Scissor–CIBERSORTx Deconvolution Reveals Functional Heterogeneity of CTAL/aTAL Cells and Associated Biomarkers in Renal Fibrosis

Response to Reviewer 1 Comments

1. Summary

Dear Reviewer,

Thank you for your thoughtful suggestions and insights, which have benefited from the manuscript, (cimb-4139540) entitled “Scissor– CIBERSORTx deconvolution reveals functional heterogeneity of CTAL/aTAL cells and associated biomarkers in renal fibrosis”. I am looking forward to working with you to move this manuscript closer to publication in “Current Issues in Molecular Biology”.

The manuscript has been rechecked and the necessary changes have been made in accordance with your suggestions. The responses to all comments have been prepared and attached below. We have tried our best to solve the problems you proposed, and we hope that the revised manuscript is now suitable for publication in the journal “Current Issues in Molecular Biology”. If you have any questions remained about this paper, please feel free to contact us.

2. Point-by-point response to Comments and Suggestions for Authors

Comments 1: [ .... samples derived from normal and fibrotic kidney allografts ....

- authors should describe the origin of these samples from kidney allografts because the study is about acute and chronic RF and not transplantation effects.]

Response 1: We sincerely appreciate the reviewer’s valuable comment. We agree that it is necessary to clearly describe the origin of the kidney allograft samples in the context of acute and chronic renal failure, rather than focusing on transplantation-related effects. The detailed information of the scRNA-seq dataset GSE195719 is shown in the following figure:

In the revised manuscript, we have clarified that a total of 9 kidney allograft biopsies were included in this study: A total of 9 biospies were included: 6 from CAD patients with IFTA and 3 from patients with stable graft function with normal/nonspecific histopathology served as controls.

The scRNA-seq dataset GSE195718 includes a total of 9 kidney allograft biopsies: 6 obtained from patients with chronic allograft dysfunction (CAD) accompanied by interstitial fibrosis and tubular atrophy (IFTA), and 3 from patients with stable graft function and normal or nonspecific histopathological findings, which served as controls. (lines 82-86)

These samples were selected specifically to reflect renal fibrosis and normal parenchymal status in the setting of chronic renal impairment, not to investigate transplantation-associated immune or surgical effects. The corresponding description has been updated in the manuscript to avoid misunderstanding. Thank you again for your professional guidance, which has significantly improved the completeness and scientific rigor of our manuscript.

Comments 2: [Scissor, CIBERSORTx analysis and WGCNA methodologies should be described in more detail, in order to be more understandable and, consequently, capable of reproducing by other researchers, preferably by a schematic diagram, for example.]

Response 2: We greatly appreciate the reviewer’s constructive suggestion. To improve the clarity and reproducibility of our study, we have added more detailed descriptions of the Scissor, CIBERSORTx, and WGCNA methodologies in the revised manuscript. Specifically, we have elaborated on the principles, analytical procedures, and key parameters of these three bioinformatics approaches. In addition, schematic diagrams illustrating the workflow of these analyses have been added to further facilitate understanding and replication by other researchers (Figure S1). The relevant modifications can be found in the "Materials and Methods" section of the revised manuscript (2.3 Scissor analysis, 2.4 CIBERSORTx analysis, and 2.5 WGCNA).

Scissor is a novel approach that utilizes the phenotypes, such as disease stage, tumor metastasis, treatment response, and survival outcomes, collected from bulk assays to identify the most highly phenotype-associated cell subpopulations from single-cell data. The workflow of Scissor is shown in Figure S1A (PMID: 41663022). (lines 105-108)

CIBERSORTx is an analytical tool developed by Newman et al. to impute gene expression profiles and provide an estimation of the abundances of member cell types in a mixed cell population, using gene expression data (PMID: 31061481 ). A typical CIBERSORTx workflow involves a serial approach (Figure S1B), in which molecular profiles of cell subsets are first obtained from a small collection of tissue samples and then repeatedly used to perform systematic analyses of cellular abundance and gene expression signatures from bulk tissue transcriptomes. This process involves: transcriptome profiling of single cells or sorted cell subpopulations to define a “signature matrix” consisting of barcode genes that can discriminate each cell subset of interest in a given tissue type; applying the signature matrix to bulk tissue RNA profiles in order to infer cell type proportions and representative cell type expression signatures; and purifying multiple transcriptomes for each cell type from a cohort of related tissue samples.(lines 125-136)

2.5 Weighted correlation network analysis (WGCNA)

WGCNA can be used for finding clusters (modules) of highly correlated genes, for summarizing such clusters using the module eigengene or an intramodular hub gene, for relating modules to one another and to external sample traits (using eigengene network methodology), and for calculating module membership measures. Correlation networks facilitate network based gene screening methods that can be used to identify candidate biomarkers or therapeutic targets (PMID:25875247). The workflow of WGCNA is shown in Figure S1C.(lines 146-153)

Figure S1. Workflow diagram of (A) Scissor (B) CIBERSORTx (C) WGCNA. (Supplementary Information)

We thank you again for your professional guidance, which has significantly improved the rigor and transparency of our research methodology.

Comments 3: [Figure S1 must be beter explained and have better resolution.]

Response 3: We sincerely thank the reviewer for pointing out this issue. We fully agree that Figure S1 requires better explanation and improved resolution to enhance readability and clarity. In the revised manuscript, We have made the following improvements to Figure S1, and Figure S1 has been updated to Figure S2.

Resolution optimization: We reprocessed the original data in Figure S1, adjusted the image parameters, increased the resolution, and ensured that all details were clearly visible, meeting the requirements of the publication.

Detailed explanation supplement: We have comprehensively explained Figure S1 in the “Results” section (3.1 Single-cell atlas reveals RF-associated renal cell heterogeneity and functional differentiation of thick ascending limb (TAL) cells) of the revised manuscript, with detailed descriptions as follows:

Following data filtering, 47,387 cells, comprising 27,263 cells from RF samples and 20,124 cells from control, as well as 30,767 genes, were retained (Figure S1A). PCA was performed on the top 2,000 HVGs, and the top 30 PCs were selected for subsequent analysis (Figure S1B-D). Through clustering analysis of cell clusters, a total of 17 categories were identified. Subsequently, cell type annotation was conducted on the 17 clusters derived from the clustering analysis, resulting in the identification of 13 distinct cell types, namely podocytes (PODO), proximal tubule (PT)/proximal convoluted tubule (PCT) cells, princal cells (PC), type A intercalated cells (AIC), type B intercalated cells (BIC), mesangial cells (MES), distal convoluted tubule cells (DCT), TAL, parietal epithelial cells (PEC), type 2 endothelial cells (Endo2), type 1 endothelial cells (Endo1), monocytes, and T cells (Figure 2A-B, Figure S1E). Following the removal of 3,554 high-confidence doublets (7.5% of the dataset), cell boundaries became more distinct with significantly reduced impact of technical noise on biological signals (Figure S1F). Among these cell types, the proportions of TAL and PT/PCT in the samples were both relatively high (Figure 2C). The Scissor algorithm demonstrated that a substantial proportion of cells in PT/PCT, TAL, Endo, Monocytes, and T populations exhibited positive associations with RF. Notably, a considerable number of cells within the PT/PCT, TAL, and AIC also displayed negative correlations with RF (Figure 2D). These findings suggest that these cell types might exhibit a certain degree of heterogeneity. This hypothesis was further supported by the ROGUE analysis, which revealed relatively lower scores for PT/PCT, TAL, and AIC (Figure 2E). Based on the integrated analysis of overall cell proportion, heterogeneity, and disease-related correlations, TAL was selected for secondary clustering. Among the 7 cell subclusturs, cluster 0 was annotated as medullary TAL (MTAL), cluster 1 as macula densa (MD), cluster 2 as ascending thin limb cell (ATL), cluster 3 as cortical TAL (CTAL), cluster 5 as adaptive TAL (aTAL). Cluster 4 was involved in protein targeted transport and specific subcellular structure localization and was named metabolically reprogrammed progenitor TAL (MRPTAL); Cluster 6 was enriched in amino acid metabolism and membrane transport-related biological processes and was named metabolically adapted transport TAL (MATTAL) (Figure 2F-G, Figure S1G, Table S1). As shown in Figure 2H, the annotation cell types and subtypes were summarized. (lines 263-293)

In the supplementary materials, we have added a more detailed legend to Figure S1. This legend clearly explains the key results depicted in Figure S1, which helps readers better understand the content and significance of Figure S1.

Figure S2. Single-cell RNA sequencing (scRNA-seq) data processing (A) scRNA-sel

data before (up) and after (down) quality control. (B) Selection of top 2,000 HVGs. The higher the height in the figure indicates a greater variance in the genes, with a larger difference. In the legend, red represents variable genes and black represents non-variable genes. The names of the top 10 genes are also displayed. (C) Permutation test for PCA. (D) Elbow plot of PCs. (E) Expression of marker genes in the main cell types. (F) The t-SNE plot of the cell types after removing the doublets. (G) Expression of marker genes for TAL subtypes. (Supplementary Information)

All revisions of Figure S1 (improved resolution and detailed legend) have been updated in the revised Supplementary Materials for the reviewer’s reference. Thank you again for your professional guidance that helps improve the formal rigor of our manuscript.

Comments 4: [The dots at the top of Fig 3B and 3C probaly represent statistical significance of differences (in addition to those marked as NS), however, this should be explained in figure caption.]

Response 4:  We appreciate the reviewer’s careful observation and valuable suggestion. The reviewer is correct that the dots at the top of Figures 3B and 3C represent the statistical significance of differences between groups, in addition to the groups marked as “NS” (not significant). Statistical significance was defined as follows: ns, not significant (P > 0.05); *P < 0.05; **P< 0.01; ***P < 0.001; ****P < 0.0001.

To avoid ambiguity, we have revised the figure caption of Figures 3B and 3C in the revised manuscript. Specifically, we have clearly added the explanation regarding statistical significance, and specified the corresponding P-value threshold.

Figure 3. Identification of key cell subtypes from TAL. (A) Heatmap of CIBERSORTx deconvolution results. (B) Box plot showing the proportion of cells in the RF group and the control group in GSE76882 dataset; (C) Box plot showing the proportion of cells in the RF group and the control group in GSE135327 dataset; (D) Clustering results of WGCNA in GSE76882 dataset; (E) Selection of soft threshold; (F) Gene dendrogram and module colors; (G) Heat map of the correlation between modules and traits. ns: not significant, * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001. (Lines 318-324)

This revision ensures that readers can clearly understand the statistical significance information reflected by the dots in Figures 3B and 3C. The updated figure captions have been included in the revised manuscript for the reviewer’s inspection. Thank you again for your professional guidance that helps improve the formal rigor of our manuscript.

Comments 5: [the concept of pseudotime analysis should be better explained.]

Response 2: We thank the reviewer for their insightful comment. We fully agree that a clearer explanation of the pseudotime analysis concept is essential to improve the readability and comprehensibility of the manuscript. In the revised manuscript, we have supplemented a detailed description of pseudotime analysis as follows:

Pseudotime analysis is a computational approach used to infer the dynamic, continuous developmental or transition trajectory of cells based on high-throughput sequencing data. It simulates a "virtual timeline" (pseudotime) that reflects the sequential changes of cells from an initial state to a terminal state, without relying on actual temporal sampling (PMID:32107427). This method can identify the ordered progression of cellular states, key genes driving the transition, and the dynamic expression patterns of genes along the trajectory, which is particularly useful for exploring the dynamic process of renal fibrosis and cell fate determination in our study.(lines 180-187)

We have incorporated this detailed explanation into the "Materials and Methods" section (2.7 Cell communication and pseudotime trajectory analyses) of the revised manuscript, ensuring that the concept of pseudotime analysis is clearly presented, making it easier for readers and other researchers to understand and follow up on.

Comments 6: [Fig. 5 G: biomarkers in cell subtypes of TAL seem to be absent in a diagram and, on the other hand, cell subtypes are marked twice. In addition, units on x-axis are missing. Please, check again.]

Response 6: We sincerely appreciate the reviewer’s careful observation and valuable suggestion. We fully agree with the reviewer’s comments regarding Figure 5G and have carefully rechecked and revised the figure to address all the raised issues.

Specifically, Figure 5G illustrates the expression profiles of all biomarkers across TAL cell subtypes. A total of 7 cell subtypes were detected within the TAL, cluster 0 was annotated as medullary TAL (MTAL), cluster 1 as macula densa (MD), cluster 2 as ascending thin limb cell (ATL), cluster 3 as cortical TAL (CTAL), Cluster 4 as metabolically reprogrammed progenitor TAL (MRPTAL)，cluster 5 as adaptive TAL (aTAL) and Cluster 6 as metabolically adapted transport TAL (MATTAL). We ensure that each TAL cell subtype is clearly associated with its corresponding specific biomarker; We have corrected the duplicate marking of TAL cell subtypes, removing the redundant labels to avoid confusion; We have added the missing "expression quantity" unit on the x-axis of Figure 5G to ensure that the presented statistical or quantitative information is complete and clear.

Figure 5. Identification of biomarkers. (A) Volcano plot of differentially expressed genes (DEGs) between the RF group and the control group in the GSE76882 dataset; (B) Heatmap of top 20 upregulated and downregulated DEGs in the GSE76882 dataset, ranked by |log2FC|; (C) Venn diagram of 378 DEGs in aTAL, 421 DEGs in GSE76882 dataset, and 4,699 key module genes; (D) Venn diagram of 174 DEGs in cTAL, 421 DEGs in GSE76882 dataset, and 4,699 key module genes; (E) Violin plot of STAT1 and PARP8 expression across samples in the GSE76882 and GSE135327 datasets; (F) Violin plot of HS6ST2, PTGER3, and TMEM207 expression across samples in the GSE76882 and GSE135327 datasets; (G) Expression of biomarkers in cell subtypes of TAL; (H) Biomarkers expression in pseudotime trajectory analysis. * P < 0.05, *** P < 0.001. (lines 360-368)

All the above revisions have been completed, and the updated Figure 5G has been included in the revised manuscript. We have also double-checked the entire figure to ensure no other similar issues exist, and the corresponding figure caption has been slightly adjusted to match the revised content for consistency. The revised Figure 5G is available for the reviewer’s inspection. Thank you again for your professional guidance that helps improve the formal rigor of our manuscript.

Comments 7: [Fig. 6: There are many different diseases mentioned in this figure that are not associated with this study and, therefore, should be mentioned in Discussion and not presented in this figure. In addition, regarding the presented gene set variation analysis (GSVA) between the RF and control groups: there is no mark which group belongs to green and which to orange colour.]

Response 7: We sincerely appreciate the reviewer’s careful observation and valuable suggestion. We fully agree with the reviewer’s comments regarding Figure 6 and have thoroughly revised the figure as required.

First, regarding the irrelevant diseases mentioned in Figure 6: We have recreated Figure 6, and all diseases not associated with this study have been removed from the figure to avoid distraction and ensure the figure content is closely related to the research focus (renal fibrosis, RF). The GSEA enrichment analysis graph is divided into two parts: the Enrichment Score line section, where the x-axis represents the sorted genes and the y-axis represents the corresponding Running ES. There is a peak in the line graph, which is the Enrichment score of this pathway gene set. The genes before the peak are the core genes of this gene set; the lines in the lower part mark the genes under this gene set.

Second, regarding the gene set variation analysis (GSVA) between the RF and control groups: We have added clear color annotations in the revised Figure 6 and its caption to specify the grouping corresponding to each color. Specifically, in the GSVA results, green represents down-regulated pathways, and orange represents up-regulated pathways. This modification ensures that readers can clearly distinguish the expression trends of each gene set between the renal fibrosis group and the control group.

Figure 6. Enrichment analysis of biomarkers. (A) Gene Set Enrichment Analysis (GSEA) of STAT1; (B) GSEA of PARP8; (C) GSEA of HS6ST2; (D) GSEA of PTGER3; (E) GSEA of TMEM207. The GSEA enrichment analysis graph is divided into two parts: the Enrichment Score line section, where the x-axis represents the sorted genes and the y-axis represents the corresponding Running ES. There is a peak in the line graph, which is the Enrichment score of this pathway gene set. The genes before the peak are the core genes of this gene set; the lines in the lower part mark the genes under this gene set. (F) Gene set variation analysis (GSVA) between the RF and control groups in the GSE76882 dataset, green represents down-regulated pathways, and orange represents up-regulated pathways. (lines 381-389)

The recreated Figure 6 (with irrelevant diseases removed and color annotations added) and the updated corresponding figure caption have been included in the revised manuscript. We have double-checked the revised figure to ensure consistency with the study content and clarity of the GSVA results. The revised Figure 6 and related descriptions are available for the reviewer’s inspection. Thank you again for your professional guidance that helps improve the formal rigor of our manuscript.

Comments 8: [The finding that "..........HS6ST2 exhibited a dynamic temporal profile, with initial downregulation during the acute phase followed by upregulation in the chronic phase ......." suggests that this biomarker might have antioxidant or anti-inflammatory characteristics and, therefore, deserves some more discussion and literature search about its role.]

Response 8: We sincerely thank the reviewer for the insightful comment regarding the potential antioxidant or anti-inflammatory properties of HS6ST2. Following a thorough review of the relevant literature, we find that current evidence indeed supports a significant role for HS6ST2 in both anti-inflammatory responses and fibrotic progression. Accordingly, we have incorporated a revised discussion into the manuscript as follows:

Notably, as a member of the heparan sulfate–modifying enzyme family, HS6ST2 exhibits differential functions in acute versus chronic disease states. The literature indicates that, in acute inflammatory conditions such as osteoarthritis, HS6ST2 expression is downregulated and is associated with reduced chondrocyte activity (PMID:29899528; PMID:30205019). Conversely, across chronic fibrotic models affecting the heart, liver, and kidney, HS6ST2 is selectively upregulated in activated fibroblasts, where it promotes extracellular matrix deposition, including collagen production, thereby driving fibrotic progression(DOI: 10.1248/bpbreports.4.3_85). Consistent with these observations, HS6ST2 upregulation during the chronic phase of the UUO model corroborates prior reports and suggests that HS6ST2 may serve as a stage-specific biomarker. Furthermore, its dynamic expression may reflect the pathological transition from acute injury to chronic fibrotic remodeling, and its context-dependent roles—anti-inflammatory in acute settings and pro-fibrotic in chronic settings—require further mechanistic validation. (lines 501-513)

We appreciate the reviewer’s constructive suggestion, which has significantly strengthened our discussion.

Comments 9: [In references, usually just one author is listed, followed by et al. This is not in agreement with journal guidelines and shoul be corrected.]

Response to Reviewer 1 Comments

1. Summary

Dear Reviewer,

Thank you for your thoughtful suggestions and insights, which have benefited from the manuscript, (cimb-4139540) entitled “Scissor– CIBERSORTx deconvolution reveals functional heterogeneity of CTAL/aTAL cells and associated biomarkers in renal fibrosis”. I am looking forward to working with you to move this manuscript closer to publication in “Current Issues in Molecular Biology”.

The manuscript has been rechecked and the necessary changes have been made in accordance with your suggestions. The responses to all comments have been prepared and attached below. We have tried our best to solve the problems you proposed, and we hope that the revised manuscript is now suitable for publication in the journal “Current Issues in Molecular Biology”. If you have any questions remained about this paper, please feel free to contact us.

2. Point-by-point response to Comments and Suggestions for Authors

Comments 1: [ .... samples derived from normal and fibrotic kidney allografts ....

- authors should describe the origin of these samples from kidney allografts because the study is about acute and chronic RF and not transplantation effects.]

Response 1: We sincerely appreciate the reviewer’s valuable comment. We agree that it is necessary to clearly describe the origin of the kidney allograft samples in the context of acute and chronic renal failure, rather than focusing on transplantation-related effects. The detailed information of the scRNA-seq dataset GSE195719 is shown in the following figure:

In the revised manuscript, we have clarified that a total of 9 kidney allograft biopsies were included in this study: A total of 9 biospies were included: 6 from CAD patients with IFTA and 3 from patients with stable graft function with normal/nonspecific histopathology served as controls.

The scRNA-seq dataset GSE195718 includes a total of 9 kidney allograft biopsies: 6 obtained from patients with chronic allograft dysfunction (CAD) accompanied by interstitial fibrosis and tubular atrophy (IFTA), and 3 from patients with stable graft function and normal or nonspecific histopathological findings, which served as controls. (lines 82-86)

These samples were selected specifically to reflect renal fibrosis and normal parenchymal status in the setting of chronic renal impairment, not to investigate transplantation-associated immune or surgical effects. The corresponding description has been updated in the manuscript to avoid misunderstanding. Thank you again for your professional guidance, which has significantly improved the completeness and scientific rigor of our manuscript.

Comments 2: [Scissor, CIBERSORTx analysis and WGCNA methodologies should be described in more detail, in order to be more understandable and, consequently, capable of reproducing by other researchers, preferably by a schematic diagram, for example.]

Response 2: We greatly appreciate the reviewer’s constructive suggestion. To improve the clarity and reproducibility of our study, we have added more detailed descriptions of the Scissor, CIBERSORTx, and WGCNA methodologies in the revised manuscript. Specifically, we have elaborated on the principles, analytical procedures, and key parameters of these three bioinformatics approaches. In addition, schematic diagrams illustrating the workflow of these analyses have been added to further facilitate understanding and replication by other researchers (Figure S1). The relevant modifications can be found in the "Materials and Methods" section of the revised manuscript (2.3 Scissor analysis, 2.4 CIBERSORTx analysis, and 2.5 WGCNA).

Scissor is a novel approach that utilizes the phenotypes, such as disease stage, tumor metastasis, treatment response, and survival outcomes, collected from bulk assays to identify the most highly phenotype-associated cell subpopulations from single-cell data. The workflow of Scissor is shown in Figure S1A (PMID: 41663022). (lines 105-108)

CIBERSORTx is an analytical tool developed by Newman et al. to impute gene expression profiles and provide an estimation of the abundances of member cell types in a mixed cell population, using gene expression data (PMID: 31061481 ). A typical CIBERSORTx workflow involves a serial approach (Figure S1B), in which molecular profiles of cell subsets are first obtained from a small collection of tissue samples and then repeatedly used to perform systematic analyses of cellular abundance and gene expression signatures from bulk tissue transcriptomes. This process involves: transcriptome profiling of single cells or sorted cell subpopulations to define a “signature matrix” consisting of barcode genes that can discriminate each cell subset of interest in a given tissue type; applying the signature matrix to bulk tissue RNA profiles in order to infer cell type proportions and representative cell type expression signatures; and purifying multiple transcriptomes for each cell type from a cohort of related tissue samples.(lines 125-136)

2.5 Weighted correlation network analysis (WGCNA)

WGCNA can be used for finding clusters (modules) of highly correlated genes, for summarizing such clusters using the module eigengene or an intramodular hub gene, for relating modules to one another and to external sample traits (using eigengene network methodology), and for calculating module membership measures. Correlation networks facilitate network based gene screening methods that can be used to identify candidate biomarkers or therapeutic targets (PMID:25875247). The workflow of WGCNA is shown in Figure S1C.(lines 146-153)

Figure S1. Workflow diagram of (A) Scissor (B) CIBERSORTx (C) WGCNA. (Supplementary Information)

We thank you again for your professional guidance, which has significantly improved the rigor and transparency of our research methodology.

Comments 3: [Figure S1 must be beter explained and have better resolution.]

Response 3: We sincerely thank the reviewer for pointing out this issue. We fully agree that Figure S1 requires better explanation and improved resolution to enhance readability and clarity. In the revised manuscript, We have made the following improvements to Figure S1, and Figure S1 has been updated to Figure S2.

Resolution optimization: We reprocessed the original data in Figure S1, adjusted the image parameters, increased the resolution, and ensured that all details were clearly visible, meeting the requirements of the publication.

Detailed explanation supplement: We have comprehensively explained Figure S1 in the “Results” section (3.1 Single-cell atlas reveals RF-associated renal cell heterogeneity and functional differentiation of thick ascending limb (TAL) cells) of the revised manuscript, with detailed descriptions as follows:

Following data filtering, 47,387 cells, comprising 27,263 cells from RF samples and 20,124 cells from control, as well as 30,767 genes, were retained (Figure S1A). PCA was performed on the top 2,000 HVGs, and the top 30 PCs were selected for subsequent analysis (Figure S1B-D). Through clustering analysis of cell clusters, a total of 17 categories were identified. Subsequently, cell type annotation was conducted on the 17 clusters derived from the clustering analysis, resulting in the identification of 13 distinct cell types, namely podocytes (PODO), proximal tubule (PT)/proximal convoluted tubule (PCT) cells, princal cells (PC), type A intercalated cells (AIC), type B intercalated cells (BIC), mesangial cells (MES), distal convoluted tubule cells (DCT), TAL, parietal epithelial cells (PEC), type 2 endothelial cells (Endo2), type 1 endothelial cells (Endo1), monocytes, and T cells (Figure 2A-B, Figure S1E). Following the removal of 3,554 high-confidence doublets (7.5% of the dataset), cell boundaries became more distinct with significantly reduced impact of technical noise on biological signals (Figure S1F). Among these cell types, the proportions of TAL and PT/PCT in the samples were both relatively high (Figure 2C). The Scissor algorithm demonstrated that a substantial proportion of cells in PT/PCT, TAL, Endo, Monocytes, and T populations exhibited positive associations with RF. Notably, a considerable number of cells within the PT/PCT, TAL, and AIC also displayed negative correlations with RF (Figure 2D). These findings suggest that these cell types might exhibit a certain degree of heterogeneity. This hypothesis was further supported by the ROGUE analysis, which revealed relatively lower scores for PT/PCT, TAL, and AIC (Figure 2E). Based on the integrated analysis of overall cell proportion, heterogeneity, and disease-related correlations, TAL was selected for secondary clustering. Among the 7 cell subclusturs, cluster 0 was annotated as medullary TAL (MTAL), cluster 1 as macula densa (MD), cluster 2 as ascending thin limb cell (ATL), cluster 3 as cortical TAL (CTAL), cluster 5 as adaptive TAL (aTAL). Cluster 4 was involved in protein targeted transport and specific subcellular structure localization and was named metabolically reprogrammed progenitor TAL (MRPTAL); Cluster 6 was enriched in amino acid metabolism and membrane transport-related biological processes and was named metabolically adapted transport TAL (MATTAL) (Figure 2F-G, Figure S1G, Table S1). As shown in Figure 2H, the annotation cell types and subtypes were summarized. (lines 263-293)

In the supplementary materials, we have added a more detailed legend to Figure S1. This legend clearly explains the key results depicted in Figure S1, which helps readers better understand the content and significance of Figure S1.

Figure S2. Single-cell RNA sequencing (scRNA-seq) data processing (A) scRNA-sel

data before (up) and after (down) quality control. (B) Selection of top 2,000 HVGs. The higher the height in the figure indicates a greater variance in the genes, with a larger difference. In the legend, red represents variable genes and black represents non-variable genes. The names of the top 10 genes are also displayed. (C) Permutation test for PCA. (D) Elbow plot of PCs. (E) Expression of marker genes in the main cell types. (F) The t-SNE plot of the cell types after removing the doublets. (G) Expression of marker genes for TAL subtypes. (Supplementary Information)

All revisions of Figure S1 (improved resolution and detailed legend) have been updated in the revised Supplementary Materials for the reviewer’s reference. Thank you again for your professional guidance that helps improve the formal rigor of our manuscript.

Comments 4: [The dots at the top of Fig 3B and 3C probaly represent statistical significance of differences (in addition to those marked as NS), however, this should be explained in figure caption.]

Response 4:  We appreciate the reviewer’s careful observation and valuable suggestion. The reviewer is correct that the dots at the top of Figures 3B and 3C represent the statistical significance of differences between groups, in addition to the groups marked as “NS” (not significant). Statistical significance was defined as follows: ns, not significant (P > 0.05); *P < 0.05; **P< 0.01; ***P < 0.001; ****P < 0.0001.

To avoid ambiguity, we have revised the figure caption of Figures 3B and 3C in the revised manuscript. Specifically, we have clearly added the explanation regarding statistical significance, and specified the corresponding P-value threshold.

Figure 3. Identification of key cell subtypes from TAL. (A) Heatmap of CIBERSORTx deconvolution results. (B) Box plot showing the proportion of cells in the RF group and the control group in GSE76882 dataset; (C) Box plot showing the proportion of cells in the RF group and the control group in GSE135327 dataset; (D) Clustering results of WGCNA in GSE76882 dataset; (E) Selection of soft threshold; (F) Gene dendrogram and module colors; (G) Heat map of the correlation between modules and traits. ns: not significant, * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001. (Lines 318-324)

This revision ensures that readers can clearly understand the statistical significance information reflected by the dots in Figures 3B and 3C. The updated figure captions have been included in the revised manuscript for the reviewer’s inspection. Thank you again for your professional guidance that helps improve the formal rigor of our manuscript.

Comments 5: [the concept of pseudotime analysis should be better explained.]

Response 2: We thank the reviewer for their insightful comment. We fully agree that a clearer explanation of the pseudotime analysis concept is essential to improve the readability and comprehensibility of the manuscript. In the revised manuscript, we have supplemented a detailed description of pseudotime analysis as follows:

Pseudotime analysis is a computational approach used to infer the dynamic, continuous developmental or transition trajectory of cells based on high-throughput sequencing data. It simulates a "virtual timeline" (pseudotime) that reflects the sequential changes of cells from an initial state to a terminal state, without relying on actual temporal sampling (PMID:32107427). This method can identify the ordered progression of cellular states, key genes driving the transition, and the dynamic expression patterns of genes along the trajectory, which is particularly useful for exploring the dynamic process of renal fibrosis and cell fate determination in our study.(lines 180-187)

We have incorporated this detailed explanation into the "Materials and Methods" section (2.7 Cell communication and pseudotime trajectory analyses) of the revised manuscript, ensuring that the concept of pseudotime analysis is clearly presented, making it easier for readers and other researchers to understand and follow up on.

Comments 6: [Fig. 5 G: biomarkers in cell subtypes of TAL seem to be absent in a diagram and, on the other hand, cell subtypes are marked twice. In addition, units on x-axis are missing. Please, check again.]

Response 6: We sincerely appreciate the reviewer’s careful observation and valuable suggestion. We fully agree with the reviewer’s comments regarding Figure 5G and have carefully rechecked and revised the figure to address all the raised issues.

Specifically, Figure 5G illustrates the expression profiles of all biomarkers across TAL cell subtypes. A total of 7 cell subtypes were detected within the TAL, cluster 0 was annotated as medullary TAL (MTAL), cluster 1 as macula densa (MD), cluster 2 as ascending thin limb cell (ATL), cluster 3 as cortical TAL (CTAL), Cluster 4 as metabolically reprogrammed progenitor TAL (MRPTAL)，cluster 5 as adaptive TAL (aTAL) and Cluster 6 as metabolically adapted transport TAL (MATTAL). We ensure that each TAL cell subtype is clearly associated with its corresponding specific biomarker; We have corrected the duplicate marking of TAL cell subtypes, removing the redundant labels to avoid confusion; We have added the missing "expression quantity" unit on the x-axis of Figure 5G to ensure that the presented statistical or quantitative information is complete and clear.

Figure 5. Identification of biomarkers. (A) Volcano plot of differentially expressed genes (DEGs) between the RF group and the control group in the GSE76882 dataset; (B) Heatmap of top 20 upregulated and downregulated DEGs in the GSE76882 dataset, ranked by |log2FC|; (C) Venn diagram of 378 DEGs in aTAL, 421 DEGs in GSE76882 dataset, and 4,699 key module genes; (D) Venn diagram of 174 DEGs in cTAL, 421 DEGs in GSE76882 dataset, and 4,699 key module genes; (E) Violin plot of STAT1 and PARP8 expression across samples in the GSE76882 and GSE135327 datasets; (F) Violin plot of HS6ST2, PTGER3, and TMEM207 expression across samples in the GSE76882 and GSE135327 datasets; (G) Expression of biomarkers in cell subtypes of TAL; (H) Biomarkers expression in pseudotime trajectory analysis. * P < 0.05, *** P < 0.001. (lines 360-368)

All the above revisions have been completed, and the updated Figure 5G has been included in the revised manuscript. We have also double-checked the entire figure to ensure no other similar issues exist, and the corresponding figure caption has been slightly adjusted to match the revised content for consistency. The revised Figure 5G is available for the reviewer’s inspection. Thank you again for your professional guidance that helps improve the formal rigor of our manuscript.

Comments 7: [Fig. 6: There are many different diseases mentioned in this figure that are not associated with this study and, therefore, should be mentioned in Discussion and not presented in this figure. In addition, regarding the presented gene set variation analysis (GSVA) between the RF and control groups: there is no mark which group belongs to green and which to orange colour.]

Response 7: We sincerely appreciate the reviewer’s careful observation and valuable suggestion. We fully agree with the reviewer’s comments regarding Figure 6 and have thoroughly revised the figure as required.

First, regarding the irrelevant diseases mentioned in Figure 6: We have recreated Figure 6, and all diseases not associated with this study have been removed from the figure to avoid distraction and ensure the figure content is closely related to the research focus (renal fibrosis, RF). The GSEA enrichment analysis graph is divided into two parts: the Enrichment Score line section, where the x-axis represents the sorted genes and the y-axis represents the corresponding Running ES. There is a peak in the line graph, which is the Enrichment score of this pathway gene set. The genes before the peak are the core genes of this gene set; the lines in the lower part mark the genes under this gene set.

Second, regarding the gene set variation analysis (GSVA) between the RF and control groups: We have added clear color annotations in the revised Figure 6 and its caption to specify the grouping corresponding to each color. Specifically, in the GSVA results, green represents down-regulated pathways, and orange represents up-regulated pathways. This modification ensures that readers can clearly distinguish the expression trends of each gene set between the renal fibrosis group and the control group.

Figure 6. Enrichment analysis of biomarkers. (A) Gene Set Enrichment Analysis (GSEA) of STAT1; (B) GSEA of PARP8; (C) GSEA of HS6ST2; (D) GSEA of PTGER3; (E) GSEA of TMEM207. The GSEA enrichment analysis graph is divided into two parts: the Enrichment Score line section, where the x-axis represents the sorted genes and the y-axis represents the corresponding Running ES. There is a peak in the line graph, which is the Enrichment score of this pathway gene set. The genes before the peak are the core genes of this gene set; the lines in the lower part mark the genes under this gene set. (F) Gene set variation analysis (GSVA) between the RF and control groups in the GSE76882 dataset, green represents down-regulated pathways, and orange represents up-regulated pathways. (lines 381-389)

The recreated Figure 6 (with irrelevant diseases removed and color annotations added) and the updated corresponding figure caption have been included in the revised manuscript. We have double-checked the revised figure to ensure consistency with the study content and clarity of the GSVA results. The revised Figure 6 and related descriptions are available for the reviewer’s inspection. Thank you again for your professional guidance that helps improve the formal rigor of our manuscript.

Comments 8: [The finding that "..........HS6ST2 exhibited a dynamic temporal profile, with initial downregulation during the acute phase followed by upregulation in the chronic phase ......." suggests that this biomarker might have antioxidant or anti-inflammatory characteristics and, therefore, deserves some more discussion and literature search about its role.]

Response 8: We sincerely thank the reviewer for the insightful comment regarding the potential antioxidant or anti-inflammatory properties of HS6ST2. Following a thorough review of the relevant literature, we find that current evidence indeed supports a significant role for HS6ST2 in both anti-inflammatory responses and fibrotic progression. Accordingly, we have incorporated a revised discussion into the manuscript as follows:

Notably, as a member of the heparan sulfate–modifying enzyme family, HS6ST2 exhibits differential functions in acute versus chronic disease states. The literature indicates that, in acute inflammatory conditions such as osteoarthritis, HS6ST2 expression is downregulated and is associated with reduced chondrocyte activity (PMID:29899528; PMID:30205019). Conversely, across chronic fibrotic models affecting the heart, liver, and kidney, HS6ST2 is selectively upregulated in activated fibroblasts, where it promotes extracellular matrix deposition, including collagen production, thereby driving fibrotic progression(DOI: 10.1248/bpbreports.4.3_85). Consistent with these observations, HS6ST2 upregulation during the chronic phase of the UUO model corroborates prior reports and suggests that HS6ST2 may serve as a stage-specific biomarker. Furthermore, its dynamic expression may reflect the pathological transition from acute injury to chronic fibrotic remodeling, and its context-dependent roles—anti-inflammatory in acute settings and pro-fibrotic in chronic settings—require further mechanistic validation. (lines 501-513)

We appreciate the reviewer’s constructive suggestion, which has significantly strengthened our discussion.

Comments 9: [In references, usually just one author is listed, followed by et al. This is not in agreement with journal guidelines and shoul be corrected.]

Response 9: We sincerely thank the reviewer for pointing out this formatting issue. We fully agree that the citation format of references (listing only one author followed by “et al.”) does not comply with the journal’s guidelines, and we have promptly corrected this problem.

In the revised manuscript, we have carefully reviewed all references and adjusted the citation format strictly in accordance with the journal’s requirements. Specifically, we have corrected the author listing format for each reference, ensuring that the number of listed authors, the use of “et al.”, and other citation details are consistent with the journal’s guidelines. We have double-checked all references to confirm that no similar formatting errors remain, ensuring the standardization and compliance of the reference section.

The revised references have been updated in the manuscript (lines 575-723), and we apologize for any inconvenience caused by the initial formatting oversight. The corrected reference section is available for the reviewer’s inspection.

Response to Reviewer 2 Comments

1. Summary

Dear Reviewer,

Thank you for your thoughtful suggestions and insights, which have benefited from the manuscript, (cimb-4139540) entitled “Scissor– CIBERSORTx deconvolution reveals functional heterogeneity of CTAL/aTAL cells and associated biomarkers in renal fibrosis”. I am looking forward to working with you to move this manuscript closer to publication in “Current Issues in Molecular Biology”.

The manuscript has been rechecked and the necessary changes have been made in accordance with your suggestions. The responses to all comments have been prepared and attached below. We have tried our best to solve the problems you proposed, and we hope that the revised manuscript is now suitable for publication in the journal “Current Issues in Molecular Biology”. If you have any questions remained about this paper, please feel free to contact us.

2. Point-by-point response to Comments and Suggestions for Authors

Comments 1: [Fix the missing negative sign in PARP8 binding energy (-7.57 kcal/mol, Section 3.6/3.7).]

Response 1: We sincerely thank the reviewer for this careful observation. The missing negative sign for the PARP8 binding energy value has been corrected. It now reads “-7.57 kcal/mol” in the revised manuscript (Line 406).

Comments 2: [Resolve duplicate section numbering (two 3.6 sections).]

Response 2: We thank the reviewer for pointing out this numbering error. The duplicate section numbering has been resolved. The second subsection originally numbered “3.6” has been correctly renumbered as “3.7” in the revised manuscript (Line 438).

Comments 3: [Correct encoding errors throughout (e.g., “principle cells” → “principal cells”).]

Response 3: We appreciate the reviewer’s attention to detail regarding terminology consistency. The encoding/typographical error has been corrected. “principle cells (PC)” has been changed to “principal cells (PC)” throughout the text, including at the indicated location (Line 270).

Comments 4: [The Abstract is overly long and repetitive—condense by removing redundant biomarker listings and splitting the final sentence.]

Response 4: Thank you sincerely for your valuable comment on the Abstract. We have carefully revised the Abstract in accordance with your suggestions to address the issues of lengthiness and repetition:

Renal fibrosis (RF) represents a major pathological outcome of chronic kidney disease, currently accompanied by extremely limited therapeutic strategies. To decipher key cellular and molecular drivers, we integrated single-cell and bulk transcriptomic profiles for comprehensive analysis. Based on the RF-related single-cell and bulk transcriptomic data, key cell subtypes were identified through Scissor analysis, custom signature matrix construction via CIBERSORTx, and Weighted gene co-expression network analysis (WGCNA). Subsequently, key subtypes-related biomarkers were identified through the expression analysis, and functional enrichment analysis for biomarkers was conducted to elucidate the potential mechanisms by which biomarkers regulate RF. Through comprehensive profiling, thick ascending limb (TAL) cells were predominant and displayed marked heterogeneity in renal fibrosis (RF), with cortical TAL (CTAL) and adaptive TAL (aTAL) identified as principal subtypes. A set of candidate biomarkers was identified. Quantitative polymerase chain reaction (qPCR) validation in mouse models confirmed aberrant expression of these biomarkers, with STAT1 and PARP8 upregulated and HS6ST2, PTGER3, and TMEM207 downregulated in RF. Furthermore, functional enrichment analyses indicated that these biomarkers were associated with pathways underlying metabolic reprogramming and immune perturbation. Our study implicates CTAL and aTAL as central cellular players in RF and identifies their associated biomarkers. These experimentally validated biomarkers provide novel targets and repurposing opportunities for RF therapeutic intervention. (Lines 10-27)

To avoid redundancy, we have removed the redundant listing of biomarkers. When first introducing the biomarkers, we simplified the relevant description to "A set of candidate biomarkers was identified"; meanwhile, we retained the key information regarding the abnormal expression of these biomarkers (STAT1 and PARP8 upregulated, HS6ST2, PTGER3, and TMEM207 downregulated) in the qPCR validation section, ensuring the core findings are not omitted.

We have split the overly long final sentence into two independent and concise sentences. One clarifies that CTAL and aTAL are central cellular players in renal fibrosis and identifies their associated biomarkers; the other elaborates on the value of these experimentally validated biomarkers for therapeutic intervention in renal fibrosis. This revision significantly improves the readability and clarity of the Abstract.

In addition, during the revision process, we have corrected minor grammatical errors in the Abstract to further ensure the accuracy and rigor of the manuscript.

Comments 5: [Unify abbreviation usage (qPCR vs. quantitative real-time PCR) and maintain consistent case for cell type abbreviations (aTAL vs. ATAL).]

Response 5: Thank you for your careful reading and constructive comments on our manuscript. We have carefully considered your suggestions and revised the manuscript accordingly. Below is our point-by-point response to the specific formatting issues you raised.

1. We agree that the inconsistent use of “qPCR” and “quantitative real-time PCR” throughout the text may cause confusion. To address this, we have standardized the abbreviation by using “qPCR” consistently in all instances, after defining it at its first occurrence. This change has been implemented in the Abstract, Methods, Results, and Figure legends.

2. Thank you for pointing out the inconsistency in the case formatting for the cell type abbreviation. We have now unified the term to “aTAL” (all uppercase) throughout the entire manuscript, including the text, tables, and figures, to adhere to standard nomenclature conventions.

Comments 6: [The Conclusion contains an excessively long final sentence—divide for clarity.]

Response 6: Thank you for your careful review and valuable comment regarding the structure of the Conclusion section. We agree that the original long sentence could affect readability and have revised it accordingly. As suggested, the final sentence has been divided into two separate sentences to improve clarity and flow. The revised Conclusion now reads:

Future studies should validate these findings in independent cohorts and perform functional assays to elucidate the regulatory mechanisms of TAL subtype transitions and biomarker functions. These efforts are essential to advance the development of cell-targeted therapies for kidney disease.(Lines 541-544)

3. Response to Comments on the Quality of English Language

Point 1: The English could be improved to more clearly express the research.

Response 1: Thank you for your careful review and for pointing out that the English expression could be further improved to more clearly convey the research. We fully agree that clear and accurate language is essential for effectively communicating scientific findings. In response to your comment, we have undertaken a systematic revision of the language throughout the manuscript, including the following measures:

We have carefully reviewed the manuscript line by line, focusing on correcting grammatical issues, simplifying overly complex sentences, and clarifying ambiguous expressions to ensure that each statement is concise and precise. In sections such as Methods, Results, and Discussion, we have strengthened the logical connections between sentences and paragraphs to present the research narrative more coherently.To further enhance the quality of the writing, we sought assistance from a colleague with expertise in English academic writing. Based on their feedback, we have revised multiple passages for clarity and readability.

We sincerely appreciate your valuable suggestion, which has helped us improve the clarity and professionalism of our work.