Stabilization of cell–cell junctions by active vitamin D ameliorates uraemia-induced loss of human endothelial barrier function

Stabilization of cell–cell junctions by active vitamin D ameliorates uraemia-induced loss of... Abstract Background Uraemia induces endothelial cell (EC) injury and impaired repair capacity, for which the underlying mechanism remains unclear. Active vitamin D (VD) may promote endothelial repair, however, the mechanism that mediates the effects of VD in chronic kidney disease are poorly understood. Thus, we investigated uraemia-induced endothelial damage and the protection against such damage by active VD. Methods We applied electric cell-substrate impedance sensing (ECIS) to study real-time responses of human ECs exposed to pooled uraemic and non-uraemic plasma with or without the addition of active VD. The effects of indoxyl sulphate and p-cresol were tested in non-uraemic plasma. Structural changes for vascular endothelial (VE)-cadherin and F-actin were assessed by immunostaining and quantified. Results The exposure of ECs to uraemic media significantly decreased endothelial barrier function after 24 h. Cell migration after electrical wounding and recovery of the barrier after thrombin-induced loss of integrity were significantly impaired in uraemic-medium stimulated cells and cells exposed to indoxyl sulphate and p-cresol. This effect on ECIS was dependent on loss of cell–cell interaction. Mechanistically, we found that EC, exposed to uraemic media, displayed disrupted VE-cadherin interactions and F-actin reorganization. VD supplementation rescued both endothelial barrier function and cell–cell interactions in ECs exposed to uraemic media. These events were associated with an increment of VE-cadherin at intercellular junctions. Conclusions Our data demonstrate a potentially clinically relevant mechanism for uraemia-induced endothelial damage. Furthermore, active VD rescued the uraemic medium-induced loss of cell–cell adhesion, revealing a novel role of active VD in preservation of endothelial integrity during uraemia. barrier function, CKD, endothelial dysfunction, VE-cadherin, vitamin D INTRODUCTION Increased mortality risk in patients with chronic kidney disease (CKD) is frequently linked to cardiovascular disease [1]. CKD impacts vascular health, and especially vascular calcification of the medial layer of the vessel wall or secondary calcifications of atheromatous plaques has received much attention [2]. The integrity of the endothelium is a somewhat neglected aspect, yet it plays a pivotal role in different aspects of the vascular function [3, 4]. Endothelial dysfunction is suggested to be an instigator of the development and progression of cardiovascular events in CKD [5]. Signs of endothelial cell (EC) injury such as impaired endothelium-dependent vasodilatation, suppression of nitric oxide production and inflammation are frequently observed in CKD patients [6–8]. However, while it is conceivable that uraemic toxins may interfere with vascular endothelial integrity [9–11], there is only limited evidence for a direct effect of uraemia on EC. Several kidney disease-related risk factors, including the progressive decline of circulating active vitamin D (VD) levels, are suggested to cause adverse changes in vascular function [12–14]. In addition, 25 hydroxyvitamin D [25(OH)D] deficiency is suggested to have a crucial contribution to the development and progression of EC dysfunction in CKD patients [15]. Moreover, active VD supplementation was suggested to reduce cardiovascular events such as vascular calcification or cardiac abnormalities in CKD animal models [16, 17]. However, these results contrast with clinical trials using active VD therapy in CKD patients and controversy remains regarding its beneficial potential, especially with regard to the attenuation of cardiac disease in CKD [18, 19]. In EC dysfunction, it was shown that uraemia impaired vasodilatory response in animals [20, 21], and disturbed artery flow-mediated dilatation in selected patients was prevented by active VD supplementation [22]. However, while it is recognized that active VD analogues modulate endothelial inflammation, cell–cell junction stability and oxidative stress in in vitro studies [23–25], the mechanisms that modulate these effects of active VD in CKD are poorly understood [26]. This lack of knowledge of molecular mechanisms that explain the effects of uraemia on endothelial barrier function and the beneficial effect of active VD, formed the rationale to use innovative in vitro measurements to study the EC behaviour in CKD. For this, we applied electric cell-substrate impedance sensing (ECIS) [27] to measure the real-time changes of the barrier function of human EC under the influence of uraemic plasma estimated glomerular filtration rate [estimated glomerular filtration rate (eGFR)  <18 mL/min/1.73 m2]. We found impaired barrier function and disrupted cell–cell contact of EC exposed to human uraemic plasma. These results were in agreement with the effects of the uraemic toxins indoxyl sulphate and p-cresol. Altered levels of the adherens junction protein vascular endothelial (VE)-cadherin were identified as potentially underlying this EC dysfunction. Finally, the active VD analogue paricalcitol (PC) attenuated the disrupted barrier function induced by uraemia and restored VE-cadherin at junctions. We suggest that targeting impaired stability of adherens junctions might form a therapeutic goal for EC dysfunction in CKD. MATERIALS AND METHODS Patients and sample collection CKD plasma samples were obtained from six patients (three male and three female) non-dialysed with CKD (Stages 4–5), age 75.8 ± 7.9 years (mean ± SD), eGFR 18.1 ± 6.7 mL/min/1.73 m2. Non-CKD plasma samples were from six healthy donors (four male and two female), age 35.9 ± 14 years with eGFR  >90 mL/min/1.73 m2. Plasma samples were obtained from each patient by centrifugation at 2500 g during 10 min at 4°C. Supernatant was centrifuged for 5 min at 13 500 g to achieve platelet-free plasma. Plasma samples were analysed for 25(OH)D by liquid–liquid extraction and liquid chromatography-tandem mass spectrometry detection [the precision of the assay is 4% mean intra-assay coefficient of variation (CV) and 7.5% mean inter-assay CV] and 1, 25-dihydroxyvitamin D [1, 25(OH)2 D] by solid-phase immunoextraction and liquid chromatography-tandem mass spectrometry detection (3.5% mean intra-assay CV and 7.5% mean inter-assay CV). Details of human plasma donors are provided in Table 1. Table 1. Clinical and biochemical characteristics of the non-CKD and CKD plasma gender is indicated Parameters non-CKD CKD Age (years) 35.9 ± 14 5.8 ± 7.9 Gender (n, m; f) (3; 3) (4; 2) eGFR (mL/min/1.73 m2) >90 18.1 ± 6.7 25(OH)D (nmol/L)a 50 56 1, 25(OH)2D (pmol/L)a 96 55 Parameters non-CKD CKD Age (years) 35.9 ± 14 5.8 ± 7.9 Gender (n, m; f) (3; 3) (4; 2) eGFR (mL/min/1.73 m2) >90 18.1 ± 6.7 25(OH)D (nmol/L)a 50 56 1, 25(OH)2D (pmol/L)a 96 55 Age and eGFR are expressed as mean ± SD (n = 6). aBiochemical measurements of 25(OH)D and 1, 25(OH)2 D are measured in plasma pools. m, male; f, female. Table 1. Clinical and biochemical characteristics of the non-CKD and CKD plasma gender is indicated Parameters non-CKD CKD Age (years) 35.9 ± 14 5.8 ± 7.9 Gender (n, m; f) (3; 3) (4; 2) eGFR (mL/min/1.73 m2) >90 18.1 ± 6.7 25(OH)D (nmol/L)a 50 56 1, 25(OH)2D (pmol/L)a 96 55 Parameters non-CKD CKD Age (years) 35.9 ± 14 5.8 ± 7.9 Gender (n, m; f) (3; 3) (4; 2) eGFR (mL/min/1.73 m2) >90 18.1 ± 6.7 25(OH)D (nmol/L)a 50 56 1, 25(OH)2D (pmol/L)a 96 55 Age and eGFR are expressed as mean ± SD (n = 6). aBiochemical measurements of 25(OH)D and 1, 25(OH)2 D are measured in plasma pools. m, male; f, female. EC culture Human umbilical vein ECs (HUVECs) were isolated as described previously [28] from umbilical cords from healthy donors, obtained from the Ziekenhuis Amstelland, Amstelveen. Cells were taken up in M199 medium, supplemented with penicillin 100 U/mL and streptomycin 100 µg/mL (Biowhittaker/Lonza, Verviers, Belgium), human serum 10% (Sanquin Blood Supply, Amsterdam, The Netherlands), newborn calf serum 10% (Gibco, Grand Island, NY, USA), crude EC growth factor 150 µg/mL (prepared from bovine brains), L-glutamine 2 mmol/L (Biowhittaker/Lonza) and heparin 5 U/mL (Leo Pharmaceutical Products, Weesp, The Netherlands). HUVECs were cultured at 37°C and 5% CO2, with a change of culture medium every 2 days. Cells were cultured up to passage 2. For experiments, HUVECs were washed with bare medium (bM199) consisting of M199 supplemented with penicillin 100 IU/mL and streptomycin 100 µg/mL, and exposed to bM199 with 20% of pooled non-CKD/CKD plasma during 24 h, combined with or without PC (100 nM; kindly provided by AbbVie, Chicago, IL, USA). Based on measurements of concentrations in CKD patients with eGFR  <30 mL/min/1.73 m2 [29, 30], uraemic toxins indoxyl sulphate and p-cresol (Sigma Aldrich, Zwijndrecht, The Netherlands) were tested at 100 µM in bM199 with 20% (v/v) of pooled non-CKD plasma. ECIS Transendothelial electrical resistance of HUVEC monolayers was measured using ECIS (Applied BioPhysics, Troy, NY, USA) as previously described [31]. Using eight-well ECIS arrays (10 electrodes per well, 0.49 mm2 electrode area) HUVECs were seeded in a 1:1 density on a gelatin-coated array. ECIS software (v1.2.65.0 PC; Applied BioPhysics) was used to calculate the level of overall resistance (4000 Hz), cell–cell contact (Rb) and cell–matrix contact (α) as depicted in Supplementary data, Figure S1A. Cells were grown to confluency (48 h) and medium was renewed 24 h after seeding. After 24 h of pretreatment, electrical wounding was applied (60 s; 100 000 Hz). For quantification of the wounding response, the slope (assessed by linear regression) and the area under the curve (AUC) of the recovery phase from each replicate (measured after electrical wound induction) was calculated. The effect of thrombin (1 U/mL; Sigma Aldrich) was quantified after 60 min pretreatment with bM199 containing 1% human serum albumin (HSA; Sanquin Blood Supply). Resistance was normalized to the initial value before the addition of thrombin. The percentage of drop after the addition of thrombin and the AUC from each replicate were quantified to assess the extent of the thrombin response. Immunofluorescence imaging HUVECs were seeded in 1:1 density on glass coverslips coated with glutaraldehyde 0.5% (Fluka, St Gallen, Switzerland) cross-linked gelatin and grown to confluency in 24–48 h. After 24 h of treatment, medium was replaced by 2% paraformaldehyde (37°C, Sigma Aldrich), followed by 15 min incubation on ice. Fixed cells were permeabilized with Triton X-100 0.05% (Sigma Aldrich) in phosphate-buffered saline (PBS), washed and incubated overnight with primary antibody against VE-cadherin (1:400; XP D87F2, Cell Signalling, Danvers, MA, USA). Afterwards, HUVECs were washed and incubated with FITC secondary antibodies (1:100; Invitrogen, Paisley, UK) and the F-actin cytoskeleton was stained with rhodamine/phalloidin (1:140; PHDH1 Cytoskeleton Inc., Denver, CO, USA) for 2 h at room temperature. Subsequently, cells were washed and nuclei were stained with DAPI (1:500; Sigma Aldrich) for 15 min. Cells were washed, and sealed with Mowiol mounting medium (Sigma Aldrich). Imaging was performed with a Leica TCS SP8 STED using a 63x Zeiss oil objective. ImageJ (National Institutes of Health, Bethesda, MD, USA) was used for analysis. For quantification of VE-cadherin/F-actin, images were equally adjusted for contrast, mean fluorescence was quantified and divided by the amount of counted nuclei (DAPI). Fluorescence-activated cell sorting VE-cadherin membrane expression was quantified in HUVECs using a specific antibody (Alexa Fluor® 647 anti-mouse CD144) for 30 min. HUVECs were rinsed with PBS twice and then detached using Accutase (GE Healthcare, Eindhoven, The Netherlands). Detached cells were resuspended in ice-cold PBS supplemented with 0.5% (w/v) bovine serum albumin (BSA), centrifuged and resuspended in PBS + 0.5% BSA. Fluorescence-activated cell sorting (FACS) analyses were performed with Calibur flow cytometer (Becton and Dickinson, San Jose, CA, USA) and FlowJo (Ashland, OR, USA) software. Statistical analysis Data were analysed using GraphPad Prism software (La Jolla, CA, USA). Statistical analysis was performed using unpaired t-test for two groups and three or more groups using one-way analysis of variance with Bonferroni correction. Data were shown as means ± SD. Differences were considered statistically significant for P < 0.05 (*P < 0.05, **P < 0.01, ***P < 0.001). RESULTS Uraemic plasma induces endothelial barrier disruption by reduced cell–cell interaction To test whether a uraemic environment affects endothelial barrier integrity, confluent HUVEC monolayers were exposed to plasma from patients with CKD (eGFR <18 mL/min/1.73 m2) or healthy subjects (eGFR >90 mL/min/1.73 m2) (v/v; 20%). Baseline characteristics of the plasma donors are reported in Table 1. Initially, endothelial resistance was reduced (1536 ± 21.8 versus 1477 ± 27.4 Ω; P = 0.0426) in uraemic plasma-treated HUVECs after 3 h when compared with non-CKD plasma (Figure 1A). Although the endothelial resistance normalized to the control values, a second decay of the resistance was observed with statistically significant differences at 24 h (1530 ± 10.1 versus 1432 ± 24.5 Ω; P = 0.0031). Subsequently, we tested recovery after thrombin-induced (1 U/mL) contraction of HUVECs inducing a loss of barrier function, 24 h after plasma exposure (Figure 1B). Increased loss of barrier function was observed for CKD-plasma-exposed EC after thrombin stimulation. This was reflected by the percentage of drop in resistance and accompanied by a significantly reduced AUC measured after data normalization: CKD [49.5%; 3.24 ± 0.01 arbitrary units (a.u.)] versus non-CKD controls (41.5%; 3.55 ± 0.01 a.u.) (Figure 1B). Similarly, HUVECs challenged with uraemic media displayed a reduced recovery capacity in the wound healing assay (Figure 1C). Quantitative analysis showed a significant reduction for both slope (1–5 h) and AUC of the recovery phase (1–8 h) of CKD-plasma-exposed cells (170 ± 1 a.u.; 7636 ± 43.1 a.u.) compared with non-CKD controls (217 ± 13.6 a.u.; 8544 ± 159.3 a.u.). Furthermore, measurements 48 h after the induction of a wound confirmed that differences in AUC between CKD and non-CKD plasma sustained without normalization of CKD condition (11.34% less of AUC in CKD after 24 h and 14% after 48 h when compared with non-CKD) (Supplementary data, Figure S2). FIGURE 1: View largeDownload slide Uraemic plasma induces endothelial barrier disruption in HUVECs. (A) Representative time-course of absolute endothelial electrical resistance measurements (Ω) of confluent HUVECs stimulated for 24 h with 20% (v/v) of non-CKD plasma (dark) or CKD plasma (grey). Insert: quantification of the absolute endothelial resistance (Ω) after 3 h and 24 h of stimulation as indicated by arrowheads. (B) Normalized resistance of thrombin (1 U/mL)-stimulated HUVECs. Upon 24 h incubation with indicated media, HUVECs were exposed to a basal medium plus 1% HSA during 1 h and challenged with thrombin (arrow). Bar graphs: (left) quantification of the thrombin response represented by the AUC (starting at time point 0) and (right) to maximum drop of resistance (%) following stimulation. (C) Resistance measurements (Ω) after electrical wounding (arrowhead) of 24 h 20% plasma pre-stimulated cells. Bar graphs: (left) quantification of the wound healing as the AUC (from 1 h to 8 h) and (right) the slope of the recovery (from 1 h to 5 h). Data show mean ± SD (n = 3, from pool of three donors). Representative data of at least three experiments. Differences were considered statistically significant for P < 0.05 using unpaired student’s t-test where *P < 0.05, **P < 0.01, ***P < 0.001. FIGURE 1: View largeDownload slide Uraemic plasma induces endothelial barrier disruption in HUVECs. (A) Representative time-course of absolute endothelial electrical resistance measurements (Ω) of confluent HUVECs stimulated for 24 h with 20% (v/v) of non-CKD plasma (dark) or CKD plasma (grey). Insert: quantification of the absolute endothelial resistance (Ω) after 3 h and 24 h of stimulation as indicated by arrowheads. (B) Normalized resistance of thrombin (1 U/mL)-stimulated HUVECs. Upon 24 h incubation with indicated media, HUVECs were exposed to a basal medium plus 1% HSA during 1 h and challenged with thrombin (arrow). Bar graphs: (left) quantification of the thrombin response represented by the AUC (starting at time point 0) and (right) to maximum drop of resistance (%) following stimulation. (C) Resistance measurements (Ω) after electrical wounding (arrowhead) of 24 h 20% plasma pre-stimulated cells. Bar graphs: (left) quantification of the wound healing as the AUC (from 1 h to 8 h) and (right) the slope of the recovery (from 1 h to 5 h). Data show mean ± SD (n = 3, from pool of three donors). Representative data of at least three experiments. Differences were considered statistically significant for P < 0.05 using unpaired student’s t-test where *P < 0.05, **P < 0.01, ***P < 0.001. To elucidate whether changes in basal EC barrier function were related to effects on the cell–cell contact or cell–matrix contact, ECIS modelling of the parameters Rb (resistance between cells) and α (resistance between cells and extracellular matrix) was performed [31] (Supplementary data, Figure S1A). As shown in Figure 2A, no significant changes were observed for cell–matrix interactions after 24 h exposure of EC to different media. In contrast, impaired cell–cell contact (9% reduction; P = 0.039) was found in HUVECs challenged with uraemic media (Figure 2B). No differences in cell–matrix interactions were detected during the recovery phase in both assays (Figure 2C and E). The changes observed in cell–cell interactions after thrombin treatment or electrical wounding mimic the EC basal alterations (Figure 2D and F). Together, these findings indicate that uraemic media decrease endothelial barrier function and limit the recovery capacity after an electrical wound healing and thrombin exposure. Abnormal cell–cell but not cell–matrix contact mediates these uraemic plasma-induced changes in resistance. FIGURE 2: View largeDownload slide Impaired endothelial barrier function in CKD is caused by reduced cell–cell interaction. Absolute endothelial electrical resistance attributable to cell–matrix interaction (α; Ω0.5cm) (A) and cell–cell interaction (Rb; Ωcm2) (B) of confluent HUVECs stimulated for 24 h with 20% (v/v) non-CKD plasma (dark) or CKD plasma (grey). Normalized values of α (C) and Rb (D) upon thrombin (1 U/mL; arrow) stimulation. Time course of endothelial α (E) and Rb (F) values after electrical wounding (arrow). Data show mean ± SD (n = 3, from a pool of three donors). Representative data of at least three experiments. *P < 0.05 using unpaired Student’s t-test. FIGURE 2: View largeDownload slide Impaired endothelial barrier function in CKD is caused by reduced cell–cell interaction. Absolute endothelial electrical resistance attributable to cell–matrix interaction (α; Ω0.5cm) (A) and cell–cell interaction (Rb; Ωcm2) (B) of confluent HUVECs stimulated for 24 h with 20% (v/v) non-CKD plasma (dark) or CKD plasma (grey). Normalized values of α (C) and Rb (D) upon thrombin (1 U/mL; arrow) stimulation. Time course of endothelial α (E) and Rb (F) values after electrical wounding (arrow). Data show mean ± SD (n = 3, from a pool of three donors). Representative data of at least three experiments. *P < 0.05 using unpaired Student’s t-test. Uraemic toxins indoxyl-sulphate and p-cresol aggravate non-CKD plasma wound recovery To investigate whether the uraemic retention solutes indoxyl sulphate and p-cresol may be involved in the uraemic media-mediated loss of the endothelial barrier, we tested how the addition of these toxins interfere with the barrier function of a non-CKD condition. The direct addition of these toxins in isolation (100 µM) resulted in no changes to resistance during measurements of basal endothelial barrier function or wound healing (data not shown). When a combination of indoxyl sulphate and p-cresol (100 µM) was tested, no differences in basal EC resistance were detected after 24 h (Supplementary data, Figure S3A). Alternatively, uraemic toxins significantly delayed the recovery of an electric wound as reflected by measurements of the slope of resistance recovery (2–6 h after wounding; 0.30 ± 0.01 a.u. for non-CKD and 0.25 ± 0.02 a.u. for indoxyl sulfate + p-cresol) (Supplementary data, Figure S3B and C). Quantitative analysis of the AUC confirmed that the combined uraemic toxin-exposure results in a declined recovery of barrier integrity at 12 h after wounding (12.58 ± 0.64 a.u. versus 10.82 ± 0.79 a.u.) and the differences where further increased after 40 h (87.67 ± 3.98 a.u. versus 71.44 ± 5.85 a.u.) (Supplementary data, Figure S3D). Similar to CKD media, the joint effects of indoxyl sulphate and p-cresol on wound recovery were mediated by reduced cell–cell interactions (Supplementary data, Figure S4B) while cell–matrix interactions remained unaffected (Supplementary data, Figure S4B). However, no significant differences were found when the AUC was measured (Supplementary data, Figure S4C) although the effect of the uraemic toxins in cell–cell interactions was more pronounced after 40 h, mirroring the AUC measurements assessed after wounding (Supplementary data, Figure S3C). Thus, the addition of uraemic toxins at 100 µM to healthy plasma showed impaired recovery after a wound with disturbed cell–cell contact. VE-cadherin localization is impaired in uraemic conditions Next, we explored the structural changes within the EC using VE-cadherin and F-actin (immuno)-staining (Figure 3A). VE-cadherin, which is essential to modulate and maintain the inter-EC contacts [32], showed a markedly disrupted peripheral localization and discontinuous organization after uraemic media exposure (Figure 3A, zoom). Consistently, after uraemic plasma stimulation, there was a significant reduction of the VE-cadherin fluorescence intensity when compared not only with non-CKD plasma (45% less) but also with a control media (42% less). Similarly, quantitative analysis using flow cytometry confirmed a decreased membrane expression of VE-cadherin in HUVECs exposed to CKD plasma (41% versus non-CKD) (Figure 3D and E). Furthermore, non-uraemic plasma-exposed EC showed a peripheral distribution of the F-actin organized as a cortical actin rim which promotes the assembly of EC–cell and cell–matrix adhesions. In contrast, exposure to uraemic plasma resulted in prominent F-actin stress fibres, characteristic of cell contraction albeit without significant changes when compared with non-CKD (0.82 ± 0.03 versus 0.64 ± 0.09 a.u.; P = 0.061) (Figure 3C). No changes in VE-cadherin expression where found in cells exposed to control media when compared with non-CKD media, although HUVECs incubated with control media displayed a significant increment of F-actin stress fibre formation. These findings suggest that disrupted adherens junction stability underlies the effect on EC dysfunction observed in uraemic plasma-exposed cells. These alterations in adherens junctions were accompanied by a reorganization of F-actin in uraemic plasma-exposed HUVECs. FIGURE 3: View largeDownload slide Membrane VE-cadherin localization is disturbed in CKD plasma exposed HUVECs. (A) Immunostaining of VE-cadherin (green), F-actin (red) and nuclei (blue) of HUVECs after 24 h of incubation in medium (bM199) plus 1% HSA (Control) or 20% (v/v) non-CKD or CKD plasma. Scale bar represents 50 µm. Zoomed images correspond to the white boxes. Normalized intensity quantifications are in the corresponding graphs for VE-cadherin (B) and F-actin (C). Surface expression of VE-cadherin was assessed by flow cytometry (D) in 24 h-stimulated cells. (E) Normalized mean fluorescence intensity measured by flow cytometry of membrane VE-cadherin is shown for the different conditions. Data are mean ± SD of three independent experiments. Differences were considered statistically significant for P < 0.05 using one-way analysis of variance (ANOVA) where *P < 0.05, **P < 0.01. FIGURE 3: View largeDownload slide Membrane VE-cadherin localization is disturbed in CKD plasma exposed HUVECs. (A) Immunostaining of VE-cadherin (green), F-actin (red) and nuclei (blue) of HUVECs after 24 h of incubation in medium (bM199) plus 1% HSA (Control) or 20% (v/v) non-CKD or CKD plasma. Scale bar represents 50 µm. Zoomed images correspond to the white boxes. Normalized intensity quantifications are in the corresponding graphs for VE-cadherin (B) and F-actin (C). Surface expression of VE-cadherin was assessed by flow cytometry (D) in 24 h-stimulated cells. (E) Normalized mean fluorescence intensity measured by flow cytometry of membrane VE-cadherin is shown for the different conditions. Data are mean ± SD of three independent experiments. Differences were considered statistically significant for P < 0.05 using one-way analysis of variance (ANOVA) where *P < 0.05, **P < 0.01. Active VD attenuates uraemia-mediated endothelial dysfunction Measurements of VD levels confirmed that, in CKD plasma, 1, 25(OH)2 D concentrations were 42% lower than those in non-CKD plasma (96 pmol/L versus 55 pmol/L), whereas 25(OH)D concentrations were similar between groups (50 nmol/L versus 56 nmol/L) (Table 1). We therefore further tested the potentially beneficial effects of the active VD form PC in the uraemic media-mediated endothelial dysfunction. Quantifying basal endothelial resistance, we found that addition of 100 nM PC rescued the significant reduction in resistance induced by uraemic plasma after 24 h when compared with non-CKD conditions (Figure 4A, insert). However, no significant differences between CKD groups were detected. No significant improvement of the barrier function was detected for healthy plasma combined with PC. Furthermore, the addition of PC to uraemic medium improved the impaired recovery after electrical wounding (Figure 4B). This was reflected by a pronounced increase of the slope (from 1 to 6 h) in CKD + PC (140 ± 9.5 a.u.) rescuing the significant differences induced by uraemic plasma when compared with non-CKD conditions (121 ± 13 a.u.), although there was no significant difference between CKD groups (Figure 4C). In addition, the AUC (from 1 to 24 h) was significantly higher (7.5%) for PC-treated cells in CKD plasma when compared with CKD alone. Moreover, the addition of PC to non-uraemic media increased the differences of the AUC and slope against CKD media when compared with non-CKD alone as shown in Figure 4C. These data show amelioration by PC of the uraemia-mediated changes on basal endothelial resistance. In addition, PC ameliorated the recovery capacity of endothelial barrier following electrical wounding in uraemic plasma. FIGURE 4: View largeDownload slide PC stabilizes uraemic plasma-induced endothelial dysfunction. (A) Representative measurements of absolute endothelial electrical resistance (Ω) of HUVECs, stimulated for 24 h as indicated. Insert: absolute endothelial resistance (Ω) after 24 h (non-CKD = dark; non-CKD + PC = striped dark; CKD = grey; CKD + PC = striped grey). (B) Electrical wound healing (arrow) of cells pre-stimulated during 24 h as indicated. (C) The quantifications of the wound healing capacity are shown as the AUC (from 1 h to 24 h) and slope of the recovery phase (a.u.; from 1 to 6 h). Representative data of at least three experiments. Data show mean ± SD (n = 3, pool of three donors). Differences were considered statistically significant for P < 0.05 using one-way analysis of variance (ANOVA) where *P < 0.05, **P < 0.01, ***P < 0.001. FIGURE 4: View largeDownload slide PC stabilizes uraemic plasma-induced endothelial dysfunction. (A) Representative measurements of absolute endothelial electrical resistance (Ω) of HUVECs, stimulated for 24 h as indicated. Insert: absolute endothelial resistance (Ω) after 24 h (non-CKD = dark; non-CKD + PC = striped dark; CKD = grey; CKD + PC = striped grey). (B) Electrical wound healing (arrow) of cells pre-stimulated during 24 h as indicated. (C) The quantifications of the wound healing capacity are shown as the AUC (from 1 h to 24 h) and slope of the recovery phase (a.u.; from 1 to 6 h). Representative data of at least three experiments. Data show mean ± SD (n = 3, pool of three donors). Differences were considered statistically significant for P < 0.05 using one-way analysis of variance (ANOVA) where *P < 0.05, **P < 0.01, ***P < 0.001. Disrupted VE-cadherin-mediated cell–cell junctions in uraemic plasma are restored by active VD After determining the potency of PC to improve the barrier function during uraemia, we explored whether this effect is the consequence of the restoration of cell–cell contact. In line with the previous experiment, PC significantly improved (9.6%) the strength of the cell–cell interaction after 24 h as reflected by the quantification of the resistance between cells (Rb) (Figure 5A and B). PC supplementation rescued the significant differences encountered in uraemic plasma-treated EC after quantification of the AUC (24 h; 12.6%) of the cell–cell quantification in the wound healing process when compared with non-CKD conditions (Figure 5C). However, no significant differences were found between uraemic conditions with or without PC treatment. No significant changes were found regarding the slope of the wound healing phase although uraemic medium alone showed a negative trend (−15 to −10%) when compared with other conditions. Further, PC induced enhanced VE-cadherin peripheral intensity under uraemic conditions as shown in Figure 6A, which is supported by significant differences of intensity measurements (38% higher; P = 0.015) when compared with uraemia alone (Figure 6B). Evaluating adherens junctions morphology, the jagged distribution of VE-cadherin was improved to a more linear distribution when HUVECs in uraemic conditions where supplemented with PC (Figure 6A, zoom). PC slightly, but not significantly, increased VE-cadherin intensity (14%) in non-uraemic medium. Finally, although cytoplasmic F-actin filaments were dominant in uraemic conditions as observed in Figure 6A, the fluorescence intensity of F-actin did not differ significantly when compared with other conditions (Figure 6C). In addition, immunostaining for F-actin showed that PC reduced the arrangement of stress fibres upon uraemic plasma while in non-uraemic conditions PC slightly enhanced F-actin cortical distribution, although no significant changes were observed after quantification. Thus, we confirmed that PC strengthens the endothelial barrier integrity under uraemic conditions by mediating the stabilization of junctional VE-cadherin. FIGURE 5: View largeDownload slide Cell–cell interactions are stabilized by PC in uraemic media. (A) Representative cell–cell interaction (Rb; Ωcm2) measurements of HUVECs incubated for 24 h as indicated (non-CKD = dark; non-CKD + PC = striped dark; CKD = grey; CKD + PC= striped grey). (B) Rb of an electric wounding response (arrow) of cells previously stimulated for 24 h (C) Quantification of the wound healing capacities are represented as the AUC (from 3 to 24 h) and the slope (a.u.; from 3 to 7 h). Data show mean ± SD (n = 3). Representative data of at least three experiments. Differences were considered statistically significant for P < 0.05 using one-way analysis of variance (ANOVA) where *P < 0.05, **P < 0.01. FIGURE 5: View largeDownload slide Cell–cell interactions are stabilized by PC in uraemic media. (A) Representative cell–cell interaction (Rb; Ωcm2) measurements of HUVECs incubated for 24 h as indicated (non-CKD = dark; non-CKD + PC = striped dark; CKD = grey; CKD + PC= striped grey). (B) Rb of an electric wounding response (arrow) of cells previously stimulated for 24 h (C) Quantification of the wound healing capacities are represented as the AUC (from 3 to 24 h) and the slope (a.u.; from 3 to 7 h). Data show mean ± SD (n = 3). Representative data of at least three experiments. Differences were considered statistically significant for P < 0.05 using one-way analysis of variance (ANOVA) where *P < 0.05, **P < 0.01. FIGURE 6: View largeDownload slide Reinforced VE-cadherin contact mediates the effect of PC in uraemic media. (A) Immunofluorescence staining of VE-cadherin (green), F-actin (red) and the nuclei (blue) of HUVEC following 24 h of incubation (scale bar, 50 µm). Zoomed images correspond to the white boxes. Normalized mean VE-cadherin (B) and F-actin (C) intensity of the immunostainings are shown (non-CKD = dark; non-CKD + PC = striped dark; CKD = grey; CKD + PC= striped grey). Data show mean ± SD of three independent experiments. Differences were considered statistically significant for P < 0.05 using one-way analysis of variance (ANOVA) where *P < 0.05, **P < 0.01, ***P < 0.001. FIGURE 6: View largeDownload slide Reinforced VE-cadherin contact mediates the effect of PC in uraemic media. (A) Immunofluorescence staining of VE-cadherin (green), F-actin (red) and the nuclei (blue) of HUVEC following 24 h of incubation (scale bar, 50 µm). Zoomed images correspond to the white boxes. Normalized mean VE-cadherin (B) and F-actin (C) intensity of the immunostainings are shown (non-CKD = dark; non-CKD + PC = striped dark; CKD = grey; CKD + PC= striped grey). Data show mean ± SD of three independent experiments. Differences were considered statistically significant for P < 0.05 using one-way analysis of variance (ANOVA) where *P < 0.05, **P < 0.01, ***P < 0.001. DISCUSSION This study demonstrates that uraemic media negatively affect the integrity of the EC barrier function and also impair the recovery following exposure to barrier-disruptive mediators. Specifically, cell–cell interactions reduced in uraemic plasma-exposed EC were driving the overall barrier dysfunction, which was corroborated by reduced peripheral VE-cadherin and cortical F-actin. The active VD compound PC attenuated the endothelial barrier-disruptive effects of uraemia in basal and electrical wounding assays by improving the cell–cell contact and restoring endothelial integrity. Prolonged exposure of uraemic toxins present in the plasma of CKD patients can affect the integrity and repair capacity of EC [8]. In our study, ECIS has been shown to be suitable to characterize in real time, the effects of uraemic media on confluent endothelial barrier function [31]. Adverse changes on EC by uraemic plasma such as limited resistance to flow, activation of nuclear factor kappa B including increased permeability, have been previously demonstrated in vitro [9–11] and in vivo [33]. Our data show that EC exposed to uraemic plasma (derived from patients with eGFR  <18 mL/min/1.73 m2) indeed displayed a reduced resistance when compared with healthy plasma (eGFR >90 mL/min/1.73 m2). Furthermore, when the barrier integrity was compromised by thrombin (inducing cell contraction and intercellular gap formation [34]) or electrical wounding [27], mimicking an injury, HUVECs exhibit a reduced recovery capacity after exposure for 24 h to CKD plasma. Based on this data, we suggest that EC that have been exposed to uraemia are more sensitive to barrier-disruptive conditions. Interestingly, this is supported by data showing altered proliferation and wound repair on HUVECs exposed to the uraemic toxins p-cresol and indoxyl sulphate [35]. These uraemic retention toxins, poorly removed by haemodialysis therapies, are kidney disease-related cardiovascular risk factors and are suggested to contribute to endothelial toxicity in CKD [29, 36, 37]. In this study, the addition of indoxyl sulphate and p-cresol in a concentration detected in serum from CKD patients with eGFR  <30 mL/min/1.73 m2 (100 μΜ) [29, 30], mimicked the deleterious effects of CKD plasma in the recovery of the EC barrier function after a wound. Alternatively, the differences found in basal EC resistance after CKD plasma stimulation were not detected in media with uraemic toxins indicating that the effects from CKD in EC dysfunction are not exclusive from indoxyl sulphate and p-cresol toxicity. There are several cellular structures that are essential to the maintenance of the EC barrier integrity [38]. By mathematically modelling our data obtained with ECIS [31], we concluded that the changes in transendothelial electrical resistance which we previously observed can be attributed to impaired cell–cell contact in uraemic media. This was confirmed after measuring of the resistance between cells (Rb), which was reduced, while the current between the cell–matrix interactions (α) remained unchanged. In good agreement with these data, our imaging analysis showed that the adherens junction protein VE-cadherin, essential for endothelial barrier integrity [32], showed disrupted localization and was less concentrated at cell boundaries upon uraemic conditions. The decrease of VE-cadherin in cells was not attributable to redistribution to the cell surface as confirmed by FACS. Furthermore, VE-cadherin is linked to the actin cytoskeleton and its association is essential to barrier function [39]. On exposure to uraemic media, EC displayed a reduced cortical ring-like F-actin distribution. A prominent cortical actin cytoskeleton structure is characteristic for quiescent endothelium, and therefore, stable barrier integrity. In addition, immunostaining suggested increased presence of stress fibres indicating a reorganization of the F-actin protein leading to a different cell shape and cell contact destabilization albeit not statistically significant. Again, this is in line with in vitro studies performed with uraemic toxins [40, 41]. HUVECs exposed to p-cresol displayed limited VE-cadherin and F-actin colocalization while indoxyl sulphate addition resulted in adherens junction disassembly and cytoskeleton reorganization [40, 41]. This mechanism is suggested to be mediated by an enhanced dissociation of VE-cadherin and actin cytoskeleton induced by the Rho/Rho kinase pathway [40, 41]. Upon activation of Rho, there is an increase of myosin light chain phosphorylation leading to the formation of stress fibres, myosin-based contraction and opening of cell junctions, which increases permeability [42]. Indeed, by measuring Rb during wound healing we confirmed that indoxyl sulphate and p-cresol combination impact cell–cell but not cell–matrix interactions, similarly as CKD plasma. Further evidence for an uraemia-induced disturbed actin cytoskeleton comes from a proteomic analysis in ECs, showing an increment of the actin depolymerizing protein dextrin upon uraemic serum stimulation [43]. Interestingly, this was accompanied by a downregulation of annexin A2, which plays a key role in the establishment of adherens junctions [43]. Our data extend these observations by comparing the effects of uraemic plasma with those of a control (bare medium). Deficiencies in kidney-specific factors such as 25(OH)D, 1, 25(OH)2 D or α-Klotho are suggested to also contribute to cardiovascular pathology [12–14]. In our study, VD measurements confirmed that 1, 25(OH)2 D in CKD plasma was lower than in plasma from healthy donors. However, when compared with CKD condition, control media showed no changes in VE-cadherin intensity suggesting that the effects observed in CKD were not related to some unidentified deficiency in the plasma, but a direct effect of uraemic toxins. Yet, our experiment does not rule out that 1, 25(OH)2 D deficiency aggravates uraemia-mediated endothelial damage. In CKD animal models [20, 21] and patients [22, 44], active VD supplementation has been shown to mitigate endothelial dysfunction. In addition, dietary VD and 1, 25(OH)2 D or PC can modulate endothelial stability by modifying inflammation, thrombosis and vasodilation [23–25]. We found that adding PC partially prevented the changes in electrical resistance induced by uraemia and improved the recovery following electrical wounding. Interestingly, Won et al. showed that 1, 25(OH)2 D attenuated the decrease of electrical resistance upon hypoxia in brain EC and restored tight junction expression, which has a prominent contribution to blood–brain barrier function [25]. This protective effect was mediated by a decrease of the Matrixmetalloproteinase-9 (MMP-9), which mediates disruption of cell–cell interaction [25]. In line with this, we found a restoration of junctional VE-cadherin in cells incubated with uraemic media by PC. Interestingly, increased MMP-9 has also been examined in vitro in EC after the application of uraemic media [45]. Although more detailed studies are necessary to elucidate the vasculo-protective mechanism of active VD upon uraemia, altered MMP-9 expression appears to be one possible mechanism of endothelial protection. This possible mode of action is illustrated in the Supplementary data, Figure S1B. In addition, it remains to be studied whether uraemia-disrupted cell–cell interaction also includes the impairment of the endothelial tight junction. In our study, however, we focused on the effects on VE-cadherin since non-brain EC exhibit less developed tight junctions. Importantly, besides being the major determinant of endothelial cell–cell interaction [32], VE-cadherin plays a key role controlling the level of expression and localization of other junctional molecules [46]. As an alternative mechanism of action, it has been proposed that targeting the uraemia-induced oxidative stress in the endothelium could be a potential strategy against endothelial dysfunction in CKD [47]. In this regard, 1, 25(OH)2 D has been shown to be protective against oxidative stress in EC [48], and therefore, this mechanism could also be related to the endothelial-protective proprieties of active VD against uraemia. Finally, it is also important to consider that EC has the ability to transform 25(OH)D to the active metabolite 1, 25(OH)2 D [49], and comparable endothelial protective effects could be achieved by 25(OH)D supplementation. This alternative therapeutic approach could be of importance in a setting with a 25(OH)D deficiency, a feature that, however, was not applicable in our study. Our study bears some limitations as the exact signalling mechanism involved in the VE-cadherin-disrupted contact and F-actin reorganization during uraemia remains to be established. Moreover, it would be interesting to determine whether the uraemia-mediated endothelial damage could be prevented by the previous addition of PC. Nevertheless, this setting would not be representative of a clinical situation. Likewise, additional EC-types need to be tested to confirm the effects observed during uraemia. Despite those disadvantages, we provide valuable in vitro data by the combination of human uraemic plasma with real-time measurements of endothelial barrier functions. In conclusion, we have extended our insight of the effects of uraemia on EC dysfunction. Although several kidney-disease-related risk factors were suggested to contribute to endothelial dysfunction in CKD, our in vitro findings show similarities with the effects of the uraemic toxins indoxyl sulphate and p-cresol. We propose that limited cell–cell interaction caused by reduced VE-cadherin and F-actin reorganization affects the integrity of the EC barrier during uraemia. These changes may carry important clinical implications since they restrict the capacity of the EC to resist and recover from barrier-disruptive conditions, aggravating vascular complications during CKD. In addition, as a therapeutic approach, we describe a novel mechanism how the active VD analogue PC modulates the uraemia-damaged endothelium by restoring cell–cell interactions. Overall, stabilizing the intercellular endothelial contact might be a crucial step to prevent the damage in the endothelium and ameliorate future vascular complications in CKD. SUPPLEMENTARY DATA Supplementary data are available at ndt online. ACKNOWLEDGEMENTS We thank Jeroen Kole and Dr. Renee Musters (VU University Medical Center Amsterdam, Advanced Optical Microscopy Core facility, The Netherlands) for their help with the microscope equipment. CONFLICT OF INTEREST STATEMENT M.V.C. reports grants from AbbVie, during the conduct of the study; grants from Amgen, personal fees from Amgen, personal fees from VFMCRP, grants from FMC, personal fees from Otsuka, personal fees from Medice, personal fees from BBraun and personal fees from Baxter, outside the submitted work. The other authors reported no conflicts of interest. REFERENCES 1 Schiffrin EL , Lipman ML , Mann JF. Chronic kidney disease: effects on the cardiovascular system . Circulation 2007 ; 116 : 85 – 97 Google Scholar CrossRef Search ADS PubMed 2 Vervloet M , Cozzolino M. Vascular calcification in chronic kidney disease: different bricks in the wall? Kidney Int 2017 ; 91 : 808 – 817 Google Scholar CrossRef Search ADS PubMed 3 Cines DB , Pollak ES , Buck CA et al. Endothelial cells in physiology and in the pathophysiology of vascular disorders . Blood 1998 ; 91 : 3527 – 3561 Google Scholar PubMed 4 Deanfield JE , Halcox JP , Rabelink TJ. Endothelial function and dysfunction: testing and clinical relevance . Circulation 2007 ; 115 : 1285 – 1295 Google Scholar PubMed 5 Goligorsky MS. Pathogenesis of endothelial cell dysfunction in chronic kidney disease: a retrospective and what the future may hold . Kidney Res Clin Pract 2015 ; 34 : 76 – 82 Google Scholar CrossRef Search ADS PubMed 6 Jacobson SH , Egberg N , Hylander B et al. Correlation between soluble markers of endothelial dysfunction in patients with renal failure . Am J Nephrol 2002 ; 22 : 42 – 47 Google Scholar CrossRef Search ADS PubMed 7 van Guldener C , Janssen MJ , Lambert J et al. Endothelium-dependent vasodilatation is impaired in peritoneal dialysis patients . Nephrol Dial Transplant 1998 ; 13 : 1782 – 1786 Google Scholar CrossRef Search ADS PubMed 8 Gris JC , Branger B , Vécina F et al. Increased cardiovascular risk factors and features of endothelial activation and dysfunction in dialyzed uremic patients . Kidney Int 1994 ; 46 : 807 – 813 Google Scholar CrossRef Search ADS PubMed 9 Caballo C , Palomo M , Cases A et al. NFκB in the development of endothelial activation and damage in uremia: an in vitro approach . PLoS One 2012 ; 7 : e43374 Google Scholar CrossRef Search ADS PubMed 10 Serradell M , Díaz-Ricart M , Cases A et al. Uremic medium causes expression, redistribution and shedding of adhesion molecules in cultured endothelial cells . Haematologica 2002 ; 87 : 1053 – 1061 Google Scholar PubMed 11 Serradell M , Díaz-Ricart M , Cases A et al. Uremic medium disturbs the hemostatic balance of cultured human endothelial cells . Thromb Haemost 2001 ; 86 : 1099 – 1105 Google Scholar CrossRef Search ADS PubMed 12 Dusso AS. Kidney disease and vitamin D levels: 25-hydroxyvitamin D, 1, 25-dihydroxyvitamin D, and VDR activation . Kidney Int (Suppl) 2011 ; 1 : 136 – 141 Google Scholar CrossRef Search ADS 13 Seiler S , Rogacev KS , Roth HJ et al. Associations of FGF-23 and sKlotho with cardiovascular outcomes among patients with CKD stages 2-4 . Clin J Am Soc Nephrol 2014 ; 9 : 1049 – 1058 Google Scholar CrossRef Search ADS PubMed 14 Kuro-O M , Matsumura Y , Aizawa H et al. Mutation of the mouse klotho gene leads to a syndrome resembling ageing . Nature 1997 ; 390 : 45 – 51 Google Scholar CrossRef Search ADS PubMed 15 Jablonski KL , Chonchol M , Pierce GL et al. 25-Hydroxyvitamin D deficiency is associated with inflammation-linked vascular endothelial dysfunction in middle-aged and older adults . Hypertension 2011 ; 57 : 63 – 69 Google Scholar CrossRef Search ADS PubMed 16 Kong J , Kim GH , Wei M et al. Therapeutic effects of vitamin D analogs on cardiac hypertrophy in spontaneously hypertensive rats . Am J Pathol 2010 ; 177 : 622 – 631 Google Scholar CrossRef Search ADS PubMed 17 Lau WL , Leaf EM , Hu MC et al. Vitamin D receptor agonists increase klotho and osteopontin while decreasing aortic calcification in mice with chronic kidney disease fed a high phosphate diet . Kidney Int 2012 ; 82 : 1261 – 1270 Google Scholar CrossRef Search ADS PubMed 18 Wang AY , Fang F , Chan J et al. Effect of paricalcitol on left ventricular mass and function in CKD–the OPERA trial . J Am Soc Nephrol 2014 ; 25 : 175 – 186 Google Scholar CrossRef Search ADS PubMed 19 Thadhani R , Appelbaum E , Pritchett Y et al. Vitamin D therapy and cardiac structure and function in patients with chronic kidney disease: the PRIMO randomized controlled trial . JAMA 2012 ; 307 : 674 – 684 Google Scholar CrossRef Search ADS PubMed 20 Wu-Wong JR , Noonan W , Nakane M et al. Vitamin d receptor activation mitigates the impact of uremia on endothelial function in the 5/6 nephrectomized rats . Int J Endocrinol 2010 ; 2010 : 625852 Google Scholar PubMed 21 Wu-Wong JR , Li X , Chen YW. Different vitamin D receptor agonists exhibit differential effects on endothelial function and aortic gene expression in 5/6 nephrectomized rats . J Steroid Biochem Mol Biol 2015 ; 148 : 202 – 209 Google Scholar CrossRef Search ADS PubMed 22 Stojanović M , Radenković M. Vitamin D versus placebo in improvement of endothelial dysfunction: a meta-analysis of randomized clinical trials . Cardiovasc Ther 2015 ; 33 : 145 – 154 Google Scholar CrossRef Search ADS PubMed 23 Gibson CC , Davis CT , Zhu W et al. Dietary vitamin D and its metabolites non-genomically stabilize the endothelium . PLoS One 2015 ; 10 : e0140370 Google Scholar CrossRef Search ADS PubMed 24 Lee AS , Jung YJ , Thanh TN et al. Paricalcitol attenuates lipopolysaccharide-induced myocardial inflammation by regulating the NF-κB signaling pathway . Int J Mol Med 2016 ; 37 : 1023 – 1029 Google Scholar CrossRef Search ADS PubMed 25 Won S , Sayeed I , Peterson BL et al. Vitamin D prevents hypoxia/reoxygenation-induced blood-brain barrier disruption via vitamin D receptor-mediated NF-kB signaling pathways . PLoS One 2015 ; 10 : e0122821 Google Scholar CrossRef Search ADS PubMed 26 Talmor-Barkan Y , Bernheim J , Green J et al. Calcitriol counteracts endothelial cell pro-inflammatory processes in a chronic kidney disease-like environment . J Steroid Biochem Mol Biol 2011 ; 124 : 19 – 24 Google Scholar CrossRef Search ADS PubMed 27 Keese CR , Wegener J , Walker SR et al. Electrical wound-healing assay for cells in vitro . Proc Natl Acad Sci USA 2004 ; 101 : 1554 – 1559 Google Scholar CrossRef Search ADS PubMed 28 van Nieuw Amerongen GP , Draijer R , Vermeer MA et al. Transient and prolonged increase in endothelial permeability induced by histamine and thrombin: role of protein kinases, calcium, and RhoA . Circ Res 1998 ; 83 : 1115 – 1123 Google Scholar CrossRef Search ADS PubMed 29 Meijers BK , Claes K , Bammens B et al. p-Cresol and cardiovascular risk in mild-to-moderate kidney disease . Clin J Am Soc Nephrol 2010 ; 5 : 1182 – 1189 Google Scholar CrossRef Search ADS PubMed 30 Kamiński TW , Pawlak K , Karbowska M et al. Indoxyl sulfate - the uremic toxin linking hemostatic system disturbances with the prevalence of cardiovascular disease in patients with chronic kidney disease . BMC Nephrol 2017 ; 18 : 35 Google Scholar CrossRef Search ADS PubMed 31 Szulcek R , Bogaard HJ , van Nieuw Amerongen GP. Electric cell-substrate impedance sensing for the quantification of endothelial proliferation, barrier function, and motility . J Vis Exp 2014 ; 85 : e51300 32 Vestweber D. VE-cadherin: the major endothelial adhesion molecule controlling cellular junctions and blood vessel formation . Arterioscler Thromb Vasc Biol 2007 ; 28 : 223 – 232 Google Scholar CrossRef Search ADS PubMed 33 Harper SJ , Tomson CR , Bates DO. Human uremic plasma increases microvascular permeability to water and proteins in vivo . Kidney Int 2002 ; 61 : 1416 – 1422 Google Scholar CrossRef Search ADS PubMed 34 Sakamoto T , Sakamoto H , Sheu SJ et al. Intercellular gap formation induced by thrombin in confluent cultured bovine retinal pigment epithelial cells . Invest Ophthalmol Vis Sci 1994 ; 35 : 720 – 729 Google Scholar PubMed 35 Dou L , Bertrand E , Cerini C et al. The uremic solutes p-cresol and indoxyl sulfate inhibit endothelial proliferation and wound repair . Kidney Int 2004 ; 65 : 442 – 451 Google Scholar CrossRef Search ADS PubMed 36 Meijers BK , Bammens B , De Moor B et al. Free p-cresol is associated with cardiovascular disease in hemodialysis patients . Kidney Int 2008 ; 73 : 1174 – 1180 Google Scholar CrossRef Search ADS PubMed 37 Lin CJ , Liu HL , Pan CF et al. Indoxyl sulfate predicts cardiovascular disease and renal function deterioration in advanced chronic kidney disease . Arch Med Res 2012 ; 43 : 451 – 456 Google Scholar CrossRef Search ADS PubMed 38 Wallez Y , Huber P. Endothelial adherens and tight junctions in vascular homeostasis, inflammation and angiogenesis . Biochim Biophys Acta 2008 ; 1778 : 794 – 809 Google Scholar CrossRef Search ADS PubMed 39 Prasain N , Stevens T. The actin cytoskeleton in endothelial cell phenotypes . Microvasc Res 2009 ; 77 : 53 – 63 Google Scholar CrossRef Search ADS PubMed 40 Peng YS , Lin YT , Chen Y et al. Effects of indoxyl sulfate on adherens junctions of endothelial cells and the underlying signaling mechanism . J Cell Biochem 2012 ; 113 : 1034 – 1043 Google Scholar CrossRef Search ADS PubMed 41 Cerini C , Dou L , Anfosso F et al. P-cresol, a uremic retention solute, alters the endothelial barrier function in vitro . Thromb Haemost 2004 ; 92 : 140 – 150 Google Scholar PubMed 42 Amado-Azevedo J , Valent ET , Van Nieuw Amerongen GP. Regulation of the endothelial barrier function: a filum granum of cellular forces, Rho-GTPase signaling and microenvironment . Cell Tissue Res 2014 ; 355 : 557 – 576 Google Scholar CrossRef Search ADS PubMed 43 Carbó C , Arderiu G , Escolar G et al. Differential expression of proteins from cultured endothelial cells exposed to uremic versus normal serum . Am J Kidney Dis 2008 ; 51 : 603 – 612 Google Scholar CrossRef Search ADS PubMed 44 Lundwall K , Jörneskog G , Jacobson SH et al. Paricalcitol, microvascular and endothelial function in non-diabetic chronic kidney disease: a randomized trial . Am J Nephrol 2015 ; 42 : 265 – 273 Google Scholar CrossRef Search ADS PubMed 45 Zafeiropoulou K , Bita T , Polykratis A et al. Hemodialysis removes uremic toxins that alter the biological actions of endothelial cells . PLoS One 2012 ; 7 : e30975 Google Scholar CrossRef Search ADS PubMed 46 Gavard J , Gutkind JS. VE-cadherin and claudin-5: it takes two to tango . Nat Cell Biol 2008 ; 10 : 883 – 885 Google Scholar CrossRef Search ADS PubMed 47 Rodrigues SD , França KC , Dallin FT et al. N-acetylcysteine as a potential strategy to attenuate the oxidative stress induced by uremic serum in the vascular system . Life Sci 2015 ; 121 : 110 – 116 Google Scholar CrossRef Search ADS PubMed 48 Jia X , Xu J , Gu Y et al. Vitamin D suppresses oxidative stress-induced microparticle release by human umbilical vein endothelial cells . Biol Reprod 2017 ; 96 : 199 – 210 Google Scholar PubMed 49 Merke J , Milde P , Lewicka S et al. Identification and regulation of 1, 25-dihydroxyvitamin D3 receptor activity and biosynthesis of 1, 25-dihydroxyvitamin D3. Studies in cultured bovine aortic endothelial cells and human dermal capillaries . J Clin Invest 1989 ; 83 : 1903 – 1915 Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nephrology Dialysis Transplantation Oxford University Press

Stabilization of cell–cell junctions by active vitamin D ameliorates uraemia-induced loss of human endothelial barrier function

Loading next page...
 
/lp/ou_press/stabilization-of-cell-cell-junctions-by-active-vitamin-d-ameliorates-ihPTWu1bVK
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.
ISSN
0931-0509
eISSN
1460-2385
D.O.I.
10.1093/ndt/gfy111
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Uraemia induces endothelial cell (EC) injury and impaired repair capacity, for which the underlying mechanism remains unclear. Active vitamin D (VD) may promote endothelial repair, however, the mechanism that mediates the effects of VD in chronic kidney disease are poorly understood. Thus, we investigated uraemia-induced endothelial damage and the protection against such damage by active VD. Methods We applied electric cell-substrate impedance sensing (ECIS) to study real-time responses of human ECs exposed to pooled uraemic and non-uraemic plasma with or without the addition of active VD. The effects of indoxyl sulphate and p-cresol were tested in non-uraemic plasma. Structural changes for vascular endothelial (VE)-cadherin and F-actin were assessed by immunostaining and quantified. Results The exposure of ECs to uraemic media significantly decreased endothelial barrier function after 24 h. Cell migration after electrical wounding and recovery of the barrier after thrombin-induced loss of integrity were significantly impaired in uraemic-medium stimulated cells and cells exposed to indoxyl sulphate and p-cresol. This effect on ECIS was dependent on loss of cell–cell interaction. Mechanistically, we found that EC, exposed to uraemic media, displayed disrupted VE-cadherin interactions and F-actin reorganization. VD supplementation rescued both endothelial barrier function and cell–cell interactions in ECs exposed to uraemic media. These events were associated with an increment of VE-cadherin at intercellular junctions. Conclusions Our data demonstrate a potentially clinically relevant mechanism for uraemia-induced endothelial damage. Furthermore, active VD rescued the uraemic medium-induced loss of cell–cell adhesion, revealing a novel role of active VD in preservation of endothelial integrity during uraemia. barrier function, CKD, endothelial dysfunction, VE-cadherin, vitamin D INTRODUCTION Increased mortality risk in patients with chronic kidney disease (CKD) is frequently linked to cardiovascular disease [1]. CKD impacts vascular health, and especially vascular calcification of the medial layer of the vessel wall or secondary calcifications of atheromatous plaques has received much attention [2]. The integrity of the endothelium is a somewhat neglected aspect, yet it plays a pivotal role in different aspects of the vascular function [3, 4]. Endothelial dysfunction is suggested to be an instigator of the development and progression of cardiovascular events in CKD [5]. Signs of endothelial cell (EC) injury such as impaired endothelium-dependent vasodilatation, suppression of nitric oxide production and inflammation are frequently observed in CKD patients [6–8]. However, while it is conceivable that uraemic toxins may interfere with vascular endothelial integrity [9–11], there is only limited evidence for a direct effect of uraemia on EC. Several kidney disease-related risk factors, including the progressive decline of circulating active vitamin D (VD) levels, are suggested to cause adverse changes in vascular function [12–14]. In addition, 25 hydroxyvitamin D [25(OH)D] deficiency is suggested to have a crucial contribution to the development and progression of EC dysfunction in CKD patients [15]. Moreover, active VD supplementation was suggested to reduce cardiovascular events such as vascular calcification or cardiac abnormalities in CKD animal models [16, 17]. However, these results contrast with clinical trials using active VD therapy in CKD patients and controversy remains regarding its beneficial potential, especially with regard to the attenuation of cardiac disease in CKD [18, 19]. In EC dysfunction, it was shown that uraemia impaired vasodilatory response in animals [20, 21], and disturbed artery flow-mediated dilatation in selected patients was prevented by active VD supplementation [22]. However, while it is recognized that active VD analogues modulate endothelial inflammation, cell–cell junction stability and oxidative stress in in vitro studies [23–25], the mechanisms that modulate these effects of active VD in CKD are poorly understood [26]. This lack of knowledge of molecular mechanisms that explain the effects of uraemia on endothelial barrier function and the beneficial effect of active VD, formed the rationale to use innovative in vitro measurements to study the EC behaviour in CKD. For this, we applied electric cell-substrate impedance sensing (ECIS) [27] to measure the real-time changes of the barrier function of human EC under the influence of uraemic plasma estimated glomerular filtration rate [estimated glomerular filtration rate (eGFR)  <18 mL/min/1.73 m2]. We found impaired barrier function and disrupted cell–cell contact of EC exposed to human uraemic plasma. These results were in agreement with the effects of the uraemic toxins indoxyl sulphate and p-cresol. Altered levels of the adherens junction protein vascular endothelial (VE)-cadherin were identified as potentially underlying this EC dysfunction. Finally, the active VD analogue paricalcitol (PC) attenuated the disrupted barrier function induced by uraemia and restored VE-cadherin at junctions. We suggest that targeting impaired stability of adherens junctions might form a therapeutic goal for EC dysfunction in CKD. MATERIALS AND METHODS Patients and sample collection CKD plasma samples were obtained from six patients (three male and three female) non-dialysed with CKD (Stages 4–5), age 75.8 ± 7.9 years (mean ± SD), eGFR 18.1 ± 6.7 mL/min/1.73 m2. Non-CKD plasma samples were from six healthy donors (four male and two female), age 35.9 ± 14 years with eGFR  >90 mL/min/1.73 m2. Plasma samples were obtained from each patient by centrifugation at 2500 g during 10 min at 4°C. Supernatant was centrifuged for 5 min at 13 500 g to achieve platelet-free plasma. Plasma samples were analysed for 25(OH)D by liquid–liquid extraction and liquid chromatography-tandem mass spectrometry detection [the precision of the assay is 4% mean intra-assay coefficient of variation (CV) and 7.5% mean inter-assay CV] and 1, 25-dihydroxyvitamin D [1, 25(OH)2 D] by solid-phase immunoextraction and liquid chromatography-tandem mass spectrometry detection (3.5% mean intra-assay CV and 7.5% mean inter-assay CV). Details of human plasma donors are provided in Table 1. Table 1. Clinical and biochemical characteristics of the non-CKD and CKD plasma gender is indicated Parameters non-CKD CKD Age (years) 35.9 ± 14 5.8 ± 7.9 Gender (n, m; f) (3; 3) (4; 2) eGFR (mL/min/1.73 m2) >90 18.1 ± 6.7 25(OH)D (nmol/L)a 50 56 1, 25(OH)2D (pmol/L)a 96 55 Parameters non-CKD CKD Age (years) 35.9 ± 14 5.8 ± 7.9 Gender (n, m; f) (3; 3) (4; 2) eGFR (mL/min/1.73 m2) >90 18.1 ± 6.7 25(OH)D (nmol/L)a 50 56 1, 25(OH)2D (pmol/L)a 96 55 Age and eGFR are expressed as mean ± SD (n = 6). aBiochemical measurements of 25(OH)D and 1, 25(OH)2 D are measured in plasma pools. m, male; f, female. Table 1. Clinical and biochemical characteristics of the non-CKD and CKD plasma gender is indicated Parameters non-CKD CKD Age (years) 35.9 ± 14 5.8 ± 7.9 Gender (n, m; f) (3; 3) (4; 2) eGFR (mL/min/1.73 m2) >90 18.1 ± 6.7 25(OH)D (nmol/L)a 50 56 1, 25(OH)2D (pmol/L)a 96 55 Parameters non-CKD CKD Age (years) 35.9 ± 14 5.8 ± 7.9 Gender (n, m; f) (3; 3) (4; 2) eGFR (mL/min/1.73 m2) >90 18.1 ± 6.7 25(OH)D (nmol/L)a 50 56 1, 25(OH)2D (pmol/L)a 96 55 Age and eGFR are expressed as mean ± SD (n = 6). aBiochemical measurements of 25(OH)D and 1, 25(OH)2 D are measured in plasma pools. m, male; f, female. EC culture Human umbilical vein ECs (HUVECs) were isolated as described previously [28] from umbilical cords from healthy donors, obtained from the Ziekenhuis Amstelland, Amstelveen. Cells were taken up in M199 medium, supplemented with penicillin 100 U/mL and streptomycin 100 µg/mL (Biowhittaker/Lonza, Verviers, Belgium), human serum 10% (Sanquin Blood Supply, Amsterdam, The Netherlands), newborn calf serum 10% (Gibco, Grand Island, NY, USA), crude EC growth factor 150 µg/mL (prepared from bovine brains), L-glutamine 2 mmol/L (Biowhittaker/Lonza) and heparin 5 U/mL (Leo Pharmaceutical Products, Weesp, The Netherlands). HUVECs were cultured at 37°C and 5% CO2, with a change of culture medium every 2 days. Cells were cultured up to passage 2. For experiments, HUVECs were washed with bare medium (bM199) consisting of M199 supplemented with penicillin 100 IU/mL and streptomycin 100 µg/mL, and exposed to bM199 with 20% of pooled non-CKD/CKD plasma during 24 h, combined with or without PC (100 nM; kindly provided by AbbVie, Chicago, IL, USA). Based on measurements of concentrations in CKD patients with eGFR  <30 mL/min/1.73 m2 [29, 30], uraemic toxins indoxyl sulphate and p-cresol (Sigma Aldrich, Zwijndrecht, The Netherlands) were tested at 100 µM in bM199 with 20% (v/v) of pooled non-CKD plasma. ECIS Transendothelial electrical resistance of HUVEC monolayers was measured using ECIS (Applied BioPhysics, Troy, NY, USA) as previously described [31]. Using eight-well ECIS arrays (10 electrodes per well, 0.49 mm2 electrode area) HUVECs were seeded in a 1:1 density on a gelatin-coated array. ECIS software (v1.2.65.0 PC; Applied BioPhysics) was used to calculate the level of overall resistance (4000 Hz), cell–cell contact (Rb) and cell–matrix contact (α) as depicted in Supplementary data, Figure S1A. Cells were grown to confluency (48 h) and medium was renewed 24 h after seeding. After 24 h of pretreatment, electrical wounding was applied (60 s; 100 000 Hz). For quantification of the wounding response, the slope (assessed by linear regression) and the area under the curve (AUC) of the recovery phase from each replicate (measured after electrical wound induction) was calculated. The effect of thrombin (1 U/mL; Sigma Aldrich) was quantified after 60 min pretreatment with bM199 containing 1% human serum albumin (HSA; Sanquin Blood Supply). Resistance was normalized to the initial value before the addition of thrombin. The percentage of drop after the addition of thrombin and the AUC from each replicate were quantified to assess the extent of the thrombin response. Immunofluorescence imaging HUVECs were seeded in 1:1 density on glass coverslips coated with glutaraldehyde 0.5% (Fluka, St Gallen, Switzerland) cross-linked gelatin and grown to confluency in 24–48 h. After 24 h of treatment, medium was replaced by 2% paraformaldehyde (37°C, Sigma Aldrich), followed by 15 min incubation on ice. Fixed cells were permeabilized with Triton X-100 0.05% (Sigma Aldrich) in phosphate-buffered saline (PBS), washed and incubated overnight with primary antibody against VE-cadherin (1:400; XP D87F2, Cell Signalling, Danvers, MA, USA). Afterwards, HUVECs were washed and incubated with FITC secondary antibodies (1:100; Invitrogen, Paisley, UK) and the F-actin cytoskeleton was stained with rhodamine/phalloidin (1:140; PHDH1 Cytoskeleton Inc., Denver, CO, USA) for 2 h at room temperature. Subsequently, cells were washed and nuclei were stained with DAPI (1:500; Sigma Aldrich) for 15 min. Cells were washed, and sealed with Mowiol mounting medium (Sigma Aldrich). Imaging was performed with a Leica TCS SP8 STED using a 63x Zeiss oil objective. ImageJ (National Institutes of Health, Bethesda, MD, USA) was used for analysis. For quantification of VE-cadherin/F-actin, images were equally adjusted for contrast, mean fluorescence was quantified and divided by the amount of counted nuclei (DAPI). Fluorescence-activated cell sorting VE-cadherin membrane expression was quantified in HUVECs using a specific antibody (Alexa Fluor® 647 anti-mouse CD144) for 30 min. HUVECs were rinsed with PBS twice and then detached using Accutase (GE Healthcare, Eindhoven, The Netherlands). Detached cells were resuspended in ice-cold PBS supplemented with 0.5% (w/v) bovine serum albumin (BSA), centrifuged and resuspended in PBS + 0.5% BSA. Fluorescence-activated cell sorting (FACS) analyses were performed with Calibur flow cytometer (Becton and Dickinson, San Jose, CA, USA) and FlowJo (Ashland, OR, USA) software. Statistical analysis Data were analysed using GraphPad Prism software (La Jolla, CA, USA). Statistical analysis was performed using unpaired t-test for two groups and three or more groups using one-way analysis of variance with Bonferroni correction. Data were shown as means ± SD. Differences were considered statistically significant for P < 0.05 (*P < 0.05, **P < 0.01, ***P < 0.001). RESULTS Uraemic plasma induces endothelial barrier disruption by reduced cell–cell interaction To test whether a uraemic environment affects endothelial barrier integrity, confluent HUVEC monolayers were exposed to plasma from patients with CKD (eGFR <18 mL/min/1.73 m2) or healthy subjects (eGFR >90 mL/min/1.73 m2) (v/v; 20%). Baseline characteristics of the plasma donors are reported in Table 1. Initially, endothelial resistance was reduced (1536 ± 21.8 versus 1477 ± 27.4 Ω; P = 0.0426) in uraemic plasma-treated HUVECs after 3 h when compared with non-CKD plasma (Figure 1A). Although the endothelial resistance normalized to the control values, a second decay of the resistance was observed with statistically significant differences at 24 h (1530 ± 10.1 versus 1432 ± 24.5 Ω; P = 0.0031). Subsequently, we tested recovery after thrombin-induced (1 U/mL) contraction of HUVECs inducing a loss of barrier function, 24 h after plasma exposure (Figure 1B). Increased loss of barrier function was observed for CKD-plasma-exposed EC after thrombin stimulation. This was reflected by the percentage of drop in resistance and accompanied by a significantly reduced AUC measured after data normalization: CKD [49.5%; 3.24 ± 0.01 arbitrary units (a.u.)] versus non-CKD controls (41.5%; 3.55 ± 0.01 a.u.) (Figure 1B). Similarly, HUVECs challenged with uraemic media displayed a reduced recovery capacity in the wound healing assay (Figure 1C). Quantitative analysis showed a significant reduction for both slope (1–5 h) and AUC of the recovery phase (1–8 h) of CKD-plasma-exposed cells (170 ± 1 a.u.; 7636 ± 43.1 a.u.) compared with non-CKD controls (217 ± 13.6 a.u.; 8544 ± 159.3 a.u.). Furthermore, measurements 48 h after the induction of a wound confirmed that differences in AUC between CKD and non-CKD plasma sustained without normalization of CKD condition (11.34% less of AUC in CKD after 24 h and 14% after 48 h when compared with non-CKD) (Supplementary data, Figure S2). FIGURE 1: View largeDownload slide Uraemic plasma induces endothelial barrier disruption in HUVECs. (A) Representative time-course of absolute endothelial electrical resistance measurements (Ω) of confluent HUVECs stimulated for 24 h with 20% (v/v) of non-CKD plasma (dark) or CKD plasma (grey). Insert: quantification of the absolute endothelial resistance (Ω) after 3 h and 24 h of stimulation as indicated by arrowheads. (B) Normalized resistance of thrombin (1 U/mL)-stimulated HUVECs. Upon 24 h incubation with indicated media, HUVECs were exposed to a basal medium plus 1% HSA during 1 h and challenged with thrombin (arrow). Bar graphs: (left) quantification of the thrombin response represented by the AUC (starting at time point 0) and (right) to maximum drop of resistance (%) following stimulation. (C) Resistance measurements (Ω) after electrical wounding (arrowhead) of 24 h 20% plasma pre-stimulated cells. Bar graphs: (left) quantification of the wound healing as the AUC (from 1 h to 8 h) and (right) the slope of the recovery (from 1 h to 5 h). Data show mean ± SD (n = 3, from pool of three donors). Representative data of at least three experiments. Differences were considered statistically significant for P < 0.05 using unpaired student’s t-test where *P < 0.05, **P < 0.01, ***P < 0.001. FIGURE 1: View largeDownload slide Uraemic plasma induces endothelial barrier disruption in HUVECs. (A) Representative time-course of absolute endothelial electrical resistance measurements (Ω) of confluent HUVECs stimulated for 24 h with 20% (v/v) of non-CKD plasma (dark) or CKD plasma (grey). Insert: quantification of the absolute endothelial resistance (Ω) after 3 h and 24 h of stimulation as indicated by arrowheads. (B) Normalized resistance of thrombin (1 U/mL)-stimulated HUVECs. Upon 24 h incubation with indicated media, HUVECs were exposed to a basal medium plus 1% HSA during 1 h and challenged with thrombin (arrow). Bar graphs: (left) quantification of the thrombin response represented by the AUC (starting at time point 0) and (right) to maximum drop of resistance (%) following stimulation. (C) Resistance measurements (Ω) after electrical wounding (arrowhead) of 24 h 20% plasma pre-stimulated cells. Bar graphs: (left) quantification of the wound healing as the AUC (from 1 h to 8 h) and (right) the slope of the recovery (from 1 h to 5 h). Data show mean ± SD (n = 3, from pool of three donors). Representative data of at least three experiments. Differences were considered statistically significant for P < 0.05 using unpaired student’s t-test where *P < 0.05, **P < 0.01, ***P < 0.001. To elucidate whether changes in basal EC barrier function were related to effects on the cell–cell contact or cell–matrix contact, ECIS modelling of the parameters Rb (resistance between cells) and α (resistance between cells and extracellular matrix) was performed [31] (Supplementary data, Figure S1A). As shown in Figure 2A, no significant changes were observed for cell–matrix interactions after 24 h exposure of EC to different media. In contrast, impaired cell–cell contact (9% reduction; P = 0.039) was found in HUVECs challenged with uraemic media (Figure 2B). No differences in cell–matrix interactions were detected during the recovery phase in both assays (Figure 2C and E). The changes observed in cell–cell interactions after thrombin treatment or electrical wounding mimic the EC basal alterations (Figure 2D and F). Together, these findings indicate that uraemic media decrease endothelial barrier function and limit the recovery capacity after an electrical wound healing and thrombin exposure. Abnormal cell–cell but not cell–matrix contact mediates these uraemic plasma-induced changes in resistance. FIGURE 2: View largeDownload slide Impaired endothelial barrier function in CKD is caused by reduced cell–cell interaction. Absolute endothelial electrical resistance attributable to cell–matrix interaction (α; Ω0.5cm) (A) and cell–cell interaction (Rb; Ωcm2) (B) of confluent HUVECs stimulated for 24 h with 20% (v/v) non-CKD plasma (dark) or CKD plasma (grey). Normalized values of α (C) and Rb (D) upon thrombin (1 U/mL; arrow) stimulation. Time course of endothelial α (E) and Rb (F) values after electrical wounding (arrow). Data show mean ± SD (n = 3, from a pool of three donors). Representative data of at least three experiments. *P < 0.05 using unpaired Student’s t-test. FIGURE 2: View largeDownload slide Impaired endothelial barrier function in CKD is caused by reduced cell–cell interaction. Absolute endothelial electrical resistance attributable to cell–matrix interaction (α; Ω0.5cm) (A) and cell–cell interaction (Rb; Ωcm2) (B) of confluent HUVECs stimulated for 24 h with 20% (v/v) non-CKD plasma (dark) or CKD plasma (grey). Normalized values of α (C) and Rb (D) upon thrombin (1 U/mL; arrow) stimulation. Time course of endothelial α (E) and Rb (F) values after electrical wounding (arrow). Data show mean ± SD (n = 3, from a pool of three donors). Representative data of at least three experiments. *P < 0.05 using unpaired Student’s t-test. Uraemic toxins indoxyl-sulphate and p-cresol aggravate non-CKD plasma wound recovery To investigate whether the uraemic retention solutes indoxyl sulphate and p-cresol may be involved in the uraemic media-mediated loss of the endothelial barrier, we tested how the addition of these toxins interfere with the barrier function of a non-CKD condition. The direct addition of these toxins in isolation (100 µM) resulted in no changes to resistance during measurements of basal endothelial barrier function or wound healing (data not shown). When a combination of indoxyl sulphate and p-cresol (100 µM) was tested, no differences in basal EC resistance were detected after 24 h (Supplementary data, Figure S3A). Alternatively, uraemic toxins significantly delayed the recovery of an electric wound as reflected by measurements of the slope of resistance recovery (2–6 h after wounding; 0.30 ± 0.01 a.u. for non-CKD and 0.25 ± 0.02 a.u. for indoxyl sulfate + p-cresol) (Supplementary data, Figure S3B and C). Quantitative analysis of the AUC confirmed that the combined uraemic toxin-exposure results in a declined recovery of barrier integrity at 12 h after wounding (12.58 ± 0.64 a.u. versus 10.82 ± 0.79 a.u.) and the differences where further increased after 40 h (87.67 ± 3.98 a.u. versus 71.44 ± 5.85 a.u.) (Supplementary data, Figure S3D). Similar to CKD media, the joint effects of indoxyl sulphate and p-cresol on wound recovery were mediated by reduced cell–cell interactions (Supplementary data, Figure S4B) while cell–matrix interactions remained unaffected (Supplementary data, Figure S4B). However, no significant differences were found when the AUC was measured (Supplementary data, Figure S4C) although the effect of the uraemic toxins in cell–cell interactions was more pronounced after 40 h, mirroring the AUC measurements assessed after wounding (Supplementary data, Figure S3C). Thus, the addition of uraemic toxins at 100 µM to healthy plasma showed impaired recovery after a wound with disturbed cell–cell contact. VE-cadherin localization is impaired in uraemic conditions Next, we explored the structural changes within the EC using VE-cadherin and F-actin (immuno)-staining (Figure 3A). VE-cadherin, which is essential to modulate and maintain the inter-EC contacts [32], showed a markedly disrupted peripheral localization and discontinuous organization after uraemic media exposure (Figure 3A, zoom). Consistently, after uraemic plasma stimulation, there was a significant reduction of the VE-cadherin fluorescence intensity when compared not only with non-CKD plasma (45% less) but also with a control media (42% less). Similarly, quantitative analysis using flow cytometry confirmed a decreased membrane expression of VE-cadherin in HUVECs exposed to CKD plasma (41% versus non-CKD) (Figure 3D and E). Furthermore, non-uraemic plasma-exposed EC showed a peripheral distribution of the F-actin organized as a cortical actin rim which promotes the assembly of EC–cell and cell–matrix adhesions. In contrast, exposure to uraemic plasma resulted in prominent F-actin stress fibres, characteristic of cell contraction albeit without significant changes when compared with non-CKD (0.82 ± 0.03 versus 0.64 ± 0.09 a.u.; P = 0.061) (Figure 3C). No changes in VE-cadherin expression where found in cells exposed to control media when compared with non-CKD media, although HUVECs incubated with control media displayed a significant increment of F-actin stress fibre formation. These findings suggest that disrupted adherens junction stability underlies the effect on EC dysfunction observed in uraemic plasma-exposed cells. These alterations in adherens junctions were accompanied by a reorganization of F-actin in uraemic plasma-exposed HUVECs. FIGURE 3: View largeDownload slide Membrane VE-cadherin localization is disturbed in CKD plasma exposed HUVECs. (A) Immunostaining of VE-cadherin (green), F-actin (red) and nuclei (blue) of HUVECs after 24 h of incubation in medium (bM199) plus 1% HSA (Control) or 20% (v/v) non-CKD or CKD plasma. Scale bar represents 50 µm. Zoomed images correspond to the white boxes. Normalized intensity quantifications are in the corresponding graphs for VE-cadherin (B) and F-actin (C). Surface expression of VE-cadherin was assessed by flow cytometry (D) in 24 h-stimulated cells. (E) Normalized mean fluorescence intensity measured by flow cytometry of membrane VE-cadherin is shown for the different conditions. Data are mean ± SD of three independent experiments. Differences were considered statistically significant for P < 0.05 using one-way analysis of variance (ANOVA) where *P < 0.05, **P < 0.01. FIGURE 3: View largeDownload slide Membrane VE-cadherin localization is disturbed in CKD plasma exposed HUVECs. (A) Immunostaining of VE-cadherin (green), F-actin (red) and nuclei (blue) of HUVECs after 24 h of incubation in medium (bM199) plus 1% HSA (Control) or 20% (v/v) non-CKD or CKD plasma. Scale bar represents 50 µm. Zoomed images correspond to the white boxes. Normalized intensity quantifications are in the corresponding graphs for VE-cadherin (B) and F-actin (C). Surface expression of VE-cadherin was assessed by flow cytometry (D) in 24 h-stimulated cells. (E) Normalized mean fluorescence intensity measured by flow cytometry of membrane VE-cadherin is shown for the different conditions. Data are mean ± SD of three independent experiments. Differences were considered statistically significant for P < 0.05 using one-way analysis of variance (ANOVA) where *P < 0.05, **P < 0.01. Active VD attenuates uraemia-mediated endothelial dysfunction Measurements of VD levels confirmed that, in CKD plasma, 1, 25(OH)2 D concentrations were 42% lower than those in non-CKD plasma (96 pmol/L versus 55 pmol/L), whereas 25(OH)D concentrations were similar between groups (50 nmol/L versus 56 nmol/L) (Table 1). We therefore further tested the potentially beneficial effects of the active VD form PC in the uraemic media-mediated endothelial dysfunction. Quantifying basal endothelial resistance, we found that addition of 100 nM PC rescued the significant reduction in resistance induced by uraemic plasma after 24 h when compared with non-CKD conditions (Figure 4A, insert). However, no significant differences between CKD groups were detected. No significant improvement of the barrier function was detected for healthy plasma combined with PC. Furthermore, the addition of PC to uraemic medium improved the impaired recovery after electrical wounding (Figure 4B). This was reflected by a pronounced increase of the slope (from 1 to 6 h) in CKD + PC (140 ± 9.5 a.u.) rescuing the significant differences induced by uraemic plasma when compared with non-CKD conditions (121 ± 13 a.u.), although there was no significant difference between CKD groups (Figure 4C). In addition, the AUC (from 1 to 24 h) was significantly higher (7.5%) for PC-treated cells in CKD plasma when compared with CKD alone. Moreover, the addition of PC to non-uraemic media increased the differences of the AUC and slope against CKD media when compared with non-CKD alone as shown in Figure 4C. These data show amelioration by PC of the uraemia-mediated changes on basal endothelial resistance. In addition, PC ameliorated the recovery capacity of endothelial barrier following electrical wounding in uraemic plasma. FIGURE 4: View largeDownload slide PC stabilizes uraemic plasma-induced endothelial dysfunction. (A) Representative measurements of absolute endothelial electrical resistance (Ω) of HUVECs, stimulated for 24 h as indicated. Insert: absolute endothelial resistance (Ω) after 24 h (non-CKD = dark; non-CKD + PC = striped dark; CKD = grey; CKD + PC = striped grey). (B) Electrical wound healing (arrow) of cells pre-stimulated during 24 h as indicated. (C) The quantifications of the wound healing capacity are shown as the AUC (from 1 h to 24 h) and slope of the recovery phase (a.u.; from 1 to 6 h). Representative data of at least three experiments. Data show mean ± SD (n = 3, pool of three donors). Differences were considered statistically significant for P < 0.05 using one-way analysis of variance (ANOVA) where *P < 0.05, **P < 0.01, ***P < 0.001. FIGURE 4: View largeDownload slide PC stabilizes uraemic plasma-induced endothelial dysfunction. (A) Representative measurements of absolute endothelial electrical resistance (Ω) of HUVECs, stimulated for 24 h as indicated. Insert: absolute endothelial resistance (Ω) after 24 h (non-CKD = dark; non-CKD + PC = striped dark; CKD = grey; CKD + PC = striped grey). (B) Electrical wound healing (arrow) of cells pre-stimulated during 24 h as indicated. (C) The quantifications of the wound healing capacity are shown as the AUC (from 1 h to 24 h) and slope of the recovery phase (a.u.; from 1 to 6 h). Representative data of at least three experiments. Data show mean ± SD (n = 3, pool of three donors). Differences were considered statistically significant for P < 0.05 using one-way analysis of variance (ANOVA) where *P < 0.05, **P < 0.01, ***P < 0.001. Disrupted VE-cadherin-mediated cell–cell junctions in uraemic plasma are restored by active VD After determining the potency of PC to improve the barrier function during uraemia, we explored whether this effect is the consequence of the restoration of cell–cell contact. In line with the previous experiment, PC significantly improved (9.6%) the strength of the cell–cell interaction after 24 h as reflected by the quantification of the resistance between cells (Rb) (Figure 5A and B). PC supplementation rescued the significant differences encountered in uraemic plasma-treated EC after quantification of the AUC (24 h; 12.6%) of the cell–cell quantification in the wound healing process when compared with non-CKD conditions (Figure 5C). However, no significant differences were found between uraemic conditions with or without PC treatment. No significant changes were found regarding the slope of the wound healing phase although uraemic medium alone showed a negative trend (−15 to −10%) when compared with other conditions. Further, PC induced enhanced VE-cadherin peripheral intensity under uraemic conditions as shown in Figure 6A, which is supported by significant differences of intensity measurements (38% higher; P = 0.015) when compared with uraemia alone (Figure 6B). Evaluating adherens junctions morphology, the jagged distribution of VE-cadherin was improved to a more linear distribution when HUVECs in uraemic conditions where supplemented with PC (Figure 6A, zoom). PC slightly, but not significantly, increased VE-cadherin intensity (14%) in non-uraemic medium. Finally, although cytoplasmic F-actin filaments were dominant in uraemic conditions as observed in Figure 6A, the fluorescence intensity of F-actin did not differ significantly when compared with other conditions (Figure 6C). In addition, immunostaining for F-actin showed that PC reduced the arrangement of stress fibres upon uraemic plasma while in non-uraemic conditions PC slightly enhanced F-actin cortical distribution, although no significant changes were observed after quantification. Thus, we confirmed that PC strengthens the endothelial barrier integrity under uraemic conditions by mediating the stabilization of junctional VE-cadherin. FIGURE 5: View largeDownload slide Cell–cell interactions are stabilized by PC in uraemic media. (A) Representative cell–cell interaction (Rb; Ωcm2) measurements of HUVECs incubated for 24 h as indicated (non-CKD = dark; non-CKD + PC = striped dark; CKD = grey; CKD + PC= striped grey). (B) Rb of an electric wounding response (arrow) of cells previously stimulated for 24 h (C) Quantification of the wound healing capacities are represented as the AUC (from 3 to 24 h) and the slope (a.u.; from 3 to 7 h). Data show mean ± SD (n = 3). Representative data of at least three experiments. Differences were considered statistically significant for P < 0.05 using one-way analysis of variance (ANOVA) where *P < 0.05, **P < 0.01. FIGURE 5: View largeDownload slide Cell–cell interactions are stabilized by PC in uraemic media. (A) Representative cell–cell interaction (Rb; Ωcm2) measurements of HUVECs incubated for 24 h as indicated (non-CKD = dark; non-CKD + PC = striped dark; CKD = grey; CKD + PC= striped grey). (B) Rb of an electric wounding response (arrow) of cells previously stimulated for 24 h (C) Quantification of the wound healing capacities are represented as the AUC (from 3 to 24 h) and the slope (a.u.; from 3 to 7 h). Data show mean ± SD (n = 3). Representative data of at least three experiments. Differences were considered statistically significant for P < 0.05 using one-way analysis of variance (ANOVA) where *P < 0.05, **P < 0.01. FIGURE 6: View largeDownload slide Reinforced VE-cadherin contact mediates the effect of PC in uraemic media. (A) Immunofluorescence staining of VE-cadherin (green), F-actin (red) and the nuclei (blue) of HUVEC following 24 h of incubation (scale bar, 50 µm). Zoomed images correspond to the white boxes. Normalized mean VE-cadherin (B) and F-actin (C) intensity of the immunostainings are shown (non-CKD = dark; non-CKD + PC = striped dark; CKD = grey; CKD + PC= striped grey). Data show mean ± SD of three independent experiments. Differences were considered statistically significant for P < 0.05 using one-way analysis of variance (ANOVA) where *P < 0.05, **P < 0.01, ***P < 0.001. FIGURE 6: View largeDownload slide Reinforced VE-cadherin contact mediates the effect of PC in uraemic media. (A) Immunofluorescence staining of VE-cadherin (green), F-actin (red) and the nuclei (blue) of HUVEC following 24 h of incubation (scale bar, 50 µm). Zoomed images correspond to the white boxes. Normalized mean VE-cadherin (B) and F-actin (C) intensity of the immunostainings are shown (non-CKD = dark; non-CKD + PC = striped dark; CKD = grey; CKD + PC= striped grey). Data show mean ± SD of three independent experiments. Differences were considered statistically significant for P < 0.05 using one-way analysis of variance (ANOVA) where *P < 0.05, **P < 0.01, ***P < 0.001. DISCUSSION This study demonstrates that uraemic media negatively affect the integrity of the EC barrier function and also impair the recovery following exposure to barrier-disruptive mediators. Specifically, cell–cell interactions reduced in uraemic plasma-exposed EC were driving the overall barrier dysfunction, which was corroborated by reduced peripheral VE-cadherin and cortical F-actin. The active VD compound PC attenuated the endothelial barrier-disruptive effects of uraemia in basal and electrical wounding assays by improving the cell–cell contact and restoring endothelial integrity. Prolonged exposure of uraemic toxins present in the plasma of CKD patients can affect the integrity and repair capacity of EC [8]. In our study, ECIS has been shown to be suitable to characterize in real time, the effects of uraemic media on confluent endothelial barrier function [31]. Adverse changes on EC by uraemic plasma such as limited resistance to flow, activation of nuclear factor kappa B including increased permeability, have been previously demonstrated in vitro [9–11] and in vivo [33]. Our data show that EC exposed to uraemic plasma (derived from patients with eGFR  <18 mL/min/1.73 m2) indeed displayed a reduced resistance when compared with healthy plasma (eGFR >90 mL/min/1.73 m2). Furthermore, when the barrier integrity was compromised by thrombin (inducing cell contraction and intercellular gap formation [34]) or electrical wounding [27], mimicking an injury, HUVECs exhibit a reduced recovery capacity after exposure for 24 h to CKD plasma. Based on this data, we suggest that EC that have been exposed to uraemia are more sensitive to barrier-disruptive conditions. Interestingly, this is supported by data showing altered proliferation and wound repair on HUVECs exposed to the uraemic toxins p-cresol and indoxyl sulphate [35]. These uraemic retention toxins, poorly removed by haemodialysis therapies, are kidney disease-related cardiovascular risk factors and are suggested to contribute to endothelial toxicity in CKD [29, 36, 37]. In this study, the addition of indoxyl sulphate and p-cresol in a concentration detected in serum from CKD patients with eGFR  <30 mL/min/1.73 m2 (100 μΜ) [29, 30], mimicked the deleterious effects of CKD plasma in the recovery of the EC barrier function after a wound. Alternatively, the differences found in basal EC resistance after CKD plasma stimulation were not detected in media with uraemic toxins indicating that the effects from CKD in EC dysfunction are not exclusive from indoxyl sulphate and p-cresol toxicity. There are several cellular structures that are essential to the maintenance of the EC barrier integrity [38]. By mathematically modelling our data obtained with ECIS [31], we concluded that the changes in transendothelial electrical resistance which we previously observed can be attributed to impaired cell–cell contact in uraemic media. This was confirmed after measuring of the resistance between cells (Rb), which was reduced, while the current between the cell–matrix interactions (α) remained unchanged. In good agreement with these data, our imaging analysis showed that the adherens junction protein VE-cadherin, essential for endothelial barrier integrity [32], showed disrupted localization and was less concentrated at cell boundaries upon uraemic conditions. The decrease of VE-cadherin in cells was not attributable to redistribution to the cell surface as confirmed by FACS. Furthermore, VE-cadherin is linked to the actin cytoskeleton and its association is essential to barrier function [39]. On exposure to uraemic media, EC displayed a reduced cortical ring-like F-actin distribution. A prominent cortical actin cytoskeleton structure is characteristic for quiescent endothelium, and therefore, stable barrier integrity. In addition, immunostaining suggested increased presence of stress fibres indicating a reorganization of the F-actin protein leading to a different cell shape and cell contact destabilization albeit not statistically significant. Again, this is in line with in vitro studies performed with uraemic toxins [40, 41]. HUVECs exposed to p-cresol displayed limited VE-cadherin and F-actin colocalization while indoxyl sulphate addition resulted in adherens junction disassembly and cytoskeleton reorganization [40, 41]. This mechanism is suggested to be mediated by an enhanced dissociation of VE-cadherin and actin cytoskeleton induced by the Rho/Rho kinase pathway [40, 41]. Upon activation of Rho, there is an increase of myosin light chain phosphorylation leading to the formation of stress fibres, myosin-based contraction and opening of cell junctions, which increases permeability [42]. Indeed, by measuring Rb during wound healing we confirmed that indoxyl sulphate and p-cresol combination impact cell–cell but not cell–matrix interactions, similarly as CKD plasma. Further evidence for an uraemia-induced disturbed actin cytoskeleton comes from a proteomic analysis in ECs, showing an increment of the actin depolymerizing protein dextrin upon uraemic serum stimulation [43]. Interestingly, this was accompanied by a downregulation of annexin A2, which plays a key role in the establishment of adherens junctions [43]. Our data extend these observations by comparing the effects of uraemic plasma with those of a control (bare medium). Deficiencies in kidney-specific factors such as 25(OH)D, 1, 25(OH)2 D or α-Klotho are suggested to also contribute to cardiovascular pathology [12–14]. In our study, VD measurements confirmed that 1, 25(OH)2 D in CKD plasma was lower than in plasma from healthy donors. However, when compared with CKD condition, control media showed no changes in VE-cadherin intensity suggesting that the effects observed in CKD were not related to some unidentified deficiency in the plasma, but a direct effect of uraemic toxins. Yet, our experiment does not rule out that 1, 25(OH)2 D deficiency aggravates uraemia-mediated endothelial damage. In CKD animal models [20, 21] and patients [22, 44], active VD supplementation has been shown to mitigate endothelial dysfunction. In addition, dietary VD and 1, 25(OH)2 D or PC can modulate endothelial stability by modifying inflammation, thrombosis and vasodilation [23–25]. We found that adding PC partially prevented the changes in electrical resistance induced by uraemia and improved the recovery following electrical wounding. Interestingly, Won et al. showed that 1, 25(OH)2 D attenuated the decrease of electrical resistance upon hypoxia in brain EC and restored tight junction expression, which has a prominent contribution to blood–brain barrier function [25]. This protective effect was mediated by a decrease of the Matrixmetalloproteinase-9 (MMP-9), which mediates disruption of cell–cell interaction [25]. In line with this, we found a restoration of junctional VE-cadherin in cells incubated with uraemic media by PC. Interestingly, increased MMP-9 has also been examined in vitro in EC after the application of uraemic media [45]. Although more detailed studies are necessary to elucidate the vasculo-protective mechanism of active VD upon uraemia, altered MMP-9 expression appears to be one possible mechanism of endothelial protection. This possible mode of action is illustrated in the Supplementary data, Figure S1B. In addition, it remains to be studied whether uraemia-disrupted cell–cell interaction also includes the impairment of the endothelial tight junction. In our study, however, we focused on the effects on VE-cadherin since non-brain EC exhibit less developed tight junctions. Importantly, besides being the major determinant of endothelial cell–cell interaction [32], VE-cadherin plays a key role controlling the level of expression and localization of other junctional molecules [46]. As an alternative mechanism of action, it has been proposed that targeting the uraemia-induced oxidative stress in the endothelium could be a potential strategy against endothelial dysfunction in CKD [47]. In this regard, 1, 25(OH)2 D has been shown to be protective against oxidative stress in EC [48], and therefore, this mechanism could also be related to the endothelial-protective proprieties of active VD against uraemia. Finally, it is also important to consider that EC has the ability to transform 25(OH)D to the active metabolite 1, 25(OH)2 D [49], and comparable endothelial protective effects could be achieved by 25(OH)D supplementation. This alternative therapeutic approach could be of importance in a setting with a 25(OH)D deficiency, a feature that, however, was not applicable in our study. Our study bears some limitations as the exact signalling mechanism involved in the VE-cadherin-disrupted contact and F-actin reorganization during uraemia remains to be established. Moreover, it would be interesting to determine whether the uraemia-mediated endothelial damage could be prevented by the previous addition of PC. Nevertheless, this setting would not be representative of a clinical situation. Likewise, additional EC-types need to be tested to confirm the effects observed during uraemia. Despite those disadvantages, we provide valuable in vitro data by the combination of human uraemic plasma with real-time measurements of endothelial barrier functions. In conclusion, we have extended our insight of the effects of uraemia on EC dysfunction. Although several kidney-disease-related risk factors were suggested to contribute to endothelial dysfunction in CKD, our in vitro findings show similarities with the effects of the uraemic toxins indoxyl sulphate and p-cresol. We propose that limited cell–cell interaction caused by reduced VE-cadherin and F-actin reorganization affects the integrity of the EC barrier during uraemia. These changes may carry important clinical implications since they restrict the capacity of the EC to resist and recover from barrier-disruptive conditions, aggravating vascular complications during CKD. In addition, as a therapeutic approach, we describe a novel mechanism how the active VD analogue PC modulates the uraemia-damaged endothelium by restoring cell–cell interactions. Overall, stabilizing the intercellular endothelial contact might be a crucial step to prevent the damage in the endothelium and ameliorate future vascular complications in CKD. SUPPLEMENTARY DATA Supplementary data are available at ndt online. ACKNOWLEDGEMENTS We thank Jeroen Kole and Dr. Renee Musters (VU University Medical Center Amsterdam, Advanced Optical Microscopy Core facility, The Netherlands) for their help with the microscope equipment. CONFLICT OF INTEREST STATEMENT M.V.C. reports grants from AbbVie, during the conduct of the study; grants from Amgen, personal fees from Amgen, personal fees from VFMCRP, grants from FMC, personal fees from Otsuka, personal fees from Medice, personal fees from BBraun and personal fees from Baxter, outside the submitted work. The other authors reported no conflicts of interest. REFERENCES 1 Schiffrin EL , Lipman ML , Mann JF. Chronic kidney disease: effects on the cardiovascular system . Circulation 2007 ; 116 : 85 – 97 Google Scholar CrossRef Search ADS PubMed 2 Vervloet M , Cozzolino M. Vascular calcification in chronic kidney disease: different bricks in the wall? Kidney Int 2017 ; 91 : 808 – 817 Google Scholar CrossRef Search ADS PubMed 3 Cines DB , Pollak ES , Buck CA et al. Endothelial cells in physiology and in the pathophysiology of vascular disorders . Blood 1998 ; 91 : 3527 – 3561 Google Scholar PubMed 4 Deanfield JE , Halcox JP , Rabelink TJ. Endothelial function and dysfunction: testing and clinical relevance . Circulation 2007 ; 115 : 1285 – 1295 Google Scholar PubMed 5 Goligorsky MS. Pathogenesis of endothelial cell dysfunction in chronic kidney disease: a retrospective and what the future may hold . Kidney Res Clin Pract 2015 ; 34 : 76 – 82 Google Scholar CrossRef Search ADS PubMed 6 Jacobson SH , Egberg N , Hylander B et al. Correlation between soluble markers of endothelial dysfunction in patients with renal failure . Am J Nephrol 2002 ; 22 : 42 – 47 Google Scholar CrossRef Search ADS PubMed 7 van Guldener C , Janssen MJ , Lambert J et al. Endothelium-dependent vasodilatation is impaired in peritoneal dialysis patients . Nephrol Dial Transplant 1998 ; 13 : 1782 – 1786 Google Scholar CrossRef Search ADS PubMed 8 Gris JC , Branger B , Vécina F et al. Increased cardiovascular risk factors and features of endothelial activation and dysfunction in dialyzed uremic patients . Kidney Int 1994 ; 46 : 807 – 813 Google Scholar CrossRef Search ADS PubMed 9 Caballo C , Palomo M , Cases A et al. NFκB in the development of endothelial activation and damage in uremia: an in vitro approach . PLoS One 2012 ; 7 : e43374 Google Scholar CrossRef Search ADS PubMed 10 Serradell M , Díaz-Ricart M , Cases A et al. Uremic medium causes expression, redistribution and shedding of adhesion molecules in cultured endothelial cells . Haematologica 2002 ; 87 : 1053 – 1061 Google Scholar PubMed 11 Serradell M , Díaz-Ricart M , Cases A et al. Uremic medium disturbs the hemostatic balance of cultured human endothelial cells . Thromb Haemost 2001 ; 86 : 1099 – 1105 Google Scholar CrossRef Search ADS PubMed 12 Dusso AS. Kidney disease and vitamin D levels: 25-hydroxyvitamin D, 1, 25-dihydroxyvitamin D, and VDR activation . Kidney Int (Suppl) 2011 ; 1 : 136 – 141 Google Scholar CrossRef Search ADS 13 Seiler S , Rogacev KS , Roth HJ et al. Associations of FGF-23 and sKlotho with cardiovascular outcomes among patients with CKD stages 2-4 . Clin J Am Soc Nephrol 2014 ; 9 : 1049 – 1058 Google Scholar CrossRef Search ADS PubMed 14 Kuro-O M , Matsumura Y , Aizawa H et al. Mutation of the mouse klotho gene leads to a syndrome resembling ageing . Nature 1997 ; 390 : 45 – 51 Google Scholar CrossRef Search ADS PubMed 15 Jablonski KL , Chonchol M , Pierce GL et al. 25-Hydroxyvitamin D deficiency is associated with inflammation-linked vascular endothelial dysfunction in middle-aged and older adults . Hypertension 2011 ; 57 : 63 – 69 Google Scholar CrossRef Search ADS PubMed 16 Kong J , Kim GH , Wei M et al. Therapeutic effects of vitamin D analogs on cardiac hypertrophy in spontaneously hypertensive rats . Am J Pathol 2010 ; 177 : 622 – 631 Google Scholar CrossRef Search ADS PubMed 17 Lau WL , Leaf EM , Hu MC et al. Vitamin D receptor agonists increase klotho and osteopontin while decreasing aortic calcification in mice with chronic kidney disease fed a high phosphate diet . Kidney Int 2012 ; 82 : 1261 – 1270 Google Scholar CrossRef Search ADS PubMed 18 Wang AY , Fang F , Chan J et al. Effect of paricalcitol on left ventricular mass and function in CKD–the OPERA trial . J Am Soc Nephrol 2014 ; 25 : 175 – 186 Google Scholar CrossRef Search ADS PubMed 19 Thadhani R , Appelbaum E , Pritchett Y et al. Vitamin D therapy and cardiac structure and function in patients with chronic kidney disease: the PRIMO randomized controlled trial . JAMA 2012 ; 307 : 674 – 684 Google Scholar CrossRef Search ADS PubMed 20 Wu-Wong JR , Noonan W , Nakane M et al. Vitamin d receptor activation mitigates the impact of uremia on endothelial function in the 5/6 nephrectomized rats . Int J Endocrinol 2010 ; 2010 : 625852 Google Scholar PubMed 21 Wu-Wong JR , Li X , Chen YW. Different vitamin D receptor agonists exhibit differential effects on endothelial function and aortic gene expression in 5/6 nephrectomized rats . J Steroid Biochem Mol Biol 2015 ; 148 : 202 – 209 Google Scholar CrossRef Search ADS PubMed 22 Stojanović M , Radenković M. Vitamin D versus placebo in improvement of endothelial dysfunction: a meta-analysis of randomized clinical trials . Cardiovasc Ther 2015 ; 33 : 145 – 154 Google Scholar CrossRef Search ADS PubMed 23 Gibson CC , Davis CT , Zhu W et al. Dietary vitamin D and its metabolites non-genomically stabilize the endothelium . PLoS One 2015 ; 10 : e0140370 Google Scholar CrossRef Search ADS PubMed 24 Lee AS , Jung YJ , Thanh TN et al. Paricalcitol attenuates lipopolysaccharide-induced myocardial inflammation by regulating the NF-κB signaling pathway . Int J Mol Med 2016 ; 37 : 1023 – 1029 Google Scholar CrossRef Search ADS PubMed 25 Won S , Sayeed I , Peterson BL et al. Vitamin D prevents hypoxia/reoxygenation-induced blood-brain barrier disruption via vitamin D receptor-mediated NF-kB signaling pathways . PLoS One 2015 ; 10 : e0122821 Google Scholar CrossRef Search ADS PubMed 26 Talmor-Barkan Y , Bernheim J , Green J et al. Calcitriol counteracts endothelial cell pro-inflammatory processes in a chronic kidney disease-like environment . J Steroid Biochem Mol Biol 2011 ; 124 : 19 – 24 Google Scholar CrossRef Search ADS PubMed 27 Keese CR , Wegener J , Walker SR et al. Electrical wound-healing assay for cells in vitro . Proc Natl Acad Sci USA 2004 ; 101 : 1554 – 1559 Google Scholar CrossRef Search ADS PubMed 28 van Nieuw Amerongen GP , Draijer R , Vermeer MA et al. Transient and prolonged increase in endothelial permeability induced by histamine and thrombin: role of protein kinases, calcium, and RhoA . Circ Res 1998 ; 83 : 1115 – 1123 Google Scholar CrossRef Search ADS PubMed 29 Meijers BK , Claes K , Bammens B et al. p-Cresol and cardiovascular risk in mild-to-moderate kidney disease . Clin J Am Soc Nephrol 2010 ; 5 : 1182 – 1189 Google Scholar CrossRef Search ADS PubMed 30 Kamiński TW , Pawlak K , Karbowska M et al. Indoxyl sulfate - the uremic toxin linking hemostatic system disturbances with the prevalence of cardiovascular disease in patients with chronic kidney disease . BMC Nephrol 2017 ; 18 : 35 Google Scholar CrossRef Search ADS PubMed 31 Szulcek R , Bogaard HJ , van Nieuw Amerongen GP. Electric cell-substrate impedance sensing for the quantification of endothelial proliferation, barrier function, and motility . J Vis Exp 2014 ; 85 : e51300 32 Vestweber D. VE-cadherin: the major endothelial adhesion molecule controlling cellular junctions and blood vessel formation . Arterioscler Thromb Vasc Biol 2007 ; 28 : 223 – 232 Google Scholar CrossRef Search ADS PubMed 33 Harper SJ , Tomson CR , Bates DO. Human uremic plasma increases microvascular permeability to water and proteins in vivo . Kidney Int 2002 ; 61 : 1416 – 1422 Google Scholar CrossRef Search ADS PubMed 34 Sakamoto T , Sakamoto H , Sheu SJ et al. Intercellular gap formation induced by thrombin in confluent cultured bovine retinal pigment epithelial cells . Invest Ophthalmol Vis Sci 1994 ; 35 : 720 – 729 Google Scholar PubMed 35 Dou L , Bertrand E , Cerini C et al. The uremic solutes p-cresol and indoxyl sulfate inhibit endothelial proliferation and wound repair . Kidney Int 2004 ; 65 : 442 – 451 Google Scholar CrossRef Search ADS PubMed 36 Meijers BK , Bammens B , De Moor B et al. Free p-cresol is associated with cardiovascular disease in hemodialysis patients . Kidney Int 2008 ; 73 : 1174 – 1180 Google Scholar CrossRef Search ADS PubMed 37 Lin CJ , Liu HL , Pan CF et al. Indoxyl sulfate predicts cardiovascular disease and renal function deterioration in advanced chronic kidney disease . Arch Med Res 2012 ; 43 : 451 – 456 Google Scholar CrossRef Search ADS PubMed 38 Wallez Y , Huber P. Endothelial adherens and tight junctions in vascular homeostasis, inflammation and angiogenesis . Biochim Biophys Acta 2008 ; 1778 : 794 – 809 Google Scholar CrossRef Search ADS PubMed 39 Prasain N , Stevens T. The actin cytoskeleton in endothelial cell phenotypes . Microvasc Res 2009 ; 77 : 53 – 63 Google Scholar CrossRef Search ADS PubMed 40 Peng YS , Lin YT , Chen Y et al. Effects of indoxyl sulfate on adherens junctions of endothelial cells and the underlying signaling mechanism . J Cell Biochem 2012 ; 113 : 1034 – 1043 Google Scholar CrossRef Search ADS PubMed 41 Cerini C , Dou L , Anfosso F et al. P-cresol, a uremic retention solute, alters the endothelial barrier function in vitro . Thromb Haemost 2004 ; 92 : 140 – 150 Google Scholar PubMed 42 Amado-Azevedo J , Valent ET , Van Nieuw Amerongen GP. Regulation of the endothelial barrier function: a filum granum of cellular forces, Rho-GTPase signaling and microenvironment . Cell Tissue Res 2014 ; 355 : 557 – 576 Google Scholar CrossRef Search ADS PubMed 43 Carbó C , Arderiu G , Escolar G et al. Differential expression of proteins from cultured endothelial cells exposed to uremic versus normal serum . Am J Kidney Dis 2008 ; 51 : 603 – 612 Google Scholar CrossRef Search ADS PubMed 44 Lundwall K , Jörneskog G , Jacobson SH et al. Paricalcitol, microvascular and endothelial function in non-diabetic chronic kidney disease: a randomized trial . Am J Nephrol 2015 ; 42 : 265 – 273 Google Scholar CrossRef Search ADS PubMed 45 Zafeiropoulou K , Bita T , Polykratis A et al. Hemodialysis removes uremic toxins that alter the biological actions of endothelial cells . PLoS One 2012 ; 7 : e30975 Google Scholar CrossRef Search ADS PubMed 46 Gavard J , Gutkind JS. VE-cadherin and claudin-5: it takes two to tango . Nat Cell Biol 2008 ; 10 : 883 – 885 Google Scholar CrossRef Search ADS PubMed 47 Rodrigues SD , França KC , Dallin FT et al. N-acetylcysteine as a potential strategy to attenuate the oxidative stress induced by uremic serum in the vascular system . Life Sci 2015 ; 121 : 110 – 116 Google Scholar CrossRef Search ADS PubMed 48 Jia X , Xu J , Gu Y et al. Vitamin D suppresses oxidative stress-induced microparticle release by human umbilical vein endothelial cells . Biol Reprod 2017 ; 96 : 199 – 210 Google Scholar PubMed 49 Merke J , Milde P , Lewicka S et al. Identification and regulation of 1, 25-dihydroxyvitamin D3 receptor activity and biosynthesis of 1, 25-dihydroxyvitamin D3. Studies in cultured bovine aortic endothelial cells and human dermal capillaries . J Clin Invest 1989 ; 83 : 1903 – 1915 Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Nephrology Dialysis TransplantationOxford University Press

Published: Apr 30, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off