Erythropoiesis-Stimulating Agent Protects Against Kidney Fibrosis by Inhibiting G2/M Cell Cycle Arrest

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Oh et al submit an original research article entitled "Erythropoiesis-Stimulating Agent Protects against Kidney Fibrosis by Inhibiting G2/M Cell Cycle Arrest".

They look at the effect of EPO-derived darbepoetin alfa on HEK epithelial cells and unilateral ureteral obstruction (UUO) mouse model. They conclude that DARB treatment demonstrated an anti-fibrotic effect in HK-2 cells by impacting G2/M cell cycle arrest.

For RT-qPCR how did the authors check the integrity of extracted RNA? Can the authors justify use of GAPDH as control? Did it vary according to experimental conditions? Same question for western blotting.

In the discussion, the authors point out that EPO treatment enhances the G1-to-S transition in erythroblasts by inhibiting regulatory factors. They could also mention that Uremic toxins block erythroblasts at the sub-G1 phase in an EPO-dependant manner. (PMID: 36596353)

A final recapitulative figure would be welcome

Author Response

Reviewer #1

Oh et al submit an original research article entitled "Erythropoiesis-Stimulating Agent Protects against Kidney Fibrosis by Inhibiting G2/M Cell Cycle Arrest".

They look at the effect of EPO-derived darbepoetin alfa on HEK epithelial cells and unilateral ureteral obstruction (UUO) mouse model. They conclude that DARB treatment demonstrated an anti-fibrotic effect in HK-2 cells by impacting G2/M cell cycle arrest.

Comment 1. For RT-qPCR how did the authors check the integrity of extracted RNA? Can the authors justify use of GAPDH as control? Did it vary according to experimental conditions? Same question for western blotting.

[Response]

We thank the reviewer for the helpful comment.

To ensure the integrity and purity of RNA used in RT-qPCR, we followed standard quality control procedures. We assessed the RNA using a NanoDrop spectrophotometer, confirming that all samples had A260/A280 ratios between 1.8 and 2.1 and A260/A230 ratios above 1.8. Additionally, we evaluated RNA integrity by electrophoresis on a 1% agarose gel to verify the presence of intact 28S and 18S rRNA bands (See below).

We used GAPDH as the reference gene for normalization in RT-qPCR, as its expression remained stable across all experimental conditions in HK-2 cells. We verified this by confirming minimal variation in Ct values between groups.

For Western blotting, β-actin was used as a loading control. We confirmed that its expression did not vary significantly between groups, ensuring reliable normalization of target protein levels.

Notably, both GAPDH and β-actin have been widely used as internal controls in previous studies involving kidney-derived cells, including HK-2 cells.^1-6 These references further support the appropriateness of our choice of normalization controls

These details have been added to the revised Methods section accordingly.

             Methods,

(page 3 line 117-118)

Primer/probes (Macrogen, Seoul, South Korea) were used to amplify p53, p21, TGF-β, cy-clin B1, cyclin D1, CTGF, MCP-1, COL1A1, and fibronectin as well as GAPDH (glycer-aldehyde-3-phosphate dehydrogenase), which was used as the normalization control.^23-25 The expression of GAPDH remained stable across all experimental conditions. The PCR

             (page 4 line 175-176)

Equal amounts of protein, determined through the Bradford method, were loaded in each lane and normalized by β-actin.^26,27 The expression of β-actin did not vary between groups, ensuring reliable normalization of target protein.

 

Comment 2. In the discussion, the authors point out that EPO treatment enhances the G1-to-S transition in erythroblasts by inhibiting regulatory factors. They could also mention that Uremic toxins block erythroblasts at the sub-G1 phase in an EPO-dependent manner. (PMID: 36596353) A final recapitulative figure would be welcome.

[Response]

We appreciate the reviewer’s insightful comment. As suggested, we have incorporated a discussion on the EPO-dependent sub-G1 phase arrest induced by uremic toxins such as indoxyl sulfate into the revised manuscript.

Additionally, we have included a new conceptual figure (Figure 7) at the end of the Discussion section to summarize our findings and proposed mechanism.

 

Discussion (page 12 line 374-383),

In addition, recent studies have identified that uremic toxins, especially indoxyl sulfate (IS), may exacerbate anemia in CKD by disrupting erythropoiesis through EPO-dependent mechanisms, primarily by inducing sub-G1 phase arrest.⁷ IS has been shown to induce apoptosis and arrest erythroid development by downregulating key genes such as GATA-1, EPO-R, and β-globin, while also increasing sub-G1 phase accumulation in erythroid progenitor cells. These effects occur even at clinically relevant concentrations and are mediated, at least in part, through EPO-dependent mechanisms. Therefore, IS not only impairs EPO signaling but also interferes with erythroid maturation and survival, compounding the challenge of managing anemia and potentially blunting the beneficial cellular effects of ESAs in CKD.

Reference

1. Ceol M, Del Prete D, Tosetto E, et al. GAPDH as housekeeping gene at renal level. Kidney Int. May 2004;65(5):1972-3; author reply 1973-4. doi:10.1111/j.1523-1755.2004.607_7.x
2. Xu LN, Sun YY, Tan YF, et al. SHROOM3 Deficiency Aggravates Adriamycin-Induced Nephropathy Accompanied by Focal Adhesion Disassembly and Stress Fiber Disorganization. Cells. Jun 13 2025;14(12)doi:10.3390/cells14120895
3. Huang PY, Juan YH, Hung TW, et al. β-Mangostin Alleviates Renal Tubulointerstitial Fibrosis via the TGF-β1/JNK Signaling Pathway. Cells. Oct 14 2024;13(20)doi:10.3390/cells13201701
4. Guh JY, Chuang TD, Chen HC, et al. Beta-hydroxybutyrate-induced growth inhibition and collagen production in HK-2 cells are dependent on TGF-beta and Smad3. Kidney Int. Dec 2003;64(6):2041-51. doi:10.1046/j.1523-1755.2003.00330.x
5. Yuan C, Jin G, Li P, et al. Tubular cell transcriptional intermediary factor 1γ deficiency exacerbates kidney injury-induced tubular cell polyploidy and fibrosis. Kidney Int. Oct 2023;104(4):769-786. doi:10.1016/j.kint.2023.07.006
6. Sun S, Ning X, Zhang Y, et al. Hypoxia-inducible factor-1alpha induces Twist expression in tubular epithelial cells subjected to hypoxia, leading to epithelial-to-mesenchymal transition. Kidney Int. Jun 2009;75(12):1278-1287. doi:10.1038/ki.2009.62
7. Hamza E, Vallejo-Mudarra M, Ouled-Haddou H, et al. Indoxyl sulfate impairs erythropoiesis at BFU-E stage in chronic kidney disease. Cell Signal. Apr 2023;104:110583. doi:10.1016/j.cellsig.2022.110583
8. Yang L, Ma L, Fu P, Nie J. Update of cellular senescence in kidney fibrosis: from mechanism to potential interventions. Front Med. Apr 2025;19(2):250-264. doi:10.1007/s11684-024-1117-z
9. Choi DE, Jeong JY, Lim BJ, Lee KW, Shin YT, Na KR. Pretreatment with darbepoetin attenuates renal injury in a rat model of cisplatin-induced nephrotoxicity. Korean J Intern Med. Sep 2009;24(3):238-46. doi:10.3904/kjim.2009.24.3.238
10. Wang J, Matsushita K, Zhong J, Ma LJ, Yang HC, Fogo AB. Low-Dose Erythropoietin Amplifies Beneficial Effects of Angiotensin II Blockade on Glomerulosclerosis. Lab Invest. Feb 2023;103(2):100015. doi:10.1016/j.labinv.2022.100015
11. Nishida A, Nishida M, Iehara T. Delayed treatment with erythropoietin attenuates renal fibrosis in mouse model of unilateral ureteral obstruction. Int J Urol. Jun 2024;31(6):685-692. doi:10.1111/iju.15427
12. Iwata Y, Sakai N, Nakajima Y, et al. Anti-fibrotic potential of erythropoietin signaling on bone marrow derived fibrotic cell. BMC Nephrol. May 31 2021;22(1):203. doi:10.1186/s12882-021-02411-0
13. Li H, Peng X, Wang Y, et al. Atg5-mediated autophagy deficiency in proximal tubules promotes cell cycle G2/M arrest and renal fibrosis. Autophagy. Sep 2016;12(9):1472-86. doi:10.1080/15548627.2016.1190071
14. Yang L, Besschetnova TY, Brooks CR, Shah JV, Bonventre JV. Epithelial cell cycle arrest in G2/M mediates kidney fibrosis after injury. Nat Med. May 2010;16(5):535-43, 1p following 143. doi:10.1038/nm.2144
15. Lin SL, Kisseleva T, Brenner DA, Duffield JS. Pericytes and perivascular fibroblasts are the primary source of collagen-producing cells in obstructive fibrosis of the kidney. Am J Pathol. Dec 2008;173(6):1617-27. doi:10.2353/ajpath.2008.080433
16. Humphreys BD, Lin SL, Kobayashi A, et al. Fate tracing reveals the pericyte and not epithelial origin of myofibroblasts in kidney fibrosis. Am J Pathol. Jan 2010;176(1):85-97. doi:10.2353/ajpath.2010.090517

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

This study investigates the effects of erythropoiesis-stimulating agent (ESA) on renal fibrosis by inhibiting G2/M cell cycle arrest, offering valuable insights into kidney protection mechanisms. Using both in vitro HK-2 cell and in vivo UUO mouse models, the researchers systematically demonstrated that darbepoetin alfa (DARB) significantly mitigates renal fibrosis and elucidated how this protection correlates with suppression of G2/M cell cycle arrest. The experimental design is robust, employing multiple techniques including flow cytometry, immunohistochemistry, RT-PCR, and immunoblotting to validate findings. Particularly commendable is the integration of in vitro and in vivo models and the exploration of molecular mechanisms, providing new therapeutic targets for nephrology research. However, several aspects could benefit from further refinement to enhance the scientific rigor and impact of the article:

1. The manuscript lacks a comprehensive conceptual diagram illustrating the relationship between G2/M cell cycle arrest, DARB intervention, and renal fibrosis. A schematic figure integrating Yang et al.'s (2010, Nature Medicine, 16(5):535-543) foundational findings with the current study's new discoveries would help readers better understand the proposed mechanism and significantly enhance the educational value and readability of the article.

2. The study inadequately distinguishes between transient G2/M arrest and long-term cellular senescence state, and does not address the role of the senescence-associated secretory phenotype (SASP) in this process. Given recent research demonstrating that senescent cells promote renal fibrosis through SASP, it would be valuable to explore whether DARB's effects involve modulation of cellular senescence and SASP.

3. The selection of 0.5 and 5 μg/mL/kg as experimental doses lacks scientific justification, with no preliminary dose-range screening data presented. It would be valuable to include the scientific basis for dose selection, preferably with preliminary experimental data, and discuss how these doses relate to clinically relevant concentrations.

4. After intraperitoneal DARB administration, no measurements of actual drug concentrations in blood and renal tissues were performed, making it impossible to determine if renal tubular cells were exposed to DARB concentrations comparable to the in vitro experiments. Pharmacokinetic assessment would enhance the comparability of in vitro and in vivo experimental results.

5. Western blot normalization methods are inconsistent in Figure 5, with p-JNK normalized to p-JNK/actin while p-p38 and p-ERK1/2 are normalized to p-p38/p38 and p-ERK1/2/ERK1/2, respectively, without explanation. Additionally, the images suggest declining trends in total p38 and total ERK expression, which are neither statistically analyzed nor discussed. Consistent normalization methods and comprehensive analysis of all proteins would be more scientifically rigorous.

6. The possibility of off-target effects deserves specific consideration in this study. DARB, as a biological agent, could potentially interact with other receptors or signaling pathways, especially at high concentrations. Previous studies have shown that some biological agents can exhibit non-specific binding or activate alternative pathways when used at pharmacological doses. The broad effects on multiple signaling pathways observed in this study (JNK, p38-MAPK, ERK) might partially result from such off-target interactions. Specific experiments such as competitive binding assays or pathway inhibition studies would be essential to confirm whether the mechanism is truly EPOR-dependent or involves alternative molecular targets.

7. The authors did not perform EPOR knockout or silencing experiments to verify that the observed effects are specifically mediated through this receptor. The use of a single agent (DARB) without receptor validation raises the possibility that at least some of the observed effects may be due to off-target binding rather than specific EPOR-mediated mechanisms. Receptor knockdown or antagonist studies would help clarify the specificity of the reported effects.

8. The article provides insufficient evidence for erythropoietin receptor (EPOR) expression in HK-2 cells and renal tubular epithelial cells. Given the controversies in literature regarding EPOR expression in renal cells, direct verification through Western blot or immunohistochemistry would strengthen the assumption that DARB exerts its effects through EPOR in the studied models.

9. The study uses only one ESA (DARB) but generalizes findings to the entire ESA class in the conclusion. A discussion of structural and functional similarities and differences between DARB and other ESAs, validation experiments with additional ESAs, or a more cautious limitation of conclusions to DARB rather than all ESAs would be appropriate.

10. The authors focus primarily on JNK and p38-MAPK pathways but fail to address other potential mechanisms that may be affected by DARB treatment. Several pathways known to influence renal fibrosis could also be modulated by ESAs, including autophagy regulation through Atg5 (Li et al., 2016), macrophage polarization (Nishida et al., 2024), and mitochondrial oxidative stress in fibrocytes (Iwata et al., 2021). Without experimental exclusion of these pathways, it remains uncertain whether DARB's protective effects are mediated exclusively through the proposed mechanisms or involve these additional signaling pathways. A more comprehensive investigation of these alternative mechanisms would strengthen the mechanistic claims and provide a more complete understanding of DARB's action in renal protection.

Author Response

Responses to the editors’ and reviewers’ comments

 

Re: Erythropoiesis-Stimulating Agent Protects against Kidney Fi-brosis by Inhibiting G2/M Cell Cycle Arrest (revised manuscript; cells-3661435)

Donghwan Oh^1†, Jong Hyun Jhee^1†, Soo Hyun Kim¹, Tae Yeon Kim¹, Hyo Jeong Kim¹, Wooram Bae¹, Hoon Young Choi^1,2 and Hyeong Cheon Park^1,2

 

We deeply appreciate the editors and reviewers for their thoughtful comments, which have provided us with valuable opportunities to improve our work. We have carefully considered all the comments and addressed them point by point.

 

Below, we have numbered and bolded the comments, followed by our responses in regular type. We hope the editors and reviewers now consider our paper suitable for publication in the Cells.

 

Reviewer #2

This study investigates the effects of erythropoiesis-stimulating agent (ESA) on renal fibrosis by inhibiting G2/M cell cycle arrest, offering valuable insights into kidney protection mechanisms. Using both in vitro HK-2 cell and in vivo UUO mouse models, the researchers systematically demonstrated that darbepoetin alfa (DARB) significantly mitigates renal fibrosis and elucidated how this protection correlates with suppression of G2/M cell cycle arrest. The experimental design is robust, employing multiple techniques including flow cytometry, immunohistochemistry, RT-PCR, and immunoblotting to validate findings. Particularly commendable is the integration of in vitro and in vivo models and the exploration of molecular mechanisms, providing new therapeutic targets for nephrology research. However, several aspects could benefit from further refinement to enhance the scientific rigor and impact of the article:

Comment 1. The manuscript lacks a comprehensive conceptual diagram illustrating the relationship between G2/M cell cycle arrest, DARB intervention, and renal fibrosis. A schematic figure integrating Yang et al.'s (2010, Nature Medicine, 16(5):535-543) foundational findings with the current study's new discoveries would help readers better understand the proposed mechanism and significantly enhance the educational value and readability of the article.

[Response]

We appreciate the reviewer’s insightful comment. We have included a new conceptual figure (Figure 7) at the end of the Discussion section to summarize our findings and proposed mechanism.

 

Comment 2. The study inadequately distinguishes between transient G2/M arrest and long-term cellular senescence state, and does not address the role of the senescence-associated secretory phenotype (SASP) in this process. Given recent research demonstrating that senescent cells promote renal fibrosis through SASP, it would be valuable to explore whether DARB's effects involve modulation of cellular senescence and SASP.

[Response]

We appreciate the reviewer’s thoughtful comment regarding the distinction between transient G2/M arrest and long-term cellular senescence, as well as the potential involvement of the senescence-associated secretory phenotype (SASP) in renal fibrosis.

While we fully agree that cellular senescence and SASP are critical contributors to chronic kidney injury and fibrosis, our study primarily focused on the early phase of kidney injury, specifically the transition from AKI to CKD. Within this context, we observed transient G2/M arrest rather than sustained cell cycle arrest or senescence-related phenotypes. Therefore, a detailed investigation into long-term senescence and SASP-related mechanisms following DARB treatment was not included in the scope of the present study.

Nonetheless, we recognize the importance of this pathway and agree that it represents a valuable direction for future research. We have added a sentence in the Limitation section to acknowledge this point and highlight it as a potential avenue for further investigation.

             Limitation (page 13 line 436-444),

Third, our study primarily focused on the early phase of kidney injury, particularly the AKI-to-CKD transition, and demonstrated transient G2/M cell cycle arrest in kidney tubu-lar epithelial cells. We did not observe features consistent with long-term cell cycle arrest or cellular senescence such as senescence-associated secretory phenotypes (SASP)57. Therefore, the potential role of cellular senescence and SASP in DARB-mediated an-ti-fibrotic effects was limited to demonstrate in the present study. Further investigations will be needed to clarify whether modulation of senescence pathways by the long-term DARB treatment contributes to chronic kidney injury.

 

Comment 3. The selection of 0.5 and 5 μg/mL/kg as experimental doses lacks scientific justification, with no preliminary dose-range screening data presented. It would be valuable to include the scientific basis for dose selection, preferably with preliminary experimental data, and discuss how these doses relate to clinically relevant concentrations.

[Response]

We thank the reviewer for pointing out this important issue.

The selection of 0.5 and 5 μg/mL/kg as experimental doses of DARB was based on previously published studies that reported pharmacologically effective concentrations in in vivo models of renal injury.^9,10. Additionally, we conducted preliminary in vitro experiments using lower concentrations of DARB (0.5 and 5 ug/mL) than those used in vivo to ensure that these doses did not induce cytotoxicity in HK-2 cells and sufficient to elicit biological responses related to cell cycle regulation. We have attached our preliminary data related to dose selection (see below).

 

 

Although we did not include a full dose-range screening in the current manuscript, we chose these concentrations to reflect the approximate range of clinically relevant plasma levels observed in patients treated with DARB. We have included this rationale in the revised Methods section and added a brief discussion of this limitation in the Discussion section.

             Methods (page 2 line 79-81),

The doses of DARB used in this study were chosen based on prior literature and preliminary in vitro data, ensuring minimal cytotoxicity while maintaining biological efficacy in kidney cells.^21,22

             Limitation (page 13 line 444-447),

Fourth, we did not perform a full dose-range screening to comprehensively evaluate the dose–response relationship of DARB. Instead, we selected two representative doses based on prior literature and preliminary in vitro findings demonstrating minimal cytotoxicity and sufficient biological activity.

 

Comment 4. After intraperitoneal DARB administration, no measurements of actual drug concentrations in blood and renal tissues were performed, making it impossible to determine if renal tubular cells were exposed to DARB concentrations comparable to the in vitro experiments. Pharmacokinetic assessment would enhance the comparability of in vitro and in vivo experimental results.

[Response]

We thank the reviewer for this important comment.

We acknowledge that we did not measure the actual concentrations of DARB in blood or renal tissue following intraperitoneal administration. This represents a limitation in our ability to directly compare in vitro and in vivo exposure levels. However, the doses used in our in vivo experiments were selected based on previously reported pharmacologically active doses in rodent models,^9,10 as well as our preliminary data from UUO mouse model (see response to comment 3). These doses fall within the range expected to exert systemic bioactivity, including renal tissues.

Although DARB is a long-acting ESA known to reach multiple tissues including the kidney, we agree that pharmacokinetic profiling, particularly drug levels in the kidney, would enhance the interpretation of our findings and the comparability with in vitro experiments. We have acknowledged this point in the revised Discussion as a limitation and as a valuable direction for future research.

             Limitation (page 13 line 447-454),

Fifth, we did not measure the actual concentrations of DARB in blood or renal tissue following intraperitoneal administration. As a result, we cannot definitively determine whether the renal tubular cells in vivo were exposed to DARB concentrations comparable to those used in vitro. Although the selected in vivo doses were based on previously reported pharmacologically active ranges, future studies incorporating pharmacokinetic analysis will be important to validate tissue-level drug exposure and enhance the translational relevance of our findings.

 

Comment 5. Western blot normalization methods are inconsistent in Figure 5, with p-JNK normalized to p-JNK/actin while p-p38 and p-ERK1/2 are normalized to p-p38/p38 and p-ERK1/2/ERK1/2, respectively, without explanation. Additionally, the images suggest declining trends in total p38 and total ERK expression, which are neither statistically analyzed nor discussed. Consistent normalization methods and comprehensive analysis of all proteins would be more scientifically rigorous.

[Response]

We thank the reviewer for this valuable comment. We agree with the reviewer that normalization to the corresponding total protein is a more accurate reflection of phosphorylation changes. Accordingly, we reanalyzed p-JNK using total JNK as the normalization control for consistency. Although slight reductions in total p38 and ERK1/2 were observed, they were not statistically significant, and the phosphorylation changes remained evident after normalization. These revisions have been incorporated into the updated Figure 5 and legend.

 

Comment 6. The possibility of off-target effects deserves specific consideration in this study. DARB, as a biological agent, could potentially interact with other receptors or signaling pathways, especially at high concentrations. Previous studies have shown that some biological agents can exhibit non-specific binding or activate alternative pathways when used at pharmacological doses. The broad effects on multiple signaling pathways observed in this study (JNK, p38-MAPK, ERK) might partially result from such off-target interactions. Specific experiments such as competitive binding assays or pathway inhibition studies would be essential to confirm whether the mechanism is truly EPOR-dependent or involves alternative molecular targets.

[Response]

We appreciate the reviewer’s thoughtful comment regarding the potential off-target effects of DARB, particularly at pharmacological doses. To address this concern, we performed additional in vitro experiments to directly assess whether the anti-fibrotic effects of DARB are EPOR-dependent. Specifically, we treated HK-2 cells under TGF-β stimulation with or without DARB (0.5 and 5 μg/mL), in the presence or absence of an EPOR-blocking peptide. Expression levels of fibrosis-related markers, including α-SMA, collagen I, fibronectin, TGF-β, and CTGF, were evaluated by Western blot. Our results showed that DARB significantly reduced the expression of these fibrotic markers in a dose-dependent manner, and importantly, these effects were attenuated when co-treated with the EPOR-blocking peptide, supporting an EPOR-dependent mechanism of action. These new findings reduce the likelihood of non-specific off-target effects and strengthen the conclusion that the anti-fibrotic actions of DARB are mediated, at least in part, through EPOR signaling. The new data have been included as Supplementary Figure 2, and the Methods and Discussion sections have been revised accordingly.

             Methods (page 2 line 81-84),

For the EPO receptor (EPOR) blockade experiment, HK-2 cells were seeded in 6-well plates and pretreated with an EPOR-blocking peptide (MyBioSource, San Diego, CA, USA, 5 μM) for 1 hour, followed by stimulation with TGF-β1 (5 ng/mL) and DARB (0.5 or 5 μg/mL) for 48 hours.

Results (page 7 line 274-279),

To further clarify whether these anti-fibrotic effects are mediated through EPOR sig-naling, we treated HK-2 cells with TGF-β and DARB in the presence or absence of an EPOR-blocking peptide. The suppressive effects of DARB on fibrosis-related markers (α-SMA, collagen I, fibronectin) were attenuated by EPOR blockade, supporting that DARB exerts its protective effects, at least in part, via EPOR-dependent mechanisms (Supplementary Figure 2).

 

Comment 7. The authors did not perform EPOR knockout or silencing experiments to verify that the observed effects are specifically mediated through this receptor. The use of a single agent (DARB) without receptor validation raises the possibility that at least some of the observed effects may be due to off-target binding rather than specific EPOR-mediated mechanisms. Receptor knockdown or antagonist studies would help clarify the specificity of the reported effects.

[Response]

We thank the reviewer for this insightful comment. While we did not perform EPOR knockout or gene silencing experiments, we addressed this concern by conducting an additional in vitro experiment using an EPOR-blocking peptide (see also Response to Comment 6 and Supplementary Figure 2). The anti-fibrotic effects of DARB on TGF-β-stimulated HK-2 cells were attenuated in the presence of the blocking peptide, suggesting that DARB exerts its effects, at least in part, through EPOR signaling. Although pharmacological inhibition may not fully replicate the genetic loss-of-function approach, our results provide supportive evidence for EPOR-mediated activity.

 

Comment 8. The article provides insufficient evidence for erythropoietin receptor (EPOR) expression in HK-2 cells and renal tubular epithelial cells. Given the controversies in literature regarding EPOR expression in renal cells, direct verification through Western blot or immunohistochemistry would strengthen the assumption that DARB exerts its effects through EPOR in the studied models.

[Response]

We appreciate the reviewer’s insightful comment regarding the need for direct evidence of EPOR expression in HK-2 cells and renal tubular epithelial cells.

To address this concern, we conducted additional Western blot analyses to assess EPOR expression. Our results confirmed that EPOR is endogenously expressed in HK-2 cells, and notably, its expression increased in a dose-dependent manner upon treatment with DARB. As a negative control, we included HeLa cells, which are known to exhibit minimal or no EPOR expression; as expected, no EPOR band was detected in HeLa cells. These findings support the validity of our use of HK-2 cells to investigate the mechanisms of DARB in renal tubular epithelial cells.

We have included the representative Western blot data as a new Supplementary Figure (Supplementary Figure 1) and added antibody details to the revised Supplementary Methods section.

 

Supplementary Methods - Antibody information used in Western blotting

Erythropoietin receptor (cat. no. ab284292; 1:1000; Abcam, Cambridge, MA, USA),

Results (page 4-5 line 193-198)

To confirm the presence of EPOR in our in vitro model, we first performed Western blot analysis in HK-2 cells. EPOR was clearly detected and showed a dose-dependent increase following DARB treatment, whereas minimal EPOR expression was observed in HeLa cells, which served as a negative control (Supplementary Figure 1). These findings validate the use of HK-2 cells as a suitable model for studying EPOR-mediated effects in renal tubular epithelium.

 

Comment 9. The study uses only one ESA (DARB) but generalizes findings to the entire ESA class in the conclusion. A discussion of structural and functional similarities and differences between DARB and other ESAs, validation experiments with additional ESAs, or a more cautious limitation of conclusions to DARB rather than all ESAs would be appropriate.

[Response]

We appreciate the reviewer’s thoughtful comment. In response, we have revised the conclusion and relevant statements throughout the manuscript to refer specifically to darbepoetin alfa (DARB), rather than the broader class of erythropoiesis-stimulating agents (ESAs). We agree that, given the structural and pharmacokinetic differences among ESAs, our findings should be interpreted within the context of DARB treatment alone. This clarification has also been added to the Discussion section as a study limitation.

             Limitation (page 13 line 454-458),

Sixth, our study investigated only one type of ESA, DARB. Given the structural and functional differences between DARB and other ESAs, such as epoetin (α, β) or methoxy polyethylene glycol-epoetin beta, the observed effects should not be generalized to the entire ESA class. Further studies are needed to determine whether similar anti-fibrotic and cell cycle-related effects are shared by other ESAs.

 

Comment 10. The authors focus primarily on JNK and p38-MAPK pathways but fail to address other potential mechanisms that may be affected by DARB treatment. Several pathways known to influence renal fibrosis could also be modulated by ESAs, including autophagy regulation through Atg5 (Li et al., 2016), macrophage polarization (Nishida et al., 2024), and mitochondrial oxidative stress in fibrocytes (Iwata et al., 2021). Without experimental exclusion of these pathways, it remains uncertain whether DARB's protective effects are mediated exclusively through the proposed mechanisms or involve these additional signaling pathways. A more comprehensive investigation of these alternative mechanisms would strengthen the mechanistic claims and provide a more complete understanding of DARB's action in renal protection.

[Response]

We thank the reviewer for this insightful comment. We agree that multiple signaling pathways including autophagy (via Atg5), macrophage polarization, and mitochondrial oxidative stress—are known to play important roles in renal fibrosis and may also be influenced by ESAs such as DARB.^11-13 In this study, we focused on the JNK and p38-MAPK pathways based on previous findings and the known involvement of these stress-activated kinases in G2/M cell cycle arrest and fibrosis.¹⁴ However, we do not exclude the possibility that other pathways may also contribute to DARB's protective effects. We acknowledge this as a limitation of the current study and have added a statement in the Discussion to reflect that further investigation is needed to explore the involvement of additional signaling pathways in DARB-mediated renoprotection.

             Limitation (page 13 line 431-436),

Second, this study primarily focused on the JNK and p38-MAPK pathways due to their well-established roles in stress-induced cell cycle arrest and fibrosis.⁷ However, other mechanisms, such as autophagy regulation, macrophage polarization, and mitochondrial oxidative stress, may also contribute to renal fibrosis and could be modulated by DARB. Further studies are needed to clarify whether DARB influences these alternative pathways as part of its protective effect.

Reference

1. Ceol M, Del Prete D, Tosetto E, et al. GAPDH as housekeeping gene at renal level. Kidney Int. May 2004;65(5):1972-3; author reply 1973-4. doi:10.1111/j.1523-1755.2004.607_7.x
2. Xu LN, Sun YY, Tan YF, et al. SHROOM3 Deficiency Aggravates Adriamycin-Induced Nephropathy Accompanied by Focal Adhesion Disassembly and Stress Fiber Disorganization. Cells. Jun 13 2025;14(12)doi:10.3390/cells14120895
3. Huang PY, Juan YH, Hung TW, et al. β-Mangostin Alleviates Renal Tubulointerstitial Fibrosis via the TGF-β1/JNK Signaling Pathway. Cells. Oct 14 2024;13(20)doi:10.3390/cells13201701
4. Guh JY, Chuang TD, Chen HC, et al. Beta-hydroxybutyrate-induced growth inhibition and collagen production in HK-2 cells are dependent on TGF-beta and Smad3. Kidney Int. Dec 2003;64(6):2041-51. doi:10.1046/j.1523-1755.2003.00330.x
5. Yuan C, Jin G, Li P, et al. Tubular cell transcriptional intermediary factor 1γ deficiency exacerbates kidney injury-induced tubular cell polyploidy and fibrosis. Kidney Int. Oct 2023;104(4):769-786. doi:10.1016/j.kint.2023.07.006
6. Sun S, Ning X, Zhang Y, et al. Hypoxia-inducible factor-1alpha induces Twist expression in tubular epithelial cells subjected to hypoxia, leading to epithelial-to-mesenchymal transition. Kidney Int. Jun 2009;75(12):1278-1287. doi:10.1038/ki.2009.62
7. Hamza E, Vallejo-Mudarra M, Ouled-Haddou H, et al. Indoxyl sulfate impairs erythropoiesis at BFU-E stage in chronic kidney disease. Cell Signal. Apr 2023;104:110583. doi:10.1016/j.cellsig.2022.110583
8. Yang L, Ma L, Fu P, Nie J. Update of cellular senescence in kidney fibrosis: from mechanism to potential interventions. Front Med. Apr 2025;19(2):250-264. doi:10.1007/s11684-024-1117-z
9. Choi DE, Jeong JY, Lim BJ, Lee KW, Shin YT, Na KR. Pretreatment with darbepoetin attenuates renal injury in a rat model of cisplatin-induced nephrotoxicity. Korean J Intern Med. Sep 2009;24(3):238-46. doi:10.3904/kjim.2009.24.3.238
10. Wang J, Matsushita K, Zhong J, Ma LJ, Yang HC, Fogo AB. Low-Dose Erythropoietin Amplifies Beneficial Effects of Angiotensin II Blockade on Glomerulosclerosis. Lab Invest. Feb 2023;103(2):100015. doi:10.1016/j.labinv.2022.100015
11. Nishida A, Nishida M, Iehara T. Delayed treatment with erythropoietin attenuates renal fibrosis in mouse model of unilateral ureteral obstruction. Int J Urol. Jun 2024;31(6):685-692. doi:10.1111/iju.15427
12. Iwata Y, Sakai N, Nakajima Y, et al. Anti-fibrotic potential of erythropoietin signaling on bone marrow derived fibrotic cell. BMC Nephrol. May 31 2021;22(1):203. doi:10.1186/s12882-021-02411-0
13. Li H, Peng X, Wang Y, et al. Atg5-mediated autophagy deficiency in proximal tubules promotes cell cycle G2/M arrest and renal fibrosis. Autophagy. Sep 2016;12(9):1472-86. doi:10.1080/15548627.2016.1190071
14. Yang L, Besschetnova TY, Brooks CR, Shah JV, Bonventre JV. Epithelial cell cycle arrest in G2/M mediates kidney fibrosis after injury. Nat Med. May 2010;16(5):535-43, 1p following 143. doi:10.1038/nm.2144
15. Lin SL, Kisseleva T, Brenner DA, Duffield JS. Pericytes and perivascular fibroblasts are the primary source of collagen-producing cells in obstructive fibrosis of the kidney. Am J Pathol. Dec 2008;173(6):1617-27. doi:10.2353/ajpath.2008.080433
16. Humphreys BD, Lin SL, Kobayashi A, et al. Fate tracing reveals the pericyte and not epithelial origin of myofibroblasts in kidney fibrosis. Am J Pathol. Jan 2010;176(1):85-97. doi:10.2353/ajpath.2010.090517

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

In the manuscript “Erythropoiesis-Stimulating Agent Protects against Kidney Fi2 brosis by Inhibiting G2/M Cell Cycle Arrest”, Oh and colleagues hypothesized that ESA agents such as DARB could have a beneficial effect in reducing renal fibrosis. To address this hypothesis, the authors used in vivo and in vitro models of fibrosis. The manuscript is generally well done, well written and somewhat novel. However, important controls are missing, as well as important points regarding molecular mechanisms of how DARB is acting. A few comments are listed below:

1- Concentration of antibodies were not completely described in Methods;

2- In Fig. 2A, IF images do not seem to match in the Control condition. According to the panel, zero signal for p53 and Ki67 are seen in the control condition in the green and red channels alone, but in the merged channel, green staining is clearly seen. Clarify.

3- Essential controls are missing in the culture and animals experiments regarding the effect of DARB treatment alone. Please include those controls, at least in the key experiments in the manuscript.

4- The statement that EMT drives renal fibrosis is at least controversial, as different groups have demonstrated that epithelial mesenchymal transition has a minor contribution to the pool of fibroblasts involved in renal fibrosis progression, as opposed to pericytes for example (PMID: 19008372; PMID: 20008127). Furthermore, no results presented support the hypothesis of EMT, as established markers of this biological effects have not been evaluated, such as loss of epithelial markers (i.e. E-cadherin) and gain of mesenchymal markers (i.e. Vimentin). Fibronectin and alfa-SMA are not reliable markers of EMT, but rather of pro-fibrotic signaling. I suggest that the authors soften those statements and include state of the art concepts in the field of fibrosis development in the kidney.

5- I have felt that a more comprehensive analysis of how DARB effect is taking place is missing. What is the receptor involved? How its expression is in diseased x health conditions? Does the use of a EPOR antagonist blocks the effect of DARB? These points should at least be addressed.

Author Response

Reviewer #3

In the manuscript “Erythropoiesis-Stimulating Agent Protects against Kidney Fibrosis by Inhibiting G2/M Cell Cycle Arrest”, Oh and colleagues hypothesized that ESA agents such as DARB could have a beneficial effect in reducing renal fibrosis. To address this hypothesis, the authors used in vivo and in vitro models of fibrosis. The manuscript is generally well done, well written and somewhat novel. However, important controls are missing, as well as important points regarding molecular mechanisms of how DARB is acting. A few comments are listed below:

Comment 1. Concentration of antibodies were not completely described in Methods;

[Response]

We thank the reviewer for the comment. The detailed information regarding the primary and secondary antibodies used in Western blotting, including their concentrations and sources, is provided in the Supplementary Methods section under “Antibody information used in Western blotting.”

 

Comment 2. In Fig. 2A, IF images do not seem to match in the Control condition. According to the panel, zero signal for p53 and Ki67 are seen in the control condition in the green and red channels alone, but in the merged channel, green staining is clearly seen. Clarify.

[Response]

We thank the reviewer for the careful observation. The weak green signal in the merged image under control conditions reflects basal nuclear expression of p53, which is consistent with its known role as a nuclear transcription factor. While the signal was below the threshold in the single-channel view, it becomes visible in the merged panel due to contrast rendering. However, we have re-examined the raw images and confirmed that all panels represent the same field. These confirm that the control group exhibits minimal baseline signal, consistent with the published figure (See below). We hope this clarifies the reviewer’s concern.

 

Comment 3. Essential controls are missing in the culture and animals experiments regarding the effect of DARB treatment alone. Please include those controls, at least in the key experiments in the manuscript.

[Response]

We thank the reviewer for highlighting the importance of appropriate controls in both in vitro and in vivo settings. In response, we have included additional control groups in the revised version of the manuscript. Specifically, we conducted an additional Western blot experiment to verify EPOR expression in HK-2 cells, in which we included a DARB-only treatment group as an internal control to assess EPOR regulation in the absence of external stimuli. This control provides further support for the specificity and relevance of DARB’s action in renal tubular epithelial cells. The data have been added to the revised manuscript and included in Supplementary Figure 1.

 

Comment 4. The statement that EMT drives renal fibrosis is at least controversial, as different groups have demonstrated that epithelial mesenchymal transition has a minor contribution to the pool of fibroblasts involved in renal fibrosis progression, as opposed to pericytes for example (PMID: 19008372; PMID: 20008127). Furthermore, no results presented support the hypothesis of EMT, as established markers of this biological effects have not been evaluated, such as loss of epithelial markers (i.e. E-cadherin) and gain of mesenchymal markers (i.e. Vimentin). Fibronectin and alfa-SMA are not reliable markers of EMT, but rather of pro-fibrotic signaling. I suggest that the authors soften those statements and include state of the art concepts in the field of fibrosis development in the kidney.

[Response]

We thank the reviewer for this important and well-informed comment. We agree that the role of epithelial-to-mesenchymal transition (EMT) in renal fibrosis remains a subject of ongoing debate, and that recent lineage tracing studies have shown that EMT contributes only minimally to the pool of matrix-producing fibroblasts in the kidney.^15,16

In our study, we did not evaluate canonical markers of EMT, such as E-cadherin or Vimentin, and therefore acknowledge that our data do not directly support EMT as a mechanism. While increased expression of fibronectin and α-SMA indicates activation of pro-fibrotic signaling, these markers alone are insufficient to confirm EMT.

In response, we have revised the manuscript to soften any conclusive statements regarding EMT and have removed or rephrased related interpretations. We have also updated the Discussion to reflect the current understanding of fibrosis pathogenesis in the kidney

 

Comment 5. I have felt that a more comprehensive analysis of how DARB effect is taking place is missing. What is the receptor involved? How its expression is in diseased x health conditions? Does the use of a EPOR antagonist blocks the effect of DARB? These points should at least be addressed.

[Response]

We thank the reviewer for raising these important mechanistic questions. In our study, we identified EPOR as the likely mediator of DARB’s anti-fibrotic effects. EPOR expression was confirmed in HK-2 cells by Western blotting, and its expression increased in response to DARB treatment in a dose-dependent manner (Supplementary Figure 1). We also examined EPOR expression under fibrotic conditions. In both TGF-β–treated HK-2 cells, EPOR expression was maintained or modestly upregulated, suggesting that the receptor remains available for ESA-mediated signaling during injury. To assess receptor specificity, we conducted additional in vitro experiments using an EPOR-blocking peptide. The presence of the blocking peptide attenuated the anti-fibrotic effects of DARB on TGF-β–stimulated HK-2 cells, supporting an EPOR-dependent mechanism (Supplementary Figure 2). These findings, while not definitive, provide supportive evidence for EPOR involvement in the observed protective effects of DARB.

 

Reference

1. Ceol M, Del Prete D, Tosetto E, et al. GAPDH as housekeeping gene at renal level. Kidney Int. May 2004;65(5):1972-3; author reply 1973-4. doi:10.1111/j.1523-1755.2004.607_7.x
2. Xu LN, Sun YY, Tan YF, et al. SHROOM3 Deficiency Aggravates Adriamycin-Induced Nephropathy Accompanied by Focal Adhesion Disassembly and Stress Fiber Disorganization. Cells. Jun 13 2025;14(12)doi:10.3390/cells14120895
3. Huang PY, Juan YH, Hung TW, et al. β-Mangostin Alleviates Renal Tubulointerstitial Fibrosis via the TGF-β1/JNK Signaling Pathway. Cells. Oct 14 2024;13(20)doi:10.3390/cells13201701
4. Guh JY, Chuang TD, Chen HC, et al. Beta-hydroxybutyrate-induced growth inhibition and collagen production in HK-2 cells are dependent on TGF-beta and Smad3. Kidney Int. Dec 2003;64(6):2041-51. doi:10.1046/j.1523-1755.2003.00330.x
5. Yuan C, Jin G, Li P, et al. Tubular cell transcriptional intermediary factor 1γ deficiency exacerbates kidney injury-induced tubular cell polyploidy and fibrosis. Kidney Int. Oct 2023;104(4):769-786. doi:10.1016/j.kint.2023.07.006
6. Sun S, Ning X, Zhang Y, et al. Hypoxia-inducible factor-1alpha induces Twist expression in tubular epithelial cells subjected to hypoxia, leading to epithelial-to-mesenchymal transition. Kidney Int. Jun 2009;75(12):1278-1287. doi:10.1038/ki.2009.62
7. Hamza E, Vallejo-Mudarra M, Ouled-Haddou H, et al. Indoxyl sulfate impairs erythropoiesis at BFU-E stage in chronic kidney disease. Cell Signal. Apr 2023;104:110583. doi:10.1016/j.cellsig.2022.110583
8. Yang L, Ma L, Fu P, Nie J. Update of cellular senescence in kidney fibrosis: from mechanism to potential interventions. Front Med. Apr 2025;19(2):250-264. doi:10.1007/s11684-024-1117-z
9. Choi DE, Jeong JY, Lim BJ, Lee KW, Shin YT, Na KR. Pretreatment with darbepoetin attenuates renal injury in a rat model of cisplatin-induced nephrotoxicity. Korean J Intern Med. Sep 2009;24(3):238-46. doi:10.3904/kjim.2009.24.3.238
10. Wang J, Matsushita K, Zhong J, Ma LJ, Yang HC, Fogo AB. Low-Dose Erythropoietin Amplifies Beneficial Effects of Angiotensin II Blockade on Glomerulosclerosis. Lab Invest. Feb 2023;103(2):100015. doi:10.1016/j.labinv.2022.100015
11. Nishida A, Nishida M, Iehara T. Delayed treatment with erythropoietin attenuates renal fibrosis in mouse model of unilateral ureteral obstruction. Int J Urol. Jun 2024;31(6):685-692. doi:10.1111/iju.15427
12. Iwata Y, Sakai N, Nakajima Y, et al. Anti-fibrotic potential of erythropoietin signaling on bone marrow derived fibrotic cell. BMC Nephrol. May 31 2021;22(1):203. doi:10.1186/s12882-021-02411-0
13. Li H, Peng X, Wang Y, et al. Atg5-mediated autophagy deficiency in proximal tubules promotes cell cycle G2/M arrest and renal fibrosis. Autophagy. Sep 2016;12(9):1472-86. doi:10.1080/15548627.2016.1190071
14. Yang L, Besschetnova TY, Brooks CR, Shah JV, Bonventre JV. Epithelial cell cycle arrest in G2/M mediates kidney fibrosis after injury. Nat Med. May 2010;16(5):535-43, 1p following 143. doi:10.1038/nm.2144
15. Lin SL, Kisseleva T, Brenner DA, Duffield JS. Pericytes and perivascular fibroblasts are the primary source of collagen-producing cells in obstructive fibrosis of the kidney. Am J Pathol. Dec 2008;173(6):1617-27. doi:10.2353/ajpath.2008.080433
16. Humphreys BD, Lin SL, Kobayashi A, et al. Fate tracing reveals the pericyte and not epithelial origin of myofibroblasts in kidney fibrosis. Am J Pathol. Jan 2010;176(1):85-97. doi:10.2353/ajpath.2010.090517

 

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

changes are ok, except that some added references do nto appear in the reference list.

Author Response

Comments: Changes are ok, except that some added references do not appear in the reference list.

Response: Thank you for your review. They were included only to respond to the comments and should not be part of the revised manuscript.

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have provided comprehensive and well-executed revisions that significantly strengthen the manuscript. The addition of new experimental data, including EPOR expression validation and blockade experiments, along with the expanded discussion of limitations, demonstrates excellent responsiveness to reviewer feedback. Most concerns have been adequately addressed.

The authors stated they would normalize p-JNK to total JNK for consistency with p-p38/p38 and p-ERK1/2/ERK1/2. However, the current Figure 5 still shows p-JNK normalized to β-actin. Could the authors please provide the revised Figure 5 with p-JNK properly normalized to total JNK as indicated in their response to Comment 5?

Author Response

Comments: The authors stated they would normalize p-JNK to total JNK for consistency with p-p38/p38 and p-ERK1/2/ERK1/2. However, the current Figure 5 still shows p-JNK normalized to β-actin. Could the authors please provide the revised Figure 5 with p-JNK properly normalized to total JNK as indicated in their response to Comment 5?

Response: Thank you for your comments. Figure 5 has been updated in the revised manuscript. Please review it.

Round 3

Reviewer 2 Report

Comments and Suggestions for Authors

After reviewing the revised Figure 5, I find the changes acceptable. The paper can proceed to acceptance.