Renal tubular senescence is a defining pathological feature of diabetic kidney disease (DKD) and a key driver of disease progression. While glucotoxic stress is recognized as a major contributor to tubular aging, the upstream regulatory mechanisms remain incompletely understood. Here, we investigated the role of C1q/TNF-related protein 1 (CTRP1) in DKD-associated tubular senescence and its mechanistic link to SGLT2-mediated glucose uptake. We conducted a retrospective observational case–control analysis of human renal biopsy specimens, complemented by mechanistic experiments in NRK-52E cells and a high-fat diet/streptozotocin/uninephrectomy (HFD–STZ–UNx) rat model. Renal cortical tissues from DKD patients and controls were analyzed via immunohistochemistry and Western blotting. CTRP1 expression increased progressively with DKD severity and was positively correlated with the senescence markers p21 (R = 0.498) and p16 (R = 0.494). Bioinformatic analysis of CTRP1-deficient mice revealed consistent SGLT2 downregulation, which was validated experimentally in NRK-52E cells, where CTRP1 overexpression increased SGLT2 expression and glucose uptake, particularly under hyperglycemic conditions. In HFD/STZ/UNx-induced DKD rats, dapagliflozin (0.1 mg/kg/day) improved metabolic abnormalities and renal dysfunction and attenuated tubular senescence. Dapagliflozin suppressed CTRP1–SGLT2 axis activation, reduced glucose uptake in renal tubular cells, and markedly decreased the proportion of CTRP1⁺SGLT2⁺ double-positive tubular cells. These findings support the presence of a previously unrecognized CTRP1–SGLT2 signaling axis associated with glucose-mediated tubular senescence in DKD and provide mechanistic insight into SGLT2 inhibitor-mediated renoprotection.

Human tissue collection

This retrospective observational case–control study included all eligible adults undergoing native kidney biopsy at a hospital during the study period. DKD tissues were obtained from 18 patients with biopsy-confirmed diabetic kidney disease, whereas control tissues were retrospectively identified from 5 non-DKD patients biopsied during the same period whose renal histology showed largely preserved architecture without evidence of DKD and whose serum creatinine and blood urea nitrogen levels were within the normal range. The cohort comprised 23 participants (18 DKD and 5 controls). Patients were classified according to KDIGO staging criteria, with representative samples from stages G1–G5 included. Those with other primary kidney diseases, active infection, malignancy, or insufficient tissue were excluded.

Sample size was determined by the number of eligible specimens available during the study period; no formal a priori sample size or power calculation was performed, and the analysis should therefore be considered exploratory. Baseline demographic and clinical characteristics are summarized in Supplementary Table 1. Clinical data, including age, eGFR, and UACR, were collected at biopsy. Tissue samples were fixed in 4% paraformaldehyde for immunohistochemistry or snap-frozen for Western blotting. For correlation analyses of CTRP1 with p16 and p21, cases with unavailable or unevaluable staining were excluded, yielding a final sample size of 18. Ethics approval (2022NL-090-01) was obtained, and written informed consent was obtained from all participants. Clinical trial number: not applicable.

Definition of variables

The primary exposure was renal cortical CTRP1 expression, quantified as the percentage of CTRP1-positive tubular cells in each biopsy core. Primary outcomes were tubular senescence markers, assessed as the percentages of p16-positive and p21-positive tubular cells. Secondary outcomes included renal function parameters (estimated glomerular filtration rate [eGFR], serum creatinine, and blood urea nitrogen), albuminuria (urinary albumin-to-creatinine ratio), and histological markers of renal cortical fibrosis. DKD cases were staged according to kidney disease: Improving Global Outcomes (KDIGO) criteria. Age was categorized as 18–35, 36–64, or ≥ 65 years, and eGFR was stratified as 30–89 or <30 mL/min/1.73 m² to reflect clinically meaningful categories of renal impairment.

Analyses were conducted using complete cases; samples with missing or unevaluable staining for a given marker were excluded from the corresponding analysis without imputation.

Immunohistochemistry and analysis

Paraffin-embedded sections (5 µm) were deparaffinized, rehydrated, treated with 3% H₂O₂, and subjected to antigen retrieval in citrate buffer (pH 6.0). After being blocked with 10% goat serum, the sections were incubated overnight at 4°C with antibodies against CTRP1, p21, or p16 (1:100), followed by incubation with HRP-conjugated secondary antibodies (1:500) for 1 h at room temperature and subsequent DAB development. Whole-slide images were acquired via a KFBIO scanner. Each whole-slide image contained one or more renal biopsy cores, and the percentage of cells positive for CTRP1, p21, or p16 in each biopsy core was quantified using QuPath software (version 0.6.0 , RRID:SCR_018257).

Western blotting

Tissues or cultured cells were lysed in RIPA buffer supplemented with protease/phosphatase inhibitors. Proteins (30–50 µg) were separated by 10% SDS‒PAGE, transferred to PVDF membranes, blocked with 5% nonfat milk, and incubated overnight with primary antibodies (1:1000) against CTRP1, SGLT2, p21, Collagen I, α-SMA, or GAPDH (1:5000). After HRP-conjugated secondary antibodies were added (1:5000, 1 h), the signals were detected via ECL and quantified via ImageJ (RRID:SCR_003070).

Antibodies and reagents

The following primary antibodies were used in this study: anti-CTRP1 (Abcam, Cat# ab25973, RRID: AB_448937), anti-SGLT2 (mouse monoclonal, Santa Cruz Biotechnology, Cat# sc-393350, RRID:AB_2814658), anti-p21 (rabbit monoclonal, Cell Signaling Technology, Cat# 2947S, RRID:AB_823586), anti-p16INK4a/CDKN2A (rabbit monoclonal, ABclonal, Cat# A11651, RRID:AB_2861619), anti-COL1A1 (rabbit monoclonal, Cell Signaling Technology, Cat# 91144S, RRID:AB_2800169), anti-α-SMA (rabbit monoclonal, Cell Signaling Technology, Cat# 19245S, RRID:AB_2734735), and anti-GAPDH (rabbit monoclonal, ABclonal, Cat# A19056, RRID:AB_2862549).

Bioinformatics analysis

Transcriptomic datasets of renal tubular epithelial cells from CTRP1-knockout and control mice were obtained from the Gene Expression Omnibus (GEO) database (GSE127332) [22] to assess glucose transporter expression, with a focus on SGLT2. Time-series transcriptomic data over 48 hours (GSE217853) [23] were further used to quantify CTRP1 and SGLT2 mRNA levels. A generalized additive mixed-effects model was applied to the time-series data to characterize their expression trajectories and evaluate parallel changes over the 48-hour period.

Cell culture and lentiviral transduction

NRK-52E cells (RRID:CVCL_0468) were cultured in DMEM supplemented with 10% FBS and 1% penicillin–streptomycin.  The cells were transduced with CTRP1-overexpressing (LV-CTRP1-OE) or control (LV-CTRP1-NC) lentivirus; GFP positivity confirmed transduction. After transduction, the cells were cultured under normal-glucose (5.5 mM) or high-glucose (30 mM) conditions for 48 h. These conditions were selected with reference to published in vitro models of diabetic hyperglycemic stress in renal tubular epithelial cells^{[24, 25]}.

Glucose uptake assay

Glucose uptake was measured via a Glucose Uptake Assay Kit-Blue (Dojindo, UP01)^[26]. The cells were washed and incubated in glucose-free DMEM (15 min) and then in probe solution (1:500, 15 min). The mean fluorescence intensity (MFI) was measured via flow cytometry.

Animal model

Male Sprague–Dawley rats (RRID:RGD_70508, 8 weeks old, 200-250 g at baseline) were fed a high-fat diet (D12462), subjected to unilateral nephrectomy, and injected with streptozotocin (50 mg/kg) to establish DKD. The HFD/STZ/UNx protocol was selected because it integrates insulin resistance, hyperglycemia, and nephron loss, thereby reproducibly accelerating DKD progression and inducing metabolic and renal abnormalities that resemble progressive DKD^{[27, 28]}. Hyperglycemic/proteinuric rats were randomized (n=5/group) into the following groups: Sham, DKD, DAPA, and DKD + DAPA. The experimental unit was a single animal. Dapagliflozin (0.1 mg/kg/day) or saline was administered orally for 12 weeks. The dapagliflozin dose and treatment duration were selected with reference to previous experimental studies and adapted to the present DKD model^[29]. No unexpected animal deaths or exclusions occurred during the study. All animal experiments were approved by the Animal Ethics Committee of Nanjing University of Chinese Medicine and carried out in strict accordance with the institutional guidelines for the care and use of laboratory animals (Approval No. 202308A014).

Metabolic and renal function assessments

After 12 weeks, glucose tolerance tests (GTTs) and insulin tolerance tests (ITTs) were performed. Serum creatinine, blood urea nitrogen, total cholesterol, and triglycerides were measured.

Histopathology

Sections were stained with PAS, HE, Masson's trichrome, and SA-β-gal. Transmission electron microscopy (TEM) was used to examine ultrastructural changes (Hitachi HT7800, ×7000). Histological image quantification was performed by investigators blinded to group allocation.

Dual immunofluorescence and colocalization analysis

The sections were incubated with primary antibodies against CTRP1 (rabbit, 1:100) and SGLT2 (mouse, 1:100), followed by Alexa Fluor-conjugated secondary antibodies and DAPI staining. Images were acquired (Pannoramic SCAN II) and analyzed (v0.6.0, RRID:SCR_018257) for colocalization and MFI. Immunofluorescence image quantification was performed by investigators blinded to group allocation.

Flow cytometry

Renal tubular cells were isolated via enzymatic digestion. The cells were incubated with a glucose uptake probe, fixed, permeabilized, and stained with anti-CTRP1 and anti-SGLT2 antibodies, followed by incubation with fluorescent secondary antibodies. The signals were analyzed by flow cytometry.

Statistical analysis

Continuous data are expressed as the mean ± SEM. Data distribution was assessed using Shapiro–Wilk normality tests, and homogeneity of variances was evaluated using Brown–Forsythe and Bartlett tests. When appropriate, right‑skewed variables were log10‑transformed to better meet model assumptions. In particular, serum creatinine, blood urea nitrogen, and serum total cholesterol and triglyceride levels were analysed on log10‑transformed values, whereas the corresponding figures display the mean ± SEM of the untransformed data.

For two‑group comparisons, unpaired Student’s t tests were used when normality and equal variance assumptions were reasonably satisfied. For comparisons among more than two groups, one‑way ANOVA followed by Tukey’s post hoc test was applied under the same assumptions. Glucose and insulin tolerance curves were analysed by two‑way repeated‑measures ANOVA (factors: group and time), and the corresponding areas under the curve were compared using one‑way ANOVA. Pearson correlation was used to assess associations between continuous variables. A two‑sided P value < 0.05 was considered statistically significant.

Given the limited sample size of the human biopsy cohort, multivariable regression modelling was not performed, and we focused on unadjusted correlations between CTRP1, senescence markers, and clinical variables.