The subendothelial extracellular matrix modulates NF-kappaB activation by flow: a potential role in atherosclerosis

Abstract

Atherosclerotic plaque forms in regions of the vasculature exposed to disturbed flow. NF-kappaB activation by fluid flow, leading to expression of target genes such as E-selectin, ICAM-1, and VCAM-1, may regulate early monocyte recruitment and fatty streak formation. Flow-induced NF-kappaB activation is downstream of conformational activation of integrins, resulting in new integrin binding to the subendothelial extracellular matrix and signaling. Therefore, we examined the involvement of the extracellular matrix in this process. Whereas endothelial cells plated on fibronectin or fibrinogen activate NF-kappaB in response to flow, cells on collagen or laminin do not. In vivo, fibronectin and fibrinogen are deposited at atherosclerosis-prone sites before other signs of atherosclerosis. Ligation of integrin alpha2beta1 on collagen prevents flow-induced NF-kappaB activation through a p38-dependent pathway that is activated locally at adhesion sites. Furthermore, altering the extracellular matrix to promote p38 activation in cells on fibronectin suppresses NF-kappaB activation, suggesting a novel therapeutic strategy for treating atherosclerosis.

Figure 1.

Shear stress–induced NF-κB activation is matrix specific. (A) BAE cells plated on Coll I, LN, FN, or FG for 4 h were sheared for 60 min, and p65 localization was assessed by immunocytochemistry. The percentage of cells showing nuclear staining for NF-κB was determined. Values are means ± SD (>100 cells counted per condition per experiment; n = 3). *, P < 0.05; **, P < 0.01. A representative p65 stain for these conditions is shown on the right. (B) Endothelial cells were grown on different matrix proteins and sheared for 0, 5, 15, 30, and 60 min. Phosphorylation of p65 on Ser536 was assessed by immunoblotting total cell lysates with a phosphorylation site-specific antibody. Values are means ± SD following normalization for total p65 (n = 4). A representative blot is shown for each condition. (C) Human umbilical vein endothelial cells were plated on different matrix proteins and sheared for 5 h, and the level of ICAM-1 expression was assessed by Western blotting. Densitometric quantification was normalized to actin. n = 3; *, P < 0.05. Immunocytometric staining for ICAM-1 surface expression is shown on the right.

Figure 2.

Matrix remodeling and inflammatory gene expression in vivo . C57/B6 mice were fed a chow diet (A), male ApoE null mice were fed a chow diet for 10 wk (B), and ApoE null mice were fed a Western diet for 10 wk (C). Mice were killed and the indicated arteries were removed and embedded in paraffin. Serial sections were stained for FN, FG, ICAM-1, VCAM-1, and Mac-2 and shown at high magnification, with lower magnification views of the entire vessels shown as insets.

Figure 3.

p38 inhibits shear stress–induced NF-κB activation on Coll. (A) BAE cells plated on Coll, LN, FN, or FG were sheared for the indicated times. Phosphorylation of p38 was assessed by Western blotting with phosphorylation site-specific antibodies, quantified by densitometry, and normalized to total p38. Values are means ± SD (n = 3). A representative Western blot is shown for each condition. (B) Cells were treated with the pharmacological p38 inhibitor SB202190 (1 μM for 1 h), and NF-κB nuclear translocation was assessed with or without shear stress for 30 min. Greater than 100 cells were counted per condition per experiment; n = 3; *, P < 0.05; **, P < 0.01. (C) Cells were preincubated with SB202190 and shear stress–induced p65 phosphorylation (Ser536) was assessed as previously described. Densitometric quantification was normalized to total p65. n = 3. (D) Cells were transiently transfected with dominant-negative p38 (p38-AGF) and sheared for 60 min, and p65 nuclear translocation was assessed. Transfected cells were identified by costaining for total p38. Greater than 100 cells were counted per condition per experiment; n = 3; *, P < 0.05. A representative stain is shown.

Figure 4.

Integrin α2β1 mediates p38 activation and NF-κB inhibition. (A) BAE cells were incubated for 30 min with either the α2β1 blocking (R2-8C8) or the α2β1 nonblocking antibody (12F1) and sheared for 5 or 30 min. Lysates were analyzed for p38 phosphorylation by immunoblotting. Bands were quantified by densitometry and normalized for total p38 protein. Values are means ± SD (n = 4). *, P < 0.05; **, P < 0.01. (B) Cells on Coll were incubated with 12F1 or R2-8C8 and sheared for 5 or 30 min, and the phosphorylation of p65 was analyzed by immunoblotting. Bands were quantified by densitometry and normalized for total p65. **, P < 0.01 (n = 5). (C) Cells plated on Coll I or FN were pretreated with the p38 inhibitor SB202190 (1 μM for 1 h) and integrin ligation was induced with the β1 integrin–activating antibody TS2/16 (15 μg/ml for 1 h). p65 localization was assessed by immunostaining, and the percentage of cells showing nuclear localization was scored (n = 3; *, P < 0.05). (D) TS2/16-induced p65 phosphorylation of p65 on Ser536 was assessed by Western blotting. Bands were quantified by densitometry and normalized to total p65 (n = 3–6).

Figure 5.

Coll/FN mixed matrices. BAE cells were plated on increasing amounts of FN with or without precoating with 10 μg/ml Coll. (A) Cells were sheared for 30 min and nuclear p65 accumulation was assessed as previously described. The abscissa represents the amount of FN adsorbed to the coverslips under these conditions. Values reflect the increase in p65 nuclear translocation relative to cells under static conditions. n = 4. (B) Cells were treated as in A, and p65 phosphorylation on Ser536 was assessed by Western blotting. Bands were quantified by densitometry and normalized to total p65. Values represent the increase in p65 phosphorylation relative to cells under static conditions. n = 3. (C) Cells were exposed to shear stress for 5 min and p38 phosphorylation was assessed by Western blotting. Bands were quantified by densitometry and normalized to total p38, and data was expressed as the fold increase in p38 phosphorylation compared with cells under static conditions. n = 3.

Figure 6.

Localized activation of p38 and IKK at adhesion sites. (A) Cells were plated on Coll or FN, sheared for 5 min or kept under static conditions, and stained for phosphorylated p38 and β1 integrin. Images are representative of four experiments. (B) BAE cells were plated on Coll, LN, FN, or FG and sheared for the indicated times, and IKK phosphorylation was assessed by Western blotting. Results are representative of four to six experiments. (C) Cells were plated on Coll or FN, sheared for 30 min or kept as static controls, and stained for phosphorylated IKK and β1 integrin. Results are representative of three experiments. (D) Cells on Coll or FN were treated with 10 U/ml TNFα and stained for phosphorylated IKK and β1 integrin. Images are representative of three experiments.

Figure 7.

FNIII-1C activates p38 and blocks shear stress–induced NF-κB activation. (A) BAE cells on FN were treated with the FNIII-1C or FNIII-2C peptides for 4 h. Then, cells were either left under static conditions or sheared for 30 min and phosphorylation of p38 was assessed by Western blotting with phosphorylation site-specific antibodies (n = 5). (B) Cells were plated on FN, treated with FNIII-1C for 4 h, and stained for phosphorylated p38 and β1 integrin (n = 3). (C) BAE cells were plated on FN, incubated overnight, and treated with 10 μM FNIII1C, 10 μM of heat-denatured FNIII1C, or 10 μM FNIII2C for 4 h. Cells were sheared for 30 min, and NF-κB nuclear translocation was assessed by immunocytochemistry (n = 3–5). (D) Cells were pretreated with FN peptides as in C, followed by incubation with the p38 inhibitor SB202190 (1 μM) for 1 h before onset of shear stress. Cells were sheared for 30 min and nuclear translocation of p65 was assessed. Values are means ± SD (>100 cells counted per condition per experiment; n = 3–8). **, P < 0.01; ***, P < 0.001. (E) Cells were pretreated with the FN peptides as indicated, incubated with SB202190, and sheared for 30 min. p65 phosphorylation at Ser536 was assessed by Western blotting. Bands were quantified by densitometry and normalized to total p65. Values are means ± SD from four to seven experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.