Inflammation plays an important role in albuminuria in type 2 diabetes mellitus (T2DM). The vitamin D receptor (VDR) has potent anti-inflammatory activities.
To investigate the correlation between VDR expression and albuminuria in T2DM.
Renal biopsies from T2DM patients with albuminuria (n = 8) and nondiabetic subjects (n = 4) were compared for VDR expression by immunohistochemistry. Recruited T2DM patients (n = 242; estimated glomerular filtration rate > 60 mL/min/1.73 m2) were divided into three groups based on urinary albumin-to-creatinine ratio (uACR): normal albuminuria (uACR < 30 mg/g; n = 85), microalbuminuria (30 mg/g ≤ uACR < 300 mg/g; n = 84), and macroalbuminuria (uACR ≥ 300 mg/g; n = 73), with healthy individuals (n = 72) as controls. Peripheral blood mononuclear cells (PBMCs) from these subjects were analyzed for VDR mRNA (n = 314), TNF-α mRNA (n = 314), microRNA (miR)-346 (n = 120; 30 for each group), and VDR protein (n = 80; 20 for each group). PBMCs from randomly selected subjects (n = 6 for each group) were cultured ex vivo to evaluate the effect of TNF-α on miR-346 and VDR, and miR-346-mediated VDR suppression was further explored in HK2 cells.
VDR expression was down-regulated in PBMCs and renal tubular epithelial cells from T2DM patients with albuminuria. VDR mRNA and protein levels were both negatively correlated with uACR, and VDR mRNA was inversely correlated with TNF-α and miR-346 in PBMCs from T2DM patients. TNF-α reduced VDR while inducing miR-346 in cultured PBMCs. TNF-α suppressed VDR by up-regulating miR-346 in HK2 cells.
VDR down-regulation in PBMCs is independently associated with the severity of albuminuria in T2DM. TNF-α suppression of VDR in PBMCs and HK2 cells is mediated by miR-346.
Type 2 diabetes mellitus (T2DM) is one of the most prevalent diseases affecting human health that can markedly reduce the life expectancy. The latest meta-analysis showed that the reduction in life span for patients with an onset of T2DM at the age of > 40 years could be up to 11.6 years for males and 22 years for females (1). Severe T2DM often leads to diabetic nephropathy (DN), the most common kidney complication that can result in end-stage renal diseases. Increasing evidence has shown that inflammation plays an important role in the development of chronic kidney disease (CKD) with albuminuria in T2DM patients (2–4). Patients with DN have a high prevalence of vitamin D deficiency or insufficiency (5, 6), which may contribute to kidney damage induced by inflammation in these patients (7), given the well-known anti-inflammatory activity of vitamin D (8, 9). Indeed, nuclear factor-κB (NF-κB) is a major target of vitamin D in the regulation of inflammation (10). Consistent with this speculation, previous studies have demonstrated that active vitamin D analogs ameliorate diabetic renal injury, reduce urine albuminuria, inhibit macrophage infiltration in the kidneys, and suppress the expression of proinflammatory molecules in animals (11–14), and treatment with vitamin D analogs attenuated systemic inflammation and renal damage in T2DM patients (15, 16). Vitamin D functions by activating the vitamin D receptor (VDR), a nuclear hormone receptor with a broad spectrum of biological activities (17). VDR is widely expressed in various tissues in the body, and its expression is influenced by many factors including genetic polymorphisms, epigenetic modifications, microRNAs, serum vitamin D levels, and inflammatory status (18–21). We have recently shown that miR-346 mediates VDR down-regulation induced by TNF-α in colonic epithelial cells (20). VDR down-regulation has been reported in a number of disorders, including systemic sclerosis, inflammatory bowel diseases, and liver fibrosis in hepatitis C patients (22–24). VDR expression was also reduced in peripheral blood mononuclear cells (PBMCs) from patients with chronic renal failure, and a significant negative correlation was found between VDR and serum creatinine values (25). VDR activation has been associated with a reduction in IL-6, monocyte chemoattractant protein-1, and TGF-β levels and ameliorated albuminuria in animal models (12, 26, 27). Diabetic VDR knockout mice developed more severe albuminuria and glomerulosclerosis compared to wild-type mice (28), whereas VDR overexpression in podocytes protected against the development of DN (27), consistent with the renoprotective property of the vitamin D-VDR signaling (29). The relationship between VDR expression and inflammation in T2DM patients with DN, however, has not been reported. Because T2DM is characterized by inflammation and VDR has anti-inflammatory property, we hypothesized that VDR expression is down-regulated in T2DM patients, which promotes the progression of diabetic renal injury. Thus, our aim in this study was to examine VDR expression in renal biopsies and PBMCs in T2DM patients and analyze its correlation with inflammation and albuminuria. We also studied the involvement of TNF-α and miR-346 in the regulation of VDR expression in T2DM patients’ PBMCs and human kidney tubular epithelial cell line HK2 cells. Our studies suggest that VDR down-regulation in PBMCs is independently associated with the severity of albuminuria in T2DM, and such down-regulation is mediated by TNF-α-induced up-regulation of miR-346.
Materials and Methods
Immunohistochemical analysis of VDR expression in renal biopsies
Kidney biopsy samples from T2DM patients diagnosed with DN pathologically by percutaneous renal biopsy and clinically by the presence of macroalbuminuria (n = 8, DN group), and from age- and gender-matched nondiabetic patients who underwent renal resection because of renal carcinoma (n = 4, Control group) were provided by the Department of Pathology of the Third Xiangya Hospital. The Control biopsies were obtained at least 5 cm apart from the lesion site and were histologically confirmed as normal renal tissues. The renal biopsies were fixed overnight with 4% formaldehyde, processed, and embedded in paraffin wax. Tissues were cut into 3-μm sections. Renal morphology was examined by standard hematoxylin and eosin staining. To assess VDR expression, immunohistochemical staining was performed using the streptavidin-peroxidase conjugated method as described (30). Sections were stained with a mouse antihuman VDR antibody (1:100; Santa Cruz Biotechnology), and mouse IgG staining in sections from the same subject served as negative controls. At least 10 fields in different sections of the same sample were captured by Motic Images Advanced 3.2 image analysis software (Motic) and analyzed via ImagePro Plus 6.0 (Media Cybernetics). For quantification, the VDR positive index was calculated using the following equation: (positive field area/target field area) × positive intensity.
Recruitment of T2DM patients and healthy controls
We recruited 242 T2DM patients with CKD (estimated glomerular filtration rate [eGFR] > 60 mL/min/1.73 m2) from the Departments of Nephrology and Endocrinology at the Third Xiangya Hospital, Central South University, China, between 2012 and 2015. Patients with the following conditions were excluded: type 1 diabetes, secondary diabetes, diabetic acute complications (such as diabetic ketoacidosis and nonketone hypertonic coma), infection, and severe cardiovascular and cerebrovascular diseases in the 3–6 months before the recruitment. We also recruited 72 age- and gender-matched healthy adults as controls. The clinical parameters of each study subject were collected and analyzed. Qualified T2DM patients were divided into three groups based on their spot urinary albumin-to-creatinine ratio (uACR) (31): the normal albuminuria group (Normo, uACR < 30 mg/g; n = 85), the microalbuminuria group (Micro, 30 mg/g ≤ uACR < 300 mg/g; n = 84), and the macroalbuminuria group (Macro, uACR ≥ 300 mg/g; n = 73). Healthy adults served as negative controls (NC; n = 72). The study was carried out in accordance with the Declaration of Helsinki (2013) of the World Medical Association. Written informed consent was obtained from all study participants, and the study protocol was approved by the Ethics Committee of the Third Xiangya Hospital of Central South University (Changsha, China).
Sample collection
Peripheral venous blood samples were collected from all 314 subjects, including 242 T2DM patients and 72 healthy controls, after overnight fasting (at least 8 hours). PBMCs, including monocytes, lymphocytes, and other leukocytes, were isolated by Percoll continuous density gradient separation from the blood samples (32). Serum biochemical indices were measured by automatic biochemical analyzers (Hitachi 7600). Spot morning urine samples were collected from the 242 T2DM patients, centrifuged, and stored at −20°C for further analyses. eGFR was estimated using the abbreviated Modification of Diet in Renal Disease (MDRD) formula: 186 × (sCr/88.4)−1.154 × age−0.203 (× 0.742 if female), where sCr is serum creatinine. uACR was calculated as urinary albumin concentration divided by urinary creatinine concentration.
Real-time RT-PCR
Total RNAs were extracted from isolated PBMCs using Trizol reagent (Invitrogen). First strand cDNAs were synthesized using a ReverTra Ace qPCR RT Kit (Toyobo) according to the manufacturer’s instruction. Real-time RT-PCR was performed using a SYBR Green PCR Master Mix (Toyobo) on an Applied Biosystems 7300 Real-time PCR system under the following conditions: predenaturation at 95°C for 10 minutes, followed by 40 cycles of denaturation at 95°C for 10 seconds, annealing at 59°C for 20 seconds, and extension at 72°C for 30 seconds, and a final extension at 72°C for 5 minutes. The melting curve from 65°C to 95°C was used to confirm the specificity of each amplified product. VDR mRNA, TNF-α mRNA, or miR-346 levels were normalized to β-actin mRNA. The relative amount was calculated using the comparative 2−ΔΔCT method (33). PCR primers (Supplemental Table 1) were designed using Oligo 6.0 software and synthesized by Shanghai Sangon Biological Engineering Technology & Service Co. Ltd.
Western blot
PBMC VDR protein levels were analyzed in 80 randomly selected study subjects, including 20 subjects from each of the Normo, Micro, Macro, and NC groups. All subjects were matched for age, sex, and disease duration. Total cellular proteins were extracted according to the procedure reported previously (34). Protein concentration was quantitated using a BCA Protein Assay kit (Pierce). Proteins were separated by SDS-PAGE and electrotransferred onto polyvinylidene difluoride membranes. The membranes were then blocked with PBS containing 5% skim milk for 2 hours before being incubated overnight at 4°C with mouse antibodies against human VDR (1:1000; Santa Cruz) and β-actin (1:4000; Santa Cruz Biotechnology). After three washes, the membranes were further incubated with horseradish peroxidase-conjugated goat anti-mouse IgG (1:10 000; Santa Cruz) at room temperature for 2 hours. The protein bands were visualized using a Chemiluminescent Staining Reagent kit (Supersignal West Femto). Protein bands were quantified using the FluorChem 8900 software (Alpha Innotech).
PBMC culture and treatment
PBMCs isolated from six subjects in each of the Normo, Micro, Macro, and NC groups were cultured ex vivo. All subjects were matched for age, sex, and disease duration. Briefly, PBMCs isolated by Percoll density gradient were suspended in DMEM (Gibco) supplemented with 10% fetal bovine serum (Gibco), 100 U/mL penicillin, and 100 μg/mL streptomycin and seeded in six-well plates at a density of 1 × 106 cells per well in triplicate. Cells were cultured at 5% CO2 and 37°C and treated with or without 10 ng/mL TNF-α (Sigma) for 12 hours. Total RNAs and cell lysates were then prepared for real-time RT-PCR and Western blot analyses.
HK2 cell studies
HK2 cells were grown in DMEM:F12 (1:1) supplemented with 10% fetal bovine serum, 100 U/mL penicillin, and 100 μg/mL streptomycin. The cells were treated with or without TNF-α (10 ng/mL) for 24 hours, followed by isolation of total RNAs or lysates for further analyses. To study the regulation of VDR expression by miR-346, HK2 cells were transfected with miR-346 oligo mimics, scramble oligonucleotides, miR-346-specific hairpin inhibitor, or scramble inhibitor (Guangzhou RiboBio Co.) using Lipofectamine 2000 (Invitrogen). After treatment with or without TNF-α (10 ng/mL) for 24 hours, the cells were lysed for VDR protein analysis by Western blotting. To study the regulation of VDR transactivating activity by miR-346, HK2 cells were cotransfected with p3xVDRE-Luc reporter and miR-346 oligo mimic, followed by treatment with or without 1,25-dihydroxyvitamin D3 (10−7m; Sigma). After 24 hours, transfected cells were lysed, and luciferase activity was measured using the Dual-Luciferase Reporter assay system (Promega). To confirm miR-346 regulation of the VDR transcript, HK2 cells were cotransfected with pGL3-Luc-3′ untranslated region (UTR) or pGL3-Luc and miR-346 oligo mimics, followed by treatment with or without TNF-α (10 ng/mL), as described previously (20). Luciferase activity was measured 24 hours later using cell lysates.
Statistical analysis
Data were analyzed using SPSS 17.0 statistical software and expressed as mean ± SD. A post hoc test in one-way ANOVA was used to determine the differences between experimental groups and the control group. The Student’s t test was used to analyze the differences between groups with or without treatment. Spearman correlation and stepwise multiple linear regression analysis were carried out to determine the correlations between clinical and biochemical data. Logistic regression analysis was performed to identify the risk factors of uACR. P < .05 was considered statistically significant.
Results
Down-regulation of VDR in renal biopsies from T2DM patients with albuminuria
Because VDR has anti-inflammatory functions and low VDR expression is found in various inflammatory diseases, we asked whether diabetic kidney disease is correlated with decreased VDR expression in T2DM patients. We compared VDR expression between renal biopsies from T2DM patients diagnosed with DN and nondiabetic patients with histologically normal renal tissues by immunohistochemical staining. VDR was detected in the normal renal biopsies, located mostly in the tubular epithelial cells (Figure 1, A–C); however, in the renal biopsies from T2DM patients with DN, the VDR protein was almost absent in the tubular epithelial cells (Figure 1, D–F). Quantitation results showed that the difference in VDR expression between these two groups was highly significant (Figure 1G).
Immunohistochemical staining of VDR protein in renal biopsies. A–C, VDR protein staining in normal renal tissues from nondiabetic kidneys (Control). D–F, VDR protein staining in renal biopsies from T2DM patients with DN. VDR-positive cells are stained brown, and nuclei are stained blue. Scale bar, 10 μm. Arrows indicate examples of VDR-positive cells. Note the marked staining differences between the Control and DN groups. A and D, 200× magnification; B and E, 400× magnification showing glomeruli; C and F, 400× magnification showing renal tubules. G, Semiquantitative data for VDR staining intensity in Control (Con; n = 4) and DN (n = 8) groups. **, P < .01 vs Con. Data represent mean ± SD.
Clinical and biochemical data of the study participants
Because inflammation is not restricted to the kidney in T2DM, VDR expression may be reduced in other tissues such as PBMCs. To examine this topic, we recruited 242 T2DM patients and 72 age- and gender-matched healthy controls. To investigate the correlation between VDR in PBMCs and the severity of albuminuria, we divided the T2DM patients into three groups based on their uACR (Normo, n = 85; Micro, n = 84; and Macro, n = 73). We minimized the variables in the study subjects by selecting patients and healthy individuals with comparable parameters (Table 1). Although body mass index (BMI) in the Micro and Normo groups was slightly higher than in the NC and Macro groups, no statistical differences in age, gender, waist-hip ratio, or calcium levels were found among all diabetic groups or between the diabetic groups and the NC group. In addition, all patients selected had no statistical differences in disease duration or hemoglobin A1c levels. However, compared to the NC group, systolic blood pressure (SBP), fasting blood glucose (FBG), and uACR levels were significantly higher in all diabetic groups, among which the Macro group had the highest level. Serum albumin (ALB) concentrations were significantly lower in the diabetic groups, among which the Macro group showed the lowest level. Moreover, diastolic blood pressure (DBP) in the Micro and Macro groups was significantly higher compared to the NC and Normo groups. Our data also showed that the Macro group had lower hemoglobin (Hb), 25-hydroxyvitamin [25(OH)D], and eGFR levels but higher total cholesterol (TC) and triglyceride (TG) levels compared to the NC, Normo, and Micro groups, whereas no significant differences were found among these latter three groups.
Clinical Parameters |$(\bar X \pm S)$| of Study Participants
| Subject Groups | NC | Normo | Micro | Macro |
|---|---|---|---|---|
| n | 72 | 85 | 84 | 73 |
| Age, y | 54.2 ± 13.3 | 55.2 ± 10.7 | 56.5 ± 10.6 | 56.0 ± 12.1 |
| Gender, M/F | 45/27 | 46/39 | 49/35 | 45/28 |
| BMI, kg/m2 | 23.34 ± 2.71 | 24.29 ± 3.02a | 24.53 ± 3.09a | 23.95 ± 2.93 |
| Waist-hip ratio | 0.95 ± 0.05 | 0.96 ± 0.06 | 0.96 ± 0.05 | 0.96 ± 0.06 |
| Duration of disease, y | — | 7.27 ± 6.43 | 8.13 ± 6.91 | 8.15 ± 5.24 |
| SBP, mm Hg | 123.9 ± 12.2 | 129.3 ± 15.5a | 134.0 ± 16.3a | 141.4 ± 16.1a,b,c |
| DBP, mm Hg | 76.9 ± 11.0 | 77.9 ± 10.0 | 81.0 ± 10.2a,b | 80.8 ± 8.5a,b |
| FBG, mmol/L | 5.19 ± 1.03 | 7.52 ± 2.23a | 7.90 ± 2.48a | 8.38 ± 2.77a,b |
| HbA1c, % | — | 8.68 ± 1.98 | 9.30 ± 2.32 | 8.62 ± 1.87 |
| ALB, g/L | 46.32 ± 4.06 | 40.68 ± 4.50a | 40.26 ± 3.87a | 38.12 ± 6.21a,b,c |
| Hb, g/L | 139.2 ± 13.7 | 137.9 ± 20.0 | 138.4 ± 15.7 | 128.8 ± 22.5a.b.c |
| TG, mmol/L | 2.02 ± 3.09 | 1.94 ± 1.27 | 2.08 ± 1.88 | 3.09 ± 4.05a,b,c |
| TC, mmol/L | 4.49 ± 0.83 | 4.86 ± 1.02 | 4.83 ± 1.24 | 5.48 ± 1.57a,b,c |
| Calcium, mmol/L | 2.378 ± 0.171 | 2.388 ± 0.170 | 2.345 ± 0.177 | 2.330 ± 0.179 |
| 25(OH)D, ng/mL | 20.38 ± 2.04 | 19.76 ± 4.20 | 18.64 ± 4.26 | 14.65 ± 6.14a,b,c |
| eGFR, mL/min | 110.18 ± 18.41 | 118.76 ± 40.72 | 111.94 ± 29.41 | 79.11 ± 28.45a,b,c |
| uACR, μg/mg | 9.24 ± 3.03 | 15.81 ± 8.93a | 94.86 ± 53.36a,b | 1064.11 ± 737.13a,b,c |
| Subject Groups | NC | Normo | Micro | Macro |
|---|---|---|---|---|
| n | 72 | 85 | 84 | 73 |
| Age, y | 54.2 ± 13.3 | 55.2 ± 10.7 | 56.5 ± 10.6 | 56.0 ± 12.1 |
| Gender, M/F | 45/27 | 46/39 | 49/35 | 45/28 |
| BMI, kg/m2 | 23.34 ± 2.71 | 24.29 ± 3.02a | 24.53 ± 3.09a | 23.95 ± 2.93 |
| Waist-hip ratio | 0.95 ± 0.05 | 0.96 ± 0.06 | 0.96 ± 0.05 | 0.96 ± 0.06 |
| Duration of disease, y | — | 7.27 ± 6.43 | 8.13 ± 6.91 | 8.15 ± 5.24 |
| SBP, mm Hg | 123.9 ± 12.2 | 129.3 ± 15.5a | 134.0 ± 16.3a | 141.4 ± 16.1a,b,c |
| DBP, mm Hg | 76.9 ± 11.0 | 77.9 ± 10.0 | 81.0 ± 10.2a,b | 80.8 ± 8.5a,b |
| FBG, mmol/L | 5.19 ± 1.03 | 7.52 ± 2.23a | 7.90 ± 2.48a | 8.38 ± 2.77a,b |
| HbA1c, % | — | 8.68 ± 1.98 | 9.30 ± 2.32 | 8.62 ± 1.87 |
| ALB, g/L | 46.32 ± 4.06 | 40.68 ± 4.50a | 40.26 ± 3.87a | 38.12 ± 6.21a,b,c |
| Hb, g/L | 139.2 ± 13.7 | 137.9 ± 20.0 | 138.4 ± 15.7 | 128.8 ± 22.5a.b.c |
| TG, mmol/L | 2.02 ± 3.09 | 1.94 ± 1.27 | 2.08 ± 1.88 | 3.09 ± 4.05a,b,c |
| TC, mmol/L | 4.49 ± 0.83 | 4.86 ± 1.02 | 4.83 ± 1.24 | 5.48 ± 1.57a,b,c |
| Calcium, mmol/L | 2.378 ± 0.171 | 2.388 ± 0.170 | 2.345 ± 0.177 | 2.330 ± 0.179 |
| 25(OH)D, ng/mL | 20.38 ± 2.04 | 19.76 ± 4.20 | 18.64 ± 4.26 | 14.65 ± 6.14a,b,c |
| eGFR, mL/min | 110.18 ± 18.41 | 118.76 ± 40.72 | 111.94 ± 29.41 | 79.11 ± 28.45a,b,c |
| uACR, μg/mg | 9.24 ± 3.03 | 15.81 ± 8.93a | 94.86 ± 53.36a,b | 1064.11 ± 737.13a,b,c |
Abbreviations: M, male; F, female, —, no data. Results are expressed as mean ± SD or ratio.
Compared with NC group, P < .05;
Compared with Normo group, P < .05;
Compared with Micro group, P < .05.
Clinical Parameters |$(\bar X \pm S)$| of Study Participants
| Subject Groups | NC | Normo | Micro | Macro |
|---|---|---|---|---|
| n | 72 | 85 | 84 | 73 |
| Age, y | 54.2 ± 13.3 | 55.2 ± 10.7 | 56.5 ± 10.6 | 56.0 ± 12.1 |
| Gender, M/F | 45/27 | 46/39 | 49/35 | 45/28 |
| BMI, kg/m2 | 23.34 ± 2.71 | 24.29 ± 3.02a | 24.53 ± 3.09a | 23.95 ± 2.93 |
| Waist-hip ratio | 0.95 ± 0.05 | 0.96 ± 0.06 | 0.96 ± 0.05 | 0.96 ± 0.06 |
| Duration of disease, y | — | 7.27 ± 6.43 | 8.13 ± 6.91 | 8.15 ± 5.24 |
| SBP, mm Hg | 123.9 ± 12.2 | 129.3 ± 15.5a | 134.0 ± 16.3a | 141.4 ± 16.1a,b,c |
| DBP, mm Hg | 76.9 ± 11.0 | 77.9 ± 10.0 | 81.0 ± 10.2a,b | 80.8 ± 8.5a,b |
| FBG, mmol/L | 5.19 ± 1.03 | 7.52 ± 2.23a | 7.90 ± 2.48a | 8.38 ± 2.77a,b |
| HbA1c, % | — | 8.68 ± 1.98 | 9.30 ± 2.32 | 8.62 ± 1.87 |
| ALB, g/L | 46.32 ± 4.06 | 40.68 ± 4.50a | 40.26 ± 3.87a | 38.12 ± 6.21a,b,c |
| Hb, g/L | 139.2 ± 13.7 | 137.9 ± 20.0 | 138.4 ± 15.7 | 128.8 ± 22.5a.b.c |
| TG, mmol/L | 2.02 ± 3.09 | 1.94 ± 1.27 | 2.08 ± 1.88 | 3.09 ± 4.05a,b,c |
| TC, mmol/L | 4.49 ± 0.83 | 4.86 ± 1.02 | 4.83 ± 1.24 | 5.48 ± 1.57a,b,c |
| Calcium, mmol/L | 2.378 ± 0.171 | 2.388 ± 0.170 | 2.345 ± 0.177 | 2.330 ± 0.179 |
| 25(OH)D, ng/mL | 20.38 ± 2.04 | 19.76 ± 4.20 | 18.64 ± 4.26 | 14.65 ± 6.14a,b,c |
| eGFR, mL/min | 110.18 ± 18.41 | 118.76 ± 40.72 | 111.94 ± 29.41 | 79.11 ± 28.45a,b,c |
| uACR, μg/mg | 9.24 ± 3.03 | 15.81 ± 8.93a | 94.86 ± 53.36a,b | 1064.11 ± 737.13a,b,c |
| Subject Groups | NC | Normo | Micro | Macro |
|---|---|---|---|---|
| n | 72 | 85 | 84 | 73 |
| Age, y | 54.2 ± 13.3 | 55.2 ± 10.7 | 56.5 ± 10.6 | 56.0 ± 12.1 |
| Gender, M/F | 45/27 | 46/39 | 49/35 | 45/28 |
| BMI, kg/m2 | 23.34 ± 2.71 | 24.29 ± 3.02a | 24.53 ± 3.09a | 23.95 ± 2.93 |
| Waist-hip ratio | 0.95 ± 0.05 | 0.96 ± 0.06 | 0.96 ± 0.05 | 0.96 ± 0.06 |
| Duration of disease, y | — | 7.27 ± 6.43 | 8.13 ± 6.91 | 8.15 ± 5.24 |
| SBP, mm Hg | 123.9 ± 12.2 | 129.3 ± 15.5a | 134.0 ± 16.3a | 141.4 ± 16.1a,b,c |
| DBP, mm Hg | 76.9 ± 11.0 | 77.9 ± 10.0 | 81.0 ± 10.2a,b | 80.8 ± 8.5a,b |
| FBG, mmol/L | 5.19 ± 1.03 | 7.52 ± 2.23a | 7.90 ± 2.48a | 8.38 ± 2.77a,b |
| HbA1c, % | — | 8.68 ± 1.98 | 9.30 ± 2.32 | 8.62 ± 1.87 |
| ALB, g/L | 46.32 ± 4.06 | 40.68 ± 4.50a | 40.26 ± 3.87a | 38.12 ± 6.21a,b,c |
| Hb, g/L | 139.2 ± 13.7 | 137.9 ± 20.0 | 138.4 ± 15.7 | 128.8 ± 22.5a.b.c |
| TG, mmol/L | 2.02 ± 3.09 | 1.94 ± 1.27 | 2.08 ± 1.88 | 3.09 ± 4.05a,b,c |
| TC, mmol/L | 4.49 ± 0.83 | 4.86 ± 1.02 | 4.83 ± 1.24 | 5.48 ± 1.57a,b,c |
| Calcium, mmol/L | 2.378 ± 0.171 | 2.388 ± 0.170 | 2.345 ± 0.177 | 2.330 ± 0.179 |
| 25(OH)D, ng/mL | 20.38 ± 2.04 | 19.76 ± 4.20 | 18.64 ± 4.26 | 14.65 ± 6.14a,b,c |
| eGFR, mL/min | 110.18 ± 18.41 | 118.76 ± 40.72 | 111.94 ± 29.41 | 79.11 ± 28.45a,b,c |
| uACR, μg/mg | 9.24 ± 3.03 | 15.81 ± 8.93a | 94.86 ± 53.36a,b | 1064.11 ± 737.13a,b,c |
Abbreviations: M, male; F, female, —, no data. Results are expressed as mean ± SD or ratio.
Compared with NC group, P < .05;
Compared with Normo group, P < .05;
Compared with Micro group, P < .05.
Inverse correlation between VDR expression in PBMCs and severity of albuminuria
To address the correlation between VDR expression in PBMCs and renal injury, we measured VDR mRNA levels in PBMCs isolated from these 242 T2DM patients and 72 healthy control subjects. In addition, we randomly selected 20 subjects from each of the four groups for determination of VDR protein levels in isolated PBMCs. Our results showed that PBMC VDR mRNA (Figure 2A) and protein (Figure 2, B and C) levels in the Normo, Micro, and Macro groups were significantly lower compared with the NC group. Among the three diabetic groups, the Normo group had the highest levels of both VDR mRNA and protein, whereas the Macro group had the lowest levels (Figure 2, A–C), suggesting that PBMC VDR expression is inversely correlated with renal injury.
VDR expression in PBMCs is down-regulated in T2DM and inversely correlated with uACR. Peripheral venous blood samples were taken from 242 T2DM patients and 72 healthy control (NC) subjects. T2DM patients were divided into Normo, Micro, and Macro groups based on their uACR. PBMCs were isolated by Percoll density gradient centrifugation. A, VDR mRNA levels in NC (n = 72), Normo (n = 85), Micro (n = 84), and Macro (n = 73) groups were quantified by real-time RT-PCR. B and C, VDR protein levels were determined by Western blotting in 20 subjects randomly selected from each of the four groups. B, Representative images of Western blot; β-actin was used as loading controls. C, Densitometric quantitation of the Western blot data. D, Scatterplot showing an inverse relationship between VDR mRNA levels and uACR (n = 242 for T2DM cohort only; r = −0.607; P < .001). E, Scatterplot showing an inverse relationship between VDR protein levels and uACR (n = 60, with 20 subjects randomly selected from each of the three T2DM groups; r = −0.658; P < .001). Data represent mean ± SD. **, P < .01 vs NC; ##, P < .01 vs Normo; ▵▵, P < .01 vs Micro.
We next used Spearman correlation analysis to study the relationship between PBMC VDR levels and uACR in T2DM patients, an indicator for the severity of albuminuria and loss of kidney function. As shown in Figure 2, D and E, PBMC VDR mRNA (n = 242) and protein (n = 60, including 20 subjects randomly selected from each of the three diabetic groups) levels were inversely correlated with uACR in these patients (r = −0.607, P < .001 for mRNA, Figure 2D; and r = −0.658, P < .001 for protein, Figure 2E). After adjusting the potential confounding factors [BMI, Hb, ALB, FBG, SBP, DBP, TC, TG, 25(OH)D, eGFR], stepwise multiple regression analysis was performed, and the results indicated that VDR mRNA (β = −0.475; P < .001) and protein (β = −0.532; P < .001) both remained inversely associated with the uACR levels. Logistic regression analysis revealed that low VDR mRNA (odds ratio, 0.107; 95% confidence interval, 0.052–0.217; P < .001) and protein (odds ratio, 0.014; 95% confidence interval, 0.001–0.277; P = .005) levels were both associated with high uACR, an indication of increased risk of kidney malfunction. Taken together, these results indicate that reduced PMBC VDR expression is independently associated with the degree of albuminuria in T2DM patients.
Increased TNF-α and miR-346 levels in PBMCs from T2DM patients with albuminuria
Chronic inflammation is a feature of T2DM. Our previous studies showed that proinflammatory cytokine TNF-α suppresses VDR expression by inducing miR-346 in colonic epithelial cells because miR-346 down-regulates human VDR via targeting the 3′UTR of the human VDR transcript (20). To assess the relationship between inflammation and VDR expression in T2DM, we measured TNF-α (n = 314) and miR-346 (n = 120, including 30 subjects randomly selected from each of the four groups) levels in freshly isolated PBMCs. Compared to the NC group, the levels of TNF-α (Figure 3A) and miR-346 (Figure 3B) were significantly increased in all three diabetic groups, among which the Macro group showed the highest level for both TNF-α and miR-346 (P < .01). Spearman correlation analysis showed that both TNF-α mRNA (n = 242 T2DM patients; r = −0.546; P < .001) (Figure 3C) and miR-346 (n = 90, including 30 subjects from each of the three T2DM groups; r = −0.286; P = .006) (Figure 3D) were inversely correlated with VDR mRNA levels in PBMCs in T2DM patients. These observations suggest that, as seen in colonic epithelial cells, VDR down-regulation in PBMCs associated with increased TNF-α is mediated by miR-346 action in T2DM patients with albuminuria.
Correlation of VDR expression with TNF-α and miR-346 in PBMCs. A, TNF-α mRNA levels in the NC (n = 72), Normo (n = 85), Micro (n = 84), and Macro (n = 73) groups were quantified by real-time RT-PCR. B, miR-346 levels in these four groups (n = 30 each group, with subjects randomly selected from each of the four groups) were quantified by real-time RT-PCR. Data represent mean ± SD. **, P < .01 vs NC; ##, P < .01 vs Normo; ▵▵, P < .01 vs Micro. C, Scatterplot showing an inverse relationship between TNF-α mRNA and VDR mRNA levels (n = 242 for T2DM cohort only; r = −0.546; P < .001). D, Scatterplot showing an inverse relationship between miR-346 and VDR mRNA levels (n = 90, with 30 subjects randomly selected from each of the three T2DM groups; r = −0.286; P = .006).
TNF-α inhibition of VDR expression in PBMCs
To address whether TNF-α indeed suppresses VDR expression in PBMCs, we cultured ex vivo PBMCs randomly selected from each of the four groups (n = 6 for each group, total n = 24) and then stimulated the cells with TNF-α. As shown in Figure 4, TNF-α treatment significantly suppressed VDR expression at both mRNA (Figure 4A) and protein (Figure 4, B and C) levels in the NC, Normo, Micro, and Macro groups compared to their nontreated counterparts (P < .05). Moreover, miR-346 levels were also significantly induced in TNF-α-treated PBMCs from all four groups. These observations suggest that TNF-α inhibits VDR expression by inducing miR-346 in human PBMCs.
TNF-α down-regulates VDR expression in cultured PBMCs. PBMCs were isolated from 24 subjects randomly selected from the four groups (n = 6 in each group, age- and gender-matched) and cultured ex vivo in triplicate. Cultured PBMCs were treated with (+) or without (−) TNF-α (10 ng/mL) for 12 hours before total RNAs or cell lysates were harvested. A, VDR mRNA levels were quantified by real-time RT-PCR in these four groups (n = 6 for each group). B and C, VDR protein levels were determined by Western blotting in these subjects. B, Representative images of Western blot; β-actin was used as loading controls. C, Densitometric quantitation of the Western blot data. D, miR-346 levels measured by real-time RT-PCR (n = 6 for each group). Data represent mean ± SD. *, P < .05 vs corresponding nontreated (NT) value.
Critical role of miR-346 in TNF-α induced down-regulation of VDR in HK2 cells
Finally, we used HK2 cells to further address whether TNF-α represses VDR by a miR-346-mediated mechanism as reported previously (20). As seen in PBMCs, TNF-α treatment (10 ng/mL for 24 hours) markedly induced miR-346 (Figure 5A) and suppressed VDR protein (Figure 5, B and C) in HK2 cells. When the cells were transfected with a miR-346 oligo mimic, but not with a scramble oligonucleotide, VDR protein levels were reduced with or without TNF-α treatment; however, transfection of HK2 cells with a miR-346-specific hairpin inhibitor, but not with a scramble inhibitor, diminished TNF-α-induced suppression of VDR (Figure 5, B and C). These results confirmed our previous finding that miR-346 mediates the down-regulation of VDR by TNF-α (20). To further solidify this notion, we measured the effect of miR-346 on VDR transactivation activity by luciferase reporter assays. As expected, transfection of HK2 cells with miR-346 mimics completely blocked 1,25-dihydroxyvitamin D-induced VDR transactivating activity when the cells were cotransfected with the 3xVDRE-luciferase reporter (Figure 5D). Moreover, when HK2 cells were cotransfected with miR-346 mimics and pGL3-Luc or pGL3-Luc-3′UTR reporter that contains the miR-346 target site within VDR 3′UTR, miR-346 suppressed the reporter activity in the presence of 3′UTR, and TNF-α further reduced the activity of these reporters (Figure 5E). These data indicate that miR-346 indeed suppresses VDR expression by targeting the 3′UTR of VDR mRNA.
TNF-α inhibits VDR expression through miR-346 induction. A, HK2 cells were treated with or without (NT) TNF-α (10 ng/mL) for 24 hours, and miR-346 levels were quantified by real-time RT-PCR. B, HK2 cells were not transfected (Con) or were transfected with miR-346, scramble oligonucleotide (Scramble), miR-346 inhibitor (miR-346 In), or scramble inhibitor (Scramble In). After 24 hours, transfected cells were treated with or without TNF-α (10 ng/mL) for 24 hours. VDR protein levels were determined by Western blot. Data from a single representative experiment are shown. β-Actin was used as loading control. C, Densitometric quantitation of the Western blot data. Experiments were repeated three times. Data represent mean ± SD. *, P < .05 vs corresponding nontreated (NT); #, P < .05. D, HK2 cells were transfected with p3xVDRE-Luc plasmid, transfected with p3xVDRE-Luc, and treated with 10 ng/mL TNF-α, or cotransfected with p3xVDRE-Luc and miR-346. After 24 hours, the cells were stimulated with dimethylsulfoxide or 1,25-dihydroxyvitamin D3 (1 × 10−7m), followed by luciferase activity assays. *, P < .05 vs corresponding dimethylsulfoxide; #, P < .05. E, HK2 cells were cotransfected with miR-346 mimic and pGL3-Luc control plasmid (pGL3-Luc) or pGL3-Luc-3′UTR. After 24 hours, the cells were treated with or without TNF-α (10 ng/mL) for 24 hours, followed by luciferase activity assays. *, P < .05 vs corresponding nontreated (NT); #, P < .05. Experiments were repeated three times, and data represent mean ± SD.
Discussion
In this report, we investigated the relationship between VDR expression and the severity of albuminuria in T2DM. First, we showed that VDR expression in renal biopsy tissues, particularly in proximal renal tubular epithelial cells, is much lower in T2DM patients compared to nondiabetic kidney tissues. We further found that VDR expression in PBMCs is significantly down-regulated in T2DM patients compared to healthy individuals. Further analyses revealed that both VDR transcript and protein levels in PBMCs are inversely correlated with the uACR levels in T2DM patients (n = 242). Stepwise multiple regression and logistic regression analyses demonstrate that the reductions in VDR mRNA and protein are independently associated with higher uACR, which is a major indicator for assessing the development of DN. This is the first report, to our knowledge, linking VDR expression levels in PBMCs to albuminuria in early-stage CKD patients. It is noteworthy, however, that due to the difficulty in obtaining renal biopsies from early-stage CKD patients, the number of renal biopsies used in this study was quite small, which did not allow for correlation analyses to assess the relationship between VDR expression in renal tissues and the severity of albuminuria, or between renal VDR expression and PBMC VDR expression. Moreover, although an inverse correlation between VDR expression in PBMCs and the severity of albuminuria was found in T2DM patients, the study was cross-sectional. Future prospective studies are needed to assess the association between the degree of VDR down-regulation and the progression of albuminuria. Another limitation is that the correlation between PBMC VDR expression and renal tissue VDR expression is unclear, but based on our data, it is tempting to speculate that overproduction of proinflammatory cytokines (such as TNF-α as discussed below) from PBMCs due to VDR down-regulation in these cells could in turn suppress VDR expression in renal tissues.
T2DM is characterized by low-grade chronic inflammation, and TNF-α has been shown to be one of the key inflammatory mediators that are up-regulated in T2DM patients (35). TNF-α expression in serum, urine, and renal tissues is dramatically increased in T2DM patients, and increased TNF-α is involved in the development of insulin resistance and various complications of T2DM. Additionally, T2DM patients with a higher level of TNF-α are prone to neuronal, retinal, or renal complications and deterioration of these complications (36). Here our study showed a significant increase in TNF-α expression in PBMCs from T2DM patients, with the highest level seen in the Macro group. This is consistent with the notion that TNF-α drives renal complications in T2DM. Our study further demonstrated that TNF-α induces VDR down-regulation in PBMCs and renal epithelial cells, and one important mechanism for VDR down-regulation is TNF-α induction of miR-346, which is known to inhibit VDR expression through a miR-346 target site in the 3′UTR of the VDR transcript (20). We showed that TNF-α simultaneously reduces VDR and increases miR-346, and we confirmed that miR-346 is able to inhibit VDR expression; when miR-346 is inhibited, TNF-α can no longer suppress VDR expression. Moreover, luciferase reporter assays further confirmed that miR-346 inhibits VDR by targeting its 3′UTR. Taken together, these data provide a logical explanation as to why VDR is down-regulated in T2DM. Therefore, it is conceivable that when TNF-α levels are elevated in T2DM, VDR expression is suppressed via a miR-346-mediated mechanism, which attenuates the anti-inflammatory and renoprotective activities of vitamin D-VDR signaling, leading to more severe inflammation and renal injury. This vicious cycle could contribute to the progression of inflammation and the development of diabetic complications including DN in T2DM. Of course, this mechanism linking TNF-α to VDR down-regulation does not exclude other potential mechanisms for VDR down-regulation in T2DM. In fact, how TNF-α induces miR-346 remains to be determined.
It is well known that the NF-κB pathway, a major inflammatory signaling pathway, is a key anti-inflammatory target of vitamin D-VDR signaling. Activated VDR inhibits p65 nuclear translocation and NF-κB activation by blocking inhibitor of NF-κB kinase complex formation and increasing inhibitor of NF-κB α (10). Because TNF-α acts through NF-κB, there appears to be a reciprocal regulatory relationship between TNF-α and VDR. When VDR is reduced in PBMCs in T2DM, its inhibitory effects on NF-κB are attenuated, and thus the inflammatory pathway prevails, which is expected to promote diabetic renal injury (37) as well as further VDR down-regulation. Therefore, therapeutic strategies that counter VDR reduction might be beneficial for controlling inflammation in T2DM. In this regard, treatment with low-calcemic vitamin D analogs can not only activate VDR signaling but also raise VDR levels in monocytes and renal cells as reported previously (8, 12). Indeed, a number of vitamin D analogs have been reported to have potent renoprotective benefits in animals and humans (29). Our data presented here suggest that future studies could be carried out to test the anti-inflammatory effects of vitamin D analogs in T2DM in laboratory and clinical settings. In fact, T2DM is usually associated with obesity and vitamin D deficiency (38), which calls for vitamin D supplement or vitamin D analog therapy. In addition, targeting miR-346 might be another strategy to raise or maintain VDR levels. There may be many challenges in this strategy, but developing drugs that target microRNAs is an ongoing effort that shows great promise (39).
Acknowledgments
We thank all of the patients for their participation in this study.
This work was funded by research grants from the National Natural Science Foundation of China (Grant 81470961) and the Natural Science Foundation of Hunan Province (Grant 2015JJ4082).
Disclosure Summary: The authors have no conflicts of interest in this work.
Abbreviations
- ALB
albumin
- BMI
body mass index
- CKD
chronic kidney disease
- DBP
diastolic blood pressure
- DN
diabetic nephropathy
- eGFR
estimated glomerular filtration rate
- FBG
fasting blood glucose
- Hb
hemoglobin
- miR
microRNA
- NC
negative control
- NF-κB
nuclear factor-κB
- 25(OH)D
25-hydroxyvitamin
- PBMC
peripheral blood mononuclear cell
- SBP
systolic blood pressure
- TC
total cholesterol
- T2DM
type 2 diabetes mellitus
- TG
triglyceride
- uACR
urinary albumin-to-creatinine ratio
- UTR
untranslated region
- VDR
vitamin D receptor.
References
Author notes
B.Y. and J.H. contributed equally to this work.
