Negative regulation of vascular smooth muscle cell migration by blood ...

Am J Physiol Heart Circ Physiol 292: H928 –H938, 2007. First published September 29, 2006; doi:10.1152/ajpheart.00821.2006.

Negative regulation of vascular smooth muscle cell migration by blood shear stress Jeremy Goldman,1* Lin Zhong,2* and Shu Q. Liu2 1

Biomedical Engineering Department, Michigan Technological University, Houghton, Michigan; and 2Biomedical Engineering Department, Northwestern University, Evanston, Illinois Submitted 31 July 2006; accepted in final form 26 September 2006

Goldman J, Zhong L, Liu SQ. Negative regulation of vascular smooth muscle cell migration by blood shear stress. Am J Physiol Heart Circ Physiol 292: H928 –H938, 2007. First published September 29, 2006; doi:10.1152/ajpheart.00821.2006.—Vortex blood flow with reduced blood shear stress in a vein graft has been hypothesized to promote smooth muscle cell (SMC) migration and intimal hyperplasia, pathological events leading to vein graft restenosis. To demonstrate that blood shear stress regulates these processes, we developed a modified vein graft model where the SMC response to reduced vortex blood flow was compared with that of control vein grafts. Vortex blood flow induced SMC migration and neointimal hyperplasia in control vein grafts, whereas reduction of vortex blood flow in the modified vein graft strongly suppressed these effects. A venous polymer implant with known fluid shear stress was employed to clarify the molecular mechanism of shear-dependent SMC migration in vivo. In the polymer implant, the phosphorylation of extracellular signal-regulated kinase (ERK1/2) and myosin light chain kinase (MLCK), found primarily in SMCs, increased from day 3 to day 5 and returned toward the control level from day 5 to day 10, with the peak phosphorylation associated with the maximal speed of SMC migration. Treatment with PD-98059 (an inhibitor specific to the ERK1/2 activator MEK1/2) significantly suppressed the phosphorylation of MLCK, suggesting a role for ERK1/2 in regulating the activity of MLCK. Treatment with PD-98059 or ML-7 (an inhibitor specific to MLCK) reduced shear stress-dependent SMC migration, resulting in an SMC distribution independent of fluid shear stress. These results suggest that fluid shear stress regulates SMC migration via the mediation of ERK1/2 and MLCK. blood flow; neointima formation; signaling transduction

smooth muscle cells (SMCs) is a process that contributes to vasculogenesis during physiological development (20). After they reach maturation, SMCs enter a quiescent state without apparent motility. However, under certain environmental stimulations, such as injury, mechanical stretch, altered fluid shear stress, and inflammatory factors, SMCs are stimulated to enter an active state, initiating migration from the media to the intima (14, 39, 53, 56). This process is not only physiologically important in terms of adaptation to an altered environment, it also contributes to focal neointima formation, facilitating the development of vascular occlusive disease (53). Mechanical stretch and blood shear stress have been hypothesized to play a role in regulating the activities of vascular cells, including SMC migration and proliferation (3, 9, 11, 12, 18, 24, 25, 28, 43, 48). This hypothesis has been supported by

THE MIGRATION OF VASCULAR

* These authors contributed equally to this work. Address for reprint requests and other correspondence: J. Goldman, Biomedical Engineering Dept., Michigan Technological Univ., Houghton, MI 49931 (e-mail: jgoldman@mtu.edu). H928

a number of experimental observations. For instance, an increase in mechanical stretch has been shown to enhance SMC proliferation (2, 27, 60, 61). SMC migration and proliferation are found more frequently in curved and bifurcating blood vessels, which are exposed to vortex blood flow, than in straight arteries, which are exposed to laminar blood flow (28, 29, 31). Several recent studies have demonstrated that fluid shear stress exerts an inhibitory effect on SMC migration (33, 38, 49). Together, these findings suggest a role for fluid shear stress and mechanical stretch in regulation of SMC migration and proliferation. Vein grafts are commonly used to replace arteries that have malfunctioned. In a vein graft subjected to arterial blood pressure, tensile stretch in the vessel wall suddenly increases (45). Such a mechanical stretch leads to vortex blood flow because of diameter mismatch between the graft and the host artery (29). It has been hypothesized that mechanical stretch and vortex blood flow may promote the pathological vascular adaptations that cause vein graft restenosis. This hypothesis is supported by numerous experimental observations. For instance, mechanical stretch has been shown to cause extensive SMC injury and death within the first several days after vein graft surgery (8, 30, 32, 42), processes that may initiate restenosis. Simultaneous restriction of mechanical stretch and minimization of vortex blood flow with an external stent have been found to significantly reduce SMC death and subsequent SMC proliferation (34, 35, 45). Unfortunately, it is difficult to decouple vortex blood flow and mechanical stretch in vivo. Largely for this reason, the roles and relative contributions of vortex blood flow and mechanical stretch during SMC migration and proliferation in the neointima of vein grafts have been difficult to clarify in vivo. Thus it remains unclear whether intimal hyperplasia in a vein graft is primarily an effect of mechanical stretch or vortex blood flow. Here, we use a novel experimental vein graft model where mechanical stretch is present but vortex blood flow is minimized. This model allows us to clarify the role of vortex blood flow and demonstrate the therapeutic effect of shear stress modulation in a vein graft. In the vascular system, SMC migration is often found in blood vessels with divergent flow and altered fluid shear stress (28, 29). However, in such blood vessels, as well as in vein grafts, it is difficult to assess the distribution of fluid shear stress and difficult to characterize SMC migration in terms of the speed and direction of migration, hindering the investigation of the mechanisms of shear stress-dependent SMC migraThe costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

0363-6135/07 $8.00 Copyright © 2007 the American Physiological Society

http://www.ajpheart.org

H929

SHEAR REGULATION OF VASCULAR SMC MIGRATION

tion. To resolve these issues, a vascular polymer implant model has been developed for the induction of SMC migration under known fluid shear stress (33, 38). The implant model allows for comparisons of SMC migration with the distribution of fluid shear stress. This model is used here, because measurement and analysis of shear stress is difficult in the vein graft model. The migration of vascular SMCs is possibly regulated by growth factor-related signaling mechanisms (16, 22, 46, 50). Several growth factors, including platelet-derived growth factor (PDGF), insulin-like growth factor, and epidermal growth factor, can activate the extracellular signal-regulated kinase (ERK1/2) signaling pathway (10, 52). The interaction of a growth factor with a cognate receptor induces autophosphorylation of the receptor tyrosine kinase, which in turn induces a sequential activation of signaling molecules, including Grb/ SOS, Ras, Raf, mitogen-activated protein kinase-ERK kinase (MEK1/2), and ERK1/2 (52). Activated ERK1/2 has been shown to induce and enhance vascular SMC migration (13, 47). Although the mechanism of ERK1/2-regulated cell migration remains a research topic, the involvement of myosin light chain kinase (MLCK) in ERK1/2-related signal transduction may play a role. A previous study has shown that ERK1/2 in cultured carcinoma cells can phosphorylate MLCK, which in turn phosphorylates myosin light chain and induces actinmyosin II interactions, enhancing cell migration (23). Previous studies have demonstrated that fluid shear stress influences the activity of growth-related factors (17, 40, 44, 51) and signaling molecules (21, 58) in cultured vascular endothelial cells. Thus it is possible that, in the present model, alterations in fluid shear stress from a physiological level may influence the activity of signaling molecules, which in turn mediate SMC migration. Here, we use a polymer implant model to investigate the role of ERK1/2 and MLCK in the regulation of shear stress-dependent SMC migration.

Vein graft model 2: increased tensile stress/strain and minimized vortex blood flow. A vein graft was created using a method similar to that described above. However, the right external, instead of the common, jugular vein was used. When subjected to arterial blood pressure, this vein graft expanded to a diameter that closely matched that of the host artery (Fig. 1, B and C). Because the development of vortex blood flow in a vein graft depends on the wall slope at the proximal anastomosis (29), development of vortex blood flow is minimized in a vein graft with a diameter that closely matches that of the host artery. Polymer implant model. Procedures for the experimental model have been described in previous reports (33, 38). Briefly, after the abdominal cavity was opened, a 2- to 3-mm-diameter segment of the inferior vena cava was separated from surrounding tissue. A 0.35-mm-OD polypropylene microcylinder coated with polyethylene glycol and silicone-urea [manufactured by 3M (Ann Arbor, MI) for human artificial lungs and kindly provided by Dr. Keith Cook (Northwestern University)] was inserted into the center of the vena cava, with the microcylinder perpendicular to blood flow (Fig. 1D). Four microcylinders were implanted along the vena cava of each rat. The abdominal cavity was closed, and the rat was allowed to recover. Observations were carried out after surgery on days 3, 5, 7, and 10 with five rats at each time point. Experimental procedures were approved by the Animal Care and Use Committees of Northwestern University and Michigan Technological University. Analysis of Fluid Shear Stress After surgery, the vascular implant was rapidly encapsulated with a ⬃50- to 100-␮m-thick layer of thrombus. The distribution of fluid shear stress on the encapsulating thrombus of the venous polymer implant due to exposure to blood flow has been analyzed using a boundary layer theory (33, 38). Briefly, blood flow in the vena cava was measured using a Transonic blood flowmeter, and the flow velocity profile was assessed using Poiseuille’s theory. The distribution of fluid shear stress on the vascular implant was assessed using the following equation (55) ␶ ⫽ 兵关共1/2兲␳u 2 兴/关ur/␯兴 1/2 其关6.973 共x/r兲 ⫺ 2.732 共x/r兲 3 ⫹ 0.292 共x/r兲 5 ⫺ 0.0183 共x/r兲 7 ⫹ 0.000043 共x/r兲 9

(1)

⫺ 0.000115 共x/r兲 ⫹ . . .兴 11

METHODS

Experimental Models Male, 3-mo-old Sprague-Dawley rats (Harlan, Indianapolis, IN) were separated into two groups of four rats each: one group was used to create vein grafts with increased tensile stress and vortex blood flow, as described previously (45), and the other group was used to create a vein graft with increased tensile stress but with minimized vortex blood flow. Observations were carried out 30 days after surgery. Vein graft model 1: increased tensile stress/strain and vortex blood flow. A rat was anesthetized with Pentothal Sodium (50 mg/kg ip) and morphine (10 mg/kg sc). The right common jugular vein was carefully isolated, harvested, and treated with heparinized (100 U/ml) culture medium. An end-to-end anastomotic technique was used to graft the vein segment into the abdominal aorta below the renal arteries, as previously described (35). For this surgical process, 10 –12 interrupted stitches (10-0 nylon suture) were placed at each end. After the grafting surgery, blood flow was reestablished, the external diameters of the vein graft and host artery were measured, the abdominal cavity was closed, and the rat was allowed to recover until day 30. The vein graft was subjected to arterial blood pressure and flow, leading to an expanded diameter that significantly exceeded the host artery diameter (Fig. 1, A and C). This diameter mismatch has been shown to result in vortex blood flow at the proximal anastomosis (29). AJP-Heart Circ Physiol • VOL

where ␶ is fluid shear stress, u is flow velocity at each selected location of the vascular implant, ␳ is blood density, ␯ is blood kinematic viscosity, x is distance in the circumferential direction of the implant from the leading edge, and r is the radius of the implant. The thickness of the encapsulating tissue was considered in shear stress analyses. Identification of SMCs In the thrombus around a vascular implant, SMC migration was found from the end to the center of the implant (33). To identify SMCs in the encapsulating tissue, specimens were fixed using 4% formaldehyde in PBS for 20 min, treated with 0.5% Triton X-100 for 20 min at 37°C, incubated with a mixture of 5 ␮g/ml anti-smooth muscle ␣-actin antibody (Roche, Indianapolis, IN) (29), 1% BSA, and PBS at 37°C for 1 h, washed in PBS, incubated with a fluorescein-conjugated anti-IgG2a antibody (Roche) at 37°C for 1 h, and washed in PBS. The encapsulating tissue was removed from the implant, flattened on a microscopic slide with a coverslip, and examined en face using a fluorescence microscope (model BX40, Olympus). Specimens incubated only with the secondary antibody were used as controls. For the vein graft experiments, vein graft specimens were collected from rats at the specified end point, fixed in 4% formaldehyde in PBS for 20 min, and cut first into longitudinal strips and then into 10-␮m-thick sections. SMCs were identified in vein graft longitudinal cryosections with a mouse anti-smooth muscle ␣-actin antibody (Chemicon International, Temecula, CA) followed by an Alexa Fluor

292 • FEBRUARY 2007 •

www.ajpheart.org

H930


Fig. 1. Experimental vein graft models. A: diameter expansion in common jugular vein (v) exposed to arterial blood pressure, resulting in mismatched venous-arterial diameters. B: diameter expansion in external jugular vein exposed to arterial blood pressure, resulting in matched venous-arterial diameters. C: external diameters of vein graft and aorta measured under a stereo microscope after exposure of the vein graft to arterial blood flow. Measured diameters were normalized to arterial diameters. A significant diameter mismatch was found between the control vein grafts (A) relative to the host artery: *P ⬍ 0.005. Engraftment of the external jugular vein (B) resulted in a diameter that was not significantly different from host artery diameter. Scale bar, 1 mm. Sample size was 4 per vein graft group. D: schematic illustration of a vascular implant in rat abdominal vena cava. Locations labeled 0° and 180° are leading and trailing edges (0 shear stress) of cylindrical implant. Flow separation occurs at 109° (0 shear stress). Maximal shear stress is found at 55°.

488 donkey anti-mouse IgG secondary antibody (Molecular Probes, Eugene, OR). Sections were mounted with Vectashield mounting medium containing 4⬘,6-diamidino-2-phenylindole for nuclear staining (Vector Laboratories, Burlingame, CA) and visualized under a fluorescence microscope (model BX51, Olympus). SMC area coverage in the vein graft neointima was measured with Metamorph Imaging software by outlining the ␣-actin-positive region in the neointima of fluorescently labeled sections.

from the end of the vascular implant to the leading migrating SMCs along the axis of the implant at circumferential locations 0°, 55°, and 180° (Fig. 1). The leading SMCs were defined as those most remote from the end of the vascular implant, with the continuous presence of positive ␣-actin between the leading SMCs and the end of the implant (33).

Measurement of SMC Migration

An immunohistochemical method was used to detect phosphorylated ERK1/2 and expressed MLCK in endothelial cells and SMCs in the thrombus of the vascular implant. Selected specimens, which were fixed as described above, were cut into 10-␮m-thick transverse sec-

SMC migration was measured in the polymer implant model. At each observation time, the distance of SMC migration was measured AJP-Heart Circ Physiol • VOL

Immunohistochemistry

292 • FEBRUARY 2007 •

www.ajpheart.org


tions using a cryomicrotome and incubated with a 1:50 dilution of anti-phosphorylated-ERK1/2 antibody (New England BioLabs, Beverly, MA) or a 1:20 dilution of anti-MLCK antibody (Convance, Princeton, NJ) and then with a fluorochrome-conjugated secondary antibody. SMCs were identified in the same specimen by positive labeling of smooth muscle ␣-actin with an anti-smooth muscle ␣-actin antibody. Cell nuclei were labeled with Hoechst 33258. Specimens were examined using a fluorescence microscope (model BX40, Olympus). Immunoprecipitation and Immunoblotting The relative expression and relative phosphorylation of ERK1/2 were detected by an immunoblotting method. At specified times, four specimens from each rat were collected and homogenized in RIPA buffer (100 mM Tris, 0.15 M NaCl, 1% deoxycholic acid, 1% Triton X-100, 0.1% SDS, 10 ␮g/ml aprotinin, 2 mM phenylmethylsulfonyl fluoride, 10 ␮g/ml leupeptin, 1 mM Na3VO4, 2 ␮g/ml pepstatin, 50 mM NaF, and 5 mM EDTA) (23). The homogenate was centrifuged at 15,000 g for 6 min at 4°C, the supernatant was collected for immunoblotting, and the concentration of total protein was determined by the Bradford method. Protein samples of equal amounts (50 ␮g), collected at specified times, were resolved by SDS-PAGE and then transferred to a polyvinylidene difluoride membrane, which was incubated with a primary antibody specific to ERK1/2 or phosphorylated ERK1/2 (New England Biolabs) and then with a horseradish peroxidaseconjugated secondary antibody. Protein signals were detected using a chemiluminescent method. The ratio of the band density of the phosphorylated ERK1/2 to the band density of the expressed total ERK1/2 from the same specimen at each observation time was calculated and used to represent the relative phosphorylation activity of ERK1/2. The relative expression and phosphorylation of MLCK were detected using immunoprecipitation and immunoblotting approaches. The encapsulating tissue of the vascular implant (4 specimens from each rat) was harvested and homogenized, and total protein was prepared as described above. Immunoprecipitation was carried out by incubation of homogenate supernatant (100 ␮g total protein) with a 1:100 dilution of anti-MLCK antibody (Convance, Princeton, NJ) for 2 h at 4°C and then with 25 ␮g of protein A-Sepharose 2B beads (Sigma, St. Louis, MO) for 12 h at 4°C with constant rotation. After the protein-bead mixture was centrifuged at 200 g for 2 min, the beads were collected, washed three times, resuspended in Laemmli sample buffer, boiled for 5 min, and centrifuged at 200 g for 2 min (23), and the supernatant was collected for immunoblotting using the method described above. An anti-MLCK antibody (1:1,500 dilution) and an antiphosphorylated threonine-proline antibody (1:1,500 dilution; Cell Signaling, Beverly, MA) were used for the detection of the relative expression and the relative phosphorylation activity of MLCK, respectively (5). A relative index of phosphorylation activity was defined as the ratio of the band density of the phosphorylated MLCK to the band density of the expressed total MLCK from the same specimen at each observation time. Administration of Pharmacological Inhibitors To dissect the role of ERK1/2 and MLCK in regulating SMC migration, we delivered PD-98059, an inhibitor specific to the ERK1/2 upstream activator MEK1/2 (1), and ML-7, an inhibitor specific to MLCK (54), to the encapsulating tissue of the vascular implant via an osmotic pump (Alza, Palo Alto, CA). Such a pump has been successfully used in our previous studies for local substance delivery (45). Briefly, a 2-ml osmotic pump was filled with 10 ␮M PD-98095 or 10 ␮M ML-7 (both from Biomol, Plymouth Meeting, PA) in culture medium and connected to a vena cava branch upstream of the polymer implant through a segment of fine polyethylene tubing AJP-Heart Circ Physiol • VOL

H931

(45). The osmotic pump was placed in the peritoneal cavity and secured to the skeletal muscle. This pump can be used continuously for 10 days with a constant delivery rate of 8 ␮l/h. Inhibitor-treated specimens were collected at days 5 and 7 for data measurements with five animals at each time. Results were compared between animals with and without an inhibitor at each time. Statistics Means and standard deviations were calculated for measured parameters. Student’s t-test (2-tailed) was used to determine the significance of difference between two selected groups. One-way analysis of variance was used to determine the significance of difference between more than two groups. Statistical significance of difference was considered at P ⬍ 0.05. RESULTS

Vortex Blood Flow Promotes SMC Migration and Intimal Hyperplasia in Vein Grafts Interposition of the external jugular vein into the abdominal aorta resulted in expansion of the vein to a diameter that closely matched the host artery (89 ⫾ 10% relative to the host artery), whereas the engraftment of the common jugular veins produced grafts with mismatched diameters (148 ⫾ 8% relative to the host artery; Fig. 1). This unique geometry allowed us to produce a vein graft with mechanical stretch ⫹ vortex blood flow (mismatched diameter grafts) or a vein graft with mechanical stretch ⫹ reduced vortex blood flow (matched diameter grafts). By comparing these two models, it is possible to clarify the role of vortex blood flow and to determine its importance relative to mechanical stretch during SMC migration and intimal hyperplasia. After vein grafting, substantial intimal hyperplasia was seen in the vein grafts with vortex blood flow (Fig. 2A). In contrast, minimal intimal hyperplasia was seen in the vein grafts with reduced vortex blood flow (Fig. 2B). The total surface area covered by the SMCs in the neointima was significantly reduced in the diameter-matched vein grafts (0.014, 0.007, 0.030, and 0.016 mm2) compared with the diameter-mismatched vein grafts (0.086, 0.083, 0.096, and 0.064 mm2; Fig. 2C). These results demonstrate that vortex blood flow promotes SMC intimal hyperplasia in a vein graft. Distribution of Fluid Shear Stress on the Vascular Implant To clarify the role of fluid shear stress in regulating SMC migration, it is necessary to assess the distribution of fluid shear stress. Such an assessment is difficult to accomplish in the vein graft model but is straightforward in the vascular implant model, which has a defined blood flow field and implant geometry. In the vascular implant model, a nonuniform distribution of fluid shear stress was found on the vascular implant in the vena cava. Shear stress was zero along the leading and trailing stagnation edges at 0° and 180°, respectively, along the flow separation line at 109°, and along the intersection of the vascular implant with the vena cava wall (Fig. 1D). A maximal shear stress was found in the central region of the vascular implant at ⬃55°. Fluid shear stress in the vortex region cannot be assessed analytically but is significantly lower than in the laminar region (see Ref. 33 for detailed distribution of shear stress).

292 • FEBRUARY 2007 •

www.ajpheart.org

H932


Fig. 2. Identification of smooth muscle cells (SMCs) in neointima of vein grafts. SMCs were identified 30 days after surgery in 10-␮m vein graft cryosections by ␣-actin labeling (green). Cell nuclei were counterstained with 4⬘,6-diamidino2-phenylindole (blue). A: pronounced intimal hyperplasia in association with a high density of SMCs in control vein grafts. B: relatively few SMCs in the neointima of diameter-matched vein grafts, demonstrating that matching of graft and host artery diameters reduces intimal hyperplasia. White arrows in each image point to suture remnants, indicating artery-vein graft juncture. Yellow arrowhead identifies ␣-actin-positive SMCs in vein graft neointima. Blood flow is from right to left. C: total surface area covered by SMCs in vein graft neointima normalized to diameter-mismatched vein grafts. Surface area occupied by SMCs was markedly reduced in diameter-matched vein grafts relative to diameter-mismatched vein grafts: *P ⬍ 0.005. Scale bar, 100 ␮m. Sample size was 4 per group.

Influence of Fluid Shear Stress on SMC Migration We assessed the influence of fluid shear stress on SMC migration by comparing the distribution of fluid shear stress with SMC migration. After surgery, the vascular implant was rapidly encapsulated with a layer of thrombus. The thickness of the thrombus varied from 50 to 100 ␮m, depending on the location on the implant, with a minimal thickness at the center and a maximal thickness near the vena cava wall. SMC migration was found in the encapsulating thrombus from the end to the center of the vascular implant within 3 days after surgery. The pattern of SMC migration at the two ends of the polymer implant was relatively symmetrical. Few SMCs were found in the region between the leading SMCs from both ends of the implant. SMC migration stopped apparently at day 10, when the leading migrating SMCs from both ends met at the center of the implant. A comparison of SMC migration with the distribution of fluid shear stress on the vascular implant demonstrated that the maximal fluid shear stress (at 55°) on the vascular implant was associated with minimal SMC migration, whereas the minimal fluid shear stress along the leading (at 0°) and trailing (at 180°) stagnation edges was associated with increased SMC migration. The pattern of SMC migration was dependent on the distribution of fluid shear stress. Representative data are shown in Fig. 3. Statistical analyses showed a significant difference in AJP-Heart Circ Physiol • VOL

SMC migration between the maximal and minimal shear stress regions at days 3, 5, and 7 (Fig. 3). Influence of Fluid Shear Stress on Signaling Molecules To assess whether fluid shear stress influences the activity of signaling molecules, we used an immunohistochemical method

Fig. 3. Changes in relative distance of SMC migration at 3 circumferential locations, 0°, 55°, and 180°, from day 3 to day 7. Percentage of SMC migration distance was calculated with respect to radius of the vena cava. At day 10, vascular implant was fully covered with SMCs. Changes from day 3 to day 10 were statistically significant (P ⬍ 0.01, by ANOVA). Values are means and SD, and sample size is 5 at each time point. **P ⬍ 0.01; ***P ⬍ 0.001, by ANOVA, between the 3 locations at each time.

292 • FEBRUARY 2007 •

www.ajpheart.org


to qualitatively assess the phosphorylation of ERK1/2 in cells on the vascular implant. The distribution of the molecular activity was compared with the map of fluid shear stress. Phosphorylation of ERK1/2 was primarily found in SMCdominant regions. The maximal fluid shear stress at ⬃55° was associated with minimal phosphorylation of ERK1/2, whereas the minimal fluid shear stress at the leading and trailing edges (0° and 180°, respectively) was associated with increased activity of ERK1/2. Typical results are presented in Fig. 4. Role of ERK 1/2 and MLCK in Regulating SMC Migration To verify whether activated ERK1/2 and MLCK regulate SMC migration, we examined the expression and phosphorylation of these molecules in the entire thrombus of the vascular implant and assessed SMC migration in response to pharmacological inhibitors for ERK1/2 and MLCK. The relative phosphorylation of ERK1/2 at day 3 was similar to or slightly higher than that in the vena cava wall, increased significantly at day 5, and returned toward the level of the vena cava wall from day 5 to day 10 (Fig. 5). Changes from day 3 to day 10 were statistically significant, as detected by analysis of variance (P ⬍ 0.01). A similar time course was found for the relative phosphorylation activity of MLCK, which peaked also at day 5 (Fig. 6, A and B). Changes from day 3 to day 10 were

H933

statistically significant (P ⬍ 0.01). Immunohistochemical assays demonstrated that MLCK was primarily expressed in SMCs (Fig. 6C). The peak time of the phosphorylation activity of ERK1/2 and MLCK was consistent with the time at which a maximal speed of SMC migration was found. Investigations using cultured cells have demonstrated that MLCK is a phosphorylation target of ERK1/2. To demonstrate whether that is the case in the present model, we locally delivered PD-98059, a pharmacological inhibitor specific to the ERK1/2 upstream activator MEK1/2, to the vascular implant and examined the influence of this substance on the phosphorylation activity of ERK1/2 and MLCK. Immunohistochemistry demonstrated that PD-98059 suppressed the phosphorylation of ERK1/2 in SMCs on the vascular implant in zero or near-zero shear regions, but not in the maximal shear region (Fig. 4C). The relative phosphorylation activity of ERK1/2, as detected using an immunoblotting method, was reduced significantly in response to PD-98059 at days 5 and 7 (Fig. 7). Furthermore, the inhibition of ERK1/2 resulted in a significant decrease in the relative phosphorylation of MLCK at days 5 and 7 (Fig. 8). To assess the role of ERK1/2 and MLCK in the regulation of SMC migration, we locally delivered the MEK1/2 inhibitor PD-98059 and an MLCK inhibitor, ML-7, to the vascular

Fig. 4. Phosphorylation of ERK1/2 in regions with different fluid shear stress in histological and immunohistochemical micrographs of encapsulating thrombus of vascular implants at day 5. Transverse cryosections were collected from encapsulating thrombus at a point between the center and the end of the vascular implant. A: transverse histological section (hematoxylin and eosin staining) showing locations for collection of immunohistochemical images. Leading and trailing stagnation edges are 0° and 180°, respectively. Note streamlined profile of encapsulating thrombus, which was a result of fluid shear stress (33). Scale bar, 100 ␮m. B and C: distribution of phosphorylated ERK1/2 (red) in the 3 regions without and with PD-98059, respectively. Green, smooth muscle ␣-actin; blue, cell nuclei. Scale bar, 10 ␮m. EC, endothelial cells. AJP-Heart Circ Physiol • VOL

292 • FEBRUARY 2007 •

www.ajpheart.org

H934


found in the control implants without PD-98059 or ML-7 treatment. However, PD-98095 or ML-7 did not significantly influence SMC migration in regions with maximal shear stress. Such an influence resulted in an SMC distribution independent of fluid shear stress (Fig. 9). DISCUSSION

Intimal Hyperplasia in Vein Grafts Is Reduced by Minimizing Vortex Blood Flow

Fig. 5. A: Western blot analysis of expression (top) and phosphorylation (bottom) of ERK1/2. VC, vena cava wall. B: relative phosphorylation activity at various times. Sample size was 5 at each time point. Changes in relative phosphorylation activity of ERK1/2 were statistically significant (P ⬍ 0.01, by ANOVA).

implant model. The influence of these substances on SMC migration is shown in Fig. 9. Treatment with PD-98059 or ML-7 suppressed SMC migration significantly in regions with zero and near-zero shear stress, including the leading and trailing stagnation edges, where maximal SMC migration was

Vein grafts are commonly used to replace arteries that have malfunctioned. However, vein grafts fail as a result of progressive intimal hyperplasia. Conventional vein grafts are often associated with vortex blood flow because of diameter mismatch between the graft and host artery (29). Vortex blood flow has been hypothesized to promote SMC migration and proliferation (9, 28, 36, 37). To reduce vortex blood flow, the diameter mismatch can be reduced by deployment of an external stent around the graft, which restricts the graft diameter expansion (29). However, the external stent also reduces mechanical stretch, a biomechanical stimulus strongly implicated in vascular remodeling (2, 4, 6, 7, 15, 19, 26, 41, 42, 59). Such an approach concurrently reduces vortex blood flow and mechanical stretch and, therefore, does not allow the role of vortex blood flow to be distinguished from that of mechanical stretch. In the present study, we controlled vortex blood flow by selecting vein grafts with lumen diameter similar to that of the host aorta under arterial blood pressure. The difference between this model and an externally stented vein graft is that mechanical stretch is significantly reduced in the wall of the

Fig. 6. A: Western blot analysis for expression (top) and phosphorylation (bottom) of myosin light chain kinase (MLCK). IP, immunoprecipitation; IB, immunoblotting. B: relative phosphorylation activity at various times. Sample size was 5 at each time. Changes in relative phosphorylation activity of MLCK were statistically significant (P ⬍ 0.01, by ANOVA). C: expression of MLCK in SMCs in an immunohistochemical micrograph of a 5-day specimen. Red, MLCK; green, smooth muscle ␣-actin; blue, cell nuclei. Scale bar, 10 ␮m. AJP-Heart Circ Physiol • VOL

292 • FEBRUARY 2007 •

www.ajpheart.org


Fig. 7. Influence of PD-98059 on phosphorylation of ERK1/2 at days 5 and 7, when maximal SMC migration was observed. A: immunoblots for total and phosphorylated ERK1/2 with and without PD-98059. Numbers 5 and 7 indicate days 5 and 7. B: relative phosphorylation activity of ERK1/2 with and without PD-98059. **P ⬍ 0.01. Sample size was 5 at each time point.

reinforced vein grafts, whereas mechanical stretch is not altered in the wall of the vein grafts in the present study. The significant reduction of mechanical stretch between the reinforced and control vein graft models does not allow the contributions of vortex blood flow and mechanical stretch to be clearly assessed. To distinguish the role of vortex blood flow from mechanical stretch in the regulation of vein graft intimal hyperplasia, vortex blood flow must be minimized without restriction of mechanical stretch. Using two experimental vein graft models, we were able to decouple mechanical stretch and vortex blood

H935

Fig. 9. SMC migration in the absence and presence of PD-98059 and ML-7. A: 5-day specimen without an inhibitor. B: 5-day specimen with PD-98059 (10 ␮M). C: 5-day specimen with ML-7 (10 ␮M). ⴱ, Location for measurement of SMC migration; arrow, direction of SMC migration. Scale bar, 100 ␮m. D: influence of PD-98059 and ML-7 on SMC migration at 3 selected locations (0°, 55°, and 180°) on the vascular implant at days 5 and 7 after surgery. Sample size is 5 for each time point. *P ⬍ 0.05; **P ⬍ 0.01; and ***P ⬍ 0.001, PD-98059 vs. control. ##P ⬍ 0.01; ###P ⬍ 0.001, ML-7 vs. control.

flow. This is possible because grafting of an external jugular vein results in a vein graft diameter that closely matches the host artery diameter. This vein graft geometry was found to reduce intimal hyperplasia by ⬃80% relative to diametermismatched vein grafts. This important result demonstrates that altered fluid shear stress in vortex blood flow, along with other factors such as mechanical stretch and activated mitogenic factors, may contribute to SMC migration and neointimal formation. The presence of mechanical stretch in both vein graft models and the reduction in anastomotic intimal hyperplasia in the diameter-matched vein grafts suggest that blood shear stress may be more important than mechanical stretch in regulating focal intimal hyperplasia in a vein graft. Role of Fluid Shear Stress in Regulating SMC Migration

Fig. 8. Influence of PD-98059 on phosphorylation of MLCK at days 5 and 7, when maximal SMC migration was observed. A: immunoblots for total and phosphorylated MLCK with and without PD-98059. B: relative phosphorylation activity of MLCK with and without PD-98059. **P ⬍ 0.01. Sample size was 5 at each time point. AJP-Heart Circ Physiol • VOL

To understand the role of blood shear stress, it is necessary to assess the level and distribution of shear stress. However, it is difficult to measure and analyze blood shear stress in the vein graft model. Therefore, a vascular implant model with a defined flow field and blood shear stress was introduced. Because of its circular geometry, the vascular implant was subject to a laminar flow of 0 to ⬃30 N/m2 in the leading region with known shear stress and a vortex flow in the trailing region. As observed in this study, blood flow-associated fluid shear stress influences the migration of vascular SMCs. The speed of SMC migration in regions with zero and near-zero shear stress (at 0° and 180°) was significantly higher than that in regions with higher shear stress (at 55°). These observations suggest that fluid shear stress suppresses SMC migration. In addition, as shown in a recent study, fluid shear stress inhibits SMC proliferation in the thrombus of a vascular implant (38). These findings support our results with minimized vortex blood flow in a vein graft.

292 • FEBRUARY 2007 •

www.ajpheart.org

H936


Role of ERK1/2 and MLCK in Mediating Shear Stress-Dependent SMC Migration The present study demonstrates that, in the vascular implant model, a zero and near-zero shear stress is associated with phosphorylation of ERK1/2 in SMC-dominant regions, whereas the maximal shear stress is associated with relatively reduced activity of ERK1/2. These results suggest that fluid shear stress inhibits the phosphorylation of ERK1/2, although it is difficult to quantitatively assess the influence of fluid shear stress. In the present model, fluid shear stress interacts directly with endothelial cells, but not with SMCs, suggesting that shear stress-dependent activity of ERK1/2 in SMCs is regulated by mediating factors between the endothelial cell and SMC. Recent studies have demonstrated that fluid shear stress regulates the expression of PDGF-BB in endothelial cells in a vascular implant model (37). PDGF-BB has been shown to play a role in the activation of ERK1/2-related signaling pathways (52). Thus it is possible that shear stress-dependent release of PDGF-BB from endothelial cells influences the activity of ERK1/2 in SMCs. A previous investigation has shown that ERK1/2 may regulate the migration of cultured carcinoma cells via the mediation of MLCK (23). To verify whether this is the case in the present model, we compared the time course between ERK1/2 phosphorylation and MLCK phosphorylation and examined the influence of ERK1/2 inhibition on the activity of MLCK. These observations showed a similarity between the time courses of ERK1/2 and MLCK phosphorylation and demonstrated a suppression of MLCK in response to the inhibition of ERK1/2, suggesting that ERK1/2 possibly regulates the activity of MLCK in the present model. To verify whether the ERK1/2-MLCK signaling mechanism plays a role in regulating shear stress-dependent SMC migration, we compared the time course of ERK1/2 and MLCK phosphorylation with that of SMC migration and showed that the peak relative phosphorylation activity of ERK1/2 and MLCK was consistent with maximal SMC migration at day 5. Furthermore, treatment with PD-98059 (an inhibitor for the ERK1/2 activator MEK1/2) or ML-7 (an inhibitor for MLCK) significantly suppressed shear stress-dependent SMC migration, suggesting that shear stress possibly regulates SMC migration via the mediation of ERK1/2 and MLCK. It is interesting to note that the inhibition of ERK1/2 and MLCK leads to a significant suppression of SMC migration in regions with zero and near-zero shear stress but not in regions with maximal shear stress. A possible reason for this phenomenon is that the maximal shear stress (⬃30 N/m2), which is 10 –30 times higher than the physiological level of shear stress in the vascular system, may completely suppress the activity of ERK1/2 and MLCK. Thus further treatment with a chemical inhibitor may no longer be effective in the maximal shear stress region. In contrast, in regions with zero or near-zero shear stress, the activity of ERK1/2 and MLCK was significantly increased. Treatment with selected chemical inhibitors can effectively suppress the activity of these molecules, leading to an apparent reduction in SMC migration. This possibility is supported by results presented in Fig. 4C. The shear-selective influence of ERK1/2 and MLCK inhibitors reduced the difference in SMC migration between regions with minimal and maximal shear stress (Fig. 9), resulting in a shear stress-independent distribuAJP-Heart Circ Physiol • VOL

tion of SMCs. It should be pointed out that, after treatment with PD-98059 or ML-7, SMC migration was still observed, with a reduced rate and altered pattern. These results suggest that factors other than ERK1/2 and MLCK are possibly involved in the regulation of SMC migration. On the basis of these observations and previous investigations, we propose the following hypothetical mechanism, by which fluid shear stress regulates SMC migration. Reduced shear stress from a physiological level may upregulate growth factors in the endothelial cells, and growth factors may be released into extracellular space, act on cognate receptors in SMCs, and activate the ERK1/2-MLCK signaling mechanism, which in turn enhances SMC migration. In contrast, an increase in fluid shear stress may exert an opposite effect. However, it remains to be determined how fluid shear stress upregulates growth factors in the vascular cells. Furthermore, it should be noted that mitogenic molecules in the thrombus-like tissue may also contribute to SMC migration and proliferation. Cell Sources of ERK1/2 and MLCK The encapsulating tissue of the vascular implant contains several cell types, including SMCs, endothelial cells, and leukocytes (33), which are possible sources of phosphorylated ERK1/2 and MLCK. As shown in this study, the phosphorylation activity of ERK1/2 and MLCK was apparently low in the encapsulating tissue at day 3, when leukocytes were the dominant cell type in the thrombus, suggesting that leukocytes were probably not a major source of activated ERK1/2 and MLCK. The increase in the phosphorylation activity of ERK1/2 and MLCK at days 5 and 7 was consistent with the increase in the content of SMCs. Although the vascular implant was gradually covered with endothelial cells after day 5, the content of the monolayered endothelial cells was much less than that of the SMCs (38). Thus SMCs were likely a major source of activated ERK1/2 and MLCK. This was further verified by immunohistochemical labeling (Figs. 4B and 5C). Origin of SMCs in the Thrombus of the Vascular Implant In the model used in this study, SMC migration was found from the end to the center of the vascular implant. There are two possible origins for the SMC: vascular wall and bloodborne SMC progenitor cells. Blood-borne SMC progenitor cells derived from the bone marrow are capable of attaching to a polymer implant during thrombus formation. These cells have a potential for transformation into SMC-like cells (57). In the present study, smooth muscle ␣-actin-positive cells initiated migration from the end of the vascular implant, with a progressive advance toward the center of the implant. This observation suggests that the vena cava wall was a possible origin of SMCs. Little evidence has been obtained for the transformation of blood-borne cells into SMC-like cells in the present model. Physiological Relevance of the Implant Model In the present study, we used a novel experimental model to study the regulatory mechanisms of vascular SMC migration. A critical question is why we need such a model. In previous studies, SMC migration in shear flow has been examined using cell culture models (49). Although these investigations provide useful information about the mechanisms of SMC migration,

292 • FEBRUARY 2007 •

www.ajpheart.org


little in vitro information has been integrated into and verified in animal experimental models. This is probably because of 1) difficulties in assessing the distribution of fluid shear stress in vivo, 2) difficulties in the measurement of vascular SMC migration in vivo, and 3) difficulties in the dissection of the role of regulatory molecules in vivo. The model presented here may be used to overcome these difficulties. The model used in this study is an artificial system. A critical question is whether this model can be used to provide insights into the mechanistic aspects of vascular SMC migration. Our recent investigations have demonstrated that the process of thrombus formation in this model is similar to that in an experimental vein graft model (29, 31). After surgery, blood cells rapidly attach to the vascular implant and form a thrombus tissue, which serves as a ground for vascular cell migration. In addition, the structure of and cell types in the thrombus and the rate of cell proliferation in various stages of the present model were similar to those in the experimental vein graft model (29, 31, 38). These results suggest that the present model can be used to simulate the remodeling process of vascular cells in a known shear stress field under realistic vascular conditions. GRANTS This work was supported by grants from the American Heart Association, the National Science Foundation, and the National Institutes of Health. REFERENCES 1. Alessi DR, Cuenda A, Cohen P, Dudley DT, Saltiel AR. PD 098059 is a specific inhibitor of the activation of mitogen-activated protein kinase kinase in vitro and in vivo. J Biol Chem 270: 27489 –27494, 1995. 2. Bunkenburg B, van Amelsvoort T, Rogg H, Wood JM. Receptormediated effects of angiotensin II on growth of vascular smooth muscle cells from spontaneously hypertensive rats. Hypertension 20: 746 –754, 1992. 3. Castier Y, Brandes RP, Leseche G, Tedgui A, Lehoux S. p47phoxDependent NADPH oxidase regulates flow-induced vascular remodeling. Circ Res 97: 533–540, 2005. 4. Cipolla MJ, Osol G. Vascular smooth muscle actin cytoskeleton in cerebral artery forced dilatation. Stroke 29: 1223–1228, 1998. 5. Clark-Lewis I, Sanghera JS, Pelech S. Definition of a consensus sequence for peptide substrate recognition by p44mapk, the meiosis-activated myelin basic protein kinase. J Biol Chem 266: 15180 –15184, 1991. 6. Clemow DB, Steers WD, Tuttle JB. Stretch-activated signaling of nerve growth factor secretion in bladder and vascular smooth muscle cells from hypertensive and hyperactive rats. J Cell Physiol 183: 289 –300, 2000. 7. Cooper GT, Kent RL, Mann DL. Load induction of cardiac hypertrophy. J Mol Cell Cardiol 21 Suppl 5: 11–30, 1989. 8. Cornelissen J, Armstrong J, Holt CM. Mechanical stretch induces phosphorylation of p38-MAPK and apoptosis in human saphenous vein. Arterioscler Thromb Vasc Biol 24: 451– 456, 2004. 9. Cunningham KS, Gotlieb AI. The role of shear stress in the pathogenesis of atherosclerosis. Lab Invest 85: 9 –23, 2005. 10. Force T, Pombo CM, Avruch JA, Bonventre JV, Kyriakis J. Stressactivated protein kinases in cardiovascular disease. Circ Res 78: 947–953, 1996. 11. Frangos SG, Gahtan V, Sumpio B. Localization of atherosclerosis: role of hemodynamics. Arch Surg 134: 1142–1149, 1999. 12. Fung MC. Biomechanics: Motion, Flow Stress, and Growth. New York: Springer-Verlag, 1990, p. 499 –546. 13. Gahtan V, Wang XJ, Willis AI, Tuszynski GP, Sumpio B. Thrombospondin-1 regulation of smooth muscle cell chemotaxis is extracellular signal-regulated protein kinases 1/2 dependent. Surgery 126: 203–207, 1999. 14. Giese NA, Marijianowski MM, McCook O, Hancock A, Ramakrishnan V, Fretto LJ, Chen C, Kelly AB, Koziol JA, Wilcox JN, Hanson SR. The role of ␣- and ␤-platelet-derived growth factor receptor in the vascular response to injury in nonhuman primates. Arterioscler Thromb Vasc Biol 19: 900 –909, 1999. AJP-Heart Circ Physiol • VOL

H937

15. Goldman J, Zhong L, Liu SQ. Degradation of ␣-actin filaments in venous smooth muscle cells in response to mechanical stretch. Am J Physiol Heart Circ Physiol 284: H1839 –H1847, 2003. 16. Grosskreutz CL, Anand-Apte B, Duplaa C, Quinn TP, Terman BI, Zetter B, D’Amore P. Vascular endothelial growth factor-induced migration of vascular smooth muscle cells in vitro. Microvasc Res 58: 128 –136, 1999. 17. Gudi S, Nolan JP, Frangos J. Modulation of GTPase activity of G proteins by fluid shear stress and phospholipid composition. Proc Natl Acad Sci USA 95: 2515–2519, 1998. 18. Gusic RJ, Myung R, Petko M, Gaynor JW, Gooch KJ. Shear stress and pressure modulate saphenous vein remodeling ex vivo. J Biomech 38: 1760 –1769, 2005. 19. Hu Y, Bock G, Wick G, Xu Q. Activation of PDGF receptor-␣ in vascular smooth muscle cells by mechanical stress. FASEB J 12: 1135– 1142, 1998. 20. Hungerford JE, Little C. Developmental biology of the vascular smooth muscle cell: building a multilayered vessel wall. J Vasc Res 36: 2–27, 1999. 21. Jalali S, Li YS, Sotoudeh M, Yuan S, Li S, Chien S, Shyy J. Shear stress activates p60src-Ras-MAPK signaling pathways in vascular endothelial cells. Arterioscler Thromb Vasc Biol 18: 227–234, 1998. 22. Kenagy RD, Hart CE, Stetler-Stevenson WG, Clowes A. Primate smooth muscle cell migration from aortic explants is mediated by endogenous platelet-derived growth factor and basic fibroblast growth factor acting through matrix metalloproteinases 2 and 9. Circulation 96: 3555– 3560, 1997. 23. Klemke RL, Cai S, Giannini AL, Gallagher PJ, de Lanerolle P, Cheresh D. Regulation of cell motility by mitogen-activated protein kinase. J Cell Biol 137: 481– 492, 1997. 24. Langille B. Arterial remodeling: relation to hemodynamics. Can J Physiol Pharmacol 74: 834 – 841, 1996. 25. Lehoux S, Castier Y, Tedgui A. Molecular mechanisms of the vascular responses to haemodynamic forces. J Intern Med 259: 381–392, 2006. 26. Li C, Hu Y, Mayr M, Xu Q. Cyclic strain stress-induced mitogenactivated protein kinase (MAPK) phosphatase 1 expression in vascular smooth muscle cells is regulated by Ras/Rac-MAPK pathways. J Biol Chem 274: 25273–25280, 1999. 27. Li C, Xu Q. Mechanical stress-initiated signal transductions in vascular smooth muscle cells. Cell Signal 12: 435– 445, 2000. 28. Liu SQ. Biomechanical basis of vascular tissue engineering. Crit Rev Biomed Eng 27: 75–148, 1999. 29. Liu SQ. Focal expression of angiotensin II type 1 receptor and smooth muscle cell proliferation in the neointima of experimental vein grafts: relation to eddy blood flow. Arterioscler Thromb Vasc Biol 19: 2630 – 2639, 1999. 30. Liu SQ. Influence of tensile strain on smooth muscle cell orientation in rat blood vessels. J Biomech Eng 120: 313–320, 1998. 31. Liu SQ. Prevention of focal intimal hyperplasia in rat vein grafts by using a tissue engineering approach. Atherosclerosis 140: 365–377, 1998. 32. Liu SQ, Fung YC. Changes in the organization of the smooth muscle cells in rat vein grafts. Ann Biomed Eng 26: 86 –95, 1998. 33. Liu SQ, Goldman J. Role of blood shear stress in the regulation of vascular smooth muscle cell migration. IEEE Trans Biomed Eng 48: 474 – 483, 2001. 34. Liu SQ, Moore MM, Glucksberg MR, Mockros LF, Grotberg JB, Mok AP. Partial prevention of monocyte and granulocyte activation in experimental vein grafts by using a biomechanical engineering approach. J Biomech 32: 1165–1175, 1999. 35. Liu SQ, Moore MM, Yap C. Prevention of mechanical stretch-induced endothelial and smooth muscle cell injury in experimental vein grafts. J Biomech Eng 122: 31–38, 2000. 36. Liu SQ, Tang D, Tieche C, Alkema PK. Pattern formation of vascular smooth muscle cells subject to nonuniform fluid shear stress: mediation by gradient of cell density. Am J Physiol Heart Circ Physiol 285: H1072– H1080, 2003. 37. Liu SQ, Tieche C, Tang D, Alkema P. Pattern formation of vascular smooth muscle cells subject to nonuniform fluid shear stress: role of PDGF-␤ receptor and Src. Am J Physiol Heart Circ Physiol 285: H1081– H1090, 2003. 38. Liu SQ, Zhong L, Goldman J. Control of the shape of a thrombusneointima-like structure by blood shear stress. J Biomech Eng 124: 30 –36, 2002.

292 • FEBRUARY 2007 •

www.ajpheart.org

H938


39. Majesky MW, Reidy MA, Bowen-Pope DF, Hart CE, Wilcox JN, Schwartz S. PDGF ligand and receptor gene expression during repair of arterial injury. J Cell Biol 111: 2149 –2158, 1990. 40. Malek AM, Gibbons GH, Dzau YJ, Izumo S. Fluid shear stress differentially modulates expression of genes encoding basic fibroblast growth factor and platelet-derived growth factor B chain in vascular endothelium. J Clin Invest 92: 2013–2021, 1993. 41. Mayr M, Hu Y, Hainaut H, Xu Q. Mechanical stress-induced DNA damage and rac-p38MAPK signal pathways mediate p53-dependent apoptosis in vascular smooth muscle cells. FASEB J 16: 1423–1425, 2002. 42. Mayr M, Li C, Zou Y, Huemer U, Hu Y, Xu Q. Biomechanical stress-induced apoptosis in vein grafts involves p38 mitogen-activated protein kinases. FASEB J 14: 261–270, 2000. 43. McIntire LV, Wagner JE, Papadaki M, Whitson PA, Eskin S. Effect of flow on gene regulation in smooth muscle cells and macromolecular transport across endothelial cell monolayers. Biol Bull 194: 394 –399, 1998. 44. Mondy JS, Lindner V, Miyashiro JK, Berk BC, Dean RH, Geary R. Platelet-derived growth factor ligand and receptor expression in response to altered blood flow in vivo. Circ Res 81: 320 –327, 1997. 45. Moore MM, Goldman J, Patel AR, Chien S, Liu SQ. Role of tensile stress and strain in the induction of cell death in experimental vein grafts. J Biomech 34: 289 –297, 2001. 46. Nelson PR, Yamamura S, Kent K. Platelet-derived growth factor and extracellular matrix proteins provide a synergistic stimulus for human vascular smooth muscle cell migration. J Vasc Surg 26: 104 –112, 1997. 47. Nelson PR, Yamamura S, Mureebe L, Itoh H, Kent KC. Smooth muscle cell migration and proliferation are mediated by distinct phases of activation of the intracellular messenger mitogen-activated protein kinase. J Vasc Surg 27: 117–125, 1998. 48. Osterberg K, Mattsson E. Intimal hyperplasia in mouse vein grafts is regulated by flow. J Vasc Res 42: 13–20, 2005. 49. Palumbo R, Gaetano C, Melillo G, Toschi E, Remuzzi A, Capogrossi MC. Shear stress downregulation of platelet-derived growth factor receptor-␤ and matrix metalloprotease-2 is associated with inhibition of smooth muscle cell invasion and migration. Circulation 102: 225–230, 2000.

AJP-Heart Circ Physiol • VOL

50. Pickering JG, Uniyal S, Ford CM, Chau T, Laurin MA, Chow LH, Ellis Christopher G, Fish J, Chan B. Fibroblast growth factor-2 potentiates vascular smooth muscle cell migration to platelet-derived growth factor: upregulation of ␣2␤1-integrin and disassembly of actin filaments. Circ Res 80: 627– 637, 1997. 51. Resnick N, Collins T, Atkinson W, Bonthron DT, Dewey CF Jr, Gimbrone MJ. Platelet-derived growth factor B chain promotor contains a cis-acting fluid shear-tress responsive element. Proc Natl Acad Sci USA 90: 4591– 4595, 1993. 52. Robinson MJ, Cobb M. Mitogen-activated protein kinase pathways. Curr Opin Cell Biol 9: 180 –186, 1997. 53. Ross R. Atherosclerosis—an inflammatory disease. N Engl J Med 340: 115–126, 1999. 54. Saitoh M, Ishikawa T, Matsushima S, Naka M, Hidaka H. Selective inhibition of catalytic activity of smooth muscle myosin light chain kinase. J Biol Chem 262: 7796 –7801, 1987. 55. Schlichting H. Boundary-Layer Theory. New York: McGraw-Hill, 1968, p. 149 –186. 56. Schwartz SM, deBlois D, O’Brien R. The intima. Soil for atherosclerosis and restenosis. Circ Res 77: 445– 465, 1995. 57. Shimizu K, Sugiyama S, Aikawa M, Fukumoto Y, Rabkin E, Libby P, Mitchell R. Host bone-marrow cells are a source of donor intimal smooth muscle-like cells in murine aortic transplant arteriopathy. Nat Med 7: 738 –741, 2001. 58. Tseng H, Peterson TE, Berk B. Fluid shear stress stimulates mitogenactivated protein kinase in endothelial cells. Circ Res 77: 869 – 878, 1995. 59. Wernig F, Mayr M, Xu Q. Mechanical stretch-induced apoptosis in smooth muscle cells is mediated by ␤1-integrin signaling pathways. Hypertension 41: 903–911, 2003. 60. Wilson E, Mai Q, Sudhir K, Weiss RH, Ives HE. Mechanical strain induces growth of vascular smooth muscle cells via autocrine action of PDGF. J Cell Biol 123: 741–747, 1993. 61. Zou Y, Hu Y, Metzler B, Xu Q. Signal transduction in arteriosclerosis: mechanical stress-activated MAP kinases in vascular smooth muscle cells. Int J Mol Med 1: 827– 834, 1998.

292 • FEBRUARY 2007 •

www.ajpheart.org