SELECTIVE INHIBITION OF VASCULAR CELL ADHESION MOLECULE-1 EXPRESSION BY VERAPAMIL IN HUMAN VASCULAR ENDOTHELIAL CELLS¹

Abstract

Up-regulated expression of adhesion molecules on vascular endothelial cells (ECs) is an initial event in leukocyte adhesion to ECs, which is a crucial step in the process of inflammatory or immune reaction leading to organ transplant rejection. Verapamil and other calcium channel blockers have been reported to improve transplantation outcome and seem to possess immunosuppressive functions. Therefore, we tested the effect of verapamil on adhesion molecule expression of ECs and adherence of leukocytes to ECs. Verapamil was added to replicate human umbilical vein EC cultures, before adding cytokines. Protein and mRNA expression of E-selectin, vascular cell adhesion molecule-1 (VCAM-1), and intercellular adhesion molecule-1 were measured by cell-ELISA and by Northern blot analysis after stimulation of ECs. Adherence of leukocytes to ECs was analyzed after stimulation of ECs by tumor necrosis factor-α, with pretreatment of verapamil. Verapamil selectively inhibited cytokine-induced protein and mRNA levels of VCAM-1, and suppressed cell adherence between tumor necrosis factor-α-stimulated ECs and mononuclear leukocytes or Molt-4 cells. These results suggest that verapamil may suppress immune response partly via inhibition of VCAM-1 expression on ECs and a certain subset of lymphocytes adherent to ECs.

Cell adhesion between leukocytes and vascular endothelial cells (ECs^*) is one of the most important aspects of the inflammatory or immune process, which can lead to organ transplant rejection ⁽¹⁾. The initial step of the adherent interactions involves local generation of mediators and initial activation of ECs adjacent to the inflammatory or immune site. Activation of ECs induces the expression of adhesion molecules, which include E-selectin, vascular cell adhesion molecule-1 (VCAM-1), and intercellular adhesion molecule-1 (ICAM-1) ⁽²⁾. This results in increased leukocyte rolling, along ECs of postcapillary venules under flow. Subsequently, the rolling leukocytes adhere firmly, diapedese between ECs, and then migrate into subendothelial tissue at the inflammatory sites. Increased expression of cell adhesion molecules may play an important role in transplant rejection. This was described for ICAM-1 on ECs, epithelial cells, and infiltrating leukocytes in human kidney allograft ⁽³⁾. Expression of E-selectin and VCAM-1 was associated with a subset of T-cell infiltrate. Anti-ICAM-1 monoclonal antibody (mAb) therapy was shown to be beneficial in patients who received renal allografts that were at high risk for delayed function (either due to prolonged preservation time or a highly sensitized recipient) ^{(4, 5)}. Anti-VCAM-1 mAb in mice can prolong cardiac allograft survival and induce down-regulation of intragraft interleukin-2 (IL-2) receptor expression on T cells and IL-2, IL-4, and interferon-γ ^{(6, 7)}.

Calcium channel blockers have been shown to prevent cyclosporine-induced renal blood flow inhibition and improve cadaveric renal transplantation outcome, in combination with conventional immunosuppressive regimens ^(8-10). Verapamil, which is one of the calcium channel blockers, has been reported to have immunosuppressive functions independent of cyclosporine, such as inhibition of lymphocyte proliferation, IL-2 production, and IL-2 responsiveness ⁽¹¹⁾. Moreover, verapamil also exhibits immunomodulatory effects on the adhesion and migration of lymphocytes ⁽¹²⁾. However, the immunosuppressive effects of calcium channel blockers are not fully understood. In the present study, we tested the effects of verapamil on adhesion molecule expression of ECs and on adherence of leukocytes to ECs, using human umbilical vein ECs.

MATERIALS AND METHODS

Reagents. Recombinant human tumor necrosis factor α (TNFα) was purchased from Research and Diagnostic Systems, Inc. (Minneapolis, MN), and added to culture media at a final concentration of 10 ng/ml as a stimulant for EC activation. Lipopolysaccharide (LPS) (from Escherichia coli serotype 055:B5; Sigma Chemical Co., St. Louis, MO) was used at a final concentration of 1 μg/ml. IL-1β (Becton Dickinson Labware, Bedford, MA) was used at a final concentration of 10 U/ml. IL-4 (Genzyme, Cambridge, MA) was used at a final concentration of 2000 U/ml. Verapamil was obtained from Sigma and added at a final concentration of 10, 50, or 100 μmol/L, 30 min before cytokine stimulation. Murine mAb against human E-selectin (CY1787: IgG3) was provided by Sumitomo Pharmaceutical Ltd. (Osaka, Japan). Murine mAbs against human VCAM-1 (E1/6: IgG1) and human ICAM-1 (LB-2: IgG2b) were purchased from Nippon Becton Dickinson Company Ltd. (Tokyo, Japan). Each of these mAbs were used for cell-ELISA, as a primary antibody at a final concentration of 10 μg/ml. The oligonucleotide primers for polymerase chain reaction (PCR), used to prepare partial cDNA probes for human E-selectin, VCAM-1, and ICAM-1 mRNA detection by Northern hybridization, were selected from published sequences for human E-selectin ⁽¹³⁾, VCAM-1 ⁽¹⁴⁾, and ICAM-1 ⁽¹⁵⁾(Table 1). The 2′,7′-bis-2-carboxyethyl-5 (and -6)-carboxyfluorescein-acetoxymethyl ester (BCECF-AM) was purchased from Molecular Probes, Inc. (Eugene, OR) and used for fluorescence labeling of leukocytes at a final concentration of 10 μM.

EC cultures. Human ECs were enzymatically harvested from umbilical cord using 1 mg/ml type II collagenase (Sigma) ⁽¹⁶⁾. ECs were grown on a 2% gelatin-coated flask in RPMI-1640 (Life Technologies, Grand Island, NY), with 10% heat-inactivated fetal calf serum (FCS) (Mitsubishi, Tokyo, Japan) supplemented with 100 μg/ml heparin, 100 U/ml penicillin, 100 μg/ml streptomycin, 2 mM L-glutamine, and 30 μg/ml EC growth supplement (Becton Dickinson Labware). Confluent EC cultures of passages 3 through 6 were used for studies. The identities of the ECs were verified by their typical cobblestone monolayer structure and acetylated low density lipoprotein uptake.

Leukocytes. Whole blood was obtained from healthy human volunteers and anticoagulated with 3.8% sodium citrate in phosphate-buffered saline (PBS). Mononuclear leukocytes were isolated by density gradient centrifugation over Ficoll-Hypaque (Pharmacia, Uppsala, Sweden), washed twice, and suspended in RPMI-1640. Molt-4 cells (American Type Culture Collection, Rockville, MD) were grown in RPMI with 10% FCS, 100 U/ml penicillin, 100 μg/ml streptomycin, and 2 mM L-glutamine.

Cell-ELISA for adhesion molecule expression. ECs were plated at 5×10⁴ cells/well in 96-well gelatin-coated plates and grown to confluence. The cells were stimulated with TNFα, LPS, or IL-1β, 30 min after verapamil addition. After stimulation for 4 hr, ECs were fixed with 3% paraformaldehyde/8% saccharose in PBS. Nonspecific binding was blocked by 2% bovine serum albumin (BSA) in PBS for 1 hr, and the cells were incubated with primary antibody for 45 min. The cells were washed with 0.1% BSA in PBS three times, and incubated with biotinylated anti-mouse IgG antibody (Amersham, Bucking-hamshire, UK) for 30 min. After washing three times, streptavidin alkaline-phosphatase was added and incubated for 30 min. The phosphatase substrate (Sigma) was added after washing with 0.1% BSA in PBS, and the optical density of each well was determined at 405 nm using a microplate reader (Bio-Rad Laboratories, Richmond, CA).

RNA extraction and analysis of mRNA by Northern blot. After stimulation with TNFα for 3 hr, total cellular RNA was extracted from ECs in replicate 10-cm dishes by a modification of the guanidinium-phenol-chloroform method described by Chomczynski and Sacchi, using ISOGEN (Nippon Gene Ltd., Toyama, Japan). Total cellular RNA (20 μg) was separated by electrophoresis on a 1% agarose gel containing 1.1% formaldehyde in 1×3-(N-morpholino)propanesulfonic acid buffer and transferred to nylon membranes. The membranes were cross-linked by ultraviolet irradiation and hybridized with the human E-selectin, VCAM-1, or ICAM-1 cDNA probe labeled with [α-³²P]dCTP by Random Primer DNA Labeling Kit, Ver. 2.0 (Takara Shuzo Co. Ltd., Shiga, Japan).

Partial human E-selectin, VCAM-1, and ICAM-1 cDNAs were isolated by PCR using cDNAs prepared from TNFα-stimulated human umbilical vein ECs. Total cellular RNA (1 μg) was reverse transcribed with Moloney murine leukemia virus-type reverse transcriptase (Life Technologies, Gaithersburg, MD). E-selectin, VCAM-1, or ICAM-1 cDNA products were amplified with 30 cycles of PCR, performed in a total volume of 50 μl with 2.5 U of Taq DNA polymerase (Life Technologies) and 50 pmol of respective sense and antisense primer with cycle settings of 94°C for 30 sec, 58°C for 30 sec, and 72°C for 90 sec. Each PCR product was subcloned into a pGEM-T vector (Promega, Madison, WI). The identity of the clone was verified as E-selectin, VCAM-1, or ICAM-1 by restriction enzyme analysis and PCR DNA sequencing.

Assay of leukocyte adherence to activated ECs. ECs were plated on 96-well gelatin-coated plates and grown to confluence. After a 4-hr stimulation by TNFα, the EC monolayers were washed once with RPMI and incubated in 100 μl/well RPMI containing 10% FCS, at 37°C for 20 min before the assay. Mononuclear leukocytes or Molt-4 cells were suspended in RPMI and incubated with BCECF-AM at 37°C for 20 min for labeling. Labeled leukocytes were washed with RPMI, suspended at 1×10⁷ cells/ml in RPMI, and incubated for 10 min before the assay. The leukocytes were added to the EC monolayers at 20 μl/well in 96-well plates and incubated at 37°C for 30 min. Nonadherent leukocytes were removed by inverting the plates for 20 min, after filling up each well with RPMI. Adherent cells were solubilized with 0.1% NP-40 in PBS. The fluorescence intensity of each well was measured using a Fluoroscan II (excitation at 490 nm, emission at 530 nm; Flow Labs., McLean, VA).

Each cell-ELISA and leukocyte adherence assay experiment contained four replicates and was repeated three times with similar results. Each data point represents the mean from a representative experiment ±SEM. Statistical analysis was performed by the Mann-Whitney U test. P values less than 0.05 were considered to be significant.

RESULTS

Effects of verapamil on the expression of E-selectin, VCAM-1, and ICAM-1 on human umbilical vein ECs activated by cytokines. Activation of ECs by incubation with 10 ng/ml of TNFα for 4 hr increased the expression of E-selectin, VCAM-1, and ICAM-1 on the surfaces of the ECs (Fig. 1). Pretreatment with 10, 50, or 100 μmol/L of verapamil did not attenuate the increased levels of either E-selectin or ICAM-1 expression on ECs (Fig. 1, A and C), but did significantly reduce VCAM-1 expression in a dose-dependent fashion (P<0.05) (Fig. 1B). Further analysis using other cytokines, such as LPS and IL-1β, also confirmed that verapamil significantly attenuated the increased levels of VCAM-1 expression on activated ECs (P<0.05) (Fig. 2, A and B), but did not affect the increased levels of E-selectin or ICAM-1 expression on ECs (data not shown). Additional experiments was performed using IL-4 stimulation for 24 hr to test the effects of verapamil on selective induction of VCAM-1 expression ⁽¹⁷⁾; these experiments also revealed the inhibitory effect of verapamil on the expression of VCAM-1 (Fig. 3).

Effects of verapamil on the mRNA levels of E-selectin, VCAM-1, and ICAM-1 in human vascular ECs stimulated by TNFα. The human umbilical vein ECs stimulated for 3 hr with TNFα contained a single 3.9-kb transcript, which hybridized with the E-selectin cDNA probe as previously reported ⁽¹³⁾ and which was absent from unstimulated ECs (Fig. 4). The VCAM-1 cDNA hybridized to an mRNA of 3.2 kb and the ICAM-1 cDNA hybridized to a 3.3-kb transcript, as reported in TNFα-stimulated ECs ^{(14, 15)}, but these transcripts were also absent from unstimulated ECs (Fig. 4). Pretreatment of verapamil at a final concentration of 50 μmol/L apparently decreased the TNFα-induced VCAM-1 mRNA level, but did not change the mRNA levels of E-selectin or ICAM-1 (Fig. 4).

Effects of verapamil on leukocyte adherence to TNFα-stimulated ECs. Mononuclear leukocytes, which include lymphocytes and monocytes, expressed very late antigen-4, and thus bind to TNFα-stimulated ECs partly via a VCAM-1-dependent mechanism ^{(18, 19)}. Moreover, cells of a human T-cell line, Molt-4, bind to TNFα-stimulated ECs, mainly via a VCAM-1-dependent mechanism ^{(18, 20)}. Adherence of mononuclear leukocytes or Molt-4 cells to ECs was increased by stimulation of ECs with TNFα for 4 hr. Pretreatment of TNFα-stimulated ECs with verapamil at various doses reduced mononuclear leukocyte adherence to ECs in a dose-dependent manner. At doses of 50 and 100 μmol/L, verapamil significantly decreased cell adherence (P<0.05). Verapamil also significantly decreased binding of Molt-4 cells to ECs that were activated by TNFα at doses of 50 and 100 μmol/L (P<0.05) (Fig. 5).

DISCUSSION

We have shown that verapamil selectively inhibited cytokine-induced protein and mRNA levels of VCAM-1 on ECs, and that it suppressed cell adherence between stimulated ECs and mononuclear leukocytes or Molt-4 cells. These results suggest that verapamil may suppress immune response partly via inhibition of VCAM-1 expression on ECs and a certain subset of lymphocytes' adherence to ECs. Although it is 50% inhibition at best, this inhibitory action against VCAM-1 expression is a novel additional immunosuppressive effect of verapamil. The selective inhibition of VCAM-1 by verapamil may act against immune reactions more effectively in transplantation, when combined with the known immunosuppressive effects of verapamil, such as inhibition of lymphocyte proliferation, IL-2 production, and IL-2 responsiveness ⁽¹¹⁾.

VCAM-1, also called INCAM-110, was first described as mediating melanoma cell adhesion to TNFα-stimulated ECs ⁽²¹⁾. It belongs to the immunoglobulin gene superfamily, which also includes ICAM-1. VCAM-1 binds to α4β1 integrin (very late antigen -4), which is expressed on lymphocytes, monocytes, eosinophils, and basophils, but not on neutrophils. It also interacts with lymphocytes that express α4β7 integrin. The interaction between VCAM-1 and α4 integrins has been shown to mediate the adhesion of lymphocytes and other leukocytes, but not the adhesion of neutrophils to cultured ECs, suggesting that the interaction could mediate a selective recruitment of leukocyte subpopulations in vivo. Localized expression of VCAM-1 in aortic endothelium, which overlies early foam cell lesions, seems to play an important role in mononuclear leukocyte recruitment during atherogenesis ⁽¹⁹⁾. Inflammatory cytokines such as LPS, TNFα, and IL-1 increase the expression of VCAM-1, as well as the expression of ICAM-1 and E-selectin on cultured ECs ⁽¹⁸⁾. The activation by these cytokines seems to occur through a common mechanism involving cytokine-mediated translocation of members of the NF-κB/Rel family from the cytoplasm to the nucleus, where they bind to κB sites in the gene promoters and activate transcription ⁽²²⁾. A study of the VCAM-1 promoter in human umbilical vein ECs has demonstrated that TNFα activates the promoter, through two NF-κB DNA binding sites located at -77 and -63 base pairs of the VCAM-1 gene ⁽²²⁾. However, evidence for a VCAM-1-specific regulatory mechanism was indicated by studies that showed that IL-4 and IL-13 selectively induced VCAM-1 expression, but not E-selectin or ICAM-1 ^{(17, 23)}. The VCAM-1-specific regulatory mechanism is yet to be clarified, but it has been reported that IL-4 specifically stabilizes an mRNA for VCAM-1 in cultured ECs ⁽¹⁷⁾. Furthermore, antioxidants selectively suppress LPS- and IL-1β-induced VCAM-1 protein and mRNA expression via inhibition of the VCAM-1 promoter, through NF-κB-like DNA enhancer elements and associated NF-κB-like DNA binding proteins ⁽²⁰⁾. This selective inhibition of VCAM-1 gene expression by antioxidants suggests that verapamil may also suppress increased VCAM-1 gene expression in stimulated ECs through the inhibition of transcriptional activation by NF-κB, although no gel mobility shift assay was performed in the present study. Our results demonstrating that verapamil did not inhibit either E-selectin or ICAM-1 gene expression may be explained by the results of studies that have indicated that NF-κB-like activation does not seem to be essential in the induction of E-selectin and ICAM-1 gene expression, despite the presence of NF-κB consensus DNA binding sites on their promoters ^{(20, 24, 25)}.

The important role of VCAM-1 expression in organ transplant was shown by the study using the anti-VCAM-1 mAb in murine cardiac transplant ^{(6, 7)}. Our result showing that verapamil suppressed increased expression of VCAM-1 on activated ECs may provide supportive evidence that verapamil improves transplantation outcome in combination with cyclosporine, and that verapamil has its own distinct immunosuppressive effects ^{(8, 11)}. However, Bergese et al. demonstrated that endothelial VCAM-1 expression can be completely blocked without interfering with acute rejection, and concluded that endothelial VCAM-1 expression might not be essential for the allograft rejection process ⁽²⁶⁾. Even though endothelial VCAM-1 expression is not essential for graft rejection, it plays a role in the graft rejection process, because anti-VCAM-1 mAb induced long-term cardiac allograft acceptance ⁽⁶⁾. Clinically, cyclosporine and other immunosuppressants are essential in preventing graft rejection. We suggest that verapamil acts synergetically with cyclosporine to prevent graft rejection, by its known immunosuppressive ability ⁽¹¹⁾ and by the newly discovered selective inhibition of VCAM-1 expression. In addition, verapamil has been reported to suppress intimal hyperplasia after vascular injury, partly via inhibition of smooth muscle cell proliferation in response to platelet-derived growth factor ⁽²⁷⁾, inhibition of lipid, calcium, and matrix accumulation ⁽²⁸⁾, and down-regulation of platelet-derived growth factor A chain mRNA levels ⁽²⁹⁾. Intimal hyperplasia in the vessels of a transplant organ causes graft failure several months after surgery, through the reduction of blood supply to the graft. In this sense, verapamil may also be effective in prolonging the survival period of the transplanted graft, by preventing intimal hyperplasia.

Acknowledgments. We are grateful to Professor J. Patrick Barron (International Medical Communications Center, Tokyo Medical College, Tokyo, Japan) for his linguistic review of the manuscript.

Figure 1: Effects of 10, 50, and 100 μmol/L verapamil on the expression of E-selectin (A), VCAM-1 (B), and ICAM-1 (C) on human umbilical vein ECs stimulated by 10 ng/ml TNFα for 4 hr. Expression of E-selectin, VCAM-1, and ICAM-1 on ECs was measured by a cell-ELISA method. Values are expressed as means±SE (n=4). *P<0.05 in verapamil pretreatment compared with TNFα stimulation without verapamil. □, TNFα negative; [square with diagonal crosshatch fill], TNFα positive.

Figure 2: Effects of 10, 50, and 100 μmol/L verapamil on the expression of VCAM-1 on human umbilical vein ECs stimulated by 1 μg/ml LPS (A) or 10 U/ml IL-1β (B) for 4 hr. Expression of VCAM-1 on ECs was measured by a cell-ELISA method. Values are expressed as means±SE (n=4). *P<0.05 in verapamil pretreatment compared with LPS or IL-1β stimulation without verapamil. (A) □, LPS negative; [square with diagonal crosshatch fill], LPS positive. (B) □, IL-1β negative; [square with upper left to lower right fill], IL-1β positive.

Figure 3: Effects of 10, 50, and 100 μmol/L verapamil on the expression of VCAM-1 on human umbilical vein ECs stimulated by 2000 U/ml IL-4 for 24 hr. Expression of VCAM-1 on ECs was measured by a cell-ELISA method. Values are expressed as means±SE (n=4). *P<0.01 in verapamil pretreatment compared with IL-4 stimulation without verapamil. □, IL-4 negative; [square with diagonal crosshatch fill], IL-4 positive.

Figure 4: Autoradiograph of Northern blot hybridization with the cDNA probes (left) to evaluate the effects of 50 μmol/L verapamil on E-selectin, VCAM-1, and ICAM-1 mRNA levels in human umbilical vein ECs stimulated by 10 ng/ml TNFα for 3 hr. Ethidium bromide staining of RNAs, showing the same densities of 18 S and 28 S ribosomal RNAs in all three lanes. Similar results were observed in three other experiments.

Figure 5: Effects of 10, 50, and 100 μmol/l verapamil on adherence of mononuclear leukocytes (A) or Molt-4 cells (B) to ECs stimulated by 10 ng/ml TNFα for 4 hr under static conditions at 37°C for 30 min. (n=4). *P<0.05 in verapamil pretreatment compared with TNFα stimulation without verapamil. □, TNFα negative; [square with diagonal crosshatch fill], TNFα positive.