Abstract
Background
Renal fibrosis is a critical factor in chronic kidney disease progression, with limited diagnostic and therapeutic options. Emerging evidence suggests RNA-binding proteins (RBPs) are pivotal in regulating cellular mechanisms underlying fibrosis.
Methods
Utilizing an extensive GEO dataset (175 renal fibrosis and 99 normal kidney samples), we identified and validated key RBPs through integrated bioinformatics and machine learning approaches, including lasso and logistic regression models. Differentially expressed genes were analyzed for pathway enrichment using Gene Ontology and KEGG. Single-cell RNA sequencing delineated cell-specific RBP expression, and a murine unilateral ureteral obstruction (UUO) model provided experimental validation.
Results
A diagnostic model incorporating five RBPs (FKBP11, DCDC2, COL6A3, PLCB4, and GNB5) achieved high accuracy (AUC = 0.899) and robust external validation. These RBPs are implicated in immune-mediated pathways such as cytokine signaling and inflammatory responses. Single-cell analysis highlighted their expression in specific renal cell populations, underscoring functional diversity. Immunofluorescence linked FKBP11 with macrophage infiltration, suggesting its potential as a therapeutic target.
Conclusion
his study identifies novel RBPs associated with renal fibrosis, advancing the understanding of its pathogenesis and offering actionable biomarkers and therapeutic targets. The integration of bioinformatics and machine learning emphasizes their translational potential in kidney care.
Keywords: RNA-binding proteins (RBPs), renal fibrosis, machine learning, bioinformatics, immune pathways, diagnostic biomarkers
1. Introduction
Post-transplant renal fibrosis is a leading cause of chronic allograft dysfunction and graft failure, significantly impacting long-term transplant outcomes [1]. Despite advances in immunosuppressive therapy, the precise molecular mechanisms driving fibrosis remain unclear, making early diagnosis and targeted intervention critical yet challenging [2].
RNA-binding proteins play an important role in the development of many diseases, such as inflammation and tumor, because they can regulate cellular RNA metabolism [3,4]. Recent technological advancements have revealed that a growing number of proteins possess RNA-binding activity, identifying them as novel candidate RNA-binding proteins (RBPs) [5,6]. These RBPs may play pivotal roles in various diseases, including post-transplant renal fibrosis [7,8]. However, the specific functions and mechanisms by which these newly identified RBPs contribute to renal fibrosis are still largely unexplored. In this study, we leverage the latest HydRA deep-learning algorithm to systematically analyze the role of newly identified candidate RBPs in post-transplant renal fibrosis. HydRA, with its enhanced specificity and sensitivity, offers unprecedented accuracy in predicting RNA-binding capacity, enabling the discovery of previously uncharacterized RBPs and their potential involvement in disease processes.
Through comprehensive differential expression analysis on a large GEO dataset, we identified RBPs that are differentially expressed in renal allografts. By integrating lasso and logistic regression techniques, we developed a robust diagnostic model to predict fibrosis in renal transplant patients, validated across multiple external datasets to ensure its reliability and clinical applicability. Moreover, we utilized Gene Ontology (GO), KEGG, and METASCAPE pathway enrichment analyses to investigate the signaling pathways associated with these candidate RBPs, particularly focusing on immune-related pathways such as cytokine signaling and inflammatory responses. Single-cell RNA sequencing further delineated the expression profiles of these RBPs across different renal cell populations, highlighting their diverse functional roles.
This study not only enhances our understanding of the molecular underpinnings of post-transplant renal fibrosis but also identifies novel RBPs as potential biomarkers and therapeutic targets. The integration of cutting-edge deep-learning techniques like HydRA in our research underscores the importance of technological innovation in unraveling complex disease mechanisms and advancing personalized medicine in renal transplantation.
2. Materials and methods
2.1. Differential expression analysis of candidate RNA-binding proteins
We identified a list of 1411 RNA-binding proteins (RBPs) from prior studies (Table S1) [9]. Differential gene expression analysis was performed using a dataset comprising 175 renal fibrosis and 99 normal kidney samples obtained from the GEO database (GSE76882). The criteria for differential expression were set at an absolute logFC value greater than 0.585 and a p value less than .05. The differentially expressed genes were then intersected with the RBP list to identify differentially expressed RBPs.
2.2. Construction of diagnostic model
Candidate genes were delineated through the intersection of differentially expressed genes. Utilizing the GSE12720 dataset, a diagnostic model was formulated, initiating with LASSO regression for the identification of pivotal RBPs, preceded by a 10-fold cross-validation to ascertain model stability. Subsequently, a logistic regression model was constructed incorporating the selected RBPs. The diagnostic model underwent validation employing the GSE65326 dataset comprising 6 normal renal and 16 renal fibrosis samples, the GSE22459 dataset consisting of 25 normal renal and 40 renal fibrosis samples, and the GSE7392 dataset with 22 normal renal and 8 renal fibrosis samples. Furthermore, a consolidated GEO-meta dataset was synthesized by amalgamating all three datasets to facilitate subsequent analysis.
2.3. Pathway enrichment analysis
Gene Ontology (GO), KEGG, and Metascape analyses were employed to identify signaling pathways related to renal fibrosis [10]. Differential expression analysis on the GEO-meta dataset was performed with a cutoff of logFC absolute value greater than 0.585 and a p value less than .05. KEGG and GO pathway analyses were conducted using R software, and Metascape analysis was performed by uploading the differential gene sets to the Metascape online platform.
2.4. Gene set enrichment analysis (GSEA)
GSEA was carried out using the ‘Enrichplot’ package in R to identify downstream signaling pathways associated with key RBPs [11].
2.5. Gene correlation analysis
Gene correlation analysis was performed using R software, with Pearson’s test employed to assess correlations. A cutoff value of correlation greater than 0.45 was applied. Cytoscape v3.9.0 was used to create correlation graphs.
2.6. Single-Cell RNA sequencing analysis
The GSE269062 single-cell RNA sequencing dataset was analyzed to examine the expression profiles of FKBP11, DCDC2, COL6A3, PLCB4, and GNB5 in renal fibrosis. UMAP was used for dimensionality reduction, and cell cluster annotation was performed based on specific markers. The expression of these RBPs in each cell population was analyzed.
2.7. RNA-binding protein domain analysis
The RNA-binding domains of the identified RBPs were predicted using the HydRA deep-learning model. Protein structures were downloaded from the UniProt database and visualized using PyMOL 1.8.6.
2.8. The hub gene expression and its relationship with macrophage
The male C57BL/6J murine model of renal fibrosis was established utilizing unilateral ureteral obstruction (UUO). Within this model, the expression levels of FKBP11, the hub gene in the LASSO model, were ascertained. Furthermore, the immunological microenvironment of fibrosis models with varying FKBP11 expression levels was characterized, with a particular focus on quantifying the infiltration of macrophages.
2.9. Statistical analysis
Statistical analyses were conducted using SPSS 24.0 and GraphPad Prism 8.0. Data are presented as means ± SEM. Two-tailed Student’s t-tests were used to compare quantitative data, and Pearson’s test was employed for correlation analysis. A p value of less than .05 was considered statistically significant. The workflow of the analysis is illustrated in Figure 1A.
Figure 1.
Identify candidate RNA binding proteins with differential expression. (A) Article analysis flowchart. (B) Gene differential analysis identified 1091 differentially expressed genes. (C) Gene differential analysis identifies 57 candidate RNA binding proteins with differential expression.
3. Results
3.1. Differential expression analysis of candidate RNA binding proteins
We identified 1,411 RNA-binding proteins from previous studies (Table S1). Utilizing one of the largest available datasets, which includes 175 renal fibrosis samples and 99 normal kidney samples, we conducted a differential gene expression analysis. This analysis revealed 1091 differentially expressed genes (Figure 1B and Table S2). By intersecting these with our candidate RNA-binding proteins, we identified 57 that were differentially expressed (Figure 1C and Table S3).
3.2. Establish and validate diagnostic models
Establishing a stable and reliable diagnostic model is crucial for early intervention in renal fibrosis patients. Through lasso regression analysis, we identified 16 key RNA-binding proteins (RBPs), including DOCK2, MYO1F, TNFAIP2, LDHD, AIM2, GPT2, FERMT3, FKBP11, ANK2, PLCB4, ANKRD36, DNAJC6, GNB5, DCDC2, MYH8, and COL6A3 (Figure 2A). Further logistic regression analysis narrowed these down to six critical features—ANKRD36, FKBP11, DCDC2, COL6A3, PLCB4, and GNB5—which were used to construct the diagnostic model (Figure 2B). The ROC curve demonstrated an AUC of 0.899, indicating strong diagnostic efficacy (Figure 2C), while the DCA curve confirmed a good safety margin and clinical value (Figure 2D). In external datasets, the model consistently achieved AUC values above 0.7, underscoring its stability (Figures 2E–G). Additionally, we combined three external validation datasets with our original experimental dataset to create a large GEO-meta dataset for further testing. This analysis yielded an AUC of 0.849 (Figure 2H), with the DCA curve further supporting the model’s reliability (Figure 2I). Notably, five of the six modeled genes—FKBP11 (the hub gene with higher coefficient), DCDC2, COL6A3, PLCB4, and GNB5—showed predictive power with AUC values above 0.5, suggesting their significant roles in renal fibrosis. Next, we will focus on functional research of the five genes mentioned above.
Figure 2.
Establishment of a diagnostic model for renal fibrosis and validation on multiple datasets. (A) LASSO regression analysis identifies important features in renal fibrosis samples. (B) Multi factor logistics regression analysis to construct a diagnostic model for renal fibrosis. (C) ROC curve shows AUC value exceeding 0.899 of the above diagnostic model in GSE76882 datasets. (D) The above diagnostic model of DCA curve has good stability and reliability in GSE76882 datasets. (E) ROC curve shows AUC value exceeding 0.792 of the above diagnostic model in GSE653326 datasets. (F) ROC curve shows AUC value exceeding 0.705 of the above diagnostic model in GSE22459 datasets. (G) ROC curve shows AUC value exceeding 0.892 of the above diagnostic model in GSE7392 datasets. (H) ROC curve shows AUC value exceeding 0.849 of the above diagnostic model in GEO-meta datasets. I. he above diagnostic model of DCA curve has good stability and reliability in GEO-meta datasets.
3.3. Molecular signal function analysis
To identify potential renal fibrosis signaling pathways, we performed differential expression analysis in a GEO-meta dataset and obtained 662 differentially expressed genes (Figures 1A and Table S4). GO analysis showed that the differential pathways were enriched in Cytokine related pathways, MHC protein complex, and other pathways (Figure 3A and Table S5). KEGG analysis showed that differential genes were enriched in Phagosome, Th1, and Th2 Cell differentiation pathways (Figure 3B and Table S6). METASCAPE analysis showed that differential genes were enriched in cell activation, inflammation response, and other pathways (Figure 3C). These results support the critical role of inflammatory pathways in renal fibrosis. Interestingly, GSEA analysis showed that the functions of FKBP11, DCDC2, COL6A3, PLCB4, and GNB5 were all enriched in immune-related pathways such as Cytokine signaling and Graft rejection, supporting the potential core functions of these three molecules in renal fibrosis (Figure 3E).
Figure 3.
Functional enrichment analysis. (A) GO Functional analysis of genes based on differentially expressed genes identified from GEO meta datasets. (B) KEGG Functional analysis of genes based on differentially expressed genes identified from GEO meta datasets. (C) Metascape Functional analysis of genes based on differentially expressed genes identified from GEO meta datasets. (D) GSEA analysis identifies the potential roles of the five key (FKBP11, DCDC2, COL6A3, PLCB4, GNB5) factors in renal fibrosis in the GEO-meta datasets.
3.4. Correlation analysis of immune cell infiltration
A large number of inflammatory immune cell aggregation and functional transformation are the core features of immune-related disease such as renal fibrosis. CIBERSORT analysis showed significant enrichment of T cell gamma delta and M1 macrophages after renal fibrosis. and decreased infiltration of NK cell and mast cell, which was consistent with previous cognition (Figure 4A,B). We conducted the above RBPs and immune cell correlation analysis, and the results showed that PLCB4 was positively correlated with M2 macrophage infiltration, GNB5 was negatively correlated with mast cell infiltration, FKBP11 and COL6A3 were significantly correlated with multiple immune cell infiltration, and DCDC2 was significantly correlated with neutrophil infiltration (Figure 4C). These results provide a direction for future research on the function of the above-mentioned genes.
Figure 4.
Immune cell related analysis. (A) Analysis of immune cell components using CIBESORT algorithm. (B) Differential analysis of immune cell components in GEO meta dataset. (C) Correlation analysis between immune cell components and the expression of the aforementioned five genes.
3.5. Single-celled sequencing analysis of gene expression profile
To identify the single-cell expression profiles of FKBP11, DCDC2, COL6A3, PLCB4 and GNB5 in renal fibrosis, we analyzed the single-cell dataset GSE269062 and identified a consensus of 9 cell subpopulations (Figure 5A,B). The expression profile analysis showed that FKBP11 was mainly expressed in Plasma cell, DCDC2A was mainly expressed in Epithelial cell and smooth muscle cell, and COL6A3 was mainly expressed in Fibroblast. PLCD4 and GNB5 are widely expressed in a variety of cells (Figure 5C,D). These results show the functional diversity of FKBP11, DCDC2, COL6A3, PLCB4, and GNB5.
Figure 5.
Identification of key RBPs in cell expression profiles based on single-cell sequencing. (A) Single cell dataset cell clustering, identifying a total of 21 subgroups. (B) Cell subpopulation annotation. (C) Analyze the expression levels of key RBPs in various cell subgroups. (D) Identify the expression distribution of key RBPs in various cell subpopulations.
3.6. Analysis of co-expressed genes
Next, we analyzed the co-expressed genes of FKBP11, DCDC2, COL6A3, PLCB4 and GNB5. In order to identify potential regulatory mechanisms, we found 1699 co-expressed genes of COL6A3, 248 co-expressed genes of DCDC2, 1059 co-expressed genes of FKBP11, 1084 co-expressed genes of GNB5, and 873 co-expressed genes of PLCB4 (Figure 6A and Table S7).
Figure 6.
Co-expression gene analysis. Identifying co-expressed genes of key RBPs, only displaying the top 25 positively and negatively correlated genes.
3.7. RNA binding domain analysis
Next, we predicted the RNA-binding domains of FKBP11, DCDC2, COL6A3, PLCB4, and GNB5 using the HydRA machine learning algorithm, and the results showed that the domain of FKBP11 may be located at 70-98 aa, the domain of PLCB4 is located at 892-918aa, and the domain of GNB5 is located at 169-210aa; though no significant RNA binding domains were identified in the DCDC2 and COL6A3 proteins, due to their high scores, DCDC2 and COL6A3 are also considered candidates for RNA binding proteins (Figure 7A and Table 1).
Figure 7.
RNA binding domain recognition. Using Hydra machine learning algorithm to identify the RNA recognition domain of key RNA binding proteins.
Table 1.
Hydra score to RBPs.
| Gene symbol | seqSVM_score | seqDNN_score | ProteinBERT_score | seqSVM_seqDNN_ProteinBERT_score |
|---|---|---|---|---|
| DCDC2 | 0.243117653 | 0.32079363 | 0.56873864 | 0.893139345 |
| COL6A3 | 0.444524509 | 0.9282898 | 0.1383371 | 0.881902066 |
| FKBP11 | 0.230450549 | 0.683325 | 0.21974936 | 0.82423609 |
| GNB5 | 0.178177853 | 0.94652075 | 0.51233387 | 0.934656238 |
| PLCB4 | 0.425274264 | 0.86373186 | 0.2391752 | 0.878517067 |
3.8. Prediction of RNA-binding target genes
We explored the RNAct database and identified potential downstream binding targets for the five genes (FKBP11, DCDC2, COL6A3, PLCB4, and GNB5) mentioned above (Figure 8A) [12]. These downstream targets, such as FOXD1, the predicted target of PLCB4, MAPK3, the target of GNB5, and TGIF1, the target of DCDC2, have all been confirmed to be closely related to the development of renal fibrosis, which can be used as a way to study the function of these genes in the future [13–15].
Figure 8.
RNA binding target prediction. Using RNAct database to identify downstream targets of RBP, only the first 50 were displayed.
3.9. UUO model validation
In the UUO model, a pronounced upregulation of FKBP11 expression was observed within renal fibrosis tissues (Figure 9A). Subsequent immunofluorescence staining of tissues exhibiting elevated and reduced Immunofluorescence showed a positive correlation between FKBP11 and macrophage infiltration (Figure 9B). These empirical findings are central to our investigation and substantiate the validity of our bioinformatics-derived predictions.
Figure 9.
FKBP11 Expression validation. (A) FKBP11 expression is elevated in renal fibrosis models. (B) FKBP11 expression was positively correlated with CD68 (macrophage infiltration).
4. Discussion
Our comprehensive study aimed to unravel the roles of RNA-binding proteins (RBPs) in the pathogenesis of post-transplant renal fibrosis, focusing on their potential as diagnostic biomarkers and therapeutic targets. We identified 57 differentially expressed RBPs from a pool of 1411 candidates, with five key RBPs—FKBP11, DCDC2, COL6A3, PLCB4, and GNB5—emerging as critical players in the fibrotic process. These findings provide new insights into the molecular mechanisms underlying renal fibrosis and highlight the significance of these RBPs in immune-related pathways. These RBPs are not only implicated in fibrosis but also play significant roles in broader inflammatory diseases. FKBP11 is involved in immunoregulation, influencing cytokine signaling that is crucial in both fibrosis and chronic inflammation [16,17]. DCDC2’s role in cellular stress responses suggests it may also modulate inflammatory processes, particularly under chronic conditions [18]. COL6A3 contributes to extracellular matrix remodeling, a key feature in both fibrosis and persistent inflammation [19,20]. GNB5 regulate immune cell signaling and balance pro- and anti-inflammatory responses, making them potential targets in treating inflammation-driven diseases [21].
The diagnostic model we developed, incorporating six RBPs, demonstrated robust diagnostic efficacy with an AUC of 0.899. This model’s performance across multiple external datasets further underscores its potential as a reliable tool for early detection of renal fibrosis. At present, there is still a lack of noninvasive diagnostic methods for renal fibrosis, and the development of new diagnostic markers is a hot research direction, and has shown the dawn of clinical application [22,23]. Our study provides directions for enriching the development of diagnostic markers for renal fibrosis. Given the progressive nature of renal fibrosis, early diagnosis is crucial for implementing timely interventions that could prevent irreversible damage and improve long-term graft outcomes. Our pathway enrichment analyses revealed that these RBPs are involved in several critical pathways, including cytokine signaling and inflammatory responses, which are well-known contributors to fibrotic processes. The strong association between these RBPs and immune-related pathways is consistent with the established understanding that immune cell infiltration and activation are pivotal in the progression of renal fibrosis. Notably, the GSEA analysis highlighted the enrichment of FKBP11, DCDC2, COL6A3, PLCB4, and GNB5 in pathways related to cytokine signaling and graft rejection, suggesting that these RBPs may play central roles in modulating immune responses during fibrosis. The single-cell RNA sequencing analysis provided a more granular view of the expression profiles of these RBPs across different renal cell populations. The identification of cell-specific expression patterns, such as FKBP11 in plasma cells, DCDC2 in epithelial and smooth muscle cells, and COL6A3 in fibroblasts, underscores the functional diversity of these RBPs. This cell-type specificity is likely to influence the fibrotic response in a context-dependent manner, with each RBP contributing to fibrosis through distinct cellular mechanisms. Our correlation analysis between RBPs and immune cell infiltration patterns further supports the involvement of these proteins in the immune regulation of fibrosis. The positive correlation between PLCB4 and M2 macrophage infiltration, for instance, suggests that PLCB4 might promote a pro-fibrotic macrophage phenotype, which is known to facilitate tissue remodeling and fibrosis [24]. Conversely, the negative correlation between GNB5 and mast cell infiltration may indicate a role for GNB5 in mitigating inflammatory responses, which could otherwise exacerbate fibrosis [25]. These findings are in line with previous studies that highlight the role of immune cells in the fibrotic process and suggest that RBPs could serve as novel therapeutic targets for modulating immune responses in renal fibrosis. Our prediction of RNA-binding domains using the HydRA machine learning algorithm provided further mechanistic insights. Although no significant RNA-binding domains were identified in DCDC2 and COL6A3, their high scores suggest that these proteins may interact with RNA in a non-canonical manner [26]. This opens up new avenues for exploring the regulatory functions of these RBPs in RNA metabolism and gene expression, particularly in the context of renal fibrosis. Additionally, the identification of potential downstream targets of these RBPs, such as FOXD1, MAPK3, and TGIF1, which are closely related to renal fibrosis, offers promising directions for future research. Finally, Within this model, we observed a pronounced upregulation of FKBP11 expression specifically within the renal fibrosis tissues. This observation is a critical piece of evidence that supports the role of FKBP11 in the fibrotic process. Furthermore, subsequent immunofluorescence staining of tissues with varying levels of FKBP11 expression revealed a marked escalation in CD68 expression in those tissues where FKBP11 levels were heightened. This correlation suggests a potential link between FKBP11 expression and the inflammatory response, as indicated by CD68, a marker associated with immune cell infiltration. These empirical findings are central to our investigation and substantiate the validity of our bioinformatics-derived predictions, thereby reinforcing the biological relevance of our computational analysis. The integration of these experimental results with our bioinformatics predictions provides a robust foundation for the further exploration of FKBP11’s role in renal fibrosis and its potential as a therapeutic target. Interestingly, previous studies have shown that FKBP11 is involved in JNK-driven inflammatory pathways and B cell function regulation [16,17]. This also provides a direction for future mechanism research.
Understanding how these RBPs influence the expression and activity of these targets could provide new strategies for therapeutic intervention, particularly in targeting the early stages of fibrosis before significant tissue damage occurs.
In conclusion, our study provides a comprehensive analysis of the roles of novel RBPs in post-transplant renal fibrosis, highlighting their potential as both diagnostic biomarkers and therapeutic targets. The integration of advanced bioinformatics and machine learning approaches allowed us to identify key RBPs involved in critical fibrotic pathways and immune regulation. Future studies should focus on validating these findings in larger cohorts and exploring the therapeutic potential of targeting these RBPs in vivo to mitigate renal fibrosis and improve transplant outcomes.
Supplementary Material
Correction Statement
This article has been corrected with minor changes. These changes do not impact the academic content of the article.
Funding Statement
This project was supported by General Project of Chongqing Natural Science Foundation (cstc2021jcyj-msxmX0968) and First Batch of Key Disciplines on Public Health in Chongqing.
Authors contributions
Jie Chen and Binghan Zhang performed the research and wrote the draft. Jie Chen performed the analysis and collected data. All authors contributed to the design and interpretation of the study as well as further drafts. Dongfang Liu is the guarantors.
Consent for publication
All authors agree to publish.
Disclosure statement
No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.
Data availability statement
Publicly available datasets were analyzed in this study. The original data can be obtained for https://www.ncbi.nlm.nih.gov/geo/.
References
- 1.Chen Y, Dai R, Cheng M, et al. Status and role of the ubiquitin-proteasome system in renal fibrosis. Biomed Pharmacother. 2024;178:117210. doi: 10.1016/j.biopha.2024.117210. [DOI] [PubMed] [Google Scholar]
- 2.Wan S, Wang S, He X, et al. Noninvasive diagnosis of interstitial fibrosis in chronic kidney disease: a systematic review and meta-analysis. Ren Fail. 2024;46(2):2367021. doi: 10.1080/0886022X.2024.2367021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Tao Y, Zhang Q, Wang H, et al. Alternative splicing and related RNA binding proteins in human health and disease. Signal Transduct Target Ther. 2024;9(1):26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Wen D, Xiao H, Gao Y, et al. N6-methyladenosine-modified SENP1, identified by IGF2BP3, is a novel molecular marker in acute myeloid leukemia and aggravates progression by activating AKT signal via de-SUMOylating HDAC2. Mol Cancer. 2024;23(1):116. doi: 10.1186/s12943-024-02013-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Zhao R, Fang X, Mai Z, et al. Transcriptome-wide identification of single-stranded RNA binding proteins. Chem Sci. 2023;14(15):4038–4047. doi: 10.1039/d3sc00957b. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Liepelt A, Naarmann-de Vries IS, Simons N, et al. Identification of RNA-binding proteins in macrophages by interactome capture. Mol Cell Proteomics. 2016;15(8):2699–2714. doi: 10.1074/mcp.M115.056564. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Correa BR, de Araujo PR, Qiao M, et al. Functional genomics analyses of RNA-binding proteins reveal the splicing regulator SNRPB as an oncogenic candidate in glioblastoma. Genome Biol. 2016;17(1):125. doi: 10.1186/s13059-016-0990-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Razpotnik R, Nassib P, Kunej T, et al. Identification of novel RNA binding proteins influencing circular RNA expression in hepatocellular carcinoma. Int J Mol Sci. 2021;22(14):7477. doi: 10.3390/ijms22147477. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Jin W, Brannan KW, Kapeli K, et al. HydRA: deep-learning models for predicting RNA-binding capacity from protein interaction association context and protein sequence. Mol Cell. 2023;83(14):2595–2611.e11. doi: 10.1016/j.molcel.2023.06.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Zhou Y, Zhou B, Pache L, et al. Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nat Commun. 2019;10(1):1523. doi: 10.1038/s41467-019-09234-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Subramanian A, Tamayo P, Mootha VK, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A. 2005;102(43):15545–15550. doi: 10.1073/pnas.0506580102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Lang B, Armaos A, Tartaglia GG.. RNAct: protein-RNA interaction predictions for model organisms with supporting experimental data. Nucleic Acids Res. 2019;47(D1):D601–D606. doi: 10.1093/nar/gky967. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ricard N, Scott RP, Booth CJ, et al. Endothelial ERK1/2 signaling maintains integrity of the quiescent endothelium. J Exp Med. 2019;216(8):1874–1890. doi: 10.1084/jem.20182151. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Gu X, Mallipattu SK, Guo Y, et al. The loss of Krüppel-like factor 15 in Foxd1+ stromal cells exacerbates kidney fibrosis. Kidney Int. 2017;92(5):1178–1193. doi: 10.1016/j.kint.2017.03.037. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ding H, Xu Y, Jiang N.. Upregulation of miR-101a suppresses chronic renal fibrosis by regulating KDM3A via Blockade of the YAP-TGF-β-Smad signaling pathway. Mol Ther Nucleic Acids. 2020;19:1276–1289. doi: 10.1016/j.omtn.2020.01.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Ruer-Laventie J, Simoni L, Schickel JN, et al. Overexpression of Fkbp11, a feature of lupus B cells, leads to B cell tolerance breakdown and initiates plasma cell differentiation. Immun Inflamm Dis. 2015;3(3):265–279. doi: 10.1002/iid3.65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Wang X, Cui X, Zhu C, et al. FKBP11 protects intestinal epithelial cells against inflammation‑induced apoptosis via the JNK‑caspase pathway in Crohn’s disease. Mol Med Rep. 2024;29(3):4428–4438. doi: 10.3892/mmr.2024.13172. [DOI] [Google Scholar]
- 18.Li LX, Chu JH, Chen XW, et al. Selenium ameliorates mercuric chloride-induced brain damage through activating BDNF/TrKB/PI3K/AKT and inhibiting NF-κB signaling pathways. J Inorg Biochem. 2022;229:111716. doi: 10.1016/j.jinorgbio.2022.111716. [DOI] [PubMed] [Google Scholar]
- 19.Gesta S, Guntur K, Majumdar ID, et al. Reduced expression of collagen VI alpha 3 (COL6A3) confers resistance to inflammation-induced MCP1 expression in adipocytes. Obesity. 2016;24(8):1695–1703. doi: 10.1002/oby.21565. [DOI] [PubMed] [Google Scholar]
- 20.Dankel SN, Grytten E, Bjune JI, et al. COL6A3 expression in adipose tissue cells is associated with levels of the homeobox transcription factor PRRX1. Sci Rep. 2020;10(1):20164. doi: 10.1038/s41598-020-77406-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Yang Y, Ding T, Xiao G, et al. Anti-inflammatory effects of allocryptopine via the target on the CX3CL1-CX3CR1 axis/GNB5/AKT/NF-κB/apoptosis in dextran sulfate-induced mice. Biomedicines. 2023;11(2):464. doi: 10.3390/biomedicines11020464. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Schmidt IM, Colona MR, Kestenbaum BR, et al. Cadherin-11, Sparc-related modular calcium binding protein-2, and Pigment epithelium-derived factor are promising non-invasive biomarkers of kidney fibrosis. Kidney Int. 2021;100(3):672–683. doi: 10.1016/j.kint.2021.04.037. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Shi J, Qin X, Sha H, et al. Identification of biomarkers for chronic renal fibrosis and their relationship with immune infiltration and cell death. Ren Fail. 2025;47(1):2449195. doi: 10.1080/0886022X.2024.2449195. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Tang PM, Nikolic-Paterson DJ, Lan HY.. Macrophages: versatile players in renal inflammation and fibrosis. Nat Rev Nephrol. 2019;15(3):144–158. doi: 10.1038/s41581-019-0110-2. [DOI] [PubMed] [Google Scholar]
- 25.Summers SA, Gan PY, Dewage L, et al. Mast cell activation and degranulation promotes renal fibrosis in experimental unilateral ureteric obstruction. Kidney Int. 2012;82(6):676–685. doi: 10.1038/ki.2012.211. [DOI] [PubMed] [Google Scholar]
- 26.Corley M, Burns MC, Yeo GW.. How RNA-binding proteins interact with RNA: molecules and mechanisms. Mol Cell. 2020;78(1):9–29. doi: 10.1016/j.molcel.2020.03.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Publicly available datasets were analyzed in this study. The original data can be obtained for https://www.ncbi.nlm.nih.gov/geo/.