Serum RAC1 as a potential biomarker in the early stages of diabetic kidney disease Authors: Yao Fang, Xianyang Mo, Nan Wang, Qunwei Ma, Ruirui Fu, Ying Sun, Tingting Yang, Jaishi Bishnu Raj, Xiaoyan Zhou, Changjiang Ying Published: 2025 ABSTRACT AbstractBackgroundPreclinical studies have suggested that Ras-related C3 botulinum toxin substrate 1 (RAC1) accelerates diabetic kidney disease (DKD) progression by triggering renal inflammation/fibrosis. Nevertheless, the clinical relevance of RAC1 with DKD is undefined. In this study, we aimed to investigate the association between circulating RAC1 levels and early DKD to evaluate its potential as a novel biomarker for early disease detection.MethodsA total of 150 participants were recruited. They were divided into three groups: the type-2 diabetes mellitus (T2DM) group, the early DKD group, and the healthy volunteers (HVs). We measured the circulating level of RAC1 with ELISAs. In addition, we measured levels of two indicators of inflammation (NOD-like receptor pyrin domain-containing 3 (NLRP3) inflammasome, and interleukin (IL)-1β) and two serological indicators reflecting renal fibrosis (fibrillin 1 (FBN1) and matrix metalloproteinase-10 (MMP-10)).ResultsThe serum level of RAC1 was increased in individuals with early DKD. The serum level of RAC1 was positively correlated with urinary clinical biomarkers, including the urine albumin-to-creatinine ratio (UACR), urine microalbumin (UMA), and α1-microglobulin (α1-MG), but inversely associated with the estimated glomerular filtration rate (eGFR). Serum levels of RAC1, NLRP3, IL-1β, and MMP-10 were strongly inter-correlated in DKD patients. In addition, after correction for pertinent clinical features and the other four biomarkers, an increased risk of early DKD was linked to a higher serum RAC1 level. Analyses of receiver operating characteristic (ROC) curves revealed the area under the ROC curve of serum levels of RAC1, NLRP3, IL-1β, and MMP-10 for identifying early DKD to be 0.785, 0.728, 0.697, and 0.714, respectively.ConclusionsThe circulating RAC1 level was associated with early DKD and might be a biomarker for the early stages of DKD.Keywords: Diabetic kidney disease, Inflammation, Fibrosis, RAC1, NLRP3 inflammasome Introduction Diabetic kidney disease (DKD) is typified by increased proteinuria and a decreased glomerular filtration rate [1]. In addition to being the primary cause of end-stage renal disease (ESRD) worldwide, it is a significant risk factor for coronary artery disease and all-cause mortality among people with diabetes mellitus (DM) [2]. Numerous intricate mechanisms are linked to DKD pathogenesis [3]. Several studies have indicated that inflammation is a key factor in the emergence and advancement of DKD [4]. Activation of the NOD-like receptor pyrin domain-containing 3 (NLRP3) inflammasome has been shown to be linked to the pathogenesis of several renal diseases, including DKD [5–7]. Activation of NLRP3 inflammatory vesicles activates caspase-1, leading to the maturation of interleukin (IL)-18 and IL-1β and the recruitment of inflammatory cells [7]. It has been demonstrated in experimental diabetic mice that activation of the NLRP3 inflammasome in podocytes leads to proteinuria and renal damage [8]. Knockout of NLRP3 expression weakened renal inflammation and enhanced renal function in mice with streptozotocin (STZ)-induced DKD [9]. Our previous study demonstrated that Ras-related C3 botulinum toxin substrate 1 (RAC1) can activate the NLRP3 inflammasome by binding directly to NLRP3, causing renal inflammation and hastening DKD development [10]. Inhibition of RAC1 expression can provide renal protection by inhibiting several inflammatory signals to a certain extent [11]. RAC1 is one of the Rho family of small guanosine triphosphate (GTP)ases. RAC1 is present in two conformational states in the circulation: the inactive guanosine diphosphate (GDP)-bound (RAC1-GDP) and active GTP-bound (RAC1-GTP) forms [12]. RAC1 plays an important part in regulating the podocyte cytoskeleton and inflammation in the kidney [13]. Hyperactivation of RAC1 in podocytes leads to podocyte loss and glomerulosclerosis, which develops rapidly into proteinuria [14]. In mice with STZ-induced diabetic nephropathy, genetic inhibition of RAC1 was shown to attenuate the destruction of slit diaphragms in podocytes as well as to inhibit the disappearance of podocyte processes and the dedifferentiation of podocytes [15, 16]. The activation and overexpression of RAC1 can also contribute to renal inflammation [17]. Moreover, the increased expression of RAC1 facilitates renal fibrosis and eventually results in chronic kidney disease (CKD) [18]. Although RAC1 is an intracellular regulator of cytoskeleton and inflammation, its presence in serum is mechanistically linked to the pathophysiology of DKD. In DKD, hyperglycemia and inflammation cause podocyte apoptosis and tubular injury. This cellular damage triggers the passive leakage of intracellular RAC1 into the circulation [19]. Moreover, RAC1 regulates the platelet actin cytoskeleton and thereby platelet aggregation [20]. Hypercoagulability promotes platelet degranulation in DKD and this process may release RAC1-containing particles into the circulation [21]. Therefore, serum RAC1 can reflect local kidney damage and systemic inflammation, providing the possibility of non-invasive diagnosis. Herein, we examined two non-invasive serum biomarkers of kidney fibrosis: fibrillin 1 (FBN1) and matrix metalloproteinase-10 (MMP-10) [22]. The former is a glycoprotein found in the extracellular matrix (ECM) and is the key constituent of the kidney fibrotic niche. FBN1 expression is upregulated in the fibrotic kidney tissue scaffold. FBN1 plays an important part in inducing injury to endothelial cells in CKD. The serum level of FBN1 is a biomarker of vascular damage [23]. MMP-10 is a protein and has an important role in ECM modification. The circulating level of MMP-10 is increased in individuals with CKD, and is inversely correlated with renal function [24, 25]. In this study, we investigated the association between circulating levels of RAC1 and early DKD in a cohort of patients, aiming to evaluate the diagnostic potential of RAC1 as a novel biomarker for early DKD detection. Materials and methods Participants and groups Participants were enrolled in the Affiliated Hospital of Xuzhou Medical University from February to April 2023. Healthy volunteers (HVs) without DM were recruited from the Health Management Center. People with type-2 diabetes mellitus (T2DM), with and without early DKD, were recruited from the Department of Endocrinology and the Department of Nephrology. Based on the following exclusion and inclusion criteria, 150 participants eventually participated in the study, including the T2DM group (n = 45), the early DKD group (n = 55), and the HVs group (n = 50). Individuals without a history of DM with glycated hemoglobin (HbA1c) ≤ 6.0% and fasting plasma glucose (FPG) ≤ 6.1 mmol/L were classified as HVs. We diagnosed T2DM according to the criteria for T2DM set by the World Health Organization in 1999 [26]. We reviewed ≥ 2 urine albumin-to-creatinine ratio (UACR) values for all patients with T2DM during the 3–6-month period prior to enrollment, and grouped them according to the median value. Early DKD was characterized by the UACR of 30–300 mg/g and an estimated glomerular filtration rate (eGFR) ≥ 90 mL/min/1.73 m2 [27]. The study protocol was approved (XYFY2022-KL483-01) by the Ethics Committee of the Affiliated Hospital of Xuzhou Medical University (Xuzhou, China). The study adhered to the Declaration of Helsinki 1964 and its later amendments. Written informed consent form was obtained from all participants. Inclusion and exclusion criteria The inclusion criteria were age > 18 years and complete medical history was available. The exclusion criteria were patients with: (1) acute complications of DM, including diabetic ketoacidosis, hyperosmolar hyperglycemia state, or lactic acidosis and hypoglycemic coma occurring during DM treatment; (2) acute cardiovascular disease, severe hepatic disease, or renal insufficiency; (3) malignant neoplasms, hematological disease, or autoimmune disease; (4) primary/secondary glomerular disease, acute/chronic renal failure, or hereditary renal disease; (5) severe malnutrition, trauma, or other stressful conditions. Variable collection and measurement We measured the height, bodyweight, and blood pressure and reviewed the medical history and medications of participants. Blood samples and urine specimens were obtained under fasting conditions at 8 am. Standard laboratory methods were used for routine analyses. The HbA1c level was measured by a fully automated glycated hemoglobin analyzer (Tosoh, Tokyo, Japan). FPG, blood uric acid (UA), blood urea nitrogen (BUN), serum creatinine (Scr), total cholesterol (TC), low-density lipoprotein (LDL-C), high-density lipoprotein (HDL-C), and triglycerides (TG) were measured on an automatic biochemistry analyzer (7600 series; Hitachi, Japan). We employed the Chronic Kidney Disease Epidemiology Collaboration formula to calculate the eGFR [28]. Levels of urine microalbumin (UMA) and α1-microglobulin (α1-MG) were measured using a fully automated biochemical analyzer (Olympus, Tokyo, Japan). The UACR was calculated by dividing the level of urinary albumin by the level of urinary creatinine. Serum samples treatment Serum was prepared by centrifugation (2000 × g, 10 min, room temperature) within 1 h of collection of whole blood, and stored at − 80 ℃ before examination. ELISA assay The RAC1 level in serum was quantified using a double-antibody sandwich enzyme-linked immunosorbent assay (ELISA) kit (Shanghai Jiang Lai Biotechnology, Shanghai, China). The limit of detection was 0.15 ng/mL, with an intra-assay and inter-assay coefficient of variability (CV) of < 10% and < 10%, respectively. Serum concentrations of NLRP3, IL-1β, FBN1, and MMP-10 were also measured by ELISA kits (Shanghai Jiang Lai Biotechnology). With regard to intra-assay and inter-assay CVs, they were, respectively: <9% and < 11% for NLRP3; <10% and < 10% for IL-1β; <10% and < 10% for FBN1; <10% and < 10% for MMP-10. The limit of detection for NLRP3, IL-1β, FBN1, and MMP-10 was 1.0 pg/mL, 2.9 pg/mL, 0.12 ng/mL, and 25.38 pg/mL, respectively. Statistical analyses Data were analyzed using SPSS 26.0 (IBM, Armonk, NY, USA). The normality of data distribution was assessed using the Kolmogorov–Smirnov test. Data are the mean ± SD or median with 25th and 75th percentiles, as appropriate. For data with a normal distribution, comparisons between two groups were made using the Student’s t-test, whereas analysis of variance was used to compare variables among multiple groups with Bonferroni’s correction. For data with a non-normal distribution, the Mann–Whitney U-test was employed to compare continuous variables between two groups, whereas the Kruskal–Wallis test was utilized to ascertain differences between three groups. Numbers and percentages have been used to represent categorical variables, and the χ2 test was employed to ascertain the significance of inter-group differences. Spearman correlation analysis was conducted to assess correlations between clinical features and biomarkers. Research on risk factors for early DKD was conducted using logistic regression models. The diagnostic value of each indicator was evaluated using receiver operating characteristic (ROC) curves. P < 0.05 was considered significant. Results Characteristics of the study cohort Table 1 presents the clinical features of the study cohort (n = 150). All groups were comparable in terms of sex, age, previous use of alcohol and/or tobacco, and medical history. Levels of Scr and eGFR did not differ significantly among the three groups. Individuals with early DKD had a higher body mass index (BMI), and higher levels of HbA1c and SBP than individuals with T2DM and HVs. Compared with the T2DM group, patients with early DKD had a longer duration of DM and higher levels of UMA, α1-MG, and UACR (p < 0.05). Table 1Clinical features of the participantsHVs (n = 50)T2DM (n = 45)Early DKD (n = 55)P valueClinical variablesMale (%)27(54.00)22(48.89)33(60.00)0.536Age (years)57.5 ± 9.3156.16 ± 14.9253.24 ± 13.940.226Smoking (%)7(14.00)8(17.78)7(12.73)0.767Drinking (%)1(2.00)3(6.67)6(10.91)0.188Duration of T2DM(years)–5.00(1.00,10.00)10.00(3.00,13.00)0.013BMI (kg/cm2)24.15 ± 3.7326.15 ± 3.55a28.45 ± 5.08ab< 0.001Systolic blood pressure (mmHg)130.18 ± 16.62131.64 ± 15.69141.2 ± 18.12ab0.002Diastolic blood pressure (mmHg)81.36 ± 10.986.8 ± 10.29a90.09 ± 12.42a0.001Medical historyCoronary heart disease (%)2(4.00)4(8.89)6(10.91)0.417Hypertension (%)19(38.00)20(44.44)30(54.54)0.229Cerebral infarction (%)3(6.00)4(8.89)5(9.09)0.807Rx (%)OHA–23(51.11)28(50.09)0.984Insulin ± OHA–19(42.22)35(63.64)b0.033ACE-Is/ARBs8(16.00)9(20.00)25(45.45)ab< 0.001Biochemical variablesHbA1c (%)5.40(5.20,5.63)6.70(5.80,9.05) a8.60(7.10,9.90) ab< 0.001FPG (mmol/L)4.84 ± 0.467.20 ± 2.27a8.39 ± 2.93ab< 0.001BUN (mmol/L)5.68 ± 1.595.99 ± 1.496.13 ± 1.790.361Scr (mmol/L)54.58 ± 12.9550.49 ± 19.4755.13 ± 16.970.332UA (mmol/L)266.96 ± 60.67277.33 ± 84.11317.05 ± 72.06ab0.001eGFR(mL/min/1.73 m2)120.00(107.45,120.00)120.00(118.13,120.00)120.00(113.52,120.00)0.686TC (mmol/L)4.77 ± 0.914.76 ± 1.284.89 ± 0.970.768TG (mmol/L)1.36(0.88,2.00)1.17(0.92,1.78)1.41(0.95,2.28)0.460HDL-C (mmol/L)1.04(0.92,1.26)1.01(0.85,1.31)1.07(0.87,1.20)0.846LDL-C (mmol/L)2.23 ± 0.682.65 ± 0.83a3.15 ± 0.80ab< 0.001UMA (mg/L)–4.15(2.01,6.95)73.47(29.52,168.00) b< 0.001α1-MG (mg/L)–3.05(2.21,6.83)7.48(3.02,19.80) b0.001UACR (mg/mmol)–0.59(0.32,0.85)8.36(4.62,17.02) b< 0.001BiomarkersRAC1 (ng/mL)6.72 ± 3.077.19 ± 3.0810.88 ± 3.48ab< 0.001NLRP3 (ng/mL)1.12(0.9,1.25)1.54(1.04,2.23) a2.45(1.62,3.10) ab< 0.001IL-1β (pg/mL)4.19(3.52,4.89)5.11(4.08,6.97) a6.47(5.15,10) ab< 0.001FBN1 (ng/mL)1.45(1.13,1.64)1.36(1.02,1.73)1.30(1.03,1.84)0.766MMP-10 (pg/mL)1021.24(641.37,1382.90)1058.41(577.98,1351.71)1285.07(1102.92,2290.94) ab< 0.001(a) compared to HVs, p < 0.05; (b) compared to T2DM, p < 0.05HVs, healthy volunteers; BMI, body mass index; Rx, prescription; OHA, oral hypoglycemic agent; ACE-Is/ARBs: angiotensin converting enzyme inhibitors or angiotensin receptor blockers; HbA1c, glycated hemoglobin A1c; FPG, fasting plasama glucose; BUN, blood urea nitrogen; Scr, serum creatinine; UA, uric acid; eGFR, estimation of glomerular filtration rate; TC, total cholesterol; TG, triglycerides; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; UMA, urine microalbumin; α1-MG, α1-Microglobulin; UACR, urine albumin-to-creatinine ratio; RAC1, Ras-related C3 botulinum toxin substrate 1; NLRP3, NOD-like receptor pyrin domain-containing 3; FBN1, fibrillin 1; MMP-10, matrix metalloproteinase-10 Clinical features of the participants Duration of T2DM (years) eGFR (mL/min/1.73 m2) (a) compared to HVs, p < 0.05; (b) compared to T2DM, p < 0.05 HVs, healthy volunteers; BMI, body mass index; Rx, prescription; OHA, oral hypoglycemic agent; ACE-Is/ARBs: angiotensin converting enzyme inhibitors or angiotensin receptor blockers; HbA1c, glycated hemoglobin A1c; FPG, fasting plasama glucose; BUN, blood urea nitrogen; Scr, serum creatinine; UA, uric acid; eGFR, estimation of glomerular filtration rate; TC, total cholesterol; TG, triglycerides; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; UMA, urine microalbumin; α1-MG, α1-Microglobulin; UACR, urine albumin-to-creatinine ratio; RAC1, Ras-related C3 botulinum toxin substrate 1; NLRP3, NOD-like receptor pyrin domain-containing 3; FBN1, fibrillin 1; MMP-10, matrix metalloproteinase-10 The early DKD group demonstrated significantly greater serum levels of RAC1 and MMP-10 (p < 0.05). Levels of the NLRP3 inflammasome and IL-1β were elevated in the T2DM group compared to HVs, and were further increased in the early DKD group compared to the T2DM group (p < 0.05). Notably, FBN1 levels in the three groups did not show a significant difference (p >0.05). Association of tertiles of the serum RAC1 level with clinical variables Table 2 summarizes the relationships between increasing tertiles of the serum RAC1 level and clinical characteristics in the T2DM group and early-DKD group. Compared with patients with a lower RAC1 level, patients with a higher RAC1 level were more likely to have a higher UACR and higher levels of UMA, and α1-MG; to have lower levels of eGFR (p < 0.05). Moreover, levels of the NLRP3 inflammasome, IL-1β, FBN1, and MMP-10 were increased in participants with a higher RAC1 level (p < 0.05). Table 2Clinical features stratified across tertiles of RAC1 in T2DM and early DKD groupsTertiles1(1.950, 7.217)Tertiles2(7.217, 10.913)Tertiles3(10.913, 18.776)P valuen333433Early DKD (%)9(27.27)18(52.94) a28(84.85) ab< 0.001UACR (mg/mmol)0.66(0.35,3.49)2.98(0.63,9.08) a8.36(3.58,18.82) ab< 0.001eGFR (mL/min/1.73 m2)120.00(120.00,120.00)120.00(119.37,120.00)120.00(99.17,120.00) a0.005Scr (mmol/L)45.76 ± 13.1249.06 ± 13.8764.42 ± 21.17ab0.001UMA (mg/L)4.33(2.26,22.13)22.38(5.14,69.47) a73.47(27.34,170.18) ab< 0.001α1-MG (mg/L)2.66(1.18,4.01)4.83(2.75,8.41) a15.81(5.43,20.05) ab< 0.001NLRP3 (ng/mL)1.33(0.80,1.81)1.83(1.33,2.73) a2.70(2.29,3.10) ab< 0.001IL-1β (pg/mL)4.76(4.08,5.46)6.28(5.06,7.29) a9.57(6.10,13.59) ab< 0.001FBN1 (ng/mL)1.13 ± 0.451.39 ± 0.46 a1.66 ± 0.46 ab< 0.001MMP-10 (pg/mL)743.69(572.51,1095.81)1228.64(1059.91,1439.24) a2193.32(1193.17,3243.30) ab< 0.001UACR, urine albumin-to-creatinine ratio; eGFR, estimation of glomerular filtration rate; Scr, serum creatinine; UMA, urine microalbumin; α1-MG, α1-Microglobulin; RAC1, Ras-related C3 botulinum toxin substrate 1; NLRP3, NOD-like receptor pyrin domain-containing 3; FBN1, fibrillin 1; MMP-10, matrix metalloproteinase-10. a p < 0.05 versus Tertiles1; b p < 0.05 versus Tertiles2 Clinical features stratified across tertiles of RAC1 in T2DM and early DKD groups UACR, urine albumin-to-creatinine ratio; eGFR, estimation of glomerular filtration rate; Scr, serum creatinine; UMA, urine microalbumin; α1-MG, α1-Microglobulin; RAC1, Ras-related C3 botulinum toxin substrate 1; NLRP3, NOD-like receptor pyrin domain-containing 3; FBN1, fibrillin 1; MMP-10, matrix metalloproteinase-10. a p < 0.05 versus Tertiles1; b p < 0.05 versus Tertiles2 Correlation between biomarkers and clinical parameters Figure 1 presents a matrix correlation between levels of biomarkers and clinical features. The eGFR was weakly or moderately correlated with levels of RCA1 (r = −0.268, p = 0.007), NLRP3 (r = −0.360, p < 0.001), and MMP-10 (r = −0.316, p = 0.001). The serum level of RAC1 was positively correlated UACR (r = 0.540, p < 0.001), UMA (r = 0.549, p < 0.001), α1-MG (r = 0.591, p < 0.001), NLRP3 (r = 0.578, p < 0.001), IL-1β (r = 0.582, p < 0.001), MMP-10 (r = 0.553, p < 0.001), and FBN1 (r = 0.391, p < 0.001). The levels of UACR, UMA, and α1-MGl were moderately or strongly correlated with levels of all biomarkers except FBN1. Serum levels of RAC1, NLRP3, IL-1β, and MMP-10 were strongly inter-correlated. There was weak correlation between these four biomarkers and the FBN1 level. Matrix correlation of clinical parameters and examined biomarkers Logistic regression analysis in the T2DM group and early-DKD group The duration of T2DM (OR 1.096 [95% CI 1.019–1.179], p = 0.014), BMI (OR 1.132 [95% CI 1.023–1.253], p = 0.016), SBP (OR 1.035 [95% CI 1.008–1.061], p = 0.009), as well as levels of HbA1c (OR 1.382 [95% CI 1.105–1.728], p = 0.005), FPG (OR 1.197 [95% CI 1.015–1.41], p = 0.032), LDL-C (OR 2.185 [95% CI 1.271–3.755], p = 0.005), UA (OR 1.007 [95% CI 1.001–1.012], p = 0.016), RAC1 (OR 1.399 [95% CI 1.204–1.625], p < 0.001), NLRP3 (OR 2.633 [95% CI 1.558–4.449], p < 0.001), IL-1β (OR 1.171 [95% CI 1.024–1.338], p = 0.021), and MMP-10 (OR 1.001 [95% CI 1.000–1.001.000.001], p = 0.002) were associated with early DKD in univariate logistic regression analysis (Table 3) and were included in multivariable logistic analysis. Table 3Univariate logistic regression analysis for the risk of early DKDOR (95% CI) P Female (%)0.638(0.288 ~ 1.413)0.268Age (years)0.986(0.958 ~ 1.014)0.313Duration of T2DM(years)1.096(1.019 ~ 1.179)0.014Smoking (%)0.674(0.224 ~ 2.029)0.483Drinking (%)1.714(0.404 ~ 7.278)0.465Coronary heart disease (%)1.255(0.331 ~ 4.752)0.738Hypertension (%)1.5(0.679 ~ 3.312)0.316Cerebral infarction (%)1.025(0.258 ~ 4.067)0.972SBP (mmHg)1.035(1.008 ~ 1.061)0.009DBP (mmHg)1.026(0.99 ~ 1.063)0.160BMI (kg/cm2)1.132(1.023 ~ 1.253)0.016HbA1c (%)1.382(1.105 ~ 1.728)0.005FPG (mmol/L)1.197(1.015 ~ 1.41)0.032BUN (mmol/L)1.053(0.828 ~ 1.34)0.671Scr (mmol/L)1.015(0.992 ~ 1.039)0.209UA (mmol/L)1.007(1.001 ~ 1.012)0.016e GFR (mmol/L)0.999(0.974 ~ 1.026)0.964TC (mmol/L)1.12(0.781 ~ 1.607)0.538TG (mmol/L)1.207(0.924 ~ 1.577)0.168HDL-C (mmol/L)1.102(0.328 ~ 3.705)0.876LDL-C (mmol/L)2.185(1.271 ~ 3.755)0.005RAC1 (ng/mL)1.399(1.204 ~ 1.625)0.000NLRP3 (ng/mL)2.633(1.558 ~ 4.449)0.000IL-1β (ng/mL)1.171(1.024 ~ 1.338)0.021FBN1 (ng/mL)1.446(0.65 ~ 3.219)0.366MMP-10 (ng/mL)1.001(1.000 ~ 1.001)0.002SBP, systolic blood pressure; DBP, diastolic blood pressure; BMI, body mass index; HbA1c, glycated hemoglobin A1c; FPG, fasting plasma glucose; BUN, blood urea nitrogen; Scr, serum creatinine; UA, uric acid; eGFR, estimation of glomerular filtration rate; TC, total cholesterol; TG, triglycerides; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; RAC1, Ras-related C3 botulinum toxin substrate 1; NLRP3, NOD-like receptor pyrin domain-containing 3; FBN1, fibrillin 1; MMP-10, matrix metalloproteinase-10 Univariate logistic regression analysis for the risk of early DKD Duration of T2DM (years) SBP, systolic blood pressure; DBP, diastolic blood pressure; BMI, body mass index; HbA1c, glycated hemoglobin A1c; FPG, fasting plasma glucose; BUN, blood urea nitrogen; Scr, serum creatinine; UA, uric acid; eGFR, estimation of glomerular filtration rate; TC, total cholesterol; TG, triglycerides; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; RAC1, Ras-related C3 botulinum toxin substrate 1; NLRP3, NOD-like receptor pyrin domain-containing 3; FBN1, fibrillin 1; MMP-10, matrix metalloproteinase-10 Multivariable logistic analysis was employed to investigate the correlation between the proteins we examined and the early DKD (Table 4). After adjustment for T2DM duration, BMI, SBP, as well as levels of HbA1c, FPG, LDL-C, and UA, we found that the RAC1 level was associated with early DKD (OR 1.394 [95% CI 1.139–1.707], p = 0.001). Moreover, after additional adjustment for levels of NLRP3, IL-1β, MMP-10, and FBN1 in individuals, RAC1 remained a risk factor for early DKD (OR 1.414 [95% CI 1.057–1.892], p = 0.020). Levels of NLRP3, IL-1β, and MMP-10 were no longer risk factors for early DKD after RAC1-based multivariate adjustment. Table 4Multivariate logistic regression analysis based on model adjustmentOR (95% CI)P valueRAC1 (ng/mL)Crude1.399(1.204 ~ 1.625)< 0.001Model 11.394(1.139 ~ 1.707)0.001Model 2b1.330(1.039 ~ 1.702)0.024Model 2c1.414(1.057 ~ 1.892)0.020NLRP3 (ng/mL)Crude2.633(1.558 ~ 4.449)< 0.001Model 12.608(1.351 ~ 5.034)0.004Model 2a1.611(0.779 ~ 3.329)0.198IL-1β (pg/mL)Crude1.171(1.024 ~ 1.338)0.021Model 11.143(0.977 ~ 1.337)0.095Model 2a0.992(0.829 ~ 1.187)0.934MMP-10 (pg/mL)Crude1.001(1.000 ~ 1.001)0.002Model 11.001(1.000 ~ 1.002)0.009Model 2a1.001(0.999 ~ 1.001)0.219Model 1: adjustment for duration of T2DM, BMI, SBP, HbA1c, FPG, LDL-C, and UAModel 2: adjustment for proteins we examined in addition to the variables in Model 1Model 2a: further adjustment for RAC1Model 2b: further adjustment for NLRP3, and IL-1βModel 2c: further adjustment for MMP-10 and FBN1 in addition to the variables in Model 2b Multivariate logistic regression analysis based on model adjustment Model 1: adjustment for duration of T2DM, BMI, SBP, HbA1c, FPG, LDL-C, and UA Model 2: adjustment for proteins we examined in addition to the variables in Model 1 Model 2a: further adjustment for RAC1 Model 2b: further adjustment for NLRP3, and IL-1β Model 2c: further adjustment for MMP-10 and FBN1 in addition to the variables in Model 2b Diagnostic value of biomarkers for identifying early DKD Table 5 displays the diagnostic values of RAC1, NLRP3, IL-1β, and MMP-10. The area under the ROC curve (AUC) for levels of RAC1, NLRP3, IL-1β, and MMP-10 was 0.785 (95% CI, 0.696–0.874; p < 0.001), 0.728 (95% CI, 0.630–0.827; p < 0.001), 0.697 (95% CI, 0.592–0.802; p = 0.001), and 0.714 (95% CI, 0.612–0.815; p < 0.001), respectively (Fig. 2). The cutoff for the RAC1 level for diagnosing early DKD was 9.410 ng/mL, with a sensitivity of 69.1% and specificity of 77.8%. The cutoff of the NLRP3 level for diagnosing early DKD was 2.133 ng/mL, with a sensitivity of 61.8% and specificity of 75.6%. The cutoff of the IL-1β level for diagnosing early DKD was 5.692 pg/mL, with a sensitivity of 72.7% and specificity of 62.2%. The cutoff of the MMP-10 level for diagnosing early DKD was 1114.483 pg/mL, with a sensitivity of 74.5% and specificity of 62.3%. Table 5Diagnostic performances of RAC1, NLRP3, IL-1β, and MMP-10 for early DKDAUC (95% CI)P valueCut-off valuesSensitivitySpecificityRAC10.785(0.696–0.874)< 0.0019.4169.177.8NLRP30.728(0.630–0.827)< 0.0012.13361.875.6IL-1β0.697(0.592–0.802)0.0015.69272.762.2MMP-100.714(0.612–0.815)< 0.0011114.48374.562.3RAC1 + NLRP30.804(0.719–0.889)< 0.0010.56476.480.0RAC1 + MMP-100.796(0.709–0.883)< 0.0010.57870.980.0RAC1 + NLRP3 + MMP-100.809(0.725–0.893)< 0.0010.52078.277.8RAC1, Ras-related C3 botulinum toxin substrate 1; NLRP3, NOD-like receptor pyrin domain-containing 3; MMP-10, matrix metalloproteinase-10 Diagnostic performances of RAC1, NLRP3, IL-1β, and MMP-10 for early DKD 0.785 (0.696–0.874) 0.728 (0.630–0.827) 0.697 (0.592–0.802) 0.714 (0.612–0.815) 0.804 (0.719–0.889) 0.796 (0.709–0.883) 0.809 (0.725–0.893) RAC1, Ras-related C3 botulinum toxin substrate 1; NLRP3, NOD-like receptor pyrin domain-containing 3; MMP-10, matrix metalloproteinase-10 Receiver operating characteristics curve analysis of circulating levels of RAC1, NLRP3, IL-1β, and MMP-10 for identifying early DKD Furthermore, combined panels showed improved diagnostic performance: the AUC for RAC1 + NLRP3 was 0.804 (95% CI, 0.719–0.889; p < 0.001) with 76.4% sensitivity and 80.0% specificity; RAC1 + MMP-10 yielded an AUC of 0.796 (95% CI, 0.709–0.883; p < 0.001) with 70.9% sensitivity and 80.0% specificity; and the triple combination RAC1 + NLRP3 + MMP-10 achieved an AUC of 0.809 (95% CI, 0.725–0.893; p < 0.001) with 78.2% sensitivity and 77.8% specificity. Discussion In our investigation of patients with T2DM and age- and sex-matched HVs, we demonstrated for the first time the clinical relevance of the circulating RAC1 level as a potential biomarker in the early stages of DKD. The serum level of RAC1 was increased in the early DKD group and was also independently associated with early DKD after adjustment of confounders. The serum RAC1 level was also positively correlated with the biomarkers of inflammation (NLRP3 inflammasome, IL-1β) and fibrosis (MMP-10) in patients with early DKD. The NLRP3 inflammasome is a megadalton complex implicated in numerous inflammation-driven diseases [29]. Evidence indicates that it can be activated by DKD-related metabolic stimuli [30, 31]. NLRP3 activation has been shown to trigger maturation of IL-1β and IL-18, which contributes to increased albumin excretion and renal unit-damage in experimental and human settings [32, 33]. Shahzad et al. reported that the levels of NLRP3 and IL-1β in the glomeruli and serum of albuminuria-diabetic patients were increased markedly in comparison with those without albuminuria, and the serum level of IL-1β was connected with the degree of albuminuria [7]. In our study, serum levels of NLRP3 and IL-1β were increased significantly in early-DKD group and the levels of NLRP3 and IL-1β were positively correlated with levels of UMA, which was consistent with previous studies. FBN1 and MMP-10 are considered to be non-invasive serological indicators of renal fibrosis [22]. Renal fibrosis is considered to be a significant factor contributing to advanced renal injury [34, 35]. Recently, a longitudinal study suggested profibrotic proteins had a profound and independent role in the early stages of renal deterioration in individuals with T2DM [36]. It has been reported that increased serum levels of FBN1 and MMP-10 are linked to renal impairment and dysfunction in patients with CKD [23, 25]. In our study, individuals with early DKD had a considerably higher serum level of MMP-10, whereas the serum level of FBN1 in the three groups did not differ significantly. This phenomenon could arise because FBN1 may be an indicator of microvascular rarefaction in advanced nephropathy [23]. In addition, serum levels of MMP-10 were negatively related to the eGFR, but positively correlated with the UACR, as well as levels of UMA and urinary α1-MG. RAC1 is a member of the RHO family of small GTPases and is ubiquitously expressed [12]. Previous studies have shown that serum RAC1 levels are associated with a variety of diseases [37, 38], which confirms the stable presence of RAC1 in serum. In this study, we observed significantly elevated circulating RAC1 levels in patients with early DKD, which correlated positively with UMA, α1-MG, and biomarkers of inflammation (NLRP3 inflammasome, IL-1β) and fibrosis (MMP-10). These findings align with established mechanisms from animal and cellular models, wherein RAC1 activation promotes renal injury by generating NADPH oxidase-derived ROS [18], stimulating inflammatory signaling [11, 39], and inducing fibrosis [18]. Furthermore, RAC1 contributes to proteinuria and renal dysfunction by disrupting the podocyte cytoskeleton and activating the TRPC5 channel [40–43]. Unlike previous studies focused on intracellular RAC1 signaling in animal or cell models, we demonstrate its detectable presence and pathological relevance in human serum. This highlights its potential as a non-invasive indicator of early DKD. Multivariate analysis confirmed that RAC1 remained an independent risk factor for early DKD after adjustment for other proteins, suggesting a central role in its pathology. Analyses of ROC curves indicated that RAC1 exhibited superior discriminatory power compared to NLRP3, IL-1β, and MMP-10 for identifying early DKD. However, its sensitivity (69.1%) and specificity (77.8%) remain moderate, suggesting that RAC1 alone may have limited utility as a standalone test. We therefore developed a combined diagnostic model incorporating RAC1 with NLRP3 and MMP-10, which exhibited improved diagnostic efficiency compared to any of these markers used individually. This study provides the first clinical evidence that circulating RAC1 is a stable and measurable biomarker in serum that is specifically linked to early DKD, thereby bridging a critical gap between preclinical mechanistic knowledge and clinical applicability. These findings bridge a critical gap between preclinical mechanistic knowledge and clinical application. UACR is a marker of established glomerular damage. However, its high variability and insensitivity to early tubular injury limit its usefulness for early detection, particularly in patients with normoalbuminuric DKD [44]. EGFR lacks sensitivity in early DKD due to renal functional reserve [45]. Our study confirmed that even in participants with preserved renal function (eGFR ≥ 90 mL/min/1.73 m²), serum RAC1 levels were inversely associated with eGFR and served as an indicator of early DKD. Therefore, RAC1 demonstrates a superior ability to detect DKD at an earlier stage than conventional clinical parameters such as UACR and eGFR. Furthermore, by combining RAC1 with markers of inflammation and fibrosis, our study proposes a novel multi-dimensional strategy for the early diagnosis of DKD, surpassing the limitations of conventional single-marker approaches. Finally, the limitations of our study deserve consideration. First, this study was conducted at a single center, and this design limits the generalizability of our findings. Future multi-center studies are essential to confirm the diagnostic and prognostic utility of RAC1. Second, it was a cross-sectional investigation. We have only demonstrated that the RAC1 level was associated with early DKD, but could not infer the causal relationship between RAC1 and early DKD. Longitudinal studies are urgently needed to determine whether RAC1 levels rise before the onset of overt renal dysfunction and whether they can predict progression to advanced DKD. Conclusions In conclusion, our findings provide the first clinical evidence that serum RAC1 is independently associated with early-stage diabetic kidney disease, underscoring its potential as a novel and stable biomarker for early detection. Importantly, we developed a multidimensional diagnostic model integrating RAC1 with NLRP3 and MMP-10, which demonstrated superior diagnostic efficiency over individual markers. Further research with a larger sample size and a longer follow-up period is required to verify our findings, and evaluate the role of RAC1 in treatment response or risk stratification. Acknowledgements The authors express their gratitude to all the volunteers who participated in the study as well as to the physicians, nurses, and technicians from the Affiliated Hospital of Xuzhou Medical University who spent their time and energy making this research possible. Author contributions Changjiang Ying and Yao Fang participated in the conception and design of the study. Yao Fang contributed to data collection, statistical analyses, and wrote the manuscript. Changjiang Ying and Xiaoyan Zhou participated in obtaining financial support and reviewed the manuscript. Qunwei Ma and Jaishi Bishnu Raj performed statistical analyses and produced the tables and figures. Xianyang Mo, Nan Wang, and Ruirui Fu contributed to data and specimen collection. Ying Sun and Tingting Yang participated in the indicator determination. All authors agreed to publish this manuscript. Funding This work was funded by the National Natural Science Foundation of China (82271205); the Jiangsu Provincial Health and Health Commission Research Project (K2023063); the Research Foundation of Medical Key Project of Jiangsu Provincial Health Commission (Grant No. K2023063); the Jiangsu Provincial Traditional Chinese Medicine Science and Technology Development Plan Project (Grant No. MS2022141); and the Medical and Health Science and Technology Project of the National Health Commission (Grant No. WKZX2022JG0111). Data availability The corresponding author will provide the study’s original data if the request is reasonable. Declarations Ethics approval and consent to participate This study was performed in line with the principles of the Declaration of Helsinki and was approved by the Ethics Committee of the Affiliated Hospital of Xuzhou Medical University (approval number: XYFY2022-KL483-01). Written informed consent was gained from all participants. Consent for publication All the authors have read and agreed to the published version of the manuscript. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The corresponding author will provide the study’s original data if the request is reasonable.