Role of Vitamin D in Maintaining Renal Epithelial Barrier Function in Uremic Conditions

As current kidney replacement therapies are not efficient enough for end-stage renal disease (ESRD) treatment, a bioartificial kidney (BAK) device, based on conditionally immortalized human proximal tubule epithelial cells (ciPTEC), could represent an attractive solution. The active transport activity of such a system was recently demonstrated. In addition, endocrine functions of the cells, such as vitamin D activation, are relevant. The organic anion transporter 1 (OAT-1) overexpressing ciPTEC line presented 1α-hydroxylase (CYP27B1), 24-hydroxylase (CYP24A1) and vitamin D receptor (VDR), responsible for vitamin D activation, degradation and function, respectively. The ability to produce and secrete 1α,25-dihydroxy-vitamin D3, was shown after incubation with the precursor, 25-hydroxy-vitamin D3. The beneficial effect of vitamin D on cell function and behavior in uremic conditions was studied in the presence of an anionic uremic toxins mixture. Vitamin D could restore cell viability, and inflammatory and oxidative status, as shown by cell metabolic activity, interleukin-6 (IL-6) levels and reactive oxygen species (ROS) production, respectively. Finally, vitamin D restored transepithelial barrier function, as evidenced by decreased inulin-FITC leakage in biofunctionalized hollow fiber membranes (HFM) carrying ciPTEC-OAT1. In conclusion, the protective effects of vitamin D in uremic conditions and proven ciPTEC-OAT1 endocrine function encourage the use of these cells for BAK application.


Introduction
It has been reported that chronic kidney disease (CKD), defined as the sustained presence of a decreased glomerular filtration rate (GFR) with or without increased albumin excretion, has a rather high global prevalence, estimated to be between 11% and 13% [1]. The progressive loss of kidney function will ultimately lead to a permanent state of end-stage renal disease (ESRD). Kidney failure is accompanied by a noticeable accumulation of a variety of endogenous uremic metabolites that are not efficiently cleared by the kidneys, leading to a broad range of pathologies, mostly cardiovascular disease and bone disorders, with reduced quality of life, as well as significantly increased mortality [2][3][4]. Although kidney transplantation is the treatment of choice for most patients with ESRD, patients who are older or have significant comorbidity are not eligible for transplantation. Table 1. Concentrations of anionic uremic toxins in healthy individuals, uremic patients, and as applied in the present study. Concentrations used are adapted from EUToX Uremic Solutes Database (http://uremic-toxins.org/DataBase.html) and Jansen et al. [33].

Expression of Vitamin D Metabolism and Function-Related Genes in ciPTEC-OAT1
In ciPTEC-OAT1, the expression of genes involved in vitamin D metabolism, activation and degradation, 1α-hydroxylase and CYP24A1, respectively, and of VDR was confirmed by real-time PCR (Figure 1a). Agarose gel electrophoresis confirmed the specificity of the primers as the size of the PCR products corresponded to the expected amplicon length ( Figure S1). The housekeeping gene used for normalization was HPRT1, whose expression levels did not change upon various stimulations ( Figure S2). Moreover, vitamin D's effect on the expression of these genes was examined after 24 h exposure to either 100 nM or 1 μM of 1,25(OH)2D3. While no significant impact on VDR expression was p-cresyl sulfate 10

Expression of Vitamin D Metabolism and Function-Related Genes in ciPTEC-OAT1
In ciPTEC-OAT1, the expression of genes involved in vitamin D metabolism, activation and degradation, 1α-hydroxylase and CYP24A1, respectively, and of VDR was confirmed by real-time PCR (Figure 1a). Agarose gel electrophoresis confirmed the specificity of the primers as the size of the PCR products corresponded to the expected amplicon length ( Figure S1). The housekeeping gene used for normalization was HPRT1, whose expression levels did not change upon various stimulations ( Figure S2). Moreover, vitamin D's effect on the expression of these genes was examined after 24 h exposure to either 100 nM or 1 μM of 1,25(OH)2D3. While no significant impact on VDR expression was

Expression of Vitamin D Metabolism and Function-Related Genes in ciPTEC-OAT1
In ciPTEC-OAT1, the expression of genes involved in vitamin D metabolism, activation and degradation, 1α-hydroxylase and CYP24A1, respectively, and of VDR was confirmed by real-time PCR (Figure 1a). Agarose gel electrophoresis confirmed the specificity of the primers as the size of the PCR products corresponded to the expected amplicon length ( Figure S1). The housekeeping gene used for normalization was HPRT1, whose expression levels did not change upon various stimulations ( Figure S2). Moreover, vitamin D's effect on the expression of these genes was examined after 24 h exposure to either 100 nM or 1 μM of 1,25(OH)2D3. While no significant impact on VDR expression was

Expression of Vitamin D Metabolism and Function-Related Genes in ciPTEC-OAT1
In ciPTEC-OAT1, the expression of genes involved in vitamin D metabolism, activation and degradation, 1α-hydroxylase and CYP24A1, respectively, and of VDR was confirmed by real-time PCR (Figure 1a). Agarose gel electrophoresis confirmed the specificity of the primers as the size of the PCR products corresponded to the expected amplicon length ( Figure S1). The housekeeping gene used for normalization was HPRT1, whose expression levels did not change upon various stimulations ( Figure S2). Moreover, vitamin D's effect on the expression of these genes was examined after 24 h exposure to either 100 nM or 1 μM of 1,25(OH)2D3. While no significant impact on VDR expression was

Expression of Vitamin D Metabolism and Function-Related Genes in ciPTEC-OAT1
In ciPTEC-OAT1, the expression of genes involved in vitamin D metabolism, activation and degradation, 1α-hydroxylase and CYP24A1, respectively, and of VDR was confirmed by real-time PCR (Figure 1a). Agarose gel electrophoresis confirmed the specificity of the primers as the size of the PCR products corresponded to the expected amplicon length ( Figure S1). The housekeeping gene used for normalization was HPRT1, whose expression levels did not change upon various stimulations ( Figure S2). Moreover, vitamin D's effect on the expression of these genes was examined after 24 h exposure to either 100 nM or 1 μM of 1,25(OH)2D3. While no significant impact on VDR expression was

Expression of Vitamin D Metabolism and Function-Related Genes in ciPTEC-OAT1
In ciPTEC-OAT1, the expression of genes involved in vitamin D metabolism, activation and degradation, 1α-hydroxylase and CYP24A1, respectively, and of VDR was confirmed by real-time PCR (Figure 1a). Agarose gel electrophoresis confirmed the specificity of the primers as the size of the PCR products corresponded to the expected amplicon length ( Figure S1). The housekeeping gene used for normalization was HPRT1, whose expression levels did not change upon various stimulations ( Figure S2). Moreover, vitamin D's effect on the expression of these genes was examined after 24 h exposure to either 100 nM or 1 μM of 1,25(OH)2D3. While no significant impact on VDR expression was

Expression of Vitamin D Metabolism and Function-Related Genes in ciPTEC-OAT1
In ciPTEC-OAT1, the expression of genes involved in vitamin D metabolism, activation and degradation, 1α-hydroxylase and CYP24A1, respectively, and of VDR was confirmed by real-time PCR (Figure 1a). Agarose gel electrophoresis confirmed the specificity of the primers as the size of the PCR products corresponded to the expected amplicon length ( Figure S1). The housekeeping gene used for normalization was HPRT1, whose expression levels did not change upon various stimulations ( Figure S2). Moreover, vitamin D's effect on the expression of these genes was examined after 24 h exposure to either 100 nM or 1 μM of 1,25(OH)2D3. While no significant impact on VDR expression was

Expression of Vitamin D Metabolism and Function-Related Genes in ciPTEC-OAT1
In ciPTEC-OAT1, the expression of genes involved in vitamin D metabolism, activation and degradation, 1α-hydroxylase and CYP24A1, respectively, and of VDR was confirmed by real-time PCR (Figure 1a). Agarose gel electrophoresis confirmed the specificity of the primers as the size of the PCR products corresponded to the expected amplicon length ( Figure S1). The housekeeping gene used for normalization was HPRT1, whose expression levels did not change upon various stimulations ( Figure S2). Moreover, vitamin D's effect on the expression of these genes was examined after 24 h exposure to either 100 nM or 1 μM of 1,25(OH)2D3. While no significant impact on VDR expression was

Expression of Vitamin D Metabolism and Function-Related Genes in ciPTEC-OAT1
In ciPTEC-OAT1, the expression of genes involved in vitamin D metabolism, activation and degradation, 1α-hydroxylase and CYP24A1, respectively, and of VDR was confirmed by real-time PCR (Figure 1a). Agarose gel electrophoresis confirmed the specificity of the primers as the size of the PCR products corresponded to the expected amplicon length ( Figure S1). The housekeeping gene used for normalization was HPRT1, whose expression levels did not change upon various stimulations ( Figure S2). Moreover, vitamin D's effect on the expression of these genes was examined after 24 h exposure to either 100 nM or 1 µM of 1,25(OH) 2 D 3. While no significant impact on VDR expression was found, the gene expression of the two enzymes was significantly affected. In fact, an almost 50% reduction in 1α-hydroxylase expression was observed after 1,25(OH) 2 D 3 treatment when used at 1 µM, and a more than 1000-fold increase in CYP24A1 expression, regardless of the vitamin D concentration used ( Figure 1b). No significant changes in gene expression were observed in the presence of a uremic toxins mixture (UT mix) at 1× or 2.5× concentrations (Figure 1c), which was used to mimic the conditions of kidney patients. found, the gene expression of the two enzymes was significantly affected. In fact, an almost 50% reduction in 1α-hydroxylase expression was observed after 1,25(OH)2D3 treatment when used at 1 μM, and a more than 1000-fold increase in CYP24A1 expression, regardless of the vitamin D concentration used ( Figure 1b). No significant changes in gene expression were observed in the presence of a uremic toxins mixture (UT mix) at 1× or 2.5× concentrations (Figure 1c), which was used to mimic the conditions of kidney patients.

Conversion of 25(OH)D 3 to 1,25(OH) 2 D 3 by ciPTEC-OAT1
To assess whether ciPTEC-OAT1 are able to produce the most active form of vitamin D, 1,25(OH) 2 D 3 , cells were exposed to 100 nM 25(OH)D 3 for 24 h, in the presence or absence of 1α-hydroxylase inhibitor-ketoconazole (10 µM). Measured levels of 1,25(OH) 2 D 3 confirmed that ciPTEC-OAT1 did produce the active form of vitamin D and the conversion was sensitive to inhibition by ketoconazole (Figure 2a). Uremic conditions (1× UT mix) did not influence the vitamin D activation (Figure 2b).

Conversion of 25(OH)D3 to 1,25(OH)2D3 by ciPTEC-OAT1
To assess whether ciPTEC-OAT1 are able to produce the most active form of vitamin D, 1,25(OH)2D3, cells were exposed to 100 nM 25(OH)D3 for 24 h, in the presence or absence of 1α-hydroxylase inhibitor-ketoconazole (10 μM). Measured levels of 1,25(OH)2D3 confirmed that ciPTEC-OAT1 did produce the active form of vitamin D and the conversion was sensitive to inhibition by ketoconazole (Figure 2a). Uremic conditions (1× UT mix) did not influence the vitamin D activation (Figure 2b). Three independent experiments were performed. ** p < 0.01 (One-way ANOVA followed by Dunnett's multiple comparison test, using as a control either untreated sample or 25(OH)D3 treated sample, as indicated).

Protective Effect of 1,25(OH)2D3 on Anionic Uremic Toxin Mix Induced Cell Toxicity
To further examine the effect of 1,25(OH)2D3 on ciPTEC-OAT1 viability in normal and uremic conditions, cells were exposed to several concentrations of 1,25(OH)2D3 in the absence or presence of increasing concentrations of UT mix. As shown in Figure 3a, the active form of vitamin D alone did not compromise cell viability. However, anionic uremic toxins did reduce cell viability after 24 h incubation, by approximately 10%, 25%, and 62% for 2.5×, 5×, and 10× concentrated mixtures, respectively (Figure 3b). Co-incubation of 1,25(OH)2D3 with UT mix could mitigate the decrease in cell viability, especially when toxicity was induced by higher concentrations of UT mix (5× and 10×; Figure  3c). Three independent experiments were performed. ** p < 0.01 (One-way ANOVA followed by Dunnett's multiple comparison test, using as a control either untreated sample or 25(OH)D 3 treated sample, as indicated).

Protective Effect of 1,25(OH) 2 D 3 on Anionic Uremic Toxin Mix Induced Cell Toxicity
To further examine the effect of 1,25(OH) 2 D 3 on ciPTEC-OAT1 viability in normal and uremic conditions, cells were exposed to several concentrations of 1,25(OH) 2 D 3 in the absence or presence of increasing concentrations of UT mix. As shown in Figure 3a, the active form of vitamin D alone did not compromise cell viability. However, anionic uremic toxins did reduce cell viability after 24 h incubation, by approximately 10%, 25%, and 62% for 2.5×, 5×, and 10× concentrated mixtures, respectively (Figure 3b). Co-incubation of 1,25(OH) 2 D 3 with UT mix could mitigate the decrease in cell viability, especially when toxicity was induced by higher concentrations of UT mix (5× and 10×; Figure 3c).

Protective Effect of 1,25(OH)2D3 on Anionic Uremic Toxin Mix Induced Oxidative Stress
To evaluate ciPTEC-OAT1 susceptibility to oxidative stress in uremic conditions and the anti-oxidative effect of 1,25(OH)2D3, intracellular reactive oxygen species (ROS) generation was measured. Cells were exposed to 5× UT mix, 1,25(OH)2D3 (500 nM or 1 μM) or a combination of UT mix and 1,25(OH)2D3 for 2 h. UT mix induced a 1.5-fold increase in ROS production, which was attenuated significantly when adding vitamin D as a co-treatment, regardless of concentration ( Figure 4). Also, the positive control H2O2 (200 μM) significantly enhanced ROS generation ( Figure  4). Using 10× UT mix, similar effects of vitamin D on intracellular ROS levels were observed ( Figure  S3a).

Protective Effect of 1,25(OH) 2 D 3 on Anionic Uremic Toxin Mix Induced Oxidative Stress
To evaluate ciPTEC-OAT1 susceptibility to oxidative stress in uremic conditions and the anti-oxidative effect of 1,25(OH) 2 D 3 , intracellular reactive oxygen species (ROS) generation was measured. Cells were exposed to 5× UT mix, 1,25(OH) 2 D 3 (500 nM or 1 µM) or a combination of UT mix and 1,25(OH) 2 D 3 for 2 h. UT mix induced a 1.5-fold increase in ROS production, which was attenuated significantly when adding vitamin D as a co-treatment, regardless of concentration ( Figure 4). Also, the positive control H 2 O 2 (200 µM) significantly enhanced ROS generation (Figure 4). Using 10× UT mix, similar effects of vitamin D on intracellular ROS levels were observed ( Figure S3a). Three independent experiments were performed in duplicate. * p < 0.05, ** p < 0.01, *** p < 0.001 (One-way ANOVA followed by Dunnett's multiple comparison test, using as a control either untreated sample or 5× UT mix, as indicated).

Anti-Inflammatory Effect of 1,25(OH)2D3 in Inflammatory and Uremic Conditions in ciPTEC-OAT1
Interleukin-6 (IL-6) levels in cell culture supernatant were measured to assess the effect of UT mix and vitamin D on the inflammatory response of ciPTEC-OAT1. Lipopolysaccharide (LPS) (10 μg/mL), which was used as a positive control, induced a 3-fold increase in IL-6 levels after 24 h exposure. Vitamin D, however, was able to reverse this pro-inflammatory effect of LPS by reducing the IL-6 levels. A 1.6-fold reduction was found for 100 nM and 500 nM, and a 1.8-fold reduction for 1 μM of 1,25(OH)2D3 (Figure 5a). A 2-fold and 2.8-fold increase in IL-6 levels was observed following the exposure to 1× and 2.5× UT mix, respectively. Similar to what was observed for LPS, a small but evident trend in IL-6 level reduction was detected after co-treatment with 1,25(OH)2D3 (Figure 5b,c). In all conditions, TNF-α levels measured were below the limit of detection. Three independent experiments were performed in duplicate. * p < 0.05, ** p < 0.01, *** p < 0.001 (One-way ANOVA followed by Dunnett's multiple comparison test, using as a control either untreated sample or 5× UT mix, as indicated).

Anti-Inflammatory Effect of 1,25(OH) 2 D 3 in Inflammatory and Uremic Conditions in ciPTEC-OAT1
Interleukin-6 (IL-6) levels in cell culture supernatant were measured to assess the effect of UT mix and vitamin D on the inflammatory response of ciPTEC-OAT1. Lipopolysaccharide (LPS) (10 µg/mL), which was used as a positive control, induced a 3-fold increase in IL-6 levels after 24 h exposure. Vitamin D, however, was able to reverse this pro-inflammatory effect of LPS by reducing the IL-6 levels. A 1.6-fold reduction was found for 100 nM and 500 nM, and a 1.8-fold reduction for 1 µM of 1,25(OH) 2 D 3 (Figure 5a). A 2-fold and 2.8-fold increase in IL-6 levels was observed following the exposure to 1× and 2.5× UT mix, respectively. Similar to what was observed for LPS, a small but evident trend in IL-6 level reduction was detected after co-treatment with 1,25(OH) 2 D 3 (Figure 5b,c). In all conditions, TNF-α levels measured were below the limit of detection. Concentration expressed as pg/mL (mean ± SEM). At least three independent experiments were performed. * p < 0.05, ** p < 0.01, *** p < 0.001 (One-way ANOVA followed by Dunnett's multiple comparison test, using as a control either untreated sample, LPS 10 μg/mL, 1× UT mix or 2.5× UT mix, as indicated).

Discussion
In the present study, we demonstrated the ability of ciPTEC-OAT1 to produce the most active form of vitamin D, 1,25(OH) 2 D 3 , and its beneficial effect on various aspects of uremic conditions in ciPTEC-OAT1, including the protective effect on epithelial monolayer tightness. Considering the importance of the vitamin D deficiency often observed in CKD and ESRD, and the fact that vitamin D production is one of the main endocrine functions of proximal tubule cells, we were interested in determining whether ciPTEC-OAT1, intended for BAK purposes, possess all the necessary enzymes responsible for vitamin D metabolism. It has been shown that proximal tubule cells express 1α-hydroxylase, which is responsible for 25(OH)D 3 conversion into 1,25(OH) 2 D 3 , as well as CYP24A1, involved in 1,25(OH) 2 D 3 degradation [34]. Besides the proximal tubule, there are other, extra-renal sites expressing these enzymes and producing vitamin D ,such as the cells of the immune system (macrophages, monocytes, dendritic cells), and epithelial cells of the gastrointestinal tract, skin, breast, and lungs [35][36][37][38][39][40][41]. However, the major part of circulating vitamin D levels is kidney-derived, hence the severe deficiency is due to kidney failure [8]. We first evaluated the baseline expression of the genes for the activating and degrading enzymes and found that both are present in ciPTEC-OAT1, with a higher basal expression of the activating enzyme compared to the inactivating one. In line with the literature, following exposure to the active form of vitamin D, we observed a significant downregulation of 1α-hydroxylase and substantial upregulation of CYP24A1, confirming the existence of the negative feedback of 1,25(OH) 2 D 3 on its circulating concentration [42,43]. Moreover, VDR is also expressed but not influenced by vitamin D treatment. Neither normal (1×) nor high (2.5×) concentrations of uremic toxins affected the expression of the enzymes and VDR, indicating that gene expression is not likely be altered in uremic conditions. In addition, we determined the actual conversion of 25(OH)D 3 into 1,25(OH) 2 D 3 in basic and uremic conditions. Cells were treated with a physiologic concentration of inactive vitamin D (100 nM, corresponding to 40 ng/mL in healthy individuals) and after 24 h the amount of active form of vitamin D generated was 32.5 pmol/L, corresponding to 13.5 pg/mL, which is slightly below the range of the active vitamin D serum levels in healthy patients [44,45]. Although speculative, this indicates that ciPTEC-OAT1 may be able to sufficiently produce the active form of vitamin D. In accordance with gene expression levels in uremic conditions, we found that the conversion of 25(OH)D 3 was not affected by uremic toxins, suggesting a normal endocrine function of ciPTEC-OAT1 in conditions relevant to BAK applications and kidney disease.
In addition to its well-described roles, such as maintenance of calcium homeostasis and mineralization, vitamin D is able to exert other, non-calciotropic effects [46]. Among the most relevant ones are certainly immunomodulatory actions [47,48], with promotion of innate immune responses and the ability of immune system to fight infections [49][50][51], but also the suppression of the adaptive immune system with generation of tolerance, as shown for various auto-immune disorders (multiple sclerosis, type 1 diabetes, systemic lupus erythematosus and rheumatoid arthritis) [52][53][54][55]. Moreover, vitamin D is also involved in modulation of cell growth and proliferation, both in benign hyperplastic conditions and various cancer types [56].
In this study, we were particularly interested in the autocrine actions of vitamin D and therefore evaluated its effects on several cellular aspects of renal PTEC in uremic conditions. Initially, we observed that uremic toxins affect cell viability in a dose-dependent manner, while vitamin D did not have any effect. However, in the presence of anionic uremic toxins, especially at higher doses, vitamin D could restore cell viability. Numerous studies have described that some uremic toxins, such as indoxyl sulfate (IS), p-cresyl sulfate (pCS), and indole-3-acetic acid (IAA), are associated with increased inflammatory responses and oxidative stress both in vitro and in vivo [57][58][59][60][61][62][63][64][65][66]. To further address this, we measured IL-6 release by cells as an indication of inflammatory response, and ROS production, as a marker of oxidative stress, in uremic conditions and in the presence or absence of 1,25(OH) 2 D 3 in ciPTEC-OAT1. We found that uremic toxins do increase IL-6 secreted levels, as well as ROS intracellular generation. Moreover, our results indicate that 1,25(OH) 2 D 3 is able to reduce this increase in IL-6 levels and ROS production, confirming that vitamin D indeed has protective effects in uremic conditions, as suggested previously by in vivo studies, evaluating therapeutic effects of paricalcitol, a VDR activator, in uremic rats and hemodialysis patients [67][68][69].
A growing body of evidence suggests that vitamin D is essential for the correct functioning and maintenance of epithelial barriers, including gut mucosal barrier, corneal, pulmonary and kidney epithelial barriers, and its deficiency has been reported to promote barrier dysfunction and increased permeability [18,[70][71][72]. The key tight junction proteins responsible for a tight monolayer formation in kidney proximal tubule are claudin 2 and ZO-1 [73,74]. We determined the effect of vitamin D on the stability of the proximal tubule epithelial monolayer in uremic conditions. For that purpose, ciPTEC-OAT1 were cultured on double-coated HFM to create kidney tubules consisting of mature epithelial cell monolayers, expressing both ZO-1 and claudin 2 ( Figure S4a). Interestingly, we observed increased barrier permeability in the presence of uremic toxins, as shown by inulin-FITC leakage. However, in the presence of vitamin D, a clear trend towards a smaller increase in inulin-FITC diffusion was detected, suggesting the protective effect of vitamin D on the proximal tubule epithelial barrier integrity. Because the gene expression levels of ZO-1 and claudin 2 were not significantly influenced by uremic toxins or by vitamin D (Figure S4b,c), we expect that the effect of vitamin D in attenuating epithelial barrier permeability might be due to a redistribution of tight junction proteins rather than an increased protein expression, as observed previously for the intestinal barrier [18].
The findings of the present study clearly support further development of BAK as a treatment modality in patients with ESRD. Extensive previous studies described an efficient way of culturing ciPTEC-OAT1 on double-coated HFM with the formation of tight epithelial monolayers, as well as the active transport activity of both OAT1 and OCT2, proteins responsible for the clearance of uremic waste metabolites [30,32]. In addition, the lack of ciPTEC induced alloimmune response in vitro [75] and the successful upscaling of the BAK device [76] further encourage the use of these cells. Our current demonstration of the ability of the cells to activate and secrete the most active form of vitamin D is an additional important asset of the system. However, future studies should further investigate the choice of membranes used to support cell attachment, growth and monolayer formation, as this could potentially abolish the vitamin D activation function of ciPTEC-OAT1. It has been shown that some membrane types, in particular highly adsorptive and high cut-off membranes, could lead to a significant reduction in VDBP and 25(OH)D 3 levels [6], potentially compromising the availability of 25(OH)D 3 for megalin uptake and conversion by 1α-hydroxylase. For that reason, the polyethersulfone membranes used to support ciPTEC-OAT1 in the current settings should be tested for its suitability for use in BAK devices.
In conclusion, the ability of ciPTEC-OAT1 to produce active vitamin D could considerably boost BAK function, thus allowing the improvement of health status of kidney patients, not only by removing the excessive amounts of protein bound uremic toxins, but also by replicating one of the key endocrine functions of the proximal tubule. Eventually, the presence of 1,25(OH) 2 D 3 would greatly contribute to the maintenance of a strong epithelial monolayers for correct and efficient BAK function, and to improved mineral homeostasis and skeletal and cardiovascular health in CKD and ESRD patients. Future experiments will be designed to evaluate the safety and efficacy of a prototype BAK device in vivo, including the assessment of the beneficial effects of vitamin D as presented in this study.

Cell Culture of ciPTEC-OAT1
The ciPTEC-OAT1 cell line was cultured as reported previously [29]. Briefly, cells were cultured in Dulbecco's Modified Eagle Medium/Nutrient Mixture F-12 (1:1 DMEM/F-12) (Gibco, Life Technologies, Paisley, UK) supplemented with 10% fetal calf serum (FCS) (Greiner Bio-One, Alphen aan den Rijn, The Netherlands), 5 µg/mL insulin, 5 µg/mL transferrin, 5 µg/mL selenium, 35 ng/mL hydrocortisone, 10 ng/mL epidermal growth factor and 40 pg/mL tri-iodothyronine to form a complete culture medium, without addition of antibiotics and up to a maximum of 60 passages. Cells were cultured at 33 • C and 5% (v/v) CO 2 to allow proliferation and prior to the experiments seeded at a density of 55,000 cell/cm 2 . Subsequently, cells were grown for one day at 33 • C, 5% (v/v) CO 2 to allow adhesion, then cultured for seven days at 37 • C, 5% (v/v) CO 2 for differentiation and maturation, refreshing the medium every other day.

ciPTEC-OAT1 Exposure to Uremic Toxins Mixture
In order to replicate the uremic conditions present in kidney patients, a specific mixture of eight known anionic uremic toxins (Table 1), predominantly derived from endogenous metabolism pathways and food digestion in the gut [33], and corresponding approximately to the concentrations found in patients (1×), or higher (2.5×, 5× and 10×) ( Table 1), was used in the present study. It was prepared as a 100× concentrated mixture in a serum-free medium and subsequently diluted to desired concentrations.

Cell Viability Assay
Cell viability was measured using PrestoBlue ® cell viability reagent (Life Technologies). After seven days of maturation, cells were exposed to increasing concentrations of 1,25(OH) 2 D 3 (100 nM, 500 nM, 1 µM), anionic UT mix (1-, 2.5-, 5-, or 10-times concentrated) and a combination of 1,25(OH) 2 D 3 and UT mix in the previously mentioned concentrations. Following 24 h incubation at 37 • C, 5% (v/v) CO 2 , ciPTEC were rinsed once with Hank's Balanced Salt Solution (HBSS; Gibco, Life Technologies) and incubated with PrestoBlue ® cell viability reagent (diluted 1:10 in complete culture medium), in the dark. After 1 h incubation at 37 • C, 5% (v/v) CO 2 , the fluorescence was measured using the Fluoroskan Ascent FL microplate reader, at excitation wavelength of 530 nm and emission wavelength of 590 nm. Data were corrected for the background, normalized to untreated cells, and presented as relative cell viability.

RNA Extraction, cDNA Synthesis, and Real-Time PCR
Total RNA from ciPTEC-OAT1 exposed to 1,25(OH) 2 D 3 (100 nM and 1 µM) and UT mix (1× and 2.5×) for 24 h, was isolated using the RNeasy Mini kit (Qiagen, Venlo, The Netherlands) according to the manufacturer's instructions and quantified using the NanoDrop ® ND-1000 spectrophotometer. Reverse transcription of RNA to complementary DNA (cDNA) was performed using the iScript TM Reverse Transcription Supermix (Bio-Rad Laboratories, Hercules, CA, USA) following manufacturer's instructions. Subsequently, Real-Time PCR was performed using the iQ SYBR ® Green Supermix (Bio-Rad Laboratories) as indicated in manufacturer's protocol and by means of CFX96 TM Real-Time PCR Detection System (Bio-Rad Laboratories). The data were analyzed using Bio-Rad CFX Manager TM Software version 3.1 (Bio-Rad Laboratories) and expressed as relative gene expression, using untreated cells as the reference sample. HPRT1 was used as a housekeeping gene for normalization. Specific sense and anti-sense primers for HPRT1 (forward: ACATCTGGAGTCCTATTGACATCG; reverse: CCGCCCAAAGGGAACTGATAG), VDR (forward: CTGACCCTGGAGACTTTGAC; reverse: TTCCTCTGCACTTCCTCATC), 1α-hydroxylase, (forward: GGCAGAGTCTGAATTGCAAAT; reverse: Measured fluorescence values were corrected for the fluorescence of the blank sample (non-stained lysed cells) and used to calculate relative ROS production, using untreated cells as the reference.

CiPTEC-OAT1 Epithelial Monolayer Integrity
To investigate the effect of vitamin D on epithelial monolayer barrier function in uremic conditions, ciPTEC-OAT1 were cultured on L-DOPA (2 mg/mL) and collagen IV (25 µg/mL) coated HFM, mounted on a tailor-made flow chamber as described previously [30,32]. HFM with untreated mature ciPTEC-OAT1 monolayers and those exposed to 1,25(OH) 2 D 3 (1 µM), 2.5× UT mix or a combination of both, were perfused with inulin-FITC (0.1 mg/mL) in Krebs-Henseleit buffer supplemented with 10 mM HEPES, pH 7.4, for 10 min. Next, aliquots from the apical compartment were collected and used to measure fluorescence by means of fluorescent microplate reader (Fluoroskan Ascent FL, Labsystems), at excitation wavelength of 492 nm and emission wavelength of 518 nm. Background values were subtracted and normalized arbitrary fluorescence unit (AFU) data were converted and plotted as nmol·min −1 ·cm −2 , as described previously [32]. From each single replicate (fiber), three different regions, with an area of 0.157 cm 2 , were analyzed.

Immunocytochemistry
To assess the expression of tight junction protein ZO-1, ciPTEC-OAT1 cultured on double-coated HFM were fixed with 4% (w/v) paraformaldehyde dissolved in PHEM buffer (120 mM PIPES, 50 mM HEPES, 4 mM MgCl 2 , 20 mM EGTA) for 15 min. After washing the samples with HBSS, block solution (2% (v/v) FCS, 2% (w/v) bovine serum albumin (BSA), 0.1% (v/v) Tween20 in HBSS) was added. The primary antibody, rabbit anti-human ZO-1 (Invitrogen, Carlsbad, CA, USA), was diluted in blocking buffer (1:200) and incubated overnight at 4 • C. Following three washing steps with HBSS, the secondary antibody, goat anti-rabbit IgG Alexa 568 (Life Technologies, Eugene, OR, USA) was added in a concentration of 1:200 and incubated for 1 h at room temperature. Finally, ProLong TM Gold antifade reagent containing DAPI (Life Technologies, Eugene, OR, USA) was used for nuclear staining, and to mount the fibers containing cells on the Willco glass bottom dishes (WillCo Wells B.V., Amsterdam, The Netherlands). Cells were imaged using confocal microscope (Leica TCS SP8 X, Leica Microsystems CMS GmbH, Wetzlar, Germany) and analyzed using Leica Application Suite X software (Leica Microsystems CMS GmbH).

Data Analysis
All data are presented as mean ± standard error of the mean (SEM). Statistical analysis was performed using one-way ANOVA followed by Dunnett's multiple comparison test. A p-value < 0.05 was considered significant. Datasets were assessed for normality and equal variances assumptions prior to one-way ANOVA, using Kolmogorov-Smirnov and Bartlett's tests, respectively. Even though some datasets did not meet one of the assumptions, due to a limited number of measurements, the expected effect on the Type I error in one-way ANOVA is minimal. Software used for statistical analysis was GraphPad Prism (version 6.07; GraphPad software, La Jolla, CA, USA). In most experiments at least three independent experiments were performed in duplicates, unless otherwise stated. The exact sample size for each experiment is indicated in the corresponding figure legend.