Expression Characteristics of SERINC1 and TDO2 in Diabetic Kidney Disease: Historical Candidate Reappraisal, Cross-Cohort Evaluation, Single-Cell Analysis, and a Mouse Model Study
- 1Division of Nephrology, Renmin Hospital of Wuhan University, Wuhan 430060, China
- 2Nephrology and Urology Research Institute, Wuhan University, Wuhan 430060, China
Abstract
Evaluating candidate molecules in diabetic kidney disease (DKD) requires evidence for both the reproducibility of expression changes and their cellular origin. Using the historically fixed SERINC1/TDO2 panel, we reappraised how these candidates were derived, evaluated both genes in human glomerular transcriptomes and kidney-cortex single-cell data, and examined their relationship with renal injury in a mouse model. The historical WGCNA, LASSO, and SVM-RFE-labelled frozen lists intersected at SERINC1 and TDO2, although the legacy RFE implementation could not be confirmed. In the current 83-sample analysis, 584 differentially expressed genes were detected; intersecting the cross-cohort direction-consistent differentially expressed genes with the disease-associated WGCNA module retained 26 candidates that did not include the two targets, whereas machine learning applied to a topology-derived subset of these candidates identified DUSP1. TDO2 was increased in both cohorts (Hedges’ g = 2.07 and 0.714; the latter Holm-adjusted p = 0.0182), whereas the SERINC1 reduction was supported only in the discovery cohort (g = −1.90) and not in GSE96804 (g = −0.252; Holm-adjusted p = 0.348). Single-cell analysis showed widespread SERINC1 distribution without donor-level significance after Benjamini–Hochberg correction, and TDO2 was detected in only 50 of 37,331 cells. In streptozotocin-induced mice, SERINC1 decreased and TDO2 increased alongside renal pathology and podocyte ultrastructural changes. Both genes remain candidates for further study, but clinical diagnostic value and shared pathogenic mechanisms are not yet established.
Related
- Clinical Evaluation of Biased μ-Opioid Receptor Agonists: A Narrative Comparative Review from Oliceridine to Tegileridine and Mechanistic Reappraisal.
- Identification of candidate selenium-associated transcriptomic biomarkers in chronic obstructive pulmonary disease: an integrated bioinformatics and expression validation study
- A comparative study for the isolation of rabbit phenotypically defined candidate HSPCs based on CD45 or CD34 marker expression
- Association of NCCRP1 expression with T category and in vitro growth-related characteristics of head and neck squamous cell carcinoma FaDu cells
- Cross-lineage anomaly detection of microRNA expression: a framework for deriving exploratory candidate biomarkers in sarcoma.
BibTeX
@article{cui2026expression,
title = {{Expression Characteristics of SERINC1 and TDO2 in Diabetic Kidney Disease: Historical Candidate Reappraisal, Cross-Cohort Evaluation, Single-Cell Analysis, and a Mouse Model Study}},
author = {Xiaofei Cui and Yijun Ren and Guohua Ding and Sen Chen},
journal = {International Journal of Molecular Sciences},
year = {2026},
month = oct,
doi = {10.3390/ijms27219418},
url = {https://www.mdpi.com/1422-0067/27/21/9418},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Expression Characteristics of SERINC1 and TDO2 in Diabetic Kidney Disease: Historical Candidate Reappraisal, Cross-Cohort Evaluation, Single-Cell Analysis, and a Mouse Model Study
AU - Xiaofei Cui
AU - Yijun Ren
AU - Guohua Ding
AU - Sen Chen
PY - 2026
DA - 2026/10/22
JO - International Journal of Molecular Sciences
DO - 10.3390/ijms27219418
UR - https://www.mdpi.com/1422-0067/27/21/9418
AB - Evaluating candidate molecules in diabetic kidney disease (DKD) requires evidence for both the reproducibility of expression changes and their cellular origin. Using the historically fixed SERINC1/TDO2 panel, we reappraised how these candidates were derived, evaluated both genes in human glomerular transcriptomes and kidney-cortex single-cell data, and examined their relationship with renal injury in a mouse model. The historical WGCNA, LASSO, and SVM-RFE-labelled frozen lists intersected at SERINC1 and TDO2, although the legacy RFE implementation could not be confirmed. In the current 83-sample analysis, 584 differentially expressed genes were detected; intersecting the cross-cohort direction-consistent differentially expressed genes with the disease-associated WGCNA module retained 26 candidates that did not include the two targets, whereas machine learning applied to a topology-derived subset of these candidates identified DUSP1. TDO2 was increased in both cohorts (Hedges’ g = 2.07 and 0.714; the latter Holm-adjusted p = 0.0182), whereas the SERINC1 reduction was supported only in the discovery cohort (g = −1.90) and not in GSE96804 (g = −0.252; Holm-adjusted p = 0.348). Single-cell analysis showed widespread SERINC1 distribution without donor-level significance after Benjamini–Hochberg correction, and TDO2 was detected in only 50 of 37,331 cells. In streptozotocin-induced mice, SERINC1 decreased and TDO2 increased alongside renal pathology and podocyte ultrastructural changes. Both genes remain candidates for further study, but clinical diagnostic value and shared pathogenic mechanisms are not yet established.
SN - 1422-0067
ER -