Shear Stress Inhibits Smooth Muscle Cell–Induced Inflammatory Gene Expression in Endothelial Cells
Role of NF-κB
Objectives— Vascular endothelial cells (ECs) are influenced by shear stress and neighboring smooth muscle cells (SMCs). We investigated the inflammation-relevant gene expression in EC/SMC cocultures under static condition and in response to shear stress.
Materials and Methods— Under static condition, DNA microarrays and reverse-transcription polymerase chain reaction identified 23 inflammation-relevant genes in ECs whose expression was significantly affected by coculture with SMCs, with 18 upregulated and 5 downregulated. Application of shear stress (12 dynes/cm2) to the EC side of the coculture for 6 hours inhibited most of the proinflammatory gene expressions in ECs induced by coculture with SMCs. Inhibition of nuclear factor-κB (NF-κB) activation by the p65-antisense, lactacystin, and N-acetyl-cysteine blocked the coculture-induced EC expression of proinflammatory genes, indicating that the NF-κB binding sites in the promoters of these genes play a significant role in their expression as a result of coculture with SMCs. Chromatin immunoprecipitation assays demonstrated the in vivo regulation of NF-κB recruitment to selected target promoters. Shear stress inhibited the SMC coculture-induced NF-κB activation in ECs and monocytic THP-1 cell adhesion to ECs.
Conclusions— Our findings suggest that shear stress plays an inhibitory role in the proinflammatory gene expression in ECs located in close proximity to SMCs.
The endothelial cells (ECs) and smooth muscle cells (SMCs) are the major cellular components of the vessel wall. The interactions between these cells play significant roles in the homeostasis of the structure and function of the blood vessel. As an interface between the blood and the vessel wall, ECs occupy a unique location directly exposed to blood flow-induced shear stress, which can influence interactions between ECs and SMCs. By incorporating the SMCs into a matrix of collagen gel, Ziegler et al1 demonstrated that ECs cocultured with SMCs aligned with the flow direction more rapidly than the EC monocultures. Imberti et al2 further demonstrated that the EC response to shear stress in this collagen coculture model was influenced by preconditioning the SMC-seeded collagen gel with cyclic strain. Redmond et al3 have developed a system in which ECs and SMCs were cocultured on opposite sides of porous polypropylene capillary tubes. A series of research studies have been conducted using this coculture system, which represents a significant advance over homogeneous culture.3–5 Nackman et al6 and Rainger and Nash7 constructed a system by combining a parallel plate flow chamber and a coculture module in which ECs and SMCs were grown on opposite sides of a 10-μm-thick permeable membrane containing 0.4-μm pores. They demonstrated that the EC/SMC coculture affected SMC proliferation6 and the EC response to tumor necrosis factor-α (TNF-α).7 Although these studies have contributed to the understanding of the effects of hemodynamic forces on the EC–SMC interactions and functional modulations, there is a lack of systematic analysis of the alteration of gene expression program of ECs and SMCs as a result of their coculture under static condition or in response to shear stress.
In the present study, we applied the high-throughput DNA microarray technology to investigate the expression of inflammation-relevant genes in ECs and SMCs in the coculture condition, using our newly developed EC/SMC coculture flow system.8,9 We used the cDNA microarray that contains >7500 known human genes to analyze the gene expression profiles of cocultured ECs and SMCs under static and shear conditions, with special attention to the 458 inflammation-relevant genes present on the array. Under static condition, the coculture of ECs with SMCs induced significant increases in mRNA expression of 18 inflammation-relevant genes and significant decreases of 5 such genes in ECs. Shear stress applied to the EC side of the coculture inhibited the expressions of most of the proinflammatory genes in ECs induced by coculture with SMCs. We showed that the modulations of the EC expression of proinflammatory genes by coculture and shear stress are attributable to their regulatory effects on the activation of EC transcription factor nuclear factor-κB (NF-κB). Our findings suggest that shear stress plays an inhibitory role in the proinflammatory gene expression in ECs located in close proximity to SMCs, thereby exerting antiinflammatory effects on the vascular wall.
Materials and Methods
The detailed materials and methods used in this study are described in the online supplement (available at http://atvb.ahajournals.org).
Coculture of ECs and SMCs Under Static Condition Alters Their Expression of a Number of Inflammation-Relevant Genes
We first identified the inflammation-relevant genes present on the array, the expression of which in ECs or SMCs was significantly different between the cocultures and monocultures under static condition. Three independent sets of experiments were performed, and each consisted of an EC/SMC coculture and the paired controls of EC and SMC monocultures. Gene expressions with a mean coculture/monoculture ratio ≥2.0 and P≤0.05 were considered positively regulated by the coculture, whereas those that had a mean ratio ≤0.5 and P≤0.05 were considered negatively regulated. Of 458 inflammation-relevant genes present on the array (Table I, available online at http://atvb.ahajournals.org), we identified 23 genes in ECs whose expression was significantly regulated by the static coculture with SMCs (Figure IIA and Table II, available online at http://atvb.ahajournals.org). Among these 23 genes, 18 showed an increase in expression by the coculture, whereas 5 showed a decrease. SMCs were less responsive than ECs in the cocultures in their inflammation-relevant gene expression. Only 2 genes (cyclooxygenase-2 [COX-2] and interleukin-11 [IL-11]) showed a significant increase, and 2 genes (IL-1β and tissue factor [TF]) showed a significant decrease in the cocultured SMCs under static condition, as compared with the monocultured SMCs (Figure IIB and Table II). Whereas the increase in COX-2 was shared with ECs, IL-11 did not change significantly in ECs. The effects of coculture on expressions of IL-1β and TF were opposite for ECs and SMCs.
The microarray results for the genes showing significant modulations by the coculture were confirmed by RT-PCR analysis. Figure 1 shows the representative blots of these genes (the primer sequences of the genes analyzed and the number of cycles of PCR reaction are provided in Table III, available online at http://atvb.ahajournals.org). The largest increases in gene expression in the cocultured ECs were observed for the growth-related oncogene-1 (GRO-1) and vascular adhesion molecule-1 (vascular cell adhesion molecule [VCAM]-1) genes, with DNA microarray mRNA ratios of 18.02±3.14 and 8.73±1.43, respectively. These results are in good agreement with the RT-PCR ratios of 15.61±1.32 and 11.53±3.01, respectively (P>0.5 between the 2 types of measurements for each gene). The caveolin-1 gene showed the largest decrease in mRNA levels in ECs by their static coculture with SMCs, with a DNA microarray mRNA ratio of 0.37±0.05. This value agrees very closely with the RT-PCR ratio of 0.40±0.04 (P>0.5).
Shear Stress Regulates the Inflammation-Relevant Gene Expression in the Cocultured ECs and SMCs
To investigate whether shear stress regulates the inflammation-relevant gene expression in the cocultured ECs and SMCs, the EC side of the coculture was subjected to a shear stress of 12 dynes/cm2 for 6 hours and the expression of the 24 genes modulated in ECs or SMCs by the coculture was examined by RT-PCR analysis. Among 18 genes whose expression in ECs was induced by the static coculture with SMCs, 15 were downregulated by shear stress (Figure IIA and Table II), with 4 of them (ie, monocyte chemotactic protein-3 [MCP-3], coagulation factor II receptor-like-1, granulocyte-macrophage colony-stimulating factor [GM-CSF] 2 receptor-β [GM-CSF2R-β], and CD44) decreased to below the basal levels in the control monocultured ECs. Only 3 of the genes upregulated by the coculture, ie, COX-2, TNF receptor superfamily member 11b (TNFRSF11B), and TF were not downregulated by shear stress. For the 5 EC genes for which expression was reduced by coculture with SMCs, intercellular adhesion molecule-2 (ICAM-2) and caveolin-1 were further downregulated by shear stress (P<0.05) but not endothelial NO synthase, scavenger receptor class A member 3, and thrombomodulin (P>1), compared with the static cocultured ECs. For the SMC genes, shear stress applied to the EC side caused an increase in expression of thrombomodulin and decreases of GRO-1, IL-11, and ICAM-2 (Figure IIB and Table II).
Coculture-Induced EC Expression of Proinflammatory Genes Is Dependent on NF-κB
To investigate whether NF-κB is involved in the inflammation-relevant gene expression in ECs induced by coculture with SMCs, we examined the effects of interference with NF-κB expression on the coculture-induced upregulation of the 18 proinflammatory genes in ECs by using the antisense oligonucleotides to NF-κB subunits p65 (p65-antisense), the transcription inhibitor lactacystin, and the antioxidant NAC. ECs cocultured with SMCs induced their expression of NF-κB subunits p65 and p50 mRNAs, as compared with the monocultured cells (Figure 2A and 2B). Pretreating ECs with p65-antisense abolished the coculture-induced p65, but not p50, mRNA expression; in contrast, p65-sense did not have an inhibitory effect on the coculture-induced p65 expression. Moreover, pretreating ECs with lactacystin and NAC for 1 hour abolished the coculture-induced p65 and p50 mRNA expressions in ECs. All 18 proinflammatory EC genes induced by the static coculture with SMCs were significantly inhibited by pretreating the ECs with p65-antisense and NAC (Figure 2C), but not p65-sense (supplemental Figure III). All but 3 of the genes (COX-2, GRO-1, and GRO-3) were downregulated by pretreating ECs with lactacystin. These results indicate that virtually all of the SMC coculture-inducible genes in ECs were regulated by NF-κB.
Promoter Analysis of the Coculture-Inducible Genes in ECs
NF-κB regulates gene expression through binding to promoter elements found in a variety of cellular genes with the DNA sequence GGGYNNRRCC.10 An attempt was made to correlate the existence of NF-κB binding sites in the promoter region of each of the 18 SMC-inducible proinflammatory genes with their regulation by the NF-κB inhibitors (ie, p65-antisense, lactacystin, and NAC). As listed in the Table, the promoter regions of most of these genes inhibited by the 3 agents contain 1 or more functional NF-κB binding sites. These genes include GRO-1, ICAM-1, VCAM-1, MCP-1, MCP-3, IL-1β, IL-6, COX-2, GM-CSF2R-β, and TF. Several other genes, including GRO-3, TNFRSF11B, CD44, coagulation factor II receptor–like-1, and natural killer cell transcript-4 (NK-4), have putative NF-κB binding sites in their promoter regions, as determined by a search of each of these sequences with the Genomatrix software (http://www.genomatix.de). Our results on the effects of NF-κB inhibitors indicate that these genes are dependent on NF-κB for their expression and suggest that the putative NF-κB promoter elements are functional. No NF-κB binding sites were found in the promoter regions of the MCP-2, MCP-4, and cathepsin C genes, despite the fact that the expression of these genes appeared to be dependent on NF-κB.
Shear Stress Inhibits the Coculture-Induced p65 and p50 mRNA Expressions and NF-κB-DNA Binding Activity in ECs
Given our findings that the coculture-induced EC expression of inflammation-relevant genes is dependent on NF-κB, and that shear stress inhibits most of the coculture-induced gene expressions, we further investigated whether shear stress modulates the expression and activation of NF-κB in ECs cocultured with SMCs. As shown in Figure 3, shear stress applied to the EC side of the coculture inhibited the SMC-induced p65 (Figure 3A) and p50 (Figure 3B) mRNA expression. In contrast, shear application had no effect on these expressions in control ECs (data not shown). The results of electrophoretic mobility shift assay showed that static coculture with SMCs caused an increase in NF-κB–DNA binding activity in the nucleus of ECs, as compared with the monocultured ECs (Figure 3C). Shear stress to the EC side of the coculture resulted in a reduction in this SMC-induced binding activity. ECs treated with TNF-α for 1 hour were used as a positive control to show increased binding activity. The specificity of this binding for NF-κB was shown by its abolition by coincubation of nuclear proteins with 20-fold unlabeled oligonucleotides and substantiated by the supershifting in gel mobility of the NF-κB–oligonucleotide complex after preincubation of nuclear proteins with an antibody to p65.
Chromatin Immunoprecipitation Assays Reveal Shear-Mediated Inhibition in Coculture-Induced In Vivo Binding of NF-κB to the ICAM-1 and MCP-1 Promoters in ECs
To further assess the in vivo regulation of NF-κB binding to the promoter regions of ICAM-1 and MCP-1 genes in ECs cocultured with SMCs, we performed chromatin immunoprecipitation assays in these ECs under static and shearing conditions by using antibodies against p65 and the promoter-specific primers. ECs cocultured with SMCs under static condition increased the in vivo NF-κB binding to their ICAM-1 and MCP-1 promoters as early as 15 minutes after the coculture, reaching maximal levels (≈7-fold for ICAM-1 and ≈4-fold for MCP-1 compared with static monocultured ECs) within 30 minutes or 1 hour (Figure 4A). The levels of NF-κB promoter binding then declined but still remained higher than the values of static monocultured ECs after 2 hours of coculture with SMCs. When ECs cocultured with SMCs under static condition for 30 minutes were subjected to a shear stress of 12 dynes/cm2 for an additional 30 minutes, the in vivo NF-κB promoter binding returned to the basal control levels (Figure 4B). These results suggest that ECs cocultured with SMCs increase their in vivo binding of NF-κB to the promoter regions of the selected genes and that shear stress applied to the EC side of the coculture inhibits this coculture-induced in vivo NF-κB recruitment to the target promoters.
Shear Stress Abolishes the Increases in Monocytic THP-1 Adhesiveness of ECs Induced by Coculture With SMCs
To investigate the functional modulations of ECs by coculture with SMCs under static condition and in response to shear stress, we examined the alterations in monocytic THP-1 cell adhesiveness of the cocultured ECs. Flow adhesion assays showed that coculture with SMCs under static condition resulted in an increase in EC adhesiveness for monocytic THP-1 cells compared with the monocultured ECs (48.21±5.2 versus 5.8±0.9 cells/mm2; P<0.05; Figure 5). This coculture-induced cell adhesion can be abolished by pre-exposing the EC/SMC coculture to a shear stress of 12 dynes/cm2 for 6 hours (P<0.05 versus static control EC/SMC). Pretreating ECs with p65-antisense, lactacystin, and NAC abolished this coculture-induced cell adhesion, implicating the involvement of NF-κB in these modulatory effects of coculture and shear stress.
Our present study has provided, for the first time to our knowledge, a systematic analysis of the inflammation-relevant gene expression in ECs and SMCs in their coculture under static condition and in response to shear stress. We demonstrated that SMCs cause an upregulation of proinflammatory gene expression in ECs located in close proximity and that shear stress acts as a negative regulator for these proinflammatory gene expression in ECs cocultured with SMCs. Under static coculture condition, DNA microarray technology identified 23 inflammation-relevant genes that exhibit significant changes (18 increases and 5 decreases) in their expression in ECs cocultured with SMCs, as compared with the control monocultured ECs. All 18 genes upregulated in ECs by the coculture encode products that serve proinflammatory and thrombogenic functions. These functions include: (1) mediation of leukocyte–EC interaction and signal transduction (ICAM-1, VCAM-1, CD44, and NK-4); (2) promotion of chemotaxis and leukocyte activation (COX-2, GRO-1, GRO-3, IL-1β, IL-6, and cathepsin C); (3) activation of cytokines and growth factors (GM-CSF2R-β and TNFRSF11B); (4) mediation of blood coagulation cascades (TF and coagulation factor II receptor-like-1); and (5) recruitment and extravasation of monocytes/macrophages (MCP-1, MCP-2, MCP-3, and MCP-4). However, the genes for which expression was reduced in ECs by the static coculture with SMCs encode proteins that serve antiinflammatory (endothelial nitric oxide synthase and caveolin-1),11,12 antioxidant (scavenger receptor class A, member 3),13 and anticoagulant (thrombomodulin) functions, or are downregulated by inflammatory agents (ICAM-2).14 These results on gene expression suggest that coculture with SMCs in a static condition may cause the ECs to be in a more proinflammatory and thrombogenic status, as compared with the ECs in monocultures.
The phenotypes of the SMCs used may have an effect on the coculture-induced gene expression. Our recent data demonstrated that the SMCs cultured in a medium containing 10% fetal bovine serum, which was used in these experiments, had lower levels of expression of SMC contractile protein markers, such as smooth muscle α-actin and myosin heavy chain, h-caldesmon, and calponin, than those cultured in the medium containing only 0.5% fetal bovine serum (Figure IVA, available online at http://atvb.ahajournals.org). Therefore, the SMCs in our present study may be in a synthetic phenotype, which may generate proinflammatory agents and exert paracrine effects to stimulate the overlying ECs.7,15 By treating ECs with the culture medium collected from 24-hour EC/SMC cocultures, we found that the expression of several proinflammatory genes, including ICAM-1, VCAM-1, MCP-1, and GRO-1, in the treated ECs increased to an extent similar to that in ECs cocultured with SMCs (Figure IVB). This indicates that there is a paracrine effect of agents released from the SMCs acting on the ECs to cause the increased expression of proinflammatory genes.
In addition to paracrine effects, cells may communicate through intercellular gap junctions. Using the same type of coculture membrane (containing 0.4-μm pores), Fillinger et al16 showed that SMCs were able to extend their cytoplasmic projections into the membrane pores. In our recent transmission electron microscopic study, we also observed such extension of the SMC cytoplasmic projections into 0.4-μm pores, but not through the entire length (unpublished observation). Moreover, by using specific inhibitors of gap junctions, such as 18β-glycyrrhetinic acid (Sigma) and the connexin mimetic peptide Gap 27 (amino acid sequence SRPTEKTIFII; Severn Bioteck, UK), we found that ECs pretreated with these inhibitors before coculture with SMCs in the presence of the respective inhibitors did not result in an inhibition in their coculture-induced proinflammatory gene expression. Thus, it is not likely that the ECs and SMCs in our coculture system could be in direct contact with each other; hence, the intercellular communication through gap junctions was probably not involved in the coculture effect on the EC gene expression.
An additional important mediator for cellular responses to inflammatory stimuli is NF-κB. In this study, we demonstrated that NF-κB may be involved in the SMC-mediated induction of the proinflammatory genes in the cocultured ECs. Several lines of evidence support this finding. First, the SMC coculture induced EC gene expression of NF-κB subunits p65 and p50 and increased the NF-κB–DNA binding activity in the nucleus. Second, NF-κB inhibitors, including p65-antisense, lactacystin, and NAC, significantly inhibited the coculture-induced proinflammatory gene expression. Third, promoter analysis of these proinflammatory genes showed the existence of NF-κB binding sites in most of their promoters. Fourth, chromatin immunoprecipitation assays demonstrated the possibility of in vivo regulation of NF-κB recruitment to selected target promoters. In addition to NF-κB, the activation of mitogen-activated protein kinases, ie, c-Jun NH2-terminal kinase, extracellular signal regulated kinase, and p38, has been shown to be involved in the expression of a number of proinflammatory genes in ECs in response to various stimuli.17,18 Our recent study has showed that ECs cocultured with SMCs under static condition induce their phosphorylation of c-Jun NH2-terminal kinase and extracellular signal regulated kinase, but not p38, as compared with the EC monocultures (Figure V, available online at http://atvb.ahajournals.org). Whether the activation of these SMC-induced c-Jun NH2-terminal kinase and extracellular signal regulated kinase plays a role in the proinflammatory gene expression in ECs cocultured with SMCs requires further studies.
An important finding of the current study is that shear stress applied to the EC side of the coculture inhibited most of the proinflammatory gene expressions in ECs induced by coculture with SMCs. The way by which shear stress inhibits these EC gene expressions in coculture is not clear. Our present study demonstrated that shear stress can inhibit the coculture-induced p65 and p50 mRNA expressions and NF-κB–DNA binding activity in ECs. Thus, the shear-mediated inhibition of NF-κB activation may play a role in this shear-mediated inhibition of the proinflammatory gene expression in the cocultured ECs.
We have demonstrated the functional consequence of the SMC coculture-induced changes in EC gene expression under static and shear conditions by assessing the EC adhesiveness for the monocytic THP-1 cells. Thus, ECs cocultured with SMCs under static condition showed an increase in THP-1 cell adhesion, as compared with the EC monoculture, which is in agreement with the findings of Rainger and Nash7 and our results on the upregulation of the adhesion-promoting genes. The current study has further shown that pre-exposure of ECs in the SMC coculture to shear stress abolished this coculture-induced increase in adhesion, indicating that shear stress exerts antiinflammatory functions on ECs cocultured with SMCs. The reduction of the SMC coculture-induced increases in adhesiveness of ECs by pretreating them with p65-antisense, lactacystin, and NAC provides evidence that NF-κB may be involved in these modulations of cell adhesiveness by coculture and shear stress.
In summary, our findings indicate that EC/SMC coculture under static condition causes the upregulation of EC proinflammatory gene expression, which is mediated by NF-κB activation and leads to an enhancement of monocytic cell adhesion. In comparison with the results obtained under static condition, the application of shear stress to the coculture causes a downregulation of EC proinflammatory gene expression, which is mediated by NF-κB de-activation and leads to a suppression of monocytic cell adhesion. These results suggest that laminar shear stress serves antiinflammatory and atheroprotective functions in vascular biology.
This work was supported by grants ME-093-PP-02 (to J.-J.C.) from the National Health Research Institutes, Taiwan, Republic of China, and HL19454 (to S.C.) from the National Heart, Lung, and Blood Institute, USA. The work was conducted in part in affiliation with the Center of Tissue Engineering and Stem Cell Research, National Chung-Hsing University, Taichung, Taiwan, Republic of China.
- Received September 28, 2004.
- Accepted February 8, 2005.
Redmond EM, Cahill PA, Sitzmann. Flow-mediated regulation of G-protein expression in cocultured vascular smooth muscle and endothelial cells. Arterioscler Thromb Vasc Biol. 1998; 18: 75–83.
Redmond EM, Cullen JP, Cahill PA, Sitzmann JV, Stefansson S, Lawrence DA, Okada SS. Endothelial cells inhibit flow-induced smooth muscle cell migration: role of plasminogen activator inhibitor-1. Circulation. 2001; 103: 597–603.
Rainger GE, Nash GB. Cellular pathology of atherosclerosis: smooth muscle cells prime cocultured endothelial cells for enhanced leukocyte adhesion. Circ Res. 2001; 88: 615–622.
Chiu JJ, Chen LJ, Lee PL, Lee CI, Lo LW, Usami S, Chien S. Shear stress inhibits adhesion molecule expression in vascular endothelial cells induced by coculture with smooth muscle cells. Blood. 2003; 101: 2667–2674.
Davis ME, Grumbach IM, Fukai T, Cutchins A, Harrison DG. Shear stress regulates endothelial nitric oxide synthase promoter activity through nuclear factor Kappa-B binding. J Biol Chem. 2004; 279: 163–168.
Han HJ, Tokino T, Nakamura Y. CSR, a scavenger receptor-like protein with a protective role against cellular damage caused by UV irradiation and oxidative stress. Hum Mol Genet. 1998; 7: 1039–1046.
Owens GK, Kumar MS, Wamhoff BR. Molecular regulation of vascular smooth muscle cell differentiation in development and disease. Physiol Rev. 2004; 84: 767–801.
Yamawaki H, Lehoux S, Berk BC. Chronic physiological shear stress inhibits tumor necrosis factor-induced proinflammatory responses in rabbit aorta perfused ex vivo. Circulation. 2003; 108: 1619–1625.
Jiang B, Xu S, Hou X, Pimentel DR, Cohen RA. Angiotensin II differentially regulates interleukin-1-beta-inducible NO synthase and vascular cell adhesion molecule-1 expression: role of p38 MAPK. J Biol Chem. 2004; 279: 20363–20368.