PMID: 27111842). Are the HSPCs in the psoriasis model more cycling? Do they show myeloid-skewed differentiation when cultured ex vivo or upon transplantation?
    3. The authors' claim that skin-derived G-CSF "induces" neutrophil infiltration warrants further clarification. Alternative explanation is that the upregulated neutrophil-attracting chemokines (Figure S1D) could induce infiltration, whereas G-CSF increase the number of neutrophils to circulate in the vessels near the psoriatic skin. This notion seems supported elsewhere (Moos et al. J Invest Dermatol. 2019, PMID: 30684554). Can the infiltration be inhibited by systemically injecting neutralizing antibody of their receptor, CXCR2?
    4. It remains unclear how skin-derived G-CSF accumulates pathogenic neutrophils. The authors state "pathogenic granulopoiesis," but are the circulating neutrophils in the psoriatic mice already "pathogenic" or do they acquire pathogenic phenotype after cutaneous infiltration? Additional RNA-seq to compare circulating and infiltrated neutrophils would answer this question.
    5. In addition, how the accumulated pathogenic neutrophils exacerbate the psoriatic changes remains obscure. Although the authors have attempted to correlate Il17a gene expression in infiltrated neutrophils with psoriatic skin changes, the data do not address to what degree it contributes to cutaneous IL-17A protein levels. The data that cutaneous neutrophil depletion leads to subtle decrease in skin IL-17A expression (Figure 2H) rather supports alternative possibilities. For instance, as indicated elsewhere, IL-17A cutaneous tone could be enhanced by neutrophil-mediated augmentation of Th17 or gamma/delta T cell function (Lambert et al. J Invest Dermatol. 2019, PMID: 30528823). Does neutrophil depletion or G-CSF neutralization alter cell numbers or function of cutaneous Th17 and gamma/delta T cells?
    6. Finally, as the above conclusions rely solely on the IMQ-induced acute psoriasis model, it would be informative if they could be derived from another psoriasis model. IMQ is known to induce unintended systemic inflammation due to grooming-associated ingestion (Gangwar et al. J Invest Dermatol. 2022, PMID: 34953514), and "pathological crosstalk between skin and BM in psoriatic inflammation" could be strengthened by an intradermal injection model.

    Minor comments:

    1. Figure 1E shows multiple elongated Ly6G+ structures in d0-2 control and d0 IMQ skins that do not appear to be neutrophils.
    2. In Figure 2C, the bottom GSEA seems to be showing type II IFN response, not type I IFN, according to the text.
    3. For the BM analysis in Figures 3, 5, S3, and S5, it would be informative if BM cellularity and numbers of committed myeloid progenitors (e.g., GMPs) are shown.
    4. In Figures 6B, in which cluster of human skin cells IL-17A expression would be enriched?

    Significance

    Although quite a few studies have reported various examples of emergency myelopoiesis (Swann et al. Nat Rev Immunol. 2024, PMID: 38467802), there is limited evidence on its occurrence and involvement in locally restricted disease, such as periodontitis (Li et al. Cell 2022, PMID: 35483374; 35483374). As an HSC biologist, I see this study is conceptually interesting as it could extend the above concept to psoriasis, a non-infectious, local inflammatory disease in the skin, and describes a potential causal link between skin-derived G-CSF and emergency myelopoiesis. That said, as detailed in the first section, the conclusions, especially that related to emergency myelopoiesis driven by skin-derived G-CSF, need to be more convincingly supported before taking its value. The findings offer additional understanding of how psoriasis is developed in concert with aberrant hematopoiesis and will be relevant to those working in the field of dermatology, immunology, and hematology.

  5. Version published to 10.1101/2025.10.10.681054 on bioRxiv