January 2010
Volume 51, Issue 1
Free
Retinal Cell Biology  |   January 2010
Inhibition of Protein Kinase C Is Not Sufficient to Prevent or Reverse Effects of VEGF165 on Claudin-1 and Permeability in Microvascular Retinal Endothelial Cells
Author Affiliations & Notes
  • Heidrun L. Deissler
    From the Departments of Ophthalmology and
  • Helmut Deissler
    Obstetrics and Gynecology, University of Ulm Medical School, Ulm, Germany.
  • Gabriele E. Lang
    From the Departments of Ophthalmology and
  • Corresponding author: Heidrun L. Deissler, Department of Ophthalmology, University of Ulm Medical School, Prittwitzstrasse 43, D-89075 Ulm, Germany; heidrun.deissler@uniklinik-ulm.de
Investigative Ophthalmology & Visual Science January 2010, Vol.51, 535-542. doi:10.1167/iovs.09-3917
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Heidrun L. Deissler, Helmut Deissler, Gabriele E. Lang; Inhibition of Protein Kinase C Is Not Sufficient to Prevent or Reverse Effects of VEGF165 on Claudin-1 and Permeability in Microvascular Retinal Endothelial Cells. Invest. Ophthalmol. Vis. Sci. 2010;51(1):535-542. doi: 10.1167/iovs.09-3917.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose.: Pathogenesis of diabetic macular edema is driven by deregulated expression of VEGF. A study of long-term exposure of immortalized bovine retinal endothelial cells (iBRECs) to VEGF165 clearly confirmed the role of the tight junction protein claudin-1, which almost completely disappeared within 24 hours, an effect that was completely reversed by addition of the VEGF-binding Fab fragment ranibizumab. This study was conducted to investigate whether the VEGF165-induced loss of claudin-1 is regulated by protein kinase C (PKC) and indeed affects the barrier function of iBRECs.

Methods.: The effects of various PKC inhibitors on claudin-1 expression and cellular localization in iBRECs treated with VEGF165 for up to 2 days were studied by Western blot analyses and immunofluorescence microscopy. The permeability of the cell layers was determined by transendothelial electrical resistance measurements.

Results.: Activation of PKC led to decreased expression of claudin-1, which was blocked by inhibitors of PKCδ. However, none of the PKC inhibitors significantly affected VEGF165-induced effects on cellular localization or expression of claudin-1. Also VEGF165-induced higher permeability of iBREC layers could be reversed or prevented by ranibizumab but not by PKC inhibitors. In addition, low claudin-1 expression and its delocalization from the plasma membrane were significantly associated with elevated permeability.

Conclusions.: In iBRECs, PKC isoforms are not crucially involved in the VEGF165-initiated signal transduction that affects permeability and expression of claudin-1. This finding is in contrast to published results concerning only short-term effects of VEGF165. The results also confirmed that claudin-1 is a highly relevant component of functional tight junctions in retinal endothelial cells.

Diabetic retinopathy (DR) is one of the major complications of diabetes mellitus and the main cause of blindness in industrialized countries. Events in the genesis and early development of DR include the loss of retinal pericytes, thickening of the capillary basement membrane, and increased leukocyte adhesion to retinal endothelial cells (RECs). It is now a generally accepted concept that the pathogenesis of DR is driven by deregulated expression of growth factors. Of these, vascular endothelial growth factor (VEGF) is considered most important because its levels are substantially elevated in the vitreous fluid and in the retinal vasculature of patients with DR. 13 Although different VEGF-binding receptors have been identified, its effects on cellular proliferation, migration, and permeability are mainly mediated through VEGF receptor 2. 4,5 VEGF is also an activator of several isoforms of serine/threonine-specific protein kinase C (PKC) in endothelial cells (ECs), and it seems possible that pathogenic processes depend on this pathway. 69 The PKC family includes several members that differ in their structure, required cofactors, and substrate specificity. They can be divided into three subgroups: conventional (α, βI/II, and γ), novel (δ, ε, and θ), and atypical (ζ and ι/λ) PKC. Activation of PKC by hyperglycemia, especially of isoform β, has been implicated in the processes deregulated in DR. 6,8,10 Hyperglycemia also results in upregulated expression of VEGF and components of the extracellular matrix that contribute to elevated permeability of RECs, probably leading to macular edema in DR. 6,1113 Several in vitro studies showed that altered permeability of ECs induced by VEGF165 was associated with changed cellular localization, expression, and/or modification of tight junction (TJ) proteins. 1117 These complexes of membrane proteins like claudins or occludin and membrane-associated proteins (e.g., ZO-1) are crucial determinants of paracellular signaling of ECs. 18,19 Site-specific phosphorylation of occludin induced by short-term treatment with VEGF165 of bovine retinal endothelial cells (BRECs) may interfere with TJ assembly. 14,16,20 However, in these previous investigations, the long-term effects of VEGF that should be more relevant in the pathologic situation in vivo were neglected. We showed that long-term treatment of immortalized (i)BRECs with VEGF165 indeed resulted in delocalization of the TJ proteins occludin and claudin-1 from the plasma membrane. This result was not associated with a change in expression or a modification of occludin, but expression of claudin-1 was markedly reduced. 17 Claudin-1 is expressed by microvascular endothelial cells of the brain and retina, although at a lower level than other claudins. 17,21,22 It is an essential part of TJ in the mammalian epidermal barrier and claudin-1-deficient mice die soon after birth. 23 In human Caco-2 cells, claudin-1 is phosphorylated at serine/threonine residues, and PKCθ is involved in the regulation of its expression and cellular localization. 24 Based on these previous observations, we studied whether PKC is involved in long-term effects of VEGF165 on claudin-1 and permeability of iBRECs, and whether the presence of claudin-1 is important in proper barrier function in these cells. 
Materials and Methods
Reagents, Antibodies, and Media
Recombinant human growth factor rhVEGF165 (SF21-expressed) was from R&D Systems (Wiesbaden, Germany). The EDTA-free protease inhibitor cocktail (Complete) was purchased from Roche Applied Science (Mannheim, Germany). Brij97, sodium deoxycholate and the phosphatase inhibitor cocktail 2 were from Sigma-Aldrich (Deisenhofen, Germany). The PKC inhibitors Gö6976, Gö6983, GF109203X (also known as BIM I or Gö6850), rottlerin (mallotoxin), and Ro 31-6045 (negative control) were obtained from Merck-Calbiochem (Darmstadt, Germany), as were the inhibitor of PKCβ (IAPD; 3-(1-(3-imidazol-1-ylpropyl)-1H-indol-3-yl)-4-anilino-1H-pyrrole-2,5-dione), the PKC activator phorbol-12-myristate-13-acetate (PMA), and the corresponding negative control substance 4α-phorbol-12,13-didecanoate (4aPDD). Ranibizumab (Lucentis; Genentech, South San Francisco, CA), the Fab fragment of a humanized VEGF-binding antibody, was purchased from Novartis Pharma GmbH (Nuremberg, Germany). 25 Rabbit polyclonal antibodies binding to human claudin-1 (JAY.8), claudin-3 (Z23.JM), claudin-5 (Z43.JK), occludin (Z-T22), and ZO-1 (Z-R1) and with AlexaFluor 594-conjugated detection antibodies (F(ab′)2) were from Invitrogen (Karlsruhe, Germany). Horseradish peroxidase–conjugated anti-rabbit secondary antibodies were included in a detection kit (Opti4CN) from Bio-Rad (Munich, Germany). Mg2+-/Ca2+-free PBS (PBSd) and PBS were obtained from Invitrogen. Complete microvascular endothelial cell growth medium (ECGM; Promocell, Heidelberg, Germany) was supplemented with co-delivered, premixed additives, resulting in final concentrations of 0.4% ECGS/H, 5% fetal calf serum, 10 ng/mL epidermal growth factor, and 103 nM hydrocortisone. Serum-reduced basal medium (EBGM; Promocell) was supplemented with 0.4% ECGS/H, 0.25% fetal calf serum, and 103 nM hydrocortisone. 
Cell Cultivation
Telomerase-immortalized microvascular endothelial cells from bovine retina (iBRECs) were cultivated on fibronectin-coated surfaces in cell culture flasks (BD Biosciences, Heidelberg, Germany) in ECGM with 2.8 μM hydrocortisone, as previously described. 26 Cultures were washed with PBSd to support enzymatic cell detachment for subcultivation. In all experiments, the concentration of hydrocortisone in the medium was 103 nM. 
Treatment of iBRECs with VEGF165 or PMA and Inhibitors of PKC
Before the experiment with confluent iBRECs, ECGM was replaced by EBGM for 24 hours. The cells were then either pretreated with inhibitors of PKC in various concentrations or 100 μg/mL ranibizumab in EBGM—supplemented with 1 μg/mL fibronectin to improve adhesion of cells, especially to glass surfaces—for 2 hours before 100 ng/mL VEGF165 was added for an additional 20 to 48 hours. Alternatively, the cells were incubated with 100 ng/mL VEGF165 for 2 days and PKC inhibitors or 100 μg/mL ranibizumab was then added to the medium (still containing VEGF165) for an additional 24 hours. 
The confluent iBRECs were pretreated with inhibitors of PKC or with ranibizumab for 2 hours, as described earlier, before 100 nM PMA (or 100 nM 4aPDD in control experiments) was added for an additional 20 hours. 
Immunofluorescence Staining
iBRECs at a density of 2 × 104 cells/cm2 were allowed to adhere to two-chamber slides (BD Biosciences or Nunc, Wiesbaden, Germany) coated with fibronectin in ECGM. After having formed a confluent monolayer, the cells were treated with the effectors as just described. The cells were fixed in methanol/acetone at −20°C, and the antigens were detected by immunofluorescence staining. 17,27 Primary antibodies against claudin-1 (5 μg/mL), claudin-5, claudin-3, ZO-1 (all 2.5 μg/mL), and occludin (20 μg/mL) were diluted in 1% fetal calf serum/PBS. Before staining of iBRECs grown on membrane inserts of multiwell plates, the cells were fixed in methanol at −20°C for 15 minutes. 
Preparation of Cell Lysates for Western Blot Analyses
To prepare extracts, 3 × 106 cells were detached by scraping, collected by centrifugation, and resuspended in 150 μL cold lysis buffer: 40 mM TrisCl (pH 7.4), 150 mM NaCl, 5 mM MgCl2, 0.5% sodium deoxycholate, and 1% Brij 97, supplemented with EDTA-free protease inhibitor and phosphatase inhibitor cocktails. After incubation on ice for 1 hour with gentle agitation, the still insoluble components were removed by centrifugation (10,000g, 4°C) for 45 minutes. For Western blot analyses of TJ proteins in the supernatant, the proteins (10 μg per lane) were separated by electrophoresis in a 4% to 15% SDS-polyacrylamide gradient gel (Bio-Rad) under reducing conditions and transferred by electroblot in transfer buffer (192 mM glycin, 25 mM Tris; pH 8.4) to a polyvinylidene difluoride membrane (Bio-Rad). After the membrane was blocked with 0.1% blocking reagent (Roche Applied Science) in PBS at 4°C overnight, it was incubated with the appropriate antibodies diluted in blocking solution. Anti-mouse or anti-rabbit IgG horseradish peroxidase conjugates (1:7500 in PBS, 0.1% Tween-20) and the substrate kit (Opti4CN; Bio-Rad) were used according to the supplier's protocol to detect antigen-bound primary antibodies. 17,26 To verify equal loading of protein, we assessed expression of actin in parallel by using a mouse monoclonal antibody (AC-40; Abcam, Cambridge, UK). 
Electrical Resistance of Cell Layers
Paracellular permeability of iBRECs was assessed with transendothelial electrical resistance (TER) measurements. 28 : Polyethylenterephthalate membrane inserts precoated with fibronectin (0.3 cm2; pore size: 0.4 μm; BioCoat; BD Biosciences) were placed in a 24-well plate and incubated with medium for 1 hour at room temperature. On each of these, 5 × 103 cells were cultivated in fresh medium to form a confluent layer. VEGF165 and/or inhibitors were then added as described earlier, with the exception that fibronectin was not added to the medium of the abluminal chamber. TER measurements were made with a resistance meter (Millicell ERS; Millipore, Schwalbach, Germany) just before and 0.5 to 4 hours after addition of inhibitors or VEGF165, and every 24 hours thereafter. Resistances (ΔTER) of iBREC layers under different conditions were calculated as the mean of five replicates from which the mean resistance of control inserts (without cells, four replicates) was subtracted. To compare independent experiments, normalized ΔTERs were calculated in relation to the ΔTER measured in low-serum medium just before the addition of effectors. 
Statistical Analyses
All experiments were repeated at least three times, and in each experiment, data were generated from multiple replicates. The Mann-Whitney U test was used to analyze experimental data, and P < 0.05 was considered indicative of significant differences. 
Results
The effects of PKC inhibitors were studied under experimental conditions, considering crucial previous observations: iBRECs were cultivated in 103 nM hydrocortisone to ensure high amounts of the TJ proteins occludin and claudin-1 in the plasma membrane. 17 The concentration of VEGF165 (∼5 nM) was in a range similar to that measured in the vitreous fluid of patients with DR and the concentration of the VEGF inhibitor ranibizumab (∼2 μM) reflected therapeutically achievable values. 13,29 Highly efficient inhibitors of different PKC isoforms were used at the lowest still-sufficient concentrations. For the anti-claudin-1 antibody used, we confirmed that it was not cross-reactive with other bovine TJ proteins (data not shown). 
Effect of PKC Inhibitors on Loss of Claudin-1 Induced by Treatment with the PKC Activator PMA
To narrow down the PKC isoforms that are involved in the regulation of claudin-1 expression or modification, we studied the effects of PKC inhibitors of various specificities (Table 1) in iBRECs. The cells were incubated with PKC inhibitors or the VEGF inhibitor ranibizumab for 24 hours before Western blot analyses of the claudins were performed. Whereas most PKC inhibitors did not significantly influence the expression or cellular localization of claudin-1, rottlerin (inhibitor of PKCδ and -θ at 1–5 μM) abolished claudin-1 expression in iBRECs (Table 1; Fig. 1A). This effect is most likely not due to inhibition of PKCδ, because it was not observed in experiments with other inhibitors of PKCδ (GF109203X and Gö6983). Treatment with rottlerin also led to changes in iBREC morphology (i.e., development of spikelike extensions). In contrast to untreated iBRECs, expression of TJ proteins in the plasma membrane was seen only at cell–cell contacts. Expression of claudin-5 was not significantly affected by the PKC inhibitors tested (Fig. 1A). 
Table 1.
 
Inhibition of PMA-Induced Lower Expression of Claudin-1 by Gö6983, GF109203X, and Rottlerin
Table 1.
 
Inhibition of PMA-Induced Lower Expression of Claudin-1 by Gö6983, GF109203X, and Rottlerin
Inhibitor Affected PKC Isoforms at Concentration Used* Effect on Basal Expression of Claudin-1? Prevention of PMA-Induced Lower Expression of Claudin-1?
α β βI βII γ δ ε ζ θ
GF109203X X X X X X No Yes
Gö6983 X X X X X No Yes
Gö6976 X X No No
IAPD X X No No
Rottlerin X X ↓↓↓ Yes
Ro 31-6045 No No
Ranibizumab† No No
Figure 1.
 
Effects of activation or inhibition of PKC on expression of claudin-1. (A) Confluent iBRECs were incubated in low-serum medium with inhibitors of PKC or ranibizumab for 24 hours before TJ proteins (and actin as control) were analyzed by Western blot. Only treatment with rottlerin, reported to inhibit PKC-δ and -θ, resulted in almost complete loss of claudin-1. Claudin-5 expression was not significantly affected by treatment with any of the inhibitors tested. (B) Confluent iBRECs were incubated with 100 nM PMA, and Western blot analyses of claudin-1 expression were performed at different time points. Claudin-1 expression was reduced by PMA (but not by the structurally related control compound 4aPDD) and completely vanished within 18 hours.
Figure 1.
 
Effects of activation or inhibition of PKC on expression of claudin-1. (A) Confluent iBRECs were incubated in low-serum medium with inhibitors of PKC or ranibizumab for 24 hours before TJ proteins (and actin as control) were analyzed by Western blot. Only treatment with rottlerin, reported to inhibit PKC-δ and -θ, resulted in almost complete loss of claudin-1. Claudin-5 expression was not significantly affected by treatment with any of the inhibitors tested. (B) Confluent iBRECs were incubated with 100 nM PMA, and Western blot analyses of claudin-1 expression were performed at different time points. Claudin-1 expression was reduced by PMA (but not by the structurally related control compound 4aPDD) and completely vanished within 18 hours.
Activation of PKC was achieved by treatment of iBRECs with 100 nM PMA for up to 24 hours. After 6 hours, expression of claudin-1 analyzed by Western blot had declined and was undetectable after 24 hours (Fig. 1B). Pretreatment of iBRECs for 2 hours with PKC inhibitors GF109203X (specific for the α, β, γ, δ and ε isoforms), Gö6983 (specific for the α, β, γ, δ and ζ isoforms), or rottlerin prevented loss of claudin-1 induced by PMA in a dose-dependent manner. This result was not observed after similar treatment with the PKC inhibitors IAPD (inhibits mainly β) and Gö6976 (specific for α and β), or with the control compound Ro 31-6045 (Table 1, Fig. 2). The VEGF-binding Fab fragment ranibizumab also failed to inhibit the PMA-induced loss of claudin-1. In contrast to claudin-1, expression of claudin-5 was not affected by PMA (Fig. 2). From these results and specificities of the inhibitors used, it can be concluded that PKCδ may play a role in regulation of claudin-1 but not PKCα and -β. 
Figure 2.
 
PKCδ is involved in signaling leading to loss of claudin-1 after PMA stimulation. Confluent iBRECs were incubated in low-serum medium with inhibitors of PKC or ranibizumab for 2 hours before PMA (100 nM) was added for an additional 20 hours. TJ proteins (and actin as control) were then assessed by Western blot analyses. Only inhibitors of PKCδ (GF109203X, Gö6983, or rottlerin) prevented loss of claudin-1 but not inhibitors of PKCα, -β, and -γ (IAPD; Gö6976) or the control compound Ro 31–6045. Claudin-5 expression was not significantly affected by PMA or any of the inhibitors tested.
Figure 2.
 
PKCδ is involved in signaling leading to loss of claudin-1 after PMA stimulation. Confluent iBRECs were incubated in low-serum medium with inhibitors of PKC or ranibizumab for 2 hours before PMA (100 nM) was added for an additional 20 hours. TJ proteins (and actin as control) were then assessed by Western blot analyses. Only inhibitors of PKCδ (GF109203X, Gö6983, or rottlerin) prevented loss of claudin-1 but not inhibitors of PKCα, -β, and -γ (IAPD; Gö6976) or the control compound Ro 31–6045. Claudin-5 expression was not significantly affected by PMA or any of the inhibitors tested.
Effect of PCK Inhibitors on Loss of Claudin-1 Induced by Long-Term Treatment with VEGF165
Based on our observation that long-term treatment of iBRECs with VEGF165 resulted in downregulation of claudin-1 (which completely disappeared from the plasma membrane, see Fig. 5) and reports on VEGF-induced activation of PKC isoforms in endothelial cells, we investigated whether long-term effects of VEGF165 on claudin-1 are mediated through PKC isoforms. 7,9,17 Loss of claudin-1 was induced by treatment of iBRECs with VEGF165 for 2 days before PKC inhibitors (or ranibizumab) were added for an additional 20 hours (Table 2). In accordance with our previous investigations, reduced expression of claudin-1, analyzed by Western blot analysis in cell extracts, was completely reversed after the addition of ranibizumab. Of all tested PKC inhibitors, only Gö6983 slightly reincreased claudin-1 expression (Fig. 3A; Table 2)—that is, in these experiments inhibition of PKC was not sufficient to reverse the VEGF165-induced loss of claudin-1. Similar results were obtained in a series of experiments in which iBRECs were pretreated with the different PKC inhibitors for 2 hours before they were incubated with VEGF165 for 1 to 2 days. Only ranibizumab prevented complete loss of claudin-1, and of the inhibitors only Gö6983 showed a very weak effect (Fig. 3B). These results confirmed that PKC isoforms are not crucially involved. 
Table 2.
 
Influence of PKC Inhibition on VEGF-Induced Effects
Table 2.
 
Influence of PKC Inhibition on VEGF-Induced Effects
Inhibitor Affected PKC Isoforms at Concentration Used* Prevention/Reversal of VEGF-Induced Loss of Claudin-1? Prevention/Reversal of VEGF-Induced Delocalization of Claudin-1? Prevention/Reversal of VEGF-Induced Elevation of Permeability ?
α β βI βII γ δ ε ζ θ
GF109203X X X X X X No No No
Gö6983 X X X X X Partly Partly No
Gö6976 X X No No ND
IAPD X X No No ND
Rottlerin X X No No ND
Ro31-6045 No No No
Ranibizumab† Yes Yes Yes
Figure 3.
 
PKC is not involved in VEGF-induced signaling leading to loss of claudin-1. Confluent iBRECs were incubated in low-serum medium with (A) VEGF165 for 2 days before inhibitors of PKC (or ranibizumab) were added for an additional 20 hours or (B) inhibitors of PKC (or ranibizumab) for 2 hours before VEGF165 was added for an additional 20 hours. TJ proteins (and actin as control) were then assessed by Western blot analyses. The VEGF-binding Fab fragment ranibizumab reversed (A) or prevented (B) loss of claudin-1, whereas the inhibitors of PKCα, -β, -γ, -δ, -ε, and θ (GF109203X, rottlerin, IAPD, and Gö6976) or the control compound Ro 31-6045 did not have any effect. Only Gö6983 weakly affected VEGF-induced loss of claudin-1. Claudin-5 expression was not significantly altered by VEGF165 or any of the inhibitors tested.
Figure 3.
 
PKC is not involved in VEGF-induced signaling leading to loss of claudin-1. Confluent iBRECs were incubated in low-serum medium with (A) VEGF165 for 2 days before inhibitors of PKC (or ranibizumab) were added for an additional 20 hours or (B) inhibitors of PKC (or ranibizumab) for 2 hours before VEGF165 was added for an additional 20 hours. TJ proteins (and actin as control) were then assessed by Western blot analyses. The VEGF-binding Fab fragment ranibizumab reversed (A) or prevented (B) loss of claudin-1, whereas the inhibitors of PKCα, -β, -γ, -δ, -ε, and θ (GF109203X, rottlerin, IAPD, and Gö6976) or the control compound Ro 31-6045 did not have any effect. Only Gö6983 weakly affected VEGF-induced loss of claudin-1. Claudin-5 expression was not significantly altered by VEGF165 or any of the inhibitors tested.
Effect of PKC Inhibitors GF109203X and Gö6983 on Increased Permeability of iBRECs Induced by Long-Term Treatment with VEGF165
To study the long-term effects of VEGF165 and PKC inhibitors on cell permeability, we performed TER measurements of iBREC layers. The inhibitors GF109203X and Gö6983 were chosen because their overlapping spectra of specificity covered most PKC isoforms. Confluent monolayers of iBRECs seeded on porous membrane filters were pretreated for 2 hours with GF109203X, Gö6983, or ranibizumab. VEGF165 was then added for 2 days, and changes in the permeability of the cell layers were measured (Fig. 4; Table 2 for summary). Inhibitors alone did not significantly influence the permeability of iBRECs (Fig. 4, t = 2 hours). Within 20 hours after addition of VEGF165, permeability of iBRECs was significantly elevated (Fig. 4, t = 22 hours). Preincubation with ranibizumab completely prevented this VEGF165-induced effect, whereas permeability was not affected by GF109203X or Gö6983 (Fig. 4, t = 22 hours and t = 46 hours). To detect potential changes in the cellular localization of TJ proteins, these were visualized after TER measurements by immunofluorescence staining (Fig. 5). Long-term treatment of iBRECs with VEGF165 led to the delocalization of claudin-1 and occludin from the plasma membrane, whereas other TJ proteins were not affected (Fig. 5, second column from the left). This effect on claudin-1 and occludin was completely reversed or prevented by addition of ranibizumab (Fig. 5, right column). Of interest, pretreatment of iBRECs with GF109203X or Gö6983 did not prevent VEGF-induced delocalization of claudin-1, but did prevent that of occludin (Fig. 5). Cellular localization of the TJ proteins claudin-3 and ZO-1 was not significantly altered (data not shown). 
Figure 4.
 
Inhibition of PKC was not sufficient to prevent VEGF-induced elevation of the permeability of iBREC layers. Confluent iBRECs, cultivated in low-serum medium on porous membrane inserts, were incubated with (A) 1 μM Gö6983 or (B) 1 μM GF109203X or 100 μg/mL ranibizumab for 2 hours, before 100 ng/mL VEGF165 was added for an additional 2 days. TER was then measured at the indicated time points. The shown TERs were normalized in relation to the TER in low-serum medium at t = 0 hours, which was set at 1. Only the VEGF-binding Fab fragment ranibizumab prevented VEGF-induced elevation of iBREC permeability, whereas both PKC inhibitors had no significant effect. *P < 0.02 compared with low-serum medium; **P < 0.02 compared with medium with VEGF165.
Figure 4.
 
Inhibition of PKC was not sufficient to prevent VEGF-induced elevation of the permeability of iBREC layers. Confluent iBRECs, cultivated in low-serum medium on porous membrane inserts, were incubated with (A) 1 μM Gö6983 or (B) 1 μM GF109203X or 100 μg/mL ranibizumab for 2 hours, before 100 ng/mL VEGF165 was added for an additional 2 days. TER was then measured at the indicated time points. The shown TERs were normalized in relation to the TER in low-serum medium at t = 0 hours, which was set at 1. Only the VEGF-binding Fab fragment ranibizumab prevented VEGF-induced elevation of iBREC permeability, whereas both PKC inhibitors had no significant effect. *P < 0.02 compared with low-serum medium; **P < 0.02 compared with medium with VEGF165.
Figure 5.
 
VEGF-induced delocalization of claudin-1 from the plasma membrane was not prevented by inhibition of PKC. Confluent iBRECs, cultivated in low-serum medium on porous membrane inserts, were incubated with inhibitors of PKC or ranibizumab for 2 hours before VEGF165 was added for an additional 2 days. TJ proteins were then visualized by immunofluorescence staining. Permeabilities of the differently treated iBRECs is provided as additional information (compare with Fig. 4) to show the correlation with expression of TJ proteins. Only the VEGF-inhibitor ranibizumab prevented delocalization of claudin-1, whereas both inhibitors of PKC did not have any significant effect. In contrast, VEGF165-induced loss of plasma membrane-localized occludin was prevented by ranibizumab or GF109203X and also partly by Gö6983. Claudin-5 localization was not significantly affected by VEGF165 or any of the inhibitors tested.
Figure 5.
 
VEGF-induced delocalization of claudin-1 from the plasma membrane was not prevented by inhibition of PKC. Confluent iBRECs, cultivated in low-serum medium on porous membrane inserts, were incubated with inhibitors of PKC or ranibizumab for 2 hours before VEGF165 was added for an additional 2 days. TJ proteins were then visualized by immunofluorescence staining. Permeabilities of the differently treated iBRECs is provided as additional information (compare with Fig. 4) to show the correlation with expression of TJ proteins. Only the VEGF-inhibitor ranibizumab prevented delocalization of claudin-1, whereas both inhibitors of PKC did not have any significant effect. In contrast, VEGF165-induced loss of plasma membrane-localized occludin was prevented by ranibizumab or GF109203X and also partly by Gö6983. Claudin-5 localization was not significantly affected by VEGF165 or any of the inhibitors tested.
Similar experiments in which iBRECs were treated with VEGF165 for 2 days before ranibizumab, GF109203X, or Gö6983 was added for another 24 hours confirmed that VEGF165-induced changes in permeability could be prevented by treatment with ranibizumab but not with any of the PKC inhibitors used. 
Discussion
Diabetic retinopathy is a major microvascular complication of diabetes. Activation of PKC, especially its β-isoform, by hyperglycemia is considered to be critically involved in some processes deregulated in this disease. 6,8,10 But despite promising results in in vitro studies and animal models, inhibitors affecting PKCβ were not particularly successful in the treatment of diabetic macular edema, perhaps because of the limited relevance of the mostly used rodent models of diabetes in which nonconserved putative phosphorylation sites of claudins could have played a role. 30,31 In addition, highly relevant long-term effects of involved growth factors like VEGF were neglected in most of the in vitro studies. It is revealing that a significant change in protein expression or modification of occludin was not found after long-term exposure of BRECs to VEGF165, although short-term exposure led to subtly changed phosphorylation at Ser490. 16,17,20 In contrast, studying long-term exposure to VEGF165 clearly confirmed the role of claudin-1, which almost completely disappeared within 24 hours, an effect that could be completely reversed by addition of the VEGF-binding Fab fragment ranibizumab. 17 In this study, we investigated whether this VEGF165-induced loss of claudin-1 is regulated by PKC and indeed affects the barrier function of iBRECs. 
Although the iBRECs used in this study were of bovine origin, these cells were shown to behave like human retinal ECs and therefore represent a relevant model system. Compared with primary cells, it is a distinct advantage of iBRECs that these are free of contaminating cells of other types, allowing more accurate and detailed studies. 17,26,27 Immortalizing primary cells by ectopic expression of human telomerase reverse transcriptase (TRT) is usually not associated with significant changes in important cellular processes, and the amount of human TRT in iBRECs is similar to that of the bovine homologue expressed in primary BRECs. iBRECs express the TJ proteins claudin-1, claudin-3, claudin-5, ZO-1, and occludin, as well as VEGF receptor 2 and PKCα, -β, and -θ, as shown by immunohistochemical staining of the bovine retina. 32 Furthermore, of the most relevant TJ protein claudin-1, most putative phosphorylation sites are conserved between the human and bovine homologues. 
It is interesting that both inhibition and activation of PKC led to lower expression of claudin-1 and its delocalization from the iBREC plasma membrane. This finding again shows that substantial additional efforts are needed to replace the current simplistic view by a more detailed knowledge of the interactions of numerous signaling molecules, including PKC isoforms, in the complex and intertwined pathways involved in the regulation of cellular functions. Effects of PKC inhibitors suggest that PKCθ and/or -δ is the PKC isoform involved in this modulation of claudin-1. This conclusion is in accordance with similar regulation of claudin-1 by PKCθ in human Caco-2 cells in which direct phosphorylation of claudin-1 was observed. 24 The effects of PKC activation on claudin-1, however, seem to depend on the type of cells analyzed. In human melanoma cells without any functional TJs, the low basal amount of claudin-1 located in the cytoplasm was increased after addition of PMA which correlated with an enhanced motility of the cells. 33  
Stimulation of PKC with PMA in macrovascular ECs led to synthesis of more VEGF mRNA. 34 Whether VEGF expression can be directly induced or enhanced by activated PKC in retinal microvascular ECs, as in retinal pericytes, remains to be shown 10,35 ; but the observation that PMA-induced effects on claudin-1 expression in iBRECs were not prevented by the VEGF-inhibitor ranibizumab supported the assumption that in these cells, production or secretion of significant amounts of VEGF were not induced. Indeed, VEGF could not be detected in cell lysates or supernatants of PMA-stimulated iBRECs (data not shown). 
To evaluate how permeability of iBRECs was affected by VEGF165, TER measurements of a confluent monolayer were performed. Although confluent cultured ECs are extremely flattened compared with their cuboidal form in vivo, this assay is considered an accurate model for assessing chronic vascular hyperpermeability associated with pathologic neoangiogenesis in malignant diseases and DR. 36 In iBRECs, the presence of plasma membrane-localized claudin-1 strongly correlated with barrier function (Figs. 4, 5). This finding is in accordance with the results obtained with canine epithelial MDKC and human CaCo-2 cells. 24,33,37 Remarkably, in contrast to claudin-1, the presence of the TJ protein occludin seemed not to be necessary for proper barrier function in iBRECs. In addition, localization of other TJ proteins claudin-3, claudin-5, or ZO-1 in the plasma membrane of iBRECs could also not compensate for the loss of claudin-1. In view of the hypothesis that PKC is crucially involved in the regulation of VEGF165-induced changes of permeability in RECs, this study now surprisingly showed that PKC inhibitors covering all relevant isoforms were not sufficient to prevent or reverse this effect in long-term experiments. In previous studies, the slightly elevated permeability of BRECs induced by treatment with VEGF165 for 1 hour could be prevented by treatment with the PKC inhibitor GF109203X. 16 In iBRECs permeability was slightly but not significantly elevated after such short treatment with VEGF165, and a significantly higher permeability was observed 20 hours after addition of VEGF165. However, this increase could not be effected with PKC inhibitors. In addition, none of the PKC inhibitors tested completely reversed or prevented the effect of VEGF165 on claudin-1 expression or localization under these conditions. The results of our investigation therefore strongly suggest that in retinal endothelial cells, PKC is most likely not involved in VEGF165-induced signaling leading to modulation of claudin-1 expression and increased permeability, at least during prolonged exposure to VEGF165, which resembles the in vivo situation in patients with DR. This result is also in accordance with established mechanisms of tumor angiogenesis in which activation of PKC by VEGF165 is assumed not to result in changed permeability of new or preexisting vasculature. 5 However, it seems possible that reported short-term effects of VEGF165 on permeability or TJ were transmitted through PKC, whereas cellular responses to prolonged treatment with VEGF165 were triggered by other signaling pathways. Cellular localization of PKCα in macrovascular ECs was also found to depend on the duration of VEGF treatment: After 3 minutes, it was localized at the plasma membrane but after 30 minutes, again in the cytoplasm. 9 Accordingly, our preliminary results of immunofluorescence staining of iBRECs treated with VEGF165 for 2 days indicated that PKCα and -β were indeed localized exclusively in the cytoplasm, where a direct interaction with TJ proteins was not possible. 
In summary, our in vitro model of immortalized retinal endothelial cells provided evidence for the assumption that PKC is most likely not involved in VEGF165-induced signaling leading to increased cellular permeability and expression of the TJ protein claudin-1. This result is in contrast to published results concerning only short-term effects of VEGF165. However, the prolonged treatment with VEGF165 in our experiments might reflect the pathologic situation in DR more closely. Our results also help to understand why the extensively studied PKCβ inhibitors were not beneficial to patients with diabetic macular edema. 
Footnotes
 Supported by Gertrud Kusen Foundation, Hamburg, Germany.
Footnotes
 Disclosure: H.L. Deissler, None; H. Deissler, None; G.E. Lang, None
The authors thank Anita Ruepp and Susanne Baur for expert technical assistance. 
References
Aiello LP Avery RL Arrigg PG . Vascular endothelial growth factor in ocular fluid of patients with diabetic retinopathy and other retinal disorders. N Engl J Med. 1994; 331: 1480–1487. [CrossRef] [PubMed]
Boulton M Foreman D Williams G McLeod D . VEGF localization in diabetic retinopathy. Br J Ophthalmol. 1998; 82: 561–568. [CrossRef] [PubMed]
Watanabe D Suzuma K Suzuma I . Vitreous levels of angiopoietin 2 and vascular endothelial growth factor in patients with proliferative diabetic retinopathy. Am J Ophthalmol. 2005; 139: 476–481. [CrossRef] [PubMed]
Ferrara N . Vascular endothelial growth factor: basic science and clinical progress. Endocr Rev. 2004; 25: 581–611. [CrossRef] [PubMed]
Kowanetz M Ferrara N . Vascular endothelial growth factor signaling pathways: therapeutic perspective. Clin Cancer Res. 2006; 12: 5018–5022. [CrossRef] [PubMed]
Aiello LP . The potential role of PKCβ in diabetic retinopathy and macular edema. Surv Ophthalmol. 2002; 47( suppl 2): S263–S269. [CrossRef] [PubMed]
Xia P Aiello LP Ishii H . Characterization of vascular endothelial growth factor's effect on the activation of protein kinase C, its isoforms, and endothelial cell growth. J Clin Invest. 1996; 98: 2018–2026. [CrossRef] [PubMed]
Evicmen ND King GL . The role of protein kinase C activation and the vascular complications of diabetes. Pharmocol Res. 2007; 55: 498–510. [CrossRef]
Rask-Madsen C King GL . Differential regulation of VEGF signaling by PKCα and PKCε in endothelial cells. Arterioscler Thromb Vasc Biol. 2008; 28: 919–924. [CrossRef] [PubMed]
Suzuma K Takahara N Suzuma I . Characterization of protein kinase C β isoform's action on retinoblastoma protein phosphorylation, vascular endothelial growth factor-induced cell proliferation, and retinal neovascularization. Proc Natl Acad Sci. 2002; 99: 721–726. [CrossRef] [PubMed]
Caldwell RB Bartoli M Behzadian MA . Vascular endothelial growth factor and diabetic retinopathy: pathophysiological mechanisms and treatment perspectives. Diabetes Metab Res Rev. 2003; 19: 442–455. [CrossRef] [PubMed]
Nguyen QD Tatlipinar S Shah SM . Vascular endothelial growth factor is a critical stimulus for diabetic macular edema. Am J Ophthalmol. 2006; 142: 961–969. [CrossRef] [PubMed]
Oshitari T Polewski P Chadda M . Effect of combined antisense oligonucleotides against high-glucose- and diabetes-induced overexpression of extracellular matrix components and increased permeability. Diabetes. 2006; 55: 86–92. [CrossRef] [PubMed]
Antonetti DA Barber AJ Hollinger LA Wolpert EB Gardner TW . Vascular endothelial growth factor induces rapid phosphorylation of tight junction proteins occludin and zonula occludens 1. J Biol Chem. 1999; 274: 23463–23467. [CrossRef] [PubMed]
Qaum T Xu Q Joussen AM . VEGF-initiated blood–retinal barrier breakdown in early diabetes. Invest Ophthalmol Vis Sci. 2001; 42: 2408–2413. [PubMed]
Harhaj NS Felinski EA Wolpert EB . VEGF activation of protein kinase C stimulates occludin phosphorylation and contributes to endothelial permeability. Invest Ophthalmol Vis Sci. 2006; 47: 5106–5115. [CrossRef] [PubMed]
Deissler H Deissler H Lang S Lang GE . VEGF-induced effects on proliferation, migration and tight junctions are restored by ranibizumab (Lucentis®) in microvascular retinal endothelial cells. Br J Ophthalmol. 2008; 92: 839–843. [CrossRef] [PubMed]
Tsukita S Furuse M Itoh M . Multifunctional strands in tight junctions. Nat Rev Mol Cell Biol. 2001; 2: 285–293. [CrossRef] [PubMed]
Bazzoni G . Endothelial tight junctions: permeable barriers of the vessel wall. Thromb Haemost. 2006; 95: 36–42. [PubMed]
Sundstrom JM Tash BR Murakami T . Identification and analysis of occludin phosphosites: a combined mass spectrometry and bioinformatics approach. J Proteome Res. 2009; 8: 808–817. [CrossRef] [PubMed]
Ohtsuki S Yamaguchi H Katsukura Y Asashima T Terasaki T . mRNA expression levels of tight junction protein genes in mouse brain capillary endothelial cells highly purified by magnetic cell sorting. J Neurochem. 2008; 104: 147–154. [PubMed]
Morcos Y Hosie MJ Bauer HC Chan-Ling T . Immunolocalization of occludin and claudin-1 to tight junctions in intact CNS vessels of mammalian retina. J Neurocytol. 2001; 30: 107–123. [CrossRef] [PubMed]
Furuse M Hata M Furuse K . Claudin-based tight junctions are crucial for the mammalian epidermal barrier: a lesson from claudin-1-deficient mice. J Cell Biol. 2002; 156: 1099–1111. [CrossRef] [PubMed]
Banan A Zhang LJ Shaikh M . θ isoform of protein kinase C alters barrier function in intestinal epithelium through modulation of distinct claudin isotypes: a novel mechanism for regulation of permeability. J Pharmacol Exp Ther. 2005; 313: 962–982. [CrossRef] [PubMed]
Ferrara N Damico L Shams N Lowman H Kim R . Development of ranibizumab, an anti-vascular endothelial growth factor antigen binding fragment, as therapy for neovascular age-related macular degeneration. Retina. 2006; 26: 859–870. [CrossRef] [PubMed]
Deissler H Deissler H Lang GK Lang GE . Generation and characterization of iBREC: novel hTERT-immortalized bovine retinal endothelial cells. Int J Mol Med. 2005; 16: 65–70. [PubMed]
Deissler H Kuhn EM Lang GE Deissler H . Tetraspanin CD9 is involved in the migration of retinal microvascular endothelial cells. Int J Mol Med. 2007; 20: 643–652. [PubMed]
Tretiach M van Driel D Gillies MC . Transendothelial electrical resistance of bovine retinal capillary endothelial cells is influenced by cell growth patterns: an ultrastructural study. Clin Exp Ophthalmol. 2003; 31: 348–353. [CrossRef]
Rosenfeld PJ Brown DM Heier JS . Ranibizumab for neovascular age-related macular degeneration. N Engl J Med. 2006; 355: 1419–1431. [CrossRef] [PubMed]
Ishii H Jirousek MR Koya D . Amelioration of vascular dysfunctions in diabetic rats by an oral PKC-β inhibitor. Science. 1996; 272: 728–731. [CrossRef] [PubMed]
The PKC-DMES Study Group. Effect of ruboxistaurin in patients with diabetic macular edema. Arch Ophthalmol. 2007; 125: 318–324. [CrossRef] [PubMed]
Shin T Kim S Ahn M Kim H . An immunohistochemical study for protein kinase C in the bovine retina. J Vet Med Sci. 2006; 68: 71–74. [CrossRef] [PubMed]
Leotlela PD Wade MS Duray PH . Claudin-1 overexpression in melanoma is regulated by PKC and contributes to melanoma cell motility. Oncogene. 2007; 26: 3846–3856. [CrossRef] [PubMed]
Xu H Czerwinski P Hortmann M Sohn H-Y Förstermann U Li H . Protein kinase C α promotes angiogenic activity of human endothelial cells via induction of vascular endothelial growth factor. Cardiovas Res. 2008; 78: 349–355. [CrossRef]
Amadio M Scapagnini G Lupo G Drago F Govoni S Pascale A . PKCβII/HuR/VEGF: A new molecular cascade in retinal pericytes for the regulation of VEGF gene expression. Pharmacol Res. 2008; 57: 60–66. [CrossRef] [PubMed]
Nagy JA Benjamin L Zeng H Dvorak AM Dvorak HF . Vascular permeability, vascular hyperpermeability and angiogenesis. Angiogenesis. 2008; 11: 109–119. [CrossRef] [PubMed]
Martínez-Estrada OM Cullerés A Soriano FX . The transcription factors Slug and Snail act as repressors of Claudin-1 expression in epithelial cells. Biochem J. 2006; 394: 449–457. [CrossRef] [PubMed]
Figure 1.
 
Effects of activation or inhibition of PKC on expression of claudin-1. (A) Confluent iBRECs were incubated in low-serum medium with inhibitors of PKC or ranibizumab for 24 hours before TJ proteins (and actin as control) were analyzed by Western blot. Only treatment with rottlerin, reported to inhibit PKC-δ and -θ, resulted in almost complete loss of claudin-1. Claudin-5 expression was not significantly affected by treatment with any of the inhibitors tested. (B) Confluent iBRECs were incubated with 100 nM PMA, and Western blot analyses of claudin-1 expression were performed at different time points. Claudin-1 expression was reduced by PMA (but not by the structurally related control compound 4aPDD) and completely vanished within 18 hours.
Figure 1.
 
Effects of activation or inhibition of PKC on expression of claudin-1. (A) Confluent iBRECs were incubated in low-serum medium with inhibitors of PKC or ranibizumab for 24 hours before TJ proteins (and actin as control) were analyzed by Western blot. Only treatment with rottlerin, reported to inhibit PKC-δ and -θ, resulted in almost complete loss of claudin-1. Claudin-5 expression was not significantly affected by treatment with any of the inhibitors tested. (B) Confluent iBRECs were incubated with 100 nM PMA, and Western blot analyses of claudin-1 expression were performed at different time points. Claudin-1 expression was reduced by PMA (but not by the structurally related control compound 4aPDD) and completely vanished within 18 hours.
Figure 2.
 
PKCδ is involved in signaling leading to loss of claudin-1 after PMA stimulation. Confluent iBRECs were incubated in low-serum medium with inhibitors of PKC or ranibizumab for 2 hours before PMA (100 nM) was added for an additional 20 hours. TJ proteins (and actin as control) were then assessed by Western blot analyses. Only inhibitors of PKCδ (GF109203X, Gö6983, or rottlerin) prevented loss of claudin-1 but not inhibitors of PKCα, -β, and -γ (IAPD; Gö6976) or the control compound Ro 31–6045. Claudin-5 expression was not significantly affected by PMA or any of the inhibitors tested.
Figure 2.
 
PKCδ is involved in signaling leading to loss of claudin-1 after PMA stimulation. Confluent iBRECs were incubated in low-serum medium with inhibitors of PKC or ranibizumab for 2 hours before PMA (100 nM) was added for an additional 20 hours. TJ proteins (and actin as control) were then assessed by Western blot analyses. Only inhibitors of PKCδ (GF109203X, Gö6983, or rottlerin) prevented loss of claudin-1 but not inhibitors of PKCα, -β, and -γ (IAPD; Gö6976) or the control compound Ro 31–6045. Claudin-5 expression was not significantly affected by PMA or any of the inhibitors tested.
Figure 3.
 
PKC is not involved in VEGF-induced signaling leading to loss of claudin-1. Confluent iBRECs were incubated in low-serum medium with (A) VEGF165 for 2 days before inhibitors of PKC (or ranibizumab) were added for an additional 20 hours or (B) inhibitors of PKC (or ranibizumab) for 2 hours before VEGF165 was added for an additional 20 hours. TJ proteins (and actin as control) were then assessed by Western blot analyses. The VEGF-binding Fab fragment ranibizumab reversed (A) or prevented (B) loss of claudin-1, whereas the inhibitors of PKCα, -β, -γ, -δ, -ε, and θ (GF109203X, rottlerin, IAPD, and Gö6976) or the control compound Ro 31-6045 did not have any effect. Only Gö6983 weakly affected VEGF-induced loss of claudin-1. Claudin-5 expression was not significantly altered by VEGF165 or any of the inhibitors tested.
Figure 3.
 
PKC is not involved in VEGF-induced signaling leading to loss of claudin-1. Confluent iBRECs were incubated in low-serum medium with (A) VEGF165 for 2 days before inhibitors of PKC (or ranibizumab) were added for an additional 20 hours or (B) inhibitors of PKC (or ranibizumab) for 2 hours before VEGF165 was added for an additional 20 hours. TJ proteins (and actin as control) were then assessed by Western blot analyses. The VEGF-binding Fab fragment ranibizumab reversed (A) or prevented (B) loss of claudin-1, whereas the inhibitors of PKCα, -β, -γ, -δ, -ε, and θ (GF109203X, rottlerin, IAPD, and Gö6976) or the control compound Ro 31-6045 did not have any effect. Only Gö6983 weakly affected VEGF-induced loss of claudin-1. Claudin-5 expression was not significantly altered by VEGF165 or any of the inhibitors tested.
Figure 4.
 
Inhibition of PKC was not sufficient to prevent VEGF-induced elevation of the permeability of iBREC layers. Confluent iBRECs, cultivated in low-serum medium on porous membrane inserts, were incubated with (A) 1 μM Gö6983 or (B) 1 μM GF109203X or 100 μg/mL ranibizumab for 2 hours, before 100 ng/mL VEGF165 was added for an additional 2 days. TER was then measured at the indicated time points. The shown TERs were normalized in relation to the TER in low-serum medium at t = 0 hours, which was set at 1. Only the VEGF-binding Fab fragment ranibizumab prevented VEGF-induced elevation of iBREC permeability, whereas both PKC inhibitors had no significant effect. *P < 0.02 compared with low-serum medium; **P < 0.02 compared with medium with VEGF165.
Figure 4.
 
Inhibition of PKC was not sufficient to prevent VEGF-induced elevation of the permeability of iBREC layers. Confluent iBRECs, cultivated in low-serum medium on porous membrane inserts, were incubated with (A) 1 μM Gö6983 or (B) 1 μM GF109203X or 100 μg/mL ranibizumab for 2 hours, before 100 ng/mL VEGF165 was added for an additional 2 days. TER was then measured at the indicated time points. The shown TERs were normalized in relation to the TER in low-serum medium at t = 0 hours, which was set at 1. Only the VEGF-binding Fab fragment ranibizumab prevented VEGF-induced elevation of iBREC permeability, whereas both PKC inhibitors had no significant effect. *P < 0.02 compared with low-serum medium; **P < 0.02 compared with medium with VEGF165.
Figure 5.
 
VEGF-induced delocalization of claudin-1 from the plasma membrane was not prevented by inhibition of PKC. Confluent iBRECs, cultivated in low-serum medium on porous membrane inserts, were incubated with inhibitors of PKC or ranibizumab for 2 hours before VEGF165 was added for an additional 2 days. TJ proteins were then visualized by immunofluorescence staining. Permeabilities of the differently treated iBRECs is provided as additional information (compare with Fig. 4) to show the correlation with expression of TJ proteins. Only the VEGF-inhibitor ranibizumab prevented delocalization of claudin-1, whereas both inhibitors of PKC did not have any significant effect. In contrast, VEGF165-induced loss of plasma membrane-localized occludin was prevented by ranibizumab or GF109203X and also partly by Gö6983. Claudin-5 localization was not significantly affected by VEGF165 or any of the inhibitors tested.
Figure 5.
 
VEGF-induced delocalization of claudin-1 from the plasma membrane was not prevented by inhibition of PKC. Confluent iBRECs, cultivated in low-serum medium on porous membrane inserts, were incubated with inhibitors of PKC or ranibizumab for 2 hours before VEGF165 was added for an additional 2 days. TJ proteins were then visualized by immunofluorescence staining. Permeabilities of the differently treated iBRECs is provided as additional information (compare with Fig. 4) to show the correlation with expression of TJ proteins. Only the VEGF-inhibitor ranibizumab prevented delocalization of claudin-1, whereas both inhibitors of PKC did not have any significant effect. In contrast, VEGF165-induced loss of plasma membrane-localized occludin was prevented by ranibizumab or GF109203X and also partly by Gö6983. Claudin-5 localization was not significantly affected by VEGF165 or any of the inhibitors tested.
Table 1.
 
Inhibition of PMA-Induced Lower Expression of Claudin-1 by Gö6983, GF109203X, and Rottlerin
Table 1.
 
Inhibition of PMA-Induced Lower Expression of Claudin-1 by Gö6983, GF109203X, and Rottlerin
Inhibitor Affected PKC Isoforms at Concentration Used* Effect on Basal Expression of Claudin-1? Prevention of PMA-Induced Lower Expression of Claudin-1?
α β βI βII γ δ ε ζ θ
GF109203X X X X X X No Yes
Gö6983 X X X X X No Yes
Gö6976 X X No No
IAPD X X No No
Rottlerin X X ↓↓↓ Yes
Ro 31-6045 No No
Ranibizumab† No No
Table 2.
 
Influence of PKC Inhibition on VEGF-Induced Effects
Table 2.
 
Influence of PKC Inhibition on VEGF-Induced Effects
Inhibitor Affected PKC Isoforms at Concentration Used* Prevention/Reversal of VEGF-Induced Loss of Claudin-1? Prevention/Reversal of VEGF-Induced Delocalization of Claudin-1? Prevention/Reversal of VEGF-Induced Elevation of Permeability ?
α β βI βII γ δ ε ζ θ
GF109203X X X X X X No No No
Gö6983 X X X X X Partly Partly No
Gö6976 X X No No ND
IAPD X X No No ND
Rottlerin X X No No ND
Ro31-6045 No No No
Ranibizumab† Yes Yes Yes
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×