Exosomal UMOD gene expression and urinary uromodulin level as early noninvasive diagnostic biomarkers for diabetic nephropathy in type 2 diabetic patients

Abstract

Background

Diabetic nephropathy (DN) is the leading cause of end-stage renal disease. Exosomes are promising biomarkers for disease diagnosis and uromodulin is a kidney-specific protein. So, this study was designed to investigate the change in the gene expression of urinary exosomal uromodulin mRNA and urinary uromodulin level and determine the diagnostic potential of these noninvasive biomarkers in the early stage of diabetic nephropathy in type 2 diabetic patients.

Method

This study included 100 participants; urinary exosomes were isolated using polyethylene glycol (PEG). Gene expression of exosomal uromodulin mRNA was determined by quantitative real-time polymerase chain reaction (q-RT-PCR). The urinary uromodulin levels were determined by an enzyme-linked immunosorbent assay (ELISA).

Result

In this study, the gene expression of exosomal uromodulin (UMOD) mRNA and the level of urinary uromodulin showed a significant increase in all diabetic groups with and without nephropathy compared to the control group. The exosomal UMOD mRNA showed a significant positive correlation with urinary uromodulin in all groups. Multiple logistic regression showed that urinary uromodulin was an independent determinant for DN. A diagnostic model of two indicators, exosomal UMOD mRNA and urinary uromodulin, can significantly predict DN. The area under the curve is 0.095, with a 95% confidence interval of 0.98–1, and 0.81, with a 95% confidence interval of 0.69–0.92, for the exosomal UMOD mRNA and urinary uromodulin, respectively.

Conclusion

Urinary exosomal mRNA of UMOD and urinary uromodulin levels are progressively elevated in an early stage of DN, even before the microalbuminuria stage, so they could be used as early predictors for DN.

Introduction

Diabetic nephropathy (DN) is one of the major microvascular complications of diabetes and most frequently results in end-stage renal disease (ESRD). DN is characterized pathophysiologically by an early phase of glomerular hypertrophy, hyperfiltration, and interstitial fibrosis associated with albuminuria (ACR≧30 mg/g creatinine) [1]. Numerous biomarkers have been proposed for DN diagnosis, such as the urinary albumin to creatinine ratio (ACR) which is conventionally used to assess the severity of DN. However, recent studies exhibit conflicting findings regarding the specificity and sensitivity of this biomarker. Notably, advanced pathological changes in glomerular and tubular structure and function may be developed in several patients without a change in urinary albumin level [2]. It can be clearly detected by a kidney biopsy, which is an invasive procedure with a risk of hemorrhage [3]. Therefore, the identification of novel non-invasive biomarkers for the early detection of DN is mandatory to reduce the progression toward ESRD [2].

Uromodulin, also known as Tamm–Horsfall protein (THP), can be considered a biomarker for kidney tubular function as it is a kidney-specific protein that is synthesized mainly in the epithelial cells of the thick ascending limb (TAL) of Henle’s loop and the initial part of the distal convoluted tubule (DCT). Structurally, uromodulin is a 105 kDa glycosyl phosphatidyl inositol (GPI) glycoprotein and the most abundant urinary protein [4]. Uromodulin has a role in the regulation of water-electrolyte balance as it controls Na⁺, K⁺, and Cl⁻ reabsorption by regulating the activity of Na⁺, K⁺, 2Cl⁻ cotransporter (NKCC2) [5], and renal outer medullary K⁺ channel (ROMK2) [6]. It also provides protection against urinary tract infection by binding type I toxin of bacteria E.coli [5], preventing calcium aggregation and kidney stone formation by upregulating calcium channel transient receptor potential cation subfamily V member 5 (TRPV5) [7]. Moreover, it exhibits an immunomodulatory effect by activating granulocytes and mediating monocyte cytotoxicity through binding to toll-like receptor 4 (TLR4) [8]; Conversely, it has an anti-inflammatory effect against ischemia–reperfusion injury (IRI) through cross-talk phenomena [9]. Uromodulin is encoded by the UMOD gene, which is located on chromosome 16p12.3, consisting of 11 exons, with the coding sequences spanning 10 exons starting from exon 2 [10].

Exosomes are specialized compartments of extracellular vesicles (EVs). EVs are lipid bilayers detached from cells and can be categorized into two major groups based on their biogenesis: exosomes (50–150 nm) and microvesicles (MVs) (150–1000 nm) [11]. Exosomes are released by the budding of endosomes into multivesicular bodies (MVBs), which contain several intraluminal vesicles (ILVs). Upon fusion with the plasma membrane, exosomes are released into extracellular space [12, 13]. MVs are formed by direct outward budding and shedding from the cell membrane [14].

Urinary exosomes have emerged as a promising biomarker, originating from diverse cell types within the kidney. These vesicles encompass an array of constituents, including proteins and genetic materials (miRNA, DNA, and RNA), carbohydrates, and lipids. This content varies based on the cell's origin, thereby reflecting the pathological state of the cells. In the early stage of DN, hypoxia-inducible factor-1 increases the production of exosomes from kidney tubules, so they could provide a diagnostic biomarker without invasive biopsy [15]. The mRNA of UMOD was extracted from the urinary exosomes because it is more stable as it is protected by the lipid bilayer membrane of the exosomes [16].

EVs have recently been studied as a new form of cellular communication; moreover, they have a critical role in disease prognosis and diagnosis [16], as they are produced from different cell types and presence in various body fluids such as blood, saliva, urine, breast milk, cerebrospinal fluid, amniotic fluid, and ascitic fluid [17]. EVs participate in cell-to-cell communication by transferring RNA from one cell to another, aiding in the design of new therapeutic protocols by modifying RNA and delivering it to the target cells [18]. Conversely, they also contribute to the development of several diseases, such as cardiovascular disease, neurodegenerative disease, and cancer metastasis, by transferring signals and programming the target cells [19]. The objective of this study was to evaluate the urinary exosomal UMOD mRNA and urinary uromodulin level as potential biomarkers for diabetic nephropathy in type 2 diabetic patients.

Subjects

This is a case–control study that was carried out on 100 subjects matched for age and sex, with ages ranged from 40 to 65 years. 25 were healthy subjects as a control group (GR I), and 75 subjects diagnosed with type 2 diabetes (T2DM) were recruited from the Internal Medicine Department, Tanta University Hospital. The T2DM patients were further categorized into three groups based on criteria that were set by the American Diabetes Association according to the albumin-to-creatinine ratio (ACR) [20], normoalbuminuria (n = 25, ACR < 30 mg/g creatinine) (GR II), microalbuminuria (n = 25, ACR 30–300 mg/g creatinine) (GR III), and macroalbuminuria (n = 25, ACR > 300 mg/g creatinine) (GR IV). DN was identified by elevated urinary albumin-to-creatinine ratio (ACR ≥ 30 mg/g creatinine). Patients with the existence of non-diabetic kidney diseases, urinary tract infections, liver diseases, autoimmune, inflammatory, or neoplastic diseases were excluded.

Samples collection

20 ml of first-morning urine samples were collected from each participant. The samples were centrifuged at 3000 xg for 10 min at 4 ℃ to remove any cell debris. 3 ml was used to urinary estimate ACR, and 15 ml of each sample was stored at – 80 ℃ until analysis. After 10 h of overnight fasting, 6 ml of blood samples were collected for the biochemical parameters assay, and 1 ml of venous blood was drawn from each subject 2 h after the meal for estimation of postprandial blood glucose level (PBG).

Biochemical assay

HbA1c, fasting and postprandial blood glucose, ACR, cystatin-C, creatinine, blood urea nitrogen (BUN), and lipid profile were estimated by commercial kits supplied by Dialab (Dialab, Austria) according to the manufacturer’s instructions. Serum Na⁺ and K⁺ were estimated by the electrolyte analyzer Rapid Chem 744 (Siemens, Germany) according to the manufacturer’s instructions. eGFR was based on creatinine and calculated by the Modification of Diet in Renal Disease equation (MDRD).

Estimation of urine uromodulin level

Urinary uromodulin was determined by a sandwich ELIZA kit supplied by (Cloud-Clone, USA, catalog No. SEG918Hu) according to the manufacturer’s instructions and it was normalized by urinary creatinine (uUCR).

Exosomes isolation

Exosomes were isolated by polymer precipitation using polyethylene glycol (PEG) with a molecular weight of 6000 Da (Sigma, USA). 25 g of PEG and 1 M NaCl were dissolved in 50 ml of distilled water and ultrasonicated for 10 min. The PEG solution was centrifuged at 5000 xg for 20 min at 4 ℃. Mother liquor (50% + 1 M NaCl).

Urine samples that were stored at − 80 °C were brought out, thawed on ice, and vortexed for 90 s. 15 ml of each sample was mixed with PEG to a final concentration of 10% of PEG. The mixture was mixed by inverting the tube several times to mix well. After overnight incubation at 4 ℃, the samples were centrifuged at 4000 xg for 1 h at 4 ℃ to pellet the exosomes, and the supernatant was removed and centrifuged again for 5 min at 4 ℃ to remove any residual supernatant. The exosome pellet was resuspended in 200 μl of phosphate buffer saline (PBS) for RNA extraction [21, 22].

Urinary exosome identification

A. Exosome morphology and size characterization

Exosome morphology and size were visualized by scanning electron microscopy (SEM). The PEG-extracted exosome pellet was fixed with 2% (v/v) glutaraldehyde in PBS overnight. After being washed by PBS three times, it was exposed to gradual dehydration with ethanol (30, 50, 70, and 95%) for 10 min at each concentration. Subsequently, the sample was air-dried at room temperature. After that, the sample was coated with gold (JFC-1100E ION SPUTTERING DEVICE, JEOL, Japan). The coated sample was observed using a scanning electron microscope, JSM - IT200 (JEOL, Japan).

b. Feasibility of exosomes for RNA analysis

The feasibility of exosomes for downstream analysis of RNA was confirmed by RNA extraction. This involved a systematic process to isolate RNA from the exosome samples, ensuring that the extracted RNA was of sufficient quality and quantity for subsequent analyses.

qRT-PCR detection

Total RNA from urinary exosomes was extracted by (Geneaid) extraction kit according to the manufacturer’s instructions. The exosomal RNA purity was evaluated spectrophotometrically by nano-volume SCANDROP 200 (Analytik-Jena, Thuringia, Germany) at absorbances of 260 and 280 nm. A 260/280 ratio was used to assess the presence of protein contaminants. cDNA was synthesized by the TOP script M-MLV RT Dry Mix Kit (Enzynomic). Primers were designed using the IDT primer quest for two distinct genes: uromodulin and beta-actin (ACTB). The primer sequences for uromodulin (forward: GACAGTGTTGACGAGGAATGA, reverse: GATGTTGAGGTCACGGATGAT). Likewise, for ACTB (forward: GACCCAGATCATGTTTGAGACC, reverse: CCAGAGGCGTACAGGGATA). Real-time PCR was performed on a Rotor-Gene Q (Qiagen, Valencia, CA) instrument, and ACTB was used as a housekeeping gene to normalize the obtained data [23]. The instrument uses the following profiles: initial denaturation at 95 ℃ for 10 s, annealing at 60 ℃ for 15 s, and elongation at 72 ℃ for 30 s. Raw threshold cycle (Ct) values were obtained by rotor-gene version 2.0.3.2 software (Qiagen, Valencia, CA). Results were analyzed by relative quantification with the comparative Ct method.

Statistical analyses

All statistical analyses were performed with SPSS software 25 (SPSS, Chicago, Il, USA). All data were analyzed for normality using the Shapiro–Wilk test. The biochemical parameters were represented as mean ± standard deviation (SD) and compared statistically using one-way ANOVA followed by Games–Howell multiple comparison tests to determine statistical significance between various groups. For non-parametric comparison of UMOD mRNA fold change expressed as median (interquartile range) (IQR), the Kruskal–Wallis test was used to evaluate the difference between groups.

Pearson’s correlation was used to determine the relationship between urinary uromodulin and other parameters, and Spearman's non-parametric rank was used to determine the correlation between urinary exosomal UMOD mRNA, urinary uromodulin level, and other parameters.

Receiver operator characteristic curve analysis (ROC) was performed, and the area under the curve (AUC) was calculated to evaluate the diagnostic accuracy of the UMOD gene and urinary uromodulin. Multivariant logistic regression was performed to determine the association between urinary uromodulin and the presence of DN.

Results

The biochemical parameters

Clinical and biochemical parameters of subjects who participated in this study are summarized in Table 2, which illustrates that fasting blood glucose (FBG), postprandial blood glucose (PBG), and glycosylated hemoglobin (HbA1C) were significantly increased in all diabetic groups with and without nephropathy compared with the healthy control group (p < 0.001). ACR was significantly elevated in all diabetic groups compared to the healthy control group (p < 0.001). Specifically, cystatin-C and creatinine levels exhibited a significant increase in individuals with microalbuminuria and macroalbuminuria compared to the healthy control group (p < 0.001), while eGFR showed a significant decrease in microalbuminuria and macroalbuminuria compared to the healthy control group (p < 0.001).

Table 2.

Correlation of urinary uromodulin and urinary uromodulin/creatinine ratio with other kidney function biomarkers

		GR II Normoalbuminuria n = 25		GR III Microalbuminuria n = 25		GR IV Macroalbuminuria n = 25
		Urinary uromodulin	Urinary uromodulin/creatinine	Urinary uromodulin	Urinary uromodulin/creatinine	Urinary uromodulin	Urinary uromodulin/creatinine
Serum Creatinine (mg/dl)	r	0.481	0.319	0.682	0.483	− 0.654	− 0.422
	p	0.007**	0.06*	0.001**	0.007**	0.001**	0.018*
BUN (mg/dl)	r	0.381	0.273	0.341	0.405	− 0.344	− 0.219
	p	0.03*	0.093	0.048*	0.022*	0.046*	0.146
eGFR (ml/min/1.73m2)	r	− 0.346	− 0.536	− 0.669	− 0.652	0.393	0.361
	p	0.045*	0.003**	0.001**	0.001**	0.025*	0.038*
Serum cystatin-C (mg/l)	r	0.642	0.344	0.593	0.663	− 0.684	− 0.358
	p	0.001**	0.046*	0.001**	0.001**	0.001**	0.039*
ACR (mg/g)	r	0.796	0.498	0.915	0.769	− 0.913	− 0.468
	p	0.001**	0.006**	0.001**	0.001**	0.001**	0.01*
Serum Na+ (mmol/l)	r	0.931	0.470	0.786	0.479	0.875	0.459
	p	0.001**	0.009**	0.001**	0.008**	0.001**	0.01**
Serum K+ (mmol/l)	r	− 0.986	− 0.425	− 0.343	− 0.122	− 0.951	− 0.636
	p	0.001**	0.017*	0.046*	0.280	0.001**	0.001**

Significance at p-value (< 0.05)

r Pearson’s correlation, ACR albumin to creatinine ratio, eGFR estimated glomerular filtration rate, BUN blood urea nitrogen

Urinary uromodulin

The results showed a significant increase across all diabetic groups with and without nephropathy compared to the healthy control group (p < 0.001), while there was a significant decrease in the macroalbuminuria group compared to normoalbuminuria and microalbuminuria (p < 0.001) (Table 1, Fig. 1a). Furthermore, upon normalization of urinary uromodulin by urinary creatinine (uUCR), a significant increase was observed in all diabetic groups, with and without nephropathy, as compared to the healthy control group (p < 0.001) while there was a significant decrease in the macroalbuminuria group compared to normoalbuminuria and microalbuminuria (p < 0.001) (Table 1, Fig. 1b).

Table 1.

Clinical and biochemical parameters of the studied groups

Parameter	GR I Control n = 25	GR II Normoalbuminuria n = 25	GR III Microalbuminuria n = 25	GR IV Macroalbuminuria n = 25	p-value
Age	50 ± 5.31	52.64 ± 6.25	52.76 ± 7.64	55.24 ± 6.25	0.051
Sex (f/m %)	15/10 (60/40)	16/9 (64/36)	14/11 (56/44)	12/13 (48/52)	0.699
Diabetes duration (years)	–	6.72 ± 1.54	8.76 ± 2b	11.2 ± 2.61bc	< 0.001*
SBP (mmHg)	120.7 ± 8	127.5 ± 11	134 ± 9ab	138 ± 10.5abc	< 0.001*
DBP (mmHg)	69.5 ± 4.9	72 ± 5.3	78.2 ± 7.2ab	82.8 ± 7.9abc	< 0.001*
FBG (mg/dl)	86.91 ± 6.98	207.56 ± 58.7a	229.5 ± 66.7a	233.4 ± 40.6a	< 0.001*
PBG (mg/dl)	113.78 ± 12.37	301 ± 68.4a	325 ± 63.8a	334.32 ± 54.54a	< 0.002*
HbA1C (%)	5.1 ± 0.2	8.7 ± 1.4a	10.1 ± 1.2ab	10.4 ± 1.09ab	< 0.001*
ACR (mg/g creatinine)	4.7 ± 1.83	18.49 ± 4.9a	115.33 ± 59.4ab	717.3 ± 163.55abc	< 0.001*
Cystatin-C (mg/l)	0.79 ± 0.1	0.86 ± 0.14	1.11 ± 0.26ab	2.3 ± 0.61abc	< 0.001*
eGFR (ml/min/1.73m2)	88.03 ± 13.02	77.61 ± 14.8	68.6 ± 15a	44.2 ± 15.8abc	< 0.001*
Creatinine (mg/dl)	1.016 ± 0.13	1.09 ± 0.15	1.25 ± 0.21ab	1.99 ± 0.65abc	< 0.001*
BUN (mg/dl)	14 ± 2.6	17.19 ± 4.21a	19.06 ± 5.66a	38 ± 16abc	< 0.001*
Na+ (mmol/l)	14 0.5 ± 2.5	142.46 ± 2a	144.52 ± 3.1ab	141.54 ± 4.77	< 0.001*
K+ (mmol/l)	4.18 ± 0.26	4.09 ± 0.29	4.36 ± 0.56	4.48 ± 0.88	0.068
Cholesterol (mg/dl)	183.26 ± 17.28	208.83 ± 32.95a	221.4 ± 45.35a	239.16 ± 64.81a	< 0.001*
Triglyceride (mg/dl)	127.48 ± 20.15	160 ± 29.6a	171.9 ± 42a	182.72 ± 59.76a	< 0.001*
HDL-C (mg/dl)	50.64 ± 6.55	45 ± 7.22a	42 ± 5.7a	37.2 ± 5.26abc	< 0.001*
LDL-C (mg/dl)	107.12 ± 20.6	131.81 ± 42.46a	145 ± 50.95a	165.41 ± 76a	< 0.001*
Urinary uromodulin (μg/ml)	18.04 ± 2.72	25.24 ± 3.54a	33.05 ± 7.98ab	20.82 ± 2.55abc	< 0.001*
uUCR (mg/g creatinine)	12.4 ± 5.5	26.3 ± 10.4a	40.6 ± 25.7ab	24.5 ± 17.6abc	< 0.001*
Exosomal UMOD gene	0.51(0.21–0.9)	2.87(2.51–3.7)a	4.75(3.82–9.7)ab	3.7(2.54–6.7)ac	< 0.001*

Clinical parameters represented as mean ± S.D., exosomal UMOD gene represented as median (IQR), significance at p-value (< 0.05), ^ap < 0.05 vs. control group, ^bp < 0.05 vs. normoalbuminuria group (GR II), ^cp < 0.05 vs. microalbuminuria group (GR III)

SBP systolic blood pressure, DBP diastolic blood pressure, ACR albumin to creatinine ratio, eGFR estimated glomerular filtration rate, BUN blood urea nitrogen, HDL high-density lipoprotein, LDL low-density lipoprotein, uUCR urinary uromodulin to creatinine ratio

Fig. 1.

a Comparison of urinary uromodulin levels (μg/ml) between the studied groups; b Comparison of urinary uromodulin to creatinine ratio (uUCR) levels (mg/g creatinine) between the studied groups. Values are expressed as mean ± SD, with significance at p-value < 0.05; ^ap < 0.05 showed significance vs. the control group; ^bp < 0.05 showed significance vs. the normoalbuminuria group; ^cp < 0.05 showed significance vs. the microalbuminuria group

Identification of urinary exosomes

Morphology and size visualization are crucial for exosome characterization. SEM images revealed that the average vesicle diameter ranged from 72 to 119 nm (Fig. 2).

Fig. 2.

Scanning electron microscopy for exosomes isolated by PEG

The efficiency and specificity of amplified RNA were confirmed by RNA extraction from urinary exosomes through the melt curve and amplification plot. The RNA concentration was 21.5 μg/ml and the OD 260/280 was 1.9, as illustrated in Fig. 3a. qRT-PCR was performed for the urinary exosomal ACTB and UMOD genes. The melt curve and amplification plot (Fig. 3b, c) provided evidence supporting the suitability of exosomes for downstream analysis.

Fig. 3.

RNA extracted from urinary exosomes and qRT-PCR of urinary exosomal ACTB and UMOD genes a RNA concentration and OD 260/280. b Amplification plot of urinary exosomal ACTB and UMOD mRNA. c The melt curve showed a single-beak and rapid curve descent, which proved the specificity of amplified

Urinary exosomal UMOD mRNA

It showed a significant upregulation in gene expression in all diabetic groups compared to the control group (p < 0.001), whereas there was a significant decrease in the macroalbuminuria group compared to normoalbuminuria group (p < 0.001) (Table 1, Fig. 4).

Fig. 4.

Fold change of exosomal UMOD gene expression in the studied groups. Values are expressed as median (IQR), with significance at p-value < 0.05; ^ap < 0.05 showed significance vs. the control group; ^bp < 0.05 showed significance vs. the normoalbuminuria group; ^cp < 0.05 showed significance vs. the microalbuminuria group

Correlation of urinary uromodulin with other biomarkers

This study revealed notable correlations of urinary uromodulin with various biomarkers. In the normoalbuminuria and microalbuminuria groups, urinary uromodulin showed a significant positive correlation with ACR, serum creatinine, BUN, cystatin-C, serum Na⁺, while exhibiting a significant negative correlation with e-GFR and serum K⁺. Whereas in the macroalbuminuria group, there was a significant negative correlation with ACR, serum creatinine, BUN, cystatin-C, serum K⁺, and a significant positive correlation with eGFR and serum Na⁺. The correlation coefficient (r) is illustrated in Table 2. Scatter plot figures are illustrated in suppl  Figs. 1, 2.

Furthermore, upon adjusting urinary uromodulin levels by urinary creatinine, a substantial positive correlation was observed with ACR, cystatin-C, and serum Na⁺ levels. Simultaneously, a significant negative correlation was noted with eGFR in both the normoalbuminuria and microalbuminuria groups. While, within the macroalbuminuria group, a significant negative correlation was found with ACR, serum creatinine, cystatin-C, and serum K⁺ levels. Conversely, there was a significant positive correlation with eGFR and serum Na⁺ levels. However, in the normoalbuminuria group, no significant correlation was detected with serum creatinine or BUN. Similarly, in the microalbuminuria group, there was no significant correlation with serum K⁺ levels (Table 2).

Correlation of the urinary exosomal UMOD gene with urinary uromodulin and other biomarkers

In the normoalbuminuria and microalbuminuria groups, the exosomal UMOD mRNA showed a significant positive correlation with urinary uromodulin, ACR, serum creatinine, cystatin-C, serum Na⁺, and a significant negative correlation with eGFR and serum K⁺, while in the macroalbuminuria group, it showed a significant positive correlation with urinary uromodulin, eGFR, serum Na⁺, and a significant negative correlation with ACR, serum creatinine, BUN, cystatin-C, and serum K⁺ (Table 3). Scatter plot figures are illustrated in suppl Fig. 3.

Table 3.

Correlation of urinary exosomal UMOD gene with urinary uromodulin and other kidney function biomarkers

		GR II Normoalbuminuria n = 25	GR III Microalbuminuria n = 25	GR IV Macroalbuminuria n = 25
uUromodulin (μg/ml)	r	0.567	0.898	0.934
	p	0.002**	0.001**	0.001**
uUCR (mg/g creatinine)	r	0.341	0.606	0.464
	p	0.48*	0.001**	0.01**
Serum Creatinine (mg/dl)	r	0.733	0.467	− 0.571
	p	0.001**	0.009**	0.001**
BUN (mg/dl)	r	0.275	0.203	− 0.577
	p	0.092	0.156	0.001**
eGFR (ml/min/1.73m2)	r	− 0.544	− 0.676	0.400
	p	0.002**	0.001**	0.024*
Serum Cystatin-C (mg/l)	r	0.690	0.714	− 0.694
	p	0.001**	0.001**	0.001**
ACR (mg/g)	r	0.682	0.806	− 0.859
	p	0.001**	0.001**	0.001**
Serum Na+ (mmol/l)	r	0.520	0.695	0.867
	p	0.004**	0.001**	0.001**
Serum K+ (mmol/l)	r	− 0.552	− 0.101	− 0.905
	p	0.002**	0.083	0.001**

Significance at p-value < 0.05

r Spearman’s correlation, uUCR urinary uromodulin to creatinine ratio, ACR albumin to creatinine ratio, eGFR estimated glomerular filtration rate, BUN blood urea nitrogen, uUromodulin urinary uromodulin.

Multivariant logistic regression

Multivariant logistic regression was performed to identify the independent predictor for DN. The model showed that there was an association between urinary uromodulin and DN, and uromodulin is an independent predictor of DN (β = 0.03, p = 0.01), as demonstrated in Table 4.

Table 4.

Multivariant logistic regression analysis of independent predictors of nephropathy among type 2 diabetic patients

Parameters	β	Exp Β	95% CI for Exp B		p-value
			Lower limit	Upper limit
Urinary uromodulin (μg/ml)	0.03	1.03	1.01	1.04	0.01*
eGFR (ml/min/1.73m2)	0.004	1	0.99	1.01	0.23
Sex	− 0.1	0.11	0.72	1.11	0.33
Age (years)	0.006	1	0.98	1.02	0.43
Diabetes duration (years)	0.035	1.035	0.99	1.07	0.1

Significance at p-value < 0.05

β Logistic regression coefficient, Exp Β estimated odds ratio, CI confidence interval, eGFR estimated glomerular filtration rate

Receiver operating curve (ROC)

ROC was performed to analyze the ability of exosomal UMOD gene expression and urinary uromodulin to diagnose diabetic patients with and without nephropathy. For UMOD mRNA showed an area under the curve (AUC) of 0.95 (p < 0.001, 95% CI 0.98–1) (sensitivity 92%, specificity 84%), urinary uromodulin showed an AUC of 0.81 (p < 0.001, 95% CI 0.69–0.92) (sensitivity 80%, specificity 60%), while cystatin-C showed an AUC of 0.77 (sensitivity 76%, specificity 56%), and eGFR showed an AUC of 0.68 (sensitivity 72%, specificity 68%) (Fig. 5).

Fig. 5.

ROC curve analysis the diagnostic ability of different parameters to diagnose diabetic patients with and without nephropathy

Discussion

Diabetic nephropathy is the most serious microvascular complication, which causes glomerular and tubular damage that results in renal failure [24]. The precise pathological mechanisms triggering DN remain not fully identified. Hyperglycemia, a hallmark of diabetes, instigates the production of advanced glycation end products (AGEs), thereby stimulating an excessive release of profibrotic cytokines and augmenting reactive oxygen species (ROS) within the renal cells. Furthermore, elevated expression of chemoattractant and adhesion molecules initiates the recruitment of inflammatory cells to kidney tissues, promoting hyperfiltration, sclerosis, and increased inflammatory responses. [25]. Early detection of DN is pivotal in impeding the progression of kidney function decline towards ESRD. Therefore, the identification of novel noninvasive biomarkers for early detection of DN before advancement toward ESRD is crucial [1].

Uromodulin, a kidney-specific protein encoded by the UMOD gene, serves as a biomarker indicative of tubular function [5]. The UMOD mRNA extracted from urine exosomes, originating from kidney cells, represents a valuable diagnostic biomarker due to their enriched genetic material and protein content [26]. Urinary exosomes were isolated using a polymer precipitation technique (PEG). This cost-effective method produces a high yield of exosomes by reducing their solubility and enabling their precipitation through low-speed centrifugation without the need for specialized equipment [27].

The data of this study revealed a notable up-regulation in the expression of urinary exosomal UMOD gene in diabetic patients with normoalbuminuria and microalbuminuria when compared to the control group. This finding aligns with the studies by Yamamoto et al. and Micanovic et al., which documented that uromodulin expression increases at an early stage of diabetes [23, 28]. We hypothesized that the observed increase in UMOD gene expression among diabetic individuals could potentially be attributed to the inflammatory response associated with interstitial fibrosis—the primary pathophysiological mechanism in DN. The precise role of urinary uromodulin during the inflammatory response is not fully elucidated. Indeed, some studies support the notion that the increase in uromodulin is associated with its pro-inflammatory effect. Research conducted by Schmid et al. and Säemann et al. provides evidence of uromodulin's ability to activate innate immunity, including monocytes, neutrophils, and dendritic cells, and facilitate neutrophil trans-epithelial migration by binding TLR4 [8, 29]. Conversely, other studies propose an immunosuppressive and protective effect of uromodulin. This protective effect was explained by EL-Achkar et al., who demonstrated uromodulin's role in mitigating inflammation by inhibiting signaling in pro-inflammatory cells. During the recovery phase of acute kidney injury (AKI), UMOD gene expression is upregulated. This upregulation is associated with the regulation of macrophage inflammatory protein-2 (MIP-2), a potent neutrophil chemoattractant, thereby reducing neutrophil infiltration and mitigating the inflammatory response [30]. Furthermore, uromodulin’s protective effects encompass its ability to bind lymphokines like interleukin-1 (IL-1) and tumor necrosis factor-α (TNF-α). Additionally, uromodulin can bind to complement 1q, inhibiting the activation of the classical complement pathway [31, 32], and it is considered an antioxidant molecule as it shows a role in the regulation of renal oxidative stress by suppressing the activity of the transient receptor potential cation channel, subfamily M, member 2 (TRPM2) [33]. Moreover, the excreted urinary uromodulin acts as a decoy receptor for the free urinary cytokines, preventing cytokines from over-activating renal epithelial cells and urothelial cells [31].

In addition, the expression of uromodulin is highly dynamic and susceptible to change due to pathological conditions, as it is controlled by several transcriptional factors [34]. In this study, exosomal mRNA expression is positively associated with urinary uromodulin levels.

In the current study, a notable elevation in the mean urinary uromodulin level was evident across all diabetic groups in comparison to the control group. Conversely, within the diabetic spectrum, the macroalbuminuria group exhibited a decrease in uromodulin levels compared to both the microalbuminuria and normoalbuminuria groups. The higher production of uromodulin is reflected by higher levels in urine, and we assume that the increase in uromodulin excretion may be due to hyperfiltration associated with the early stage of diabetes This hyperfiltration is linked to tubular injury caused by hyperglycemia, resulting in extracellular matrix (ECM) accumulation and subsequent loss of basement membrane integrity, whereas the decreased level reflects the advanced disease stage and reduction of the renal cell mass. Consistent with our findings, Lou et al. reported elevated level of uromodulin with DKD [35]. Similarly, Köttgen et al. observed an upsurge in urinary uromodulin concentration preceding the onset of chronic kidney disease (CKD) [36]; and El-Achkar et al. illustrated that 24-h excretion of uromodulin increased in patients with early kidney disease without eGFR impairment, but the excretion decreased at an advanced stage when eGFR was significantly reduced [37]. On the contrary, some studies showed contradictory results, as Chakraborty et al. reported a decrease in uromodulin excretion with kidney damage [38]. However, these studies were all constrained by the absence of ROC curve analysis, presenting a deficiency in the data related to diagnostic accuracy.

The present study revealed a strong correlation of exosomal UMOD mRNA and urinary uromodulin with other kidney function biomarkers such as ACR, cystatin-C, and eGFR. These findings are aligned with the studies conducted by Lou et al. and Yamamoto et al. [23, 35]. Furthermore, our data showed high diagnostic potency for exosomal UMOD mRNA and urinary uromodulin, as estimated by the ROC curve, with high sensitivity and specificity. For exosomal UMOD mRNA, the AUC was 0.95 (sensitivity 92%, specificity 84%), and for urinary uromodulin, the AUC was 0.81 (sensitivity 80%, specificity 60%) compared to cystatin-C and eGFR, which showed lower AUC, sensitivity, and specificity.

Conclusion

The present study found an increased level of uromodulin excretion and increased expression of the urinary exosomal UMOD gene at an early stage of diabetes, even before the microalbuminuria stage, with high diagnostic performance and high sensitivity as shown by the ROC curve, so we hypothesized that patients with early kidney damage have an elevated level of urinary uromodulin and have a higher chance of developing CKD, so urinary uromodulin and the urinary exosomal UMOD gene may be used as early predictors of DN and monitor the progression of CKD.

Author contributions

SIB: performed the practical part, data analysis, and wrote the manuscript. TMM: supervision, guidance throughout the work, reviewing, and editing. SSB: supervision, reviewing, and editing. EMA: supervision, reviewing, and editing. All authors have read and approved the final manuscript.

Funding

This research did not receive any specific grant from any funding agencies.

Data availability

The datasets used during the current study are available from the corresponding author upon reasonable request.

Declarations

Conflict of interest

The authors declared that there is no conflict of interest.

Ethics approval and consent to participate

This study was approved by the Ethical Committee of Tanta University Hospital number (33229/07/19). Informed consent was obtained from each participant.

Consent for publication

Not applicable.