Association of significantly elevated plasma levels of NGAL and IGFBP4 in patients with diabetic nephropathy

A total of 159 Kuwaiti participants were enrolled in this study. The study involved three main groups: 67 participants with diabetic nephropathy (DN), 50 with Type 2 diabetes (T2D) and no kidney disease and 42 non-diabetic individuals matched for age and body mass index (BMI). Clinical diagnosis of T2D was based on having persistent hyperglycemia (fasting glucose level > 7 mmol/L and 2-h FBG > 11 mmol/L) with normal kidney function. People with DN were clinically diagnosed by a nephrologist according to the American Diabetes Association criteria [21]. These patients showed pronounced T2D and persistent elevation in ACR > 30 mg/g and/or sustained reduction in estimated glomerular filtration rate (eGFR; < 60 ml/min per 1.73 m2). Patients’ exclusion criteria included: 1) Non-diabetic kidney disease, 2) Chronic liver disease, 3) Heart failure, 4) Current/recent infection, 5) Acute/chronic inflammatory disease, 6) Allergic condition, 7) Autoimmune disease, 8) Malignancy. Non- diabetic participants, who underwent routine medical check-ups at the Amiri Hospital, Kuwait, and had no history of T2D or DN and were not diagnosed with any other medical conditions were recruited as controls. This study was approved by the Ethical Review Committee of Dasman Diabetes Institute and were in accordance with the guidelines of the Declaration of Helsinki. All participants were recruited at Dasman Diabetes Institute (Dasman, Kuwait) and gave written informed consents before enrollment in the study.

Blood and urine samples were obtained from participants under fasting conditions (8–12 h). Blood samples were collected in vacutainers containing Ethylenediaminetetraacetic acid (EDTA) and centrifuged at 400×g for 10 min. The Plasma was separated, aliquoted and stored at − 80 °C for further analysis. First void urine samples were collected in 120 mL urine collection tubes in the morning. The blood pressure readings presented are the average of three consecutive measurements, with a 10-min rest period between each reading (digital sphygmomanometer, Omron HEM-907XL Digital). Fasting blood glucose (FBG), triglyceride (TG), total cholesterol (TC), low-density lipoprotein (LDL), and high-density lipoprotein (HDL) were quantified using the chemistry analyzer (Siemens Dimension RXL; Diamond Diagnostics, Holliston, MA). Concentrations of urinary albumin “spot” and creatinine in addition to urinary albumin-to-creatinine ratio were measured using an automated analyzer (CLINITEK Novus Automated Urine Chemistry Analyzer; Siemens Healthineers, Erlangen, Germany). Fully automated renal function tests (RFTs) were performed using a VITROS 250 automatic analyzer. The estimated glomerular filtration rate (eGFR) was calculated using the modification of diet in renal disease study (MDRD) equation [22].

Stored plasma samples were thawed and centrifuged at 10,000×g for 5 min at 4 °C to remove any debris. The level of NGAL, IGFBP-1, − 3, and − 4 were measured using customized Human Magnetic Luminex Assays (LXSAHM-11 and LXSAHM-13; R&D Systems Europe, Ltd., Abingdon, UK). As per the manufacturer’s assurance, the Luminex assay kits are accurate and consistent. They perform rigorous assay validation for specificity, precision, stability, linearity and recovery to ensure high quality results. The plasma samples were diluted 50× with sample diluent for the multiplex assay to detect NGAL. To detect the different IGFBPs, plasma was diluted 2x with sample diluent. The assays were performed following manufacturer’s instructions. The results were obtained by running the assays on the Bio-Plex 200 system (Bio-Rad, CA, USA), and the Bio-Plex manager software was used to quantify the concentration of each analyte through the generated standard curve. It has been reported that the reference range for NGAL in healthy male population aged 41–60 yrs. is 103.95 ± 13.39 to 123.7 ± 15.94 μg/L. While, the reference range in healthy females is 158.37 ± 20.44 to 163.62 ± 21.12 μg/L [23]. The detection limits for IGFBP4 are 70–51,300 pg/mL and for NGAL the detection limits are 38–28,130 pg/mL. For both assays, results were calculated using the 5 PL non-linear setting in the Bioplex software with a confidence range between 75 and 110%. The standard values in the assays show 95–100% confidence range. The inter and intra-assay CVs for IGFBP4 was < 3% and the inter and intra-assay CV for NGAL was ≤5%.

In this study, the level of NGAL in plasma was investigated. The data showed a significant increase in plasma level of NGAL in people with DN (92.76 ± 7.5 mg/L, Table 1) compared to people with T2D (57.22 ± 8.7 mg/L; p = 0.002, Table 1) and non-diabetic individuals (52.47 ± 2.9 mg/L; p = 0.001, Table 1) (Fig. 1A). The levels of different IGFBP proteins were also analyzed. The results showed a significant increase in the level of IGFBP-4 in people with DN (795.61 ± 130 ng/ml) compared to both, people with T2D (374.56 ± 86.8 ng/ml, p = 0.013) and non-diabetic group (273.06 ± 27.8 ng/ml, p = 0.003) (Table 1, Fig. 1D). On the other hand, a significant increase in the level of IGFBP-1was seen among the DN (34.85 ± 3.3 mg/l) when compared to people from the non-diabetic group (19.01 ± 2.2, p = 0.008). However, no significant difference was observed between DN and T2D groups (31.38 ± 4.7 mg/l) (Table 1, Fig. 1B) IGFBP-3 expression levels showed no significant differences between the three groups (Table 1, Fig. 1C).

Spearman’s correlation analysis showed a significant association between NGAL and various markers of kidney function (Table 2) within the DN group. The data showed that NGAL was positively correlated with SCr (ρ = 0.53, p <  0.001, Fig. 2A), while it correlated negatively with eGFR (ρ = − 0.552, p <  0.001, Fig. 2B). A positive correlation was also observed between NGAL and the blood urea nitrogen (BUN) (ρ = 0.608, p <  0.001, Fig. 2C). Interestingly, the analysis revealed a significant correlation between NGAL and IGFBP-4 in people with DN (ρ = 0.62, p <  0.001, Fig. 3C). No significant associations between NGAL and IGFBP1/3 (Fig. 3A and B respectively). In a similar manner, NGAL showed a significant association with parameters of kidney function in people with T2D (Table 2). The correlation was positive with SCr (ρ = 0.63, p <  0.001), and BUN (ρ = 0.577, p <  0.001) but negative with eGFR (ρ = − 0.707, p <  0.001). There was no significant correlation between NGAL and IGFBP-4 in people with T2D (Table 2). Furthermore, no significant association between NGAL and urinary proteins was found.

Spearman’s rank correlation analysis was performed to study the correlation between the IGFBPs and the various indicators of renal activity. A significant correlation was observed between IGFBP-4 and the indicators of renal activity within the DN group. This was presented through a significant positive correlation with serum creatinine (ρ = 0.39, p <  0.001), BUN (ρ = 0.32, p <  0.05), and a negative correlation with eGFR (ρ = − 0.44, p <  0.001). In the case of IGFBP-1, the data showed a significant negative correlation with both sCr (ρ = − 0.249, p <  0.05) and BUN (ρ = − 0.371, p <  0.001) in people with DN, whereas eGFR showed no association with IGFBP-1 in the same group (Table 3). No association was found between IGFBP-3 and the parameters reflecting kidney function in people with DN (Table 3). Furthermore, the relationship between IGFBP4 and urinary protein excretion (urine creatinine, microalbumin and ACR) was assessed. IGFBP4 was found to associate significantly only with urine Creatinine in the DN group (ρ = − 0.289, p = 0.021, Supplementary Table 1).

Receiver Operating Characteristic (ROC) analysis was performed to determine the probable use of NGAL and IGFBP4 markers to differentiate between diabetic patients with or without diabetic nephropathy. NGAL and IGFBP4 markers show moderate diagnostic accuracy considering the ROC analysis results and the fact that AUC for both markers were 76.2 and 74.3 respectively (Fig. 4A and B). Moreover, ROC curve analyses showed significant predictive power of the ACR and SCr for NGAL and IGFBP4 markers, with SCr exhibiting high accuracy (supplementary Fig. 1A and B) and ACR exhibiting low accuracy (Supplementary Fig. 1C and D). The AUCs, p-values, cut-off points optimized for sensitivity and specificity, 95% CIs, sensitivities, and specificities obtained for SCr and ACR are reported in Supplementary Table 2. The ROC analysis results were interpreted as follows: AUC < 0.70, low diagnostic accuracy; AUC in the range of 0.70–0.90, moderate diagnostic accuracy; and AUC ≥0.90, high diagnostic accuracy. We also created ROC curve for serum creatinine as one of the conventional circulatory markers to assess the diagnostic utilities of the NGAL and IGFBP4 circulatory molecules (Supplementary Fig. 2).

The study groups namely non-diabetic, T2D and DN were further categorized based on creatinine levels into 2 sub-categories as normal (Male <=119.3; Female <=91.9) and high (Male > 119.3; Female > 91.9). Plasma levels for both NGAL and IGFBP4 seem to be increased in both subgroups. However, their increased in levels were more pronounced in the high creatinine subgroup (Supplementary Table 3). Moreover, and within the normal creatinine subgroup, it was evident the increased of both NGAL and IGFBP4 levels in the DN group compared to both DM and non-diabetic ones supporting the utility of both markers for early diagnosis of DN.