Abstract Diabetic kidney disease (DKD) has become a major cause of chronic kidney disease and end-stage renal disease. Numerous studies have indicated that exosomal miRNAs play a crucial role in the pathophysiological processes of DKD. We screened differentially expressed miRNAs in urinary exosomes from patients with DKD using the GEO database and performed PCR validation both in patients who were pathologically diagnosed and clinically diagnosed. We assessed the clinical diagnostic value of urinary exosomal miRNAs using ROC curves, analyzed the correlation between miRNAs and clinical indicators, predicted the target genes using the miRTarBase database and explored the potential signaling pathways involved through functional enrichment analysis. Screening results showed that the expression of miR-136-5p was significantly elevated in the DKD group compared to the DM group, with significance maintained in validation samples. The sensitivity and specificity of miR-136-5p for diagnosing DKD were 72.2% and 78.4%, respectively, with an area under the curve of 0.722. Additionally, the expression of miR-136-5p was positively correlated with UACR, urea nitrogen, cystatin C, chronic kidney disease progression risk stratification indicators, and negatively correlated with eGFR. 152 target genes of miR-136-5p were predicted and enriched in 25 pathways, including insulin secretion, cAMP signaling pathway, and sphingolipid signaling pathway. Supplementary Information The online version contains supplementary material available at 10.1038/s41598-025-06810-3. Keywords: Urinary exosomes, miR-136-5p, Diabetic kidney disease, Biomarkers, Diagnostic value Subject terms: Biomarkers, Nephrology Introduction Diabetic kidney disease (DKD) is a common severe microvascular complication of diabetes and has become a major cause of chronic kidney disease and end-stage renal disease^[46]1,[47]2, as well as one of the leading causes of death among diabetic patients. Epidemiological data suggest that approximately 30–40% of diabetic patients would develop DKD^[48]3. Although renal biopsy remains the gold standard for DKD diagnosis, its invasive nature carries risks of complications (e.g., infection, arteriovenous fistula, and hemorrhage), and serial monitoring is impractical due to procedural constraints. Currently, the clinical diagnosis of DKD is primarily based on the history of diabetes, concurrent diabetic retinopathy, and the presence of persistent proteinuria and progressive decline in renal function, which are exclusionary diagnoses that lack effective and sensitive early diagnostic markers. The early predictive ability of urinary microalbumin is limited; transient increases in urinary proteins can also occur under conditions such as fever, infection, heart failure, and intense exercise, which limits its clinical utility. Additionally, the concept of non-proteinuric DKD has also been widely validated^[49]4,[50]5. Although dietary control and pharmacotherapy for DKD somewhat improve patient conditions, they cannot completely stop the progression of DKD to end-stage renal disease. Therefore, enhancing early screening, diagnosis, and intervention is crucial for reducing cardiovascular events, improving patient survival rates, and enhancing quality of life. In recent years, multiple studies have shown that exosome contents are abundant and participate in the pathophysiological processes of various kidney diseases. Exosomes, with a diameter of about 40-160 nm, are membrane vesicles formed through processes such as "endocytosis-fusion-excretion" by active cells and have a phospholipid bilayer structure^[51]6. Exosomes are naturally secreted into biofluids, including blood, urine, saliva, cerebrospinal fluid, and other physiological fluids (e.g., amniotic fluid, semen, pleural/ascitic effusions). Exosomes, carrying a diverse array of biologically active molecules including proteins/peptides, nucleic acids (DNA, mRNA, miRNA, lncRNA), and lipids^[52]7, play pivotal roles in multiple pathophysiological processes such as cellular communication^[53]8, migration^[54]9, angiogenesis^[55]10, inflammatory responses^[56]11, and tumor growth^[57]12. MiRNAs are a class of endogenous non-coding single-stranded small molecular RNAs widely present in eukaryotes, primarily regulating post-transcriptional gene expression by binding to target mRNAs. Increasing evidence indicates that exosomal miRNAs play a crucial role in the development of human kidney diseases. Viñas et al.^[58]13 found that injection miR-486-5p mimic could prevent ischemic damage in mice with acute kidney injury by reduced apoptosis, inhibition of phosphatase and tensin homolog genes, and enhanced phosphorylation of protein kinase B. MiR-486-5p significantly downregulated the expression of genes related to acute kidney injury in proximal tubules (KIM-1, NGAL, Krt20, Sox9, clusterin) and endothelial cell ischemic damage (Cxcl11, Cxcl10), suggesting miR-486-5p may have therapeutic potential in ischemic acute kidney injury. Lv et al.^[59]11 found that exosomal miR-19b-3p derived from renal tubular epithelial cells in kidney injury promotes activation of M1 macrophages by targeting SOCS-1, participating in tubulointerstitial inflammation. Eissa et al.^[60]14 reported that compared to the healthy control group, type 2 DKD patients (DM with UACR ≥ 30 mg/g) had increased expressions of urinary exosomal miR-30a, miR-342, and miR-133b, and these miRNAs were also upregulated in some DM cases (UACR < 30 mg/g), suggesting that follow-up of these cases might help in early detection of DKD before microalbuminuria occurs. Urinary exosomal miRNAs, protected by the exosome membrane, are resistant to degradation by RNase and are stably enriched. Compared to blood samples, urinary exosomes are non-invasively obtained and easier to handle and store. Originating from the cells of the urinary reproductive system, urinary exosomes could directly reflect the pathophysiological state of the kidneys, making them an ideal source for diagnosing kidney diseases. Due to the high stability and specificity, urinary exosomal miRNAs are being extensively studied as emerging biomarkers for diagnosing kidney diseases. In this study, differentially expressed miRNAs were initially identified through GEO database ([61]http://www.ncbi.nlm.nih.gov/geo, [62]GSE51674), followed by systematic literature curation. The candidate miRNAs were subsequently validated in clinical cohorts. Research indicates that the expression of urinary exosomal miR-136-5p is significantly elevated in the diabetic kidney disease (DKD) group (1.51[0.57, 3.13] vs. 1.00[0.73, 1.03], P = 0.017). Demonstrating good diagnostic performance for DKD. Furthermore, the dysregulation of urinary exosomal miR-136-5p is closely associated with indicators of renal function damage, urine protein levels, and risk stratification markers for the progression of chronic kidney disease. Additionally, we hypothesize that urinary exosomal miR-136-5p may be involved in mediating renal pathological damage in DKD through its interactions with target genes and regulation of multiple signaling pathways, including insulin secretion, the cAMP signaling pathway, the AGE-RAGE signaling pathway, the sphingolipid signaling pathway, and the NF-κB signaling pathway. Therefore, we hypothesize that urinary exosomal miR-136-5p may serve as a novel non-invasive biomarker for DKD, with potential mechanistic links to renal injury through modulation of insulin signaling and inflammatory pathways risk stratification markers. Materials and methods Screening of differentially expressed miRNAs in DKD Preliminary literature review identified dataset [63]GSE51674 for miRNA expression profiling. Differentially expressed miRNAs between DKD and controls were identified using the GEO2R online tool. Adjusted P values < 0.01 and differential expression fold change |logFC|≥ 2 were set as the cutoff for differentially expressed miRNAs. Study subjects This study included 64 patients hospitalized in the Department of Nephrology and Endocrinology at the First Affiliated Hospital of Shandong First Medical University from December 2020 to June 2022. Inclusion criteria followed the 2020 ADA Standards of Medical Care in Diabetes^[64]15. According to the "Microvascular Complications and Foot Care" section of the 2020 ADA Standards^[65]16–[66]19, DKD patients met the same criteria as the DM group and additionally had a urinary albumin/creatinine ratio (UACR) > 300 mg/g or confirmed by renal biopsy. Exclusion criteria included refusal to participate, patients with acute diabetic complications or other critical illnesses, type 1 diabetes or special types of diabetes, recent infections, concurrent urogenital diseases, malignancies, immune-related diseases, transplant recipients, recent cerebrovascular diseases, and those whose laboratory tests (ANA/dsDNA, ENA/body fluid immunity, anti-PLA2R antibodies, anti-glomerular basement membrane antibodies, ANCA, complement C3, C4, urine light chains κ, λ, oncology markers, hepatitis B, hepatitis C, HIV, syphilis infections markers, thyroid function, immunofixation electrophoresis) or renal biopsy excluded other primary or secondary glomerular diseases and systemic diseases. Patient urine samples were collected and basic information registered. The study protocol was approved by the Ethics Committee of the First Affiliated Hospital of Shandong First Medical University and registered clinically (Registration No. [67]NCT06123871). We confirm that all methods were performed in accordance with the relevant guidelines and regulations. Sample processing and urinary exosome extraction Morning urine samples of 50 mL from eligible patients were collected in centrifuge tubes within 2 h. Urine samples were centrifuged at 3000 × g at 4 °C for 10 min, discarding the precipitate to remove the impurities and dead cells. The supernatant was transferred to sterilized RNase-Free centrifuge tubes and further centrifuged at 13,000 × g at 4 °C for 2 min to remove cellular debris. The supernatant was then ultracentrifuged at 200,000 × g at 4 °C for 120 min(Thermo, Sorvall WX 80 + , USA). The pellet was resuspended in PBS (pH 7.2), ultracentrifuged again at 200,000 × g at 4 °C for 60 min, and the final pellet, containing exosomes, was transferred to a new RNase-Free EP tube and stored at -80 °C. Identification of urinary exosomes Transmission electron microscopy observation of exosome structure Exosomes were fixed in 1% glutaraldehyde, and 10 µL of exosome solution was applied to a formvar-coated copper grid, left to stand at room temperature for 10 min, and excess liquid was carefully removed with filter paper. Samples were stained with 2% uranyl acetate at room temperature for 5 min and air-dried. Exosomes were observed under a transmission electron microscope at 80 kV. Nanoparticle Tracking Analysis of Exosome Size Exosomes were resuspended in PBS buffer and analyzed for particle size using a Flow NanoAnalyzer (NanoFCM, China). Data were analyzed using NanoFCM software (NanoFCM Profession V1.0). Western blot identification of exosome marker proteins RIPA lysis buffer (Beyotime Institute of Biotechnology, China) and Protease Inhibitor Cocktail (APExBIO, USA) were mixed at a 100:1 ratio and added to the extracted exosomes for lysis. Exosomal proteins were separated using 10% SDS-PAGE (Yeasen Biotechnology, China) and transferred onto a PVDF membrane (Beijing Land Bridge Technology), which was blocked with 5% skim milk at room temperature for 1 h. Membranes were washed and incubated overnight at 4 °C with primary antibodies against CD63, TSG101, and the negative control Calnexin (Abcam, ab275018, USA) (diluted 1:1000). After washing three times with TBST buffer (Monad, China) for 5 min each, membranes were incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG secondary antibody (Beijing Zhongshan Jinqiao Biotechnology, China) (diluted 1:5000) at room temperature for 1 h. Membranes were washed three times with TBST for 10 min each, and proteins were detected using the ChemiDoc Imaging System (BIO-RAD, USA). MiRNA extraction, reverse transcription, and quantitative real-time PCR analyses (qRT-PCR) Total RNA was extracted and purified using the Exosomal RNA Isolation Kit (Norgen Biotek, Canada) according to the manufacturer’s instructions. RNA was reverse transcribed to cDNA using the Hifair® III 1st Strand cDNA Synthesis Kit (gDNA digester plus) (Yeasen Biotechnology) and Gene Specific Primers (RIBIOBIO, China). Primers for miR-136-3p, miR-136-5p, miR-454-5p, and miR-39-3p were synthesized by RIBIOBIO, China. Gene expression was determined using the SYBR Green method according to the manufacturer’s instructions. Relative gene expression was calculated using the 2^-ΔΔCt method. Relative quantification of target miRNAs was based on the expression of the external reference gene, Caenorhabditis elegans miR-39-3p (miDETECT™ miRNA External Control, RIBIOBIO, China). Statistical analysis Statistical analyses were performed using SPSS® version 25.0 software (IBM® SPSS® Statistics, Armonk, New York), and graphs were generated using GraphPad Prism (Version 9.5.0). Data normality was assessed by the Shapiro–Wilk test. Continuous variables that followed a normal distribution were analyzed using the two-sample independent t-test and represented as mean ± standard deviation; non-normally distributed variables were analyzed using the Mann–Whitney U test and represented as median and interquartile ranges. Statistical significance was defined as P < 0.05. Diagnostic accuracy of study parameters was assessed using the receiver operating characteristic curve (ROC). Spearman rank correlation analysis was used to evaluate the relationship between two variables. Differentially expressed miRNAs were predicted using the MiRTarBase ([68]https://mirtarbase.cuhk.edu.cn/) database, and target genes were subjected to KEGG^[69]17–[70]19 enrichment analysis using the clusterProfiler package (version 3.14.3) in R software. The Venn Diagram was constructed using SRplot ([71]https://bioinformatics.com.cn/). Results Differentially expressed exosomal miRNAs in DKD We analyzed differentially expressed miRNAs between DKD patients, healthy controls, and patients with membranous nephropathy using the GEO database ([72]GSE51674), identifying an overlap of 13 differentially expressed miRNAs. Further literature review led to the selection of three miRNAs—miR-136-3p, miR-136-5p, and miR-454-5p as potential biomarkers for subsequent validation (Fig. [73]1; Supplementary Fig. 1). Fig. 1. Fig. 1 [74]Open in a new tab GEO database predicts differentially expressed miRNAs. ([75]GSE51674, analysed by GEO2R). Adjusted P values < 0.01, differential expression fold change |logFC|≥ 2. Identification of urinary exosomes We successfully isolated urinary exosomes from DKD and T2DM patients and verified their identity at three levels. Under the transmission electron microscope, exosomes displayed a hemispherical or cup-shaped vesicular structure with one side indented (Fig. [76]2a). Nanoparticle tracking analysis revealed that the diameters of the isolated particles predominantly ranged between 50 and 150 nm, consistent with the size of exosomes (Fig. [77]2b). Further analysis by Western blot confirmed the successful expression of exosomal marker proteins CD63 and TSG101 in the extracted vesicles, while the negative control protein calnexin was only expressed in cell lysates from the podocytes (Fig. [78]2c). Fig. 2. [79]Fig. 2 [80]Open in a new tab Characterization of exosomes. (a)Transmission electron microscope of exosomes. The arrow indicates the exosome. magnification: 100,000. (b) Nanoparticle tracking analysis of exosomes. (c) Western blot. Calnexin, TSG101 and CD63 levels in exosomes and podocyte lysis. Calnexin protein was used as a negative control. Participants This study consisted of 41 patients with type 2 diabetes mellitus (T2DM) and 23 patients with diabetic kidney disease (DKD). The cohort was divided into a screening cohort (DM/ DKD -A group) and a validation cohort (DM/ DKD -B group). In the screening cohort, four DKD patients (DKD -A group) were selected based on renal biopsy results and four T2DM patients (DM-A group) was matched with them. The validation cohort further included 19 DKD patients (DKD -B group) and 37 T2DM patients (DM-B group). Clinical characteristics of patients in screening cohort In the screening cohort, there were no statistically significant differences between the DKD and DM groups in terms of age, duration of diabetes, systolic blood pressure, diastolic blood pressure, glycated hemoglobin, fasting blood glucose, serum creatinine, urea nitrogen, cystatin C, serum albumin, total serum protein, cholesterol, triglycerides, and low-density lipoprotein. The DKD -A group had significantly lower estimated glomerular filtration rate (eGFR) (12.62 [4.78, 51.82] vs. 111.51 [183.52, 113.97] ml/min/1.73m2, P = 0.034) and significantly higher UACR (6083.37 [4059.64, 9872.04] vs. 19.53 [0.00, 19.53] mg/g, P = 0.034) compared to the DM-A group. Baseline data and clinical characteristics of the patients are shown in Table [81]1. Table 1. Clinical parameters of screening cohort patients (DM-A and DKD-A). Variable DM-A(n = 4) DKD-A (n = 4) P-value Age 64.00 ± 12.73 53.50 ± 5.20 0.177 Diabetes duration (years) 10.00 (3.25, 10.00) 6.5 (3.00, 10.00) 0.739 SBP (mmHg) 136.50 ± 16.68 152.00 ± 34.94 0.454 DBP (mmHg) 84.00 ± 10.23 83.00 ± 22.41 0.938 HbA1C (%) 10.78 ± 3.46 6.63 ± 1.37 0.112 FBG (mmol/L) 6.81 ± 1.42 5.71 ± 2.69 0.495 eGFR(ml/min/1.73 m^2) 111.51 (183.52, 113.97) 12.62 (4.78, 51.82) 0.034* Serum creatinine (μmol/L) 85.333 ± 40.53 584.50 ± 479.10 0.125 BUN(mmol/L) 6.83 ± 4.15 19.40 ± 10.27 0.107 Cys C (mg/L) 0.97 (0.87, 10.97) 3.89 (1.48, 4.356) 0.289 Alb (g/L) 35.90 (35.90, 35.90) 30.10 (20.50, 38.05) 0.285 total protein (g/L) 61.57 ± 4.63 50.78 ± 9.57 0.137 TC (mmol/L) 4.63 ± 1.46 5.68 ± 1.93 0.450 TG (mmol/L) 1.10 ± 0.66 1.47 ± 0.29 0.420 LDL-C (mmol/L) 2.86 (2.05, 3.90) 2.93 (2.88, 2.93) 0.480 UACR (mg/g) 19.53 (0.00, 19.53) 6083.37 (4059.64, 9872.04) 0.034* [82]Open in a new tab Continuous variable results in normal distribution are expressed as the means ± SD; continuous variable results in nonnormal distribution are expressed as median (interquartile range). Discontinuous variable results are expressed as n (%). SBP, systolic blood pressure; DBP, Diastole blood pressure; %HbA1c, glycated haemoglobin A1C; FBG, fasting blood glucose; BUN, blood urea nitrogen; Cys C, Cystatin C; eGFR, estimated glomerular filtration; Alb, albumin; TC, total cholesterol; TG, Triglyceride; LDL, low-density lipoprotein cholesterol; UACR, urine albumin creatine ratio. *Significant differences against DM-A and DKD-A group are indicated by * (P < 0.05). qRT-PCR validation of differentially expressed urinary exosomal miRNAs in screening cohort qRT-PCR results showed that in the screening cohort, urinary exosomal miR-136-5p expression was 2.21-fold higher in the DKD-A group compared to DM-A controls (2.21[2.12,2.73]vs.1.00[0.71,1.21], P = 0.021), indicating a statistically significant difference (P < 0.05) (Fig. [83]3). Fig. 3. [84]Fig. 3 [85]Open in a new tab Differential expression of urinary exosomal microRNAs were verified by qRT-PCR in screening cohort. *Significant differences against DM-A and DKD-A group are indicated by * (P < 0.05). Clinical characteristics of patients in validation cohort In the validation cohort, there were no statistically significant differences between the DKD -B and DM-B groups in terms of gender, systolic blood pressure, diastolic blood pressure, glycated hemoglobin, fasting blood glucose, fasting insulin, total cholesterol, triglycerides, and low-density lipoprotein cholesterol. Patients in the DKD -B group were older on average (60.53 ± 11.97 vs. 53.05 ± 11.96 years, P = 0.031) and had a longer history of diabetes (14 [10, 20] vs. 5 [0.75, 14.50] years, P = 0.026). They also had higher levels of fasting C-peptide (3.09 [1.47, 4.49] vs. 2.00 [1.32, 2.62] ng/ml, P = 0.021), Serum creatinine (85[77, 162] vs. 69.5[56.75, 78.75] μmol/L, P < 0.001), urea nitrogen (7.50 [6.12, 12.80] vs. 5.15 [4.30, 6.10] mmol/L, P < 0.001), cystatin C (1.10 [0.95, 1.99] vs. 0.75 [0.67, 0.82] mg/L, P < 0.001), and UACR (1080.77 [593.39, 2027.52] vs. 5.78 [1.60, 11.86] mg/g, P < 0.001) compared to the DM-B group. The eGFR was significantly lower (67.23 ± 32.22 vs. 107.16 ± 9.03 ml/min/1.73m2, P < 0.001), and serum albumin levels were also lower (39.75 ± 4.06 vs. 43.73 ± 4.42 g/L, P = 0.002) in the DKD -B group compared to the DM-B group. Baseline data and clinical characteristics of the patients are shown in Table [86]2. Table 2. Clinical parameters of validation cohort patients (DM-B and DKD-B). Variable DM-B (n = 37) DKD-B (n = 19) P-value Age 53.05 ± 11.96 60.53 ± 11.97 0.031* Male, n (%) 29 (78.40) 12 (63.20) 0.223 Diabetes duration (years) 5 (0.75, 14.50) 14 (10, 20) 0.026* SBP (mmHg) 135.35 ± 14.39 147.16 ± 24.95 0.068 DBP (mmHg) 81.81 ± 11.42 83.79 ± 12.12 0.550 HbA1C (%) 8.30 ± 1.59 8.92 ± 2.40 0.253 FBG (mmol/L) 7.44 (6.57, 8.61) 7.60 (6.49, 11.28) 0.516 Fasting C-peptide (ng/mL) 2.00 (1.32, 2.62) 3.09 (1.47, 4.49) 0.021* Fasting insulin (UIu/mL) 8.16 (3.77, 10.77) 8.81 (4.16, 17.54) 0.451 Serum creatinine (μmol/L) 69.50 (58.25, 79.00) 85 (77, 162) < 0.001** BUN(mmol/L) 5.15 (4.30, 6.10) 7.50 (6.12, 12.80) < 0.001** Cys C (mg/L) 0.75 (0.67, 0.82) 1.10 (0.95, 1.99) < 0.001** eGFR (ml/min/1.73m2) 107.16 ± 9.03 67.23 ± 32.22 < 0.001** Alb (g/L) 43.73 ± 4.42 39.75 ± 4.06 0.002* TC (mmol/L) 4.79 (4.09, 5.19) 4.78 (4.04, 5.84) 0.359 TG (mmol/L) 1.51 (1.15, 2.19) 1.62 (1.23, 2.08) 0.883 LDL-C (mmol/L) 2.89 (2.21, 3.19) 2.82 (2.28, 3.38) 0.716 [87]Open in a new tab Continuous variable results in normal distribution are expressed as the means ± SD; continuous variable results in nonnormal distribution are expressed as median (interquartile range). Discontinuous variable results are expressed as n (%). SBP, systolic blood pressure; DBP, Diastole blood pressure; %HbA1c, glycated haemoglobin A1C; FBG, fasting blood glucose; BUN, blood urea nitrogen; Cys C, Cystatin C; eGFR, estimated glomerular filtration; Alb, albumin; TC, total cholesterol; TG, Triglyceride; LDL, low-density lipoprotein cholesterol; UACR, urine albumin creatine ratio. *Significant differences against DM-B and DKD-B group are indicated by * (P < 0.05). **Significant differences against DM-B and DKD-B group are indicated by ** (P < 0.001). qRT-PCR validation of differentially expressed urinary exosomal miRNAs in validation cohort Expanding the sample size for further study, we found that miR-136-5p was also significantly elevated in the DKD -B group compared to the DM-B group (1.51 [0.57, 3.13] vs. 1.00 [0.73, 1.03], P = 0.017) in the validation cohort. This finding was consistent with earlier data screening and results from the screening cohort (Fig. [88]4). Fig. 4. Fig. 4 [89]Open in a new tab Differential expression of urinary exosomal miR-136-5p was verified by qRT-PCR in validation cohort. *Significant difference against DM-B and DKD-B group is indicated by * (P < 0.05). Analysis of baseline data for patients in the validation cohort revealed statistically significant differences in age and diabetes history between the DKD -B and DM-B groups. To eliminate the effects of confounding factors, we performed binary logistic regression adjustments for these variables. The results showed that after adjusting for age and diabetes history, the expression of miR-136-5p in DKD remained statistically significant (P < 0.05) (Table [90]3). Table 3. Binary logistic regression to adjust of age and diabetes duration. Variable B Wals P Exp (B) 95% Confidence interval Lower limit Upper limit 3–1 miR-136-5p 0.838 4.495 0.034* 2.311 1.065 5.013 Age 0.030 1.043 0.307 1.030 0.973 1.091 3–2 miR-136-5p 0.813 4.453 0.035* 2.254 1.060 4.796 Diabetes duration (years) 0.060 2.539 0.111 1.061 0.986 1.142 [91]Open in a new tab *Significant differences against DM-B and DKD-B group are indicated by * (P<0.05). Diagnostic value of urinary exosomal miR-136-5p in DKD We assessed the clinical diagnostic value of urinary exosomal miR-136-5p in DKD using ROC curves (Table [92]4; Fig. [93]5; Supplementary Fig. 2). The optimal cutoff value for diagnosing DKD with urinary exosomal miR-136-5p was 0.141, with a sensitivity of 72.2%, specificity of 78.4%, positive predictive value of 61.9%, negative predictive value of 85.2%, and Youden’s index of 0.506. The ROC analysis yielded an AUC of 0.722 (95% CI 0.552–0.892; P = 0.008). Table 4. ROC analysis of miRNAs in DKD and T2DM patients. Variable AUC Cutoff Sensitivity, % Specificity, % PPV, % NPV, % Youden index P 95% Confidence interval Lower limit Upper limit miR-136-5p 0.722 0.141 72.2 78.4 61.9 85.2 0.506 0.008* 0.552 0.892 [94]Open in a new tab Best cutoff point of miRNA-136-5p was 0.141, [sensitivity = 72.2%, specificity = 78.4%, AUC = 0.722, 95% CI = 0.552–0.892, P = 0.008]. PPV: Positive Predicted Value, NPV: Negative Predictive Value. Youden index = Sensitivity + Specificity-1. *Significant difference is indicated by * (P<0.05). Fig. 5. [95]Fig. 5 [96]Open in a new tab ROC curve analysis of miRNA-136-5p in DKD and T2DM groups. Correlation analysis of urinary exosomal miR-136-5p with clinical parameters Correlation analysis showed that the expression of urinary exosomal miR-136-5p was positively correlated with UACR (ρ = 0.380, P = 0.004, Fig. [97]6a; Supplementary Fig. 3a), urea nitrogen (ρ = 0.277, P = 0.040, Fig. [98]6b; Supplementary Fig. 3b), cystatin C (ρ = 0.279, P = 0.041, Fig. [99]6c; Supplementary Fig. 3c), and negatively correlated with eGFR (ρ = − 0.300, P = 0.026, Fig. [100]6d; Supplementary Fig. 3d). Fig. 6. [101]Fig. 6 [102]Open in a new tab Spearman’s correlation analysis of urinary exosomal miR-136-5p expression and clinical parameters. UACR, urine albumin creatine ratio; BUN, blood urea nitrogen; Cys C, Cystatin C; eGFR, estimated glomerular filtration. Correlation analysis of urinary exosomal miR-136-5p with chronic kidney disease progression risk stratification indicators Correlation analysis showed that the expression of urinary exosomal miR-136-5p was positively correlated with chronic kidney disease progression risk stratification indicators (ρ = 0.339, P = 0.011, Table [103]5). Table 5. Spearman correlation analysis between miR136-5p and risk of progression of chronic kidney disease. Spearman correlation analysis Chronic kidney disease risk stratification Correlation coefficient ρ P-value miR-136-5p 0.339 0.011* [104]Open in a new tab *Significant difference is indicated by * (P<0.05). Functional prediction of urinary exosomal miR-136-5p Using the miRTarBase database, we predicted 152 target genes for miR-136-5p and conducted KEGG enrichment analysis. The results revealed enrichment in pathways including insulin secretion, glutamatergic synapse, cAMP signaling pathway, sphingolipid signaling pathway, and renin secretion, among a total of 25 pathways. Based on previous studies and literature review, we believe that miR-136-5p may be involved in the insulin secretion pathway by binding to its target genes SNAP25, PCLO and GNAS, in the cAMP signaling pathway by binding to RAC1, SOX9 and GNAS, in the sphingolipid signaling pathway by binding to BCL2, RAC1 and PPP2R2A, BCL2, RAC1, VCAM1 participation AGE—RAGE signal pathway, BCL2 CSNK2A1, combined VCAM1 involved in the NF-κB pathway and multiple signaling pathways, mediated pathological injury of diabetic kidney disease (Table [105]6; Fig. [106]7). These pathway associations suggest miR-136-5p may exacerbate DKD progression through dual mechanisms of metabolic dysregulation (impaired insulin secretion) and inflammatory activation (NF-κB signaling). Table 6. The KEGG analysis of urinary exosomal miR-136-5p. KEGG pathway Count Genes Insulin secretion 6 SNAP25, PCLO, GNAS, CACNA1D, ABCC8, ADCYAP1R1 Circadian entrainment 5 GNAS, CACNA1D, PER1, ADCYAP1R1, GRIN2A Viral myocarditis 4 RAC1, EIF4G3, CD55, HLA-DRB5 cAMP signaling pathway 7 RAC1, SOX9, GNAS, CACNA1D, ADCYAP1R1, GRIN2A, ROCK1 Protein processing in endoplasmic reticulum 5 BCL2, UBQLN1, UBE2D3, PPP1R15A, GANAB Glutamatergic synapse 4 GNAS, CACNA1D, SLC38A1, GRIN2A Sphingolipid signaling pathway 4 BCL2, RAC1, PPP2R2A, ROCK1 Renin secretion 3 GNAS, CACNA1D, ADCYAP1R1 Amphetamine addiction 3 GNAS, CACNA1D, GRIN2A Renal cell carcinoma 3 RAC1, PAK6, EGLN2 Dopaminergic synapse 4 PPP2R2A, GNAS, CACNA1D, GRIN2A Spinocerebellar ataxia 4 MCU, ATXN1, KCNC3, GRIN2A Adrenergic signaling in cardiomyocytes 4 BCL2, PPP2R2A, GNAS, CACNA1D Phagosome 4 RAC1, TFRC, FCGR3B, HLA-DRB5 GABAergic synapse 3 CACNA1D, SLC38A1, GABRG2 Pathways of neurodegeneration—multiple diseases 8 BCL2, RAC1, CACNA1D, CSNK2A1, MCU, ATXN1, GRIN2A, CSNK1A1 Nicotine addiction 2 GABRG2, GRIN2A mRNA surveillance pathway 3 PPP2R2A, CSTF2T, RBM8A Fc gamma R-mediated phagocytosis 3 RAC1, ASAP1, FCGR3B Hematopoietic cell lineage 3 TFRC, CD55, HLA-DRB5 AGE-RAGE signaling pathway in diabetic complications 3 BCL2, RAC1, VCAM1 Tight junction 4 RAC1, PPP2R2A, CACNA1D, ROCK1 NF-kappa B signaling pathway 3 BCL2, CSNK2A1, VCAM1 Type II diabetes mellitus 2 CACNA1D, ABCC8 HIF-1 signaling pathway 3 BCL2, TFRC, EGLN2 [107]Open in a new tab Fig. 7. [108]Fig. 7 [109]Open in a new tab KEGG Enrichment Analysis Circle Plot for miR-136-5p. Discussion Diabetic kidney disease (DKD) is a clinical disease characterized by persistent proteinuria and/ or progressive decline in kidney function. It is the leading cause of end-stage kidney disease worldwide, closely associated with cardiovascular events, and significantly increases cardiovascular mortality and all-cause mortality, imposing a substantial economic burden on global healthcare systems. In this study, we included 64 clinical patients, consisting of 23 DKD patients and 41 DM patients. Clinical data revealed a slightly higher prevalence of DKD in males, with most DKD patients having a diabetes history of over 10 years. Markers of kidney damage such as urea nitrogen and cystatin C were significantly elevated in DKD patients, accompanied by significant proteinuria and decreased eGFR. These findings are consistent with the traditional characteristics of DKD, meanwhile we found that fasting C-peptide levels were higher in patients with DKD. As research into diabetes and its complications deepens, it is increasingly recognized that C-peptide, not only an inert peptide within the proinsulin molecule with structural efficacy, possesses potent physiological functions. It can regulate various intracellular signaling pathways in the kidneys, potentially playing a critical role in delaying the onset and progression of chronic diabetic complications^[110]20. Physiological levels of C-peptide have protective effects, including antioxidation, vascular protection, anti-apoptosis, anti-inflammation, and NF-κB modulation^[111]21. Maintaining physiological C-peptide levels in Type 2 diabetes patients may delay the progression of DKD^[112]22, while excessive C-peptide (> 2.84 ng/mL) may promote the onset of DKD^[113]23. Janowska et al.^[114]24 confirmed that high concentrations of C-peptide might exacerbate inflammation and atherosclerosis processes in obese Type 2 diabetes patients. Kim et al.^[115]25 reported that C-peptide levels are closely related to microvascular complications in Type 2 diabetes, suggesting that C-peptide levels might directly influence the occurrence or progression of microvascular complications. Clinical manifestations and course of DKD are diverse. DKD has a stealthy onset with atypical early symptoms, and many patients are diagnosed when kidney damage is irreversible. Despite significant advances in DKD treatment, it remains the most common cause of end-stage kidney disease. Therefore, identifying new biomarkers for diabetic kidney diseases beyond proteinuria is crucial for the early diagnosis and treatment of DKD. Recent years have seen a surge in interest in extracellular vesicles, especially exosomes, for their extensive role in the intercellular transfer of genetic material, lipids, and cell-specific proteins. Numerous studies have shown that kidney-released exosomes and their contents are involved in the pathophysiological mechanisms underlying DKD. Exosomes are vesicles released by cells that participate in intercellular communication under both physiological and pathological conditions^[116]7,[117]26. miRNAs, short non-coding RNAs of 18–22 nucleotides, regulate gene expression post-transcriptionally by binding to the 3’ untranslated regions (3’UTRs) of mRNAs, inhibiting translation and affecting cellular growth, differentiation, apoptosis, and proliferation^[118]27. Several studies have indicated that miRNA dysregulation is involved in kidney disease processes, closely associated with mechanisms such as autophagy^[119]28, fibrosis^[120]29, inflammation^[121]30, and epithelial-mesenchymal transition^[122]31. Given the non-invasive and stable nature of urine samples, urinary exosomes are promising as ideal diagnostic markers for DKD. With the development of high-throughput sequencing and bioinformatics, numerous miRNAs have been identified and proven to play significant roles in various diseases. In our study, we identified differentially expressed miRNAs in DKD patients using the GEO database and further validated the differential expression of urinary exosomal miR-136-5p between DKD and DM groups via qRT-PCR in a screening cohort with confirmed DKD pathology and an expanded validation cohort. Previous studies have shown that miR-136-5p is regulated by circ-0003928 and is targeted to bind PAQR3. Overexpression of miR-136-5p can significantly improve the apoptosis, oxidative stress and fibrosis of HK-2 cells in the state of hyperglycemia^[123]32. It has also been found that miR-136 can be targeted to bind SYK to mediate TGF-β1/Smad3 signaling, thereby affecting DKD fibrosis, and overexpression of miR-136 can improve STZ-induced renal fibrosis in SD rats and NRK-52E inflammation induced by high sugar^[124]33. Li X et al.^[125]34 discovered that miR-136-5p, enriched in extracellular vesicles derived from human umbilical cord mesenchymal stem cells (hUC-MSCs), inhibits the Akirin1/EGR1/TP53/SLC7A11 axis by targeting Akirin1, thereby exerting potent protective effects against ferroptosis and renal ischemia–reperfusion injury in transplanted kidneys. Zhu S et al.^[126]35 demonstrated that miR-136-5p targets AKT3, thereby suppressing the PI3K/AKT3 signaling pathway. Jiang X et al.^[127]36 revealed that the miR-136-5p mimic directly targets GNAS, resulting in marked downregulation of key inflammatory mediators such as PI3K, STAT3, ERK, and pro-inflammatory cytokines (IL-6, IL-1β, IL-10, TNF-α). As a regulatory factor, miR-136-5p is involved in the development and progression of various types of cancer, including gastric cancer^[128]37, kidney cancer^[129]12, liver cancer^[130]38, bladder cancer^[131]39, and non-small cell lung cancer^[132]40. Gao et al.^[133]41 found that miR-136-5p was significantly lower in thyroid cancer tissues compared to adjacent non-cancerous tissues and was significantly associated with tumor pathological grading, TNM staging, lymph node metastasis, and distant metastasis. The low expression of miR-136-5p might promote the development and progression of thyroid cancer by upregulating metadherin (MTDH). Chen et al.^[134]42 found that miR-136-5p was significantly downregulated in renal cell carcinoma tissues, and overexpression of miR-136-5p significantly inhibited cell proliferation, migration, and invasion, inducing apoptosis in renal cell carcinoma cells. The tumor-suppressive miR-136-5p could potentially serve as a prognostic biomarker and a potential therapeutic target for renal cell carcinoma. Beyond oncological diseases, miR-136-5p is also associated with various non-oncological diseases. MiR-136-5p can inhibit pyroptosis by participating in the ATXN1L/CIC axis, upregulating PYDC1 and thus mitigating cardiac damage caused by coronary thrombosis^[135]43. Ji et al.^[136]44 demonstrated that high levels of miR-136, by targeting the B-cell lymphoma-2 gene (BCL2), inhibit cell proliferation and promote apoptosis in mesenchymal stem cells. Additionally, a prospective pregnancy cohort case–control study^[137]45 found that 10 miRNAs, including miR-136-5p, were highly expressed in patients with gestational diabetes. Bioinformatics analysis indicated that these differentially expressed miRNAs are involved in the proliferation and differentiation of trophoblasts, as well as the insulin secretion, regulation, and glucose transport in pregnant women. In this study, we confirmed for the first time via qRT-PCR that urinary exosomal miR-136-5p is significantly upregulated in DKD, and further validated its diagnostic value in DKD through ROC analysis, suggesting that urinary exosomal miR-136-5p holds promise as a novel biomarker for diagnosing DKD. Additionally, we observed that urinary exosomal miR-136-5p was significantly positively correlated with the risk stratification indicators used to assess chronic kidney disease progression, suggesting that urinary exosomal miR-136-5p is not only a potential marker for diagnosing DKD but may also be closely associated with the progression of the disease, potentially serving as a new therapeutic target for DKD. A systematic study on miRNA dysregulation in patients with Type 2 diabetes^[138]46 indicated that dysregulated miRNAs in the pancreas, including miR-136-5p, are crucial genes involved in the disruption of insulin signaling and pathways related to Type 2 diabetes, linked to various events related to pancreatic β-cell development and glucose-stimulated insulin secretion. Furthermore, research by Deng et al.^[139]47 found that miR-136-5p also plays a significant role in regulating inflammatory responses. In a rat model of spinal cord injury, upregulated expression of miR-136-5p, by targeting the I-kB kinase β gene and A20 gene, upregulated NF-κB expression, subsequently promoting the production of pro-inflammatory cytokines such as IFN-α, IL-1β, TNF-α, and IL-6. Silencing miR-136-5p improved spinal cord injury and inflammation cell infiltration, confirming the significant role of miR-136-5p in inflammatory responses induced by spinal cord injury. Whether miR-136-5p participates in the development and progression of DKD by regulating insulin signaling and inflammatory responses requires further mechanistic studies. To further explore the potential pathological mechanisms of urinary exosomal miR-136-5p in DKD, we predicted 152 target genes of miR-136-5p (such as CACNA1D, RAC1, GRIN2A, GNAS, BCL2, PPP2R2A) using the miTarBase database and conducted pathway enrichment analysis. This analysis yielded 25 signaling pathways including insulin secretion, cAMP signaling pathway, sphingolipid signaling pathway, AGE-RAGE signaling pathway, NF-κB signaling pathway, and HIF-1 signaling pathway. Several of these pathways are closely related to diabetes, such as insulin secretion, Type 2 diabetes signaling, intercellular communication (tight junctions), and the NF-κB signaling pathway. In the context of diabetes or hyperlipidemia, reduced adenylate cyclase activity and cAMP levels can lead to increased cell proliferation and vascular complications. A study using a streptozotocin-induced diabetic rat model^[140]48 first demonstrated that berberine could mediate the G protein-adenylate cyclase-cAMP signaling pathway, upregulate Gas expression, downregulate Gai expression, elevate cAMP levels, and inhibit mesangial cell proliferation, thereby improving hyperglycemia and reducing renal tissue damage and inflammation cell infiltration in rats. Fox et al.^[141]49 reported that although sphingolipids represent only a small portion of total lipids, there is substantial evidence that dysfunctions in sphingolipid metabolism or specific sphingolipid metabolites are potent inducers of diabetic complications and are implicated in the pathogenesis of DKD. Therapies targeting dysfunctional sphingolipid metabolism represent a new strategy to reduce diabetic complications. Numerous studies have highlighted the critical role of the AGE-RAGE signaling pathway in diabetes and its microvascular complications, making it a key factor in the development and progression of DKD^[142]50. Under high glucose conditions, the interaction of AGEs with RAGE generates a large amount of reactive oxygen species, amplifying the inflammatory response, causing oxidative stress and chronic inflammation in renal tissues, ultimately leading to loss of kidney function; it can also activate the NF-κB pathway, enhancing the expression and production of various inflammatory cells, chemokines, and adhesion molecules in DKD, causing chronic cell activation and tissue damage; it stimulates VEGF expression, increasing vascular permeability, leading to the occurrence of proteinuria. Additionally, the NF-κB signaling pathway, a key pro-inflammatory pathway, serves as a hub in multiple interconnected pathways and can mediate the fibrotic processes of DKD progression^[143]51. Jiang et al.^[144]52 demonstrated that hypoxia-inducible factor-1α (HIF-1α) could improve tubular damage in DKD by mediating mitochondrial dynamics control through heme oxygenase-1 (HO-1); in tubular epithelial cells, overexpression of HIF-1α ameliorated mitochondrial morphological damage and dysfunction induced by hypoxia, suggesting that enhancing the HIF-1 signaling pathway could improve DKD progression. They proposed a new mechanism—the HIF-1α/HO-1 pathway could be a key pathway mediating mitochondrial dynamics in DKD tubular cells. Based on previous literature and research findings, we believe that urinary exosomal miR-136-5p could be involved in the development and progression of DKD by regulating its target genes and affecting various signaling pathways such as insulin secretion, cAMP signaling pathway, AGE-RAGE signaling pathway, sphingolipid signaling pathway, HIF-1α/HO-1 pathway, and NF-κB signaling pathway, but its specific mechanisms still require further in-depth research and validation. In this study, we first time confirmed the elevated expression of miR-136-5p in urinary exosomes of DKD patients, demonstrating good diagnostic efficacy and a close correlation with the progression of chronic kidney disease, thus holding clinical significance and potential for translational applications. However, this study has certain limitations. First, due to the lack of follow-up, the precise relationship between the level changes of urinary exosomal miR-136-5p and the progression of DKD remains unclear. Secondly, the sample size of this study is limited, and further validation is needed through larger randomized controlled studies. In conclusion, we demonstrated that miR-136-5p is upregulated in the urinary exosomes of patients with diabetic kidney disease, and we have evaluated the diagnostic efficacy of urinary exosomal miR-136-5p, finding it closely correlated with kidney damage indicators and risk stratification markers for the progression of chronic kidney disease. Furthermore, functional prediction analysis suggests that it may be involved in the development and progression of DKD through multiple pathways, including insulin secretion, the cAMP signaling pathway, the AGE-RAGE signaling pathway, the sphingolipid signaling pathway, and the NF-κB signaling pathway. Urinary exosomal miR-136-5p holds promise as a novel non-invasive biomarker for DKD, and further research into its regulatory mechanisms could provide potential therapeutic targets for the prevention and treatment of DKD, offering promising prospects for clinical application. Electronic supplementary material Below is the link to the electronic supplementary material. [145]Supplementary Material 1^ (296.9KB, jpg) [146]Supplementary Material 2^ (261KB, jpg) [147]Supplementary Material 3^ (279.4KB, jpg) [148]Supplementary Material 4^ (566.8KB, tif) [149]Supplementary Material 5^ (814.9KB, tif) [150]Supplementary Material 6^ (709.8KB, tif) [151]Supplementary Material 7^ (10.5MB, tif) [152]Supplementary Material 8^ (10.5MB, tif) [153]Supplementary Material 9^ (10.5MB, tif) [154]Supplementary Material 10^ (10.9KB, xlsx) [155]Supplementary Material 11^ (456.7KB, docx) Abbreviations DKD Diabetic kidney disease miRNA MicroRNA T2DM Type 2 diabetes mellitus UACR Urinary albumin/creatinine ratio ROC Receiver operating characteristic curve eGFR Estimated glomerular filtration rate Author contributions Conceptualization, Liu.Y. and Zheng.S.; methodology, Zheng.S.; software, Ma.C. and Lin.S.; validation, Zheng.S. and Li.J.; formal analysis, Wang.B., Chen.P. and Wang.P.; investigation, Ma.C. and Li.W; resources, Wang.Z.; data curation, Li.H. and Wu.J.; writing—original draft preparation, Li.J.; writing—review and editing, Zheng.S.; visualization, Li.J.; supervision, Liu.Y.; project administration, Liu.Y. and Zheng.S.. All authors have read and agreed to the published version of the manuscript. Funding This study was supported by grants from the the National Natural Science Foundation of China (82170742) and Taishan Scholar Project of Shandong Province (tsqn202408366). Data availability The datasets during and/or analysed during the current study available from the corresponding author on reasonable request. Declarations Competing interest The authors declare no competing interests. Ethical approval and consent to participate All experimental procedures were approved by the Ethics Committee of the First Affiliated Hospital of Shandong First Medical University (approval code and date:YXLL-KY-2023(127), October 2023) and registered clinically (Registration No. and date: [156]NCT06123871, November 2023). Informed consent Informed consent was obtained from all individual participants included in the study. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Shanshan Zheng and Jiao Li equally contributed to this work. References