Protective Effects of 1,25 Dihydroxyvitamin D₃ against High-Glucose-Induced Damage in Human Umbilical Vein Endothelial Cells Involve Activation of Nrf2 Antioxidant Signaling

Diabetes is a serious global epidemic with a major impact on public health [1]. Atherosclerotic cardiovascular complications are the leading cause of death for diabetic individuals, and endothelial dysfunction is considered as the first step of the atherosclerotic process associated with diabetes [2, 3]. It is supposed that long-term high glucose exposure contributes to endothelial dysfunction through increased oxidative stress and activation of the proinflammatory phenotype, which is characterized by elevated reactive oxygen species (ROS), enhanced expression of adhesion molecules and chemokines, increased sensitivity to apoptosis, and deficiency of nitric oxide (NO) bioavailability [4-7].

In recent years, the cardiovascular protective effect of active vitamin D, 1,25 dihydroxyvitamin D₃ (1,25(OH)₂D₃), has received much attention. Epidemiological and experimental studies have shown that vitamin D deficiency is highly prevalent in patients with type 2 diabetes, and supplementation of vitamin D, especially 1,25(OH)₂D₃, can alleviate endothelial dysfunction and prevent diabetic vascular complications partly through its antioxidative stress/anti-inflammatory activities [8-11]. Our previous studies have found that 1,25(OH)₂D₃ can prevent human umbilical vein endothelial cells (HUVECs) against ROS production derived from mitochondria and reduced nicotinamide adenine dinucleotide phosphate (NADPH) oxidase in high-glucose settings and improve vascular endothelial function [12, 13]. However, whether the endothelial protective effects of 1,25(OH)₂D₃ can be achieved by promoting antioxidant production and the mechanisms involved remain unclear.

Cellular oxidative stress is caused by the imbalance between antioxidants and pro-oxidants in the body [14, 15]. Nuclear factor erythroid 2-related factor 2 (Nrf2) belongs to the cap'n'collar leucine zipper transcription activator family and is an important nuclear transcription factor that regulates multiple antioxidant genes’ expression and maintains redox homeostasis [16, 17]. In the physiological state, Nrf2 is anchored in the cytoplasm by kelch-like epichlorohydrin-related protein-1 and is degraded by ubiquitination. When ROS accumulates, Nrf2 dissociates from kelch-like epichlorohydrin-related protein-1 and translocates to the nucleus to bind with the antioxidant reaction element (ARE), inducing the transcription and expression of a series of downstream antioxidant enzymes, such as superoxide dismutase (SOD), catalase, and glutathione reductase (GR), thus protecting the body from oxidative substances [18, 19]. Studies have found that reduced Nrf2 activity is associated with vascular endothelial dysfunction and atherosclerosis development, while activation of the Nrf2/ARE pathway has vascular endothelial protection and possibly ameliorates diabetes complications [20-23]. Therefore, this study intends to observe whether 1,25(OH)₂D₃ can promote the generation of antioxidants by HUVECs in a high-glucose environment and antagonize oxidative injury through activation of the Nrf2 antioxidant signaling pathway.

The HUVECs were cultured and identified as previously -described [13]. HUVECs were pretreated with or without 1,25(OH)₂D₃ for 45 min and then incubated with a high concentration of glucose (33 mM) for 48 h. HUVECs incubated with a normal concentration of glucose (5.5 mM) served as control. All the experimental operations were approved by the Ethics Committee of the Affiliated Hospital of Putian University and were performed following the Declaration of Helsinki.

MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay was used to evaluate the viability of HUVECs treated with different concentrations of glucose and the effect of 1,25(OH)₂D₃ on cell viability. After treatment, MTT (Biosharp, China) was added to each well, and the wells were incubated for 4 h at 37°C. Subsequently, the medium was removed, and dimethyl sulfoxide (Sigma-Aldrich, St. Louis, MO, USA) was added (150 μL/well) to solubilize the formazan crystals. Optical density was measured at 570 nm with a Varioskan Flash microplate reader (Thermo Scientific, Waltham, MA, USA), and cell viability was calculated as a percentage of the control optical density.

The Annexin V/propidium iodide (PI; BD Clontech) was used to quantify the number of apoptotic cells. Cells were washed twice with PBS and stained with Annexin V and PI for 20 min at room temperature. The level of apoptosis cells was determined by flow cytometry analysis with Annexin V-fluorescein isothiocyanate/propidium iodide double dye assay.

ROS production was measured in 2′,7′-dichlorodihydrofluorescein diacetate (H2-DCFDA)-treated HUVECs by flow cytometry using Coulter Epics XL (Beckman Coulter, Brea, CA, USA) [24]. On oxidation, H2-DCFDA becomes highly green fluorescent 2′,7′-dichlorofluorescein. HUVECs were incubated with a medium containing H2-DCFDA (10 μM) for 1 h at 37°C, washed with PBS, and collected. Signals were obtained using a 525-nm band-pass filter (FL1 channel). Dihydroethidium (Sigma, St Louis, MO, USA) can react with O₂₋ in cells to form fluorescent ethidium [25]. HUVECs were incubated in a medium containing dihydroethidium (10 μM) for 30 min at 37°C. Ethidium red fluorescence was detected using a Leica TCS-SP2 confocal microscope (Wetzlar, Giessen, Germany).

To determine the activity of SOD, catalase, and GR, commercial assay kits obtained from Beyotime Biotechnology (S0101, S0051, and S0055) were used as per the manufacturer’s instructions. The protein concentration was measured with a Bradford protein assay kit (Beyotime, China).

Production of NO by HUVECs was measured as its stable oxidation product, nitrite, using the Bioxytech nitric oxide assay kit. In brief, 50 μL of the culture medium was diluted with 35 μL assay buffer and mixed with 10 μL nitrate reductase and 10 μL NADH. Following 20 min of incubation to convert nitrate to nitrite, total nitrite was measured at 540-nm absorbance by reaction with Griess reagents (sulfanilamide and naphthalene-ethylenediamine dihydrochloride).

Cells (5 × 10⁵) were cultured in 6-well cell culture plates and were treated as indicated. Total RNA was isolated by using TriZol reagent (Life Technologies). The cDNA was synthesized with the High-Capacity cDNA Reverse Transcription Kit (Life Technologies). TaqMan gene expression assays for Homo sapiens (Hs) (Applied Biosystems) were used to detect the monocyte chemoattractant protein 1 (MCP1; Hs00234140_m1) and vascular cell adhesion molecule 1 (VCAM1) mRNA expression. PCR amplification was carried out by using the ABI Prism 7.500 fast (Applied Biosystems) and standard cycling conditions. The expression of each target gene was normalized to the relative expression of 18S-ribosomal RNA as an internal efficiency control. The mRNA fold change was calculated by using the 2 (−delta delta C[T]) method.

The NE-PER^TM Nuclear and Cytoplasmic Extraction Reagents kit (ThermoFisher, Waltham, MA, USA) was utilized to extract nuclear and cytosolic protein fractions. Immunoblotting was conducted as previously described [13], with the following antibodies: anti-Nrf2 antibody (abcam ab137550), anti-Bax antibody (abcam ab220180), anti-Bcl-2 antibody (abcam ab31394), and anti-caspase-3 antibody (abcam ab115183).

Predesigned small interfering RNAs (siRNAs) against human Nrf2 and vitamin D receptor (VDR) were synthesized by GenePharma (Shanghai, China). The dsRNA sequences of Nrf2 and VDR used in the current experiments were as follows: for Nrf2, sense and antisense siRNAs were 5′-CCCUGUGUAAAGCUUUCAATT-3′ and 5′-GAAAGCUUUACACAGGGTT-3′, respectively. For VDR, sense and antisense siRNAs were 5′-CAAUCUGGAUCUGAGUGAAdTdT-3′ and 5′-UUCACUCAGAUCCAG-AUUGdTdT-3′, respectively. HUVECs were transfected using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. After transfection, the cells were incubated with complete medium for 48 h before assaying for target gene inhibition.

Values are expressed as mean ± SD. Results are compared by one-way factorial ANOVA followed by a post hoc Scheffé comparison test. p value <0.05 was considered to be statistically significant.

High glucose reduced cell viability of HUVECs in a time-dependent manner. The viability of HUVECs was significantly decreased after 48 h of high-glucose incubation (p < 0.01), while there was no significant difference between the viability of HUVECs after 72 h of high-glucose incubation and that of HUVECs after 48 h of high-glucose incubation (p > 0.05). Therefore, 48 h was selected as the time of high-glucose exposure for subsequent experiments. After pretreatment with 1,25(OH)₂D₃ (10⁻⁹–10⁻⁷ M) for 45 min, HUVECs were coincubated with high glucose for 48 h. The results showed that both 1,25(OH)₂D₃ 10⁻⁸ and 10⁻⁷ M could significantly inhibit the decline of HUVECs viability under hyperglycemic conditions, and 1,25(OH)₂D₃ 10⁻⁷ M was more significant (p < 0.01). Therefore, 10⁻⁷ M was selected as the working concentration of 1,25(OH)₂D₃ for subsequent experiments (see online suppl. Fig. 1; for all online suppl. material, see www.karger.com/doi/10.1159/000515512).

To further assess the damage of high glucose to HUVECs and the protective effect of 1,25(OH)₂D₃, we quantified ROS production and proportion of apoptotic cells and detected the expression of apoptosis-associated proteins. Besides, to clarify the role of Nrf2 protein in this process, siRNA targeting Nrf2 protein expression was designed. Immunoblotting analysis showed that high glucose increased the protein expression of Nrf2 in HUVECs compared with normal glucose, and Nrf2-specific siRNA significantly reduced the protein expression of Nrf2 compared with scrambled siRNA (online suppl. Fig. 2). As to the effects of 1,25(OH)₂D₃ on high-glucose-induced damage to HUVECs, fluorescence microscopy displayed a strikingly increased fluorescence intensity of intracellular ROS in high-glucose-cultured HUVECs compared with those in normal-glucose-cultured HUVECs. 1,25(OH)₂D₃ treatment inhibited the increased fluorescence intensity induced by high glucose, which was abrogated by Nrf2 siRNA. Flow cytometry further confirmed that the intracellular ROS levels in high-glucose-cultured HUVECs were significantly higher than those in normal-glucose-cultured HUVECs, which were effectively prevented by 1,25(OH)₂D₃ treatment. However, when Nrf2 was blocked by siRNA, the inhibitory effect of 1,25(OH)₂D₃ on ROS production was impaired (Fig. 1a, b).

Similarly, flow cytometry showed that high glucose increased the apoptosis percentage of HUVECs compared with normal glucose, and pretreatment with 1,25(OH)₂D₃ significantly attenuated the increased apoptosis induced by high glucose. However, when Nrf2 was downregulated by siRNA in advance, the antiapoptotic protective effect of 1,25(OH)₂D₃ was weakened. Immunoblotting demonstrated that the increased protein expression of the proapoptotic gene BAX and apoptosis effector caspase 3 and the decreased protein expression of the antiapoptotic gene BCL2 driven by high glucose were all significantly alleviated by 1,25(OH)₂D₃ treatment. However, when Nrf2 was downregulated by siRNA, the beneficial regulatory effects of 1,25(OH)₂D₃ on apoptosis-related proteins were significantly compromised (Fig. 1c, d).

To further explore the mechanism by which 1,25(OH)₂D₃ protects endothelial damage in high-glucose environment, we investigated whether 1,25(OH)₂D₃ can increase the activities of intracellular antioxidant substances. As shown in Figure 2a–c, respectively, the activities of SOD, GR, and catalase, all of which are downstream genes regulated by Nrf2, were significantly lower in the high-glucose-cultured HUVECs than those in the normal-glucose-cultured HUVECs. 1,25(OH)₂D₃ treatment significantly improved the above 3 antioxidant enzyme activities compared with those in the high-glucose-cultured HUVECs. Specifically, 1,25(OH)₂D₃ restored the hyperglycemic-induced decreased activities of SOD and catalase to the levels under normal-glucose environment, but only partially reversed the decline of GR activity caused by hyperglycemia. When Nrf2 protein expression was downregulated, the role of 1,25(OH)₂D₃ in promoting the production of antioxidant enzyme substances was eliminated.

The common characteristics of vascular endothelial dysfunction are increased expression of inflammatory cytokines and decreased synthesis of NO. As shown in Figure 3, the mRNA levels of MCP1 and VCAM1 in the high-glucose-cultured HUVECs were dramatically higher than those in the normal-glucose-cultured HUVECs. 1,25(OH)₂D₃ treatment significantly abrogated the increased mRNA expression of MCP1 and VCAM1 induced by high glucose. In line with the improved inflammatory cytokines observed by 1,25(OH)₂D₃ treatment, the supernatant NO generation in the active drug-treated HUVECs was also significantly higher than those in the high-glucose-treated HUVECs, but still lower than those in the normal-glucose-treated HUVECs. Similarly, when Nrf2 protein was downregulated, the antagonistic effect of 1,25(OH)₂D₃ on the increased mRNA expression of inflammatory factors and decreased NO generation induced by high glucose was significantly weakened.

To observe whether high glucose or high glucose plus 1,25(OH)₂D₃ can affect the subcellular distribution of Nrf2, we detected the expression levels of cytoplasm and nucleus protein of Nrf2, respectively. Besides, to observe the possible role of VDR in this process, we designed -siRNA targeting VDR. As shown in online suppl. Figure 3, high glucose decreased the protein expression of VDR in HUVECs compared with normal glucose, and VDR-specific siRNA significantly reduced the protein expression of VDR compared with scrambled siRNA. As shown in Figure 4, high glucose slightly and significantly promoted Nrf2 nuclear translocation compared with normal glucose. The cytoplasmic Nrf2 protein expression in high-glucose-cultured HUVECs was not significantly different from that in normal-glucose-cultured HUVECs, while the nucleus Nrf2 protein content in high-glucose-cultured HUVECs was slightly and significantly higher than that in normal-glucose-cultured HUVECs. 1,25(OH)₂D₃ intervention can significantly enhance Nrf2 nuclear translocation under hyperglycemic conditions, with the Nrf2 protein expression in the cytoplasm significantly -decreased and in the nucleus significantly increased, respectively. By blocking VDR in advance, the effect of 1,25(OH)₂D₃ on stimulating Nrf2 nuclear translocation was significantly abolished, with the subcellular distribution of Nrf2 similar to that of high-glucose-cultured HUVECs.

Hyperglycemia aggravates vascular endothelial dysfunction by inducing oxidative stress, which is the initial step of diabetic macrovascular disease. Our previous studies on animal and cellular levels found that vitamin D can inhibit oxidative stress and inflammation induced by high glucose and improve the diastolic response of the thoracic aortic endothelium to acetylcholine in diabetic mice [12, 13]. On this basis, this study further demonstrated that, by activating the Nrf2/ARE pathways and increasing downstream antioxidants including SOD, GR, and catalase activities, vitamin D can antagonize ROS generation and release of proinflammatory cytokines by HUVECs, increase cell activity and reduce apoptosis, and promote synthesis of NO, and thus may be a potential drug for the prevention and treatment of diabetic endothelial dysfunction and vascular complications.

As a classic antioxidant pathway in vivo, downregulation of Nrf2/ARE signaling has been proved to be associated with endothelial dysfunction and atherosclerosis [20]. Berridge [26] has proposed a hypothesis that vitamin D may be involved in the regulation of transcriptional activity of multiple genes by inducing the expression of 2 key regulatory proteins Nrf2 and Klotho, thereby maintaining redox signaling and calcium homeostasis. Later, Chen L and his colleagues [27] have confirmed the above hypothesis and documented that mice with targeted deletion of 25-hydroxyvitamin D3-1α-hydroxylase, a key enzyme involved in the synthesis of 1,25(OH)₂D₃, exhibited increased levels of oxidative stress markers in the skin and kidney tissue, enhanced aging-related protein p16, p53, and p21 expression, and attenuated cell proliferation activity, and vitamin D supplements can block the aging process by activating the VDR and Nrf2/ARE pathway. In this study, we found that preconditioning with vitamin D can promote Nrf2 nuclear translocation, induce the generation of downstream antioxidant substances, suppress the release of inflammatory factors, and ultimately alleviate the oxidative damage of the vascular endothelium induced by high glucose. Moreover, the antioxidative, anti-inflammatory, and antiapoptotic effects of 1,25(OH)₂D₃ observed in high-glucose-cultured HUVECs were abolished when cells were treated with siRNA to Nrf2. Consistent with our findings, Nakai et al. [29] revealed that vitamin D can inhibit local renal NADPH oxidase subunits and NF-κB protein expression by activating the Nrf2/ARE signaling pathway and reduce proteinuria and mesangial matrix proliferation in diabetic mice. Teixeira et al. [28] demonstrated that vitamin D pretreatment of HUVECs can prevent leptin-induced oxidative stress and inflammation by activating the Nrf2 antioxidant pathway. These studies suggest that vitamin D can antagonize the metabolic diseases-associated damage of target organs, including the heart and kidney, by upregulating the expression of multiple downstream antioxidant enzymes controlled by the Nrf2/ARE pathway.

Vitamin D works by binding to VDR, which includes nuclear and membrane receptors, and mediates the nongenetic and genetic effects of vitamin D, respectively. Our previous studies have found that vitamin D can exert atheroprotective effects through activating VDR and regulating PKC protein phosphorylation and NADPH oxidase subunit p22phox and prolyl isomerase-1 (Pin1) protein expression [12, 13]. This study further suggested that vitamin D promotes Nrf2 nucleus translocation in high-glucose-cultured HUVECs in a VDR-dependent manner. There are few reports on how to promote Nrf2 translocation to the nucleus after VDR activation. Kim et al. [30] found that the expression of Pin1 was upregulated in vascular smooth muscle cells in the intima of the femoral artery following guidewire injury, which inhibited nuclear translocation of Nrf2, downregulated Nrf2/ARE pathway-dependent heme oxygenase 1 expression, and promoted ROS-mediated vascular smooth muscle cell proliferation. Therefore, Pin1 may be the upstream negative regulatory molecule of Nrf2 and may become a therapeutic target for a variety of vascular diseases. It is worth noting that we have proved that the expression and activity of Pin1 protein decreased after activation of VDR at the level of vascular endothelial cells. Therefore, whether vitamin D can inhibit the expression and activity of Pin1 protein by activating VDR and then stimulate Nrf2 translocation to the nucleus and consequently play its role in maintaining vascular endothelial redox homeostasis under high-glucose environment needs further study in the future.

There are also some limitations in this study. Firstly, this study lacks further validation in animal experiments on whether vitamin D can improve vascular endothelial function through upregulation of the Nrf2/ARE signaling axis, such as vascular relaxation in vitro. Second, this study can only conclude that Nrf2 nucleus translocation is related to the vascular endothelial protection of vitamin D, but cannot directly prove its causal correlation. In the future, it is necessary to observe whether the endothelial protection of vitamin D is weakened under Nrf2 inhibition. Third, the vitamin D dose of 10⁻⁷ mmol/L used in this study corresponds to a supraphysiological dose in vivo, of which the long-term safety needs further evaluation.

Numerous epidemiological studies have shown that vitamin D deficiency is associated with diabetic macrovascular complications, and experimental studies have suggested that vitamin D supplementation may have cardioprotective effects. However, recent clinical trials of vitamin D in the primary prevention of cardiovascular disease and type 2 diabetes in the general population have yielded negative results [31-33]. Notably, participants in these clinical studies were not complicated with vitamin D deficiency, or benefits were observed in vitamin D-deficient subgroups [34, 35]. Therefore, the use of vitamin D in the prevention and treatment of cardiovascular disease and type 2 diabetes remains inconclusive [36].

In conclusion, this study first confirmed at the cellular level that vitamin D can protect the vascular endothelium dysfunction from oxidative stress induced by high glucose by activating the Nrf2/ARE pathway, which provides a certain theoretical basis for the potential use of vitamin D in the future for the prevention and treatment of major vascular complications in type 2 diabetes patients accompanied with vitamin D deficiency.

The authors thank Changsheng Xu for his technical support.

All participants signed an informed consent form, and all research procedures were approved by the Medical Research Ethics Committee of the Affiliated Hospital of Putian University.

The authors have no conflicts of interest to declare.

This study was supported by grants from the National Natural Science Foundation of China (Grant No. 81800278), Natural Science Foundation of Fujian Province (Grant No. 13171570), Youth Project of Fujian Provincial Health and Family Planning Commission (Grant No. 2017-1-93), and Young/Middle-aged Talent Cultivation Project of Fujian Provincial Health and Family Planning Commission (Grant No. 2018-ZQN-79) to Dr. Liming Lin and a grant from the Natural Science Foundation of Fujian Province (Grant No. 13181097) to Dr. Meifang Wu.

Kaizu Xu and Liming Lin conceived and designed the research. Meifang Wu and Ying Wu performed the experiments. Meifang Wu was responsible for data analysis and for writing the manuscript. All authors listed approved the final manuscript.

Meifang Wu and Ying Wu contributed equally to this article.