The involvement of vascular fibroblasts (FBs) and smooth muscle (SM)-like cells in physiological and pathological processes in large vessels (intimal hyperplasia) and microvessels (capillary arterialization), and the realization that these cells are exposed to interstitial flow shear stress (SS), motivate this study of SS on FB migratory activity. Rat adventitial FBs were grown to either 30–50% confluence (subconfluent FBs; SFBs) or full confluence (confluent FBs; CFBs) in culture. Immunofluorescence and Western blotting assays were conducted to evaluate the expression of two phenotype markers: SM α-actin and SM myosin heavy chain (MHC). Both assays indicated a significant increase in SM α-actin expression in CFBs compared with SFBs, suggesting a phenotype difference between the two cell populations. SFBs and CFBs both expressed minimal SM MHC. Both cell populations were seeded on Matrigel-coated cell culture inserts and exposed to 4 h of either 1 or 20 dyn/cm2 SS via a rotating disk apparatus in the presence of the chemoattractant platelet-derived growth factor-BB to quantify the effect of SS on SFB and CFB migration. Four hours of 20 dyn/cm2 SS significantly enhanced SFB migration while it suppressed CFB migratory activity. Four hours of 1 dyn/cm2 SS did not significantly alter either SFB or CFB migration levels. Because of the distinct migratory responses of SFBs and CFBs in response to SS, phenotype modulation appears to be one way to regulate their involvement in both physiological and pathological remodeling processes.
- vascular fibroblasts
- interstitial flow
the influences of mechanical stimuli on vascular fibroblast (FB) function have not been as well characterized as on endothelial cells (ECs) and smooth muscle cells (SMCs) because FBs are contained in the outer adventitial layer of the blood vessel, which has historically been considered a supporting tissue (18). Recent in vivo work, however, has shown that, in addition to SMCs, adventitial FBs contribute to intimal lesion formation following vascular injury in large vessels (6, 17, 20, 21, 24). A majority of these cells transform to smooth muscle (SM)-like cells, myofibroblasts, in part through the acquisition of SM-specific α-actin (20, 21, 32), one of the proteins displayed earliest during SM differentiation (9), but not SM myosin heavy chain (MHC), a protein found only in fully differentiated, mature SMCs. FBs are sparsely populated in the adventitia but come closer to other FBs and medial SMCs as they migrate to the intima following injury. We hypothesize that the proximity of FBs to neighboring cells plays a role in modulating their conversion to myofibroblasts through the expression of SM α-actin.
Intimal lesion development is pronounced in regions where the endothelium has been damaged by vascular procedures such as angioplasty or at the anastomoses of vascular grafts (8, 19). Recent experimental work has shown that endothelial denudation results in an ∼75% increase in aortic hydraulic conductivity (22). This increase leads to elevated transvascular, interstitial flow that we hypothesize contributes to intimal hyperplasia. Modeling studies have shown that 1) interstitial flow driven by the transmural pressure gradient imposes fluid shear stresses on SMCs that are on the same order of magnitude as those experienced by ECs due to luminal blood flow, and 2) the SMC layer adjacent to the porous internal elastic lamina is exposed to shear stresses an order of magnitude greater than underlying SMC layers (27–29, 31). Although, to our knowledge, there have been no experimental or computational studies conducted to evaluate the order of shear stress (SS) that adventitial FBs are subjected to in an artery, we hypothesize that in the case of endothelial injury, FBs adjacent to the porous external elastic lamina are exposed to elevated shear stresses that affect their migratory activity. SMCs have displayed reduced migratory activity in response to elevated blood flow in a balloon catheter injury model in vivo (5) and have also demonstrated inhibition of migration in response to fluid SS in vitro (3, 10). There have been, to our knowledge, no such in vitro studies conducted with FBs to determine the role of interstitial flow SS on their migration, potentially a contributing factor to their accumulation in the intima following vascular injury.
Studies conducted at the microcirculatory level have reported that interstitial FBs participate in the arterialization of capillaries that connect two high pressure arteriolar trees, allowing for regulation of local blood flow to tissues during processes such as exercise hypertrophy and wound healing (12–16, 25, 26). The involvement of FBs in arterialization is thought to be regulated in part by elevated circumferential wall stresses in the arterializing vessel due to elevated luminal pressure compared with other capillaries. These elevated stresses can induce secretion of growth factors by ECs of the vessel wall, which recruit FBs located in the interstitium adjacent to neighboring arterioles. Once in the vicinity of the arterializing capillary, these FBs differentiate to a SM-like phenotype through the expression of SM α-actin (12–16, 25, 26). An alternate hypothesis relating hemodynamics to FB migration during arterialization is that elevated luminal pressure in the arterializing vessel drives enhanced interstitial flow imposing elevated fluid SS on adjacent FBs, and it is this force that directly stimulates these cells to migrate to the vicinity of the vessel. The expression of SM α-actin by FBs as they approach the arterializing vessel and come into closer contact with neighboring cells suggests that the degree of “confluence” of FBs may affect their expression of SM-specific phenotype markers.
The involvement of different vascular fibroblast phenotypes, distinguished in part through the expression of SM α-actin, in arterialization and intimal hyperplasia motivated the present study. Adventitial FBs were isolated from rat aortas and maintained in either subconfluent or confluent culture conditions. Both cell populations were examined for their expression of two SM phenotype markers, SM α-actin and SM MHC. The migratory response of both cell populations to SS was then determined by using a protocol established previously to study rat aortic SMC migration (3). In this manner, we were able to test our working hypothesis that two distinct populations of adventitial FBs display markedly different migratory responses to SS. We observed that subconfluent FBs (SFBs) were deficient in SM α-actin and SM MHC while their migration was enhanced by SS. Confluent FBs (CFBs) expressed abundant SM α-actin and minimal SM MHC, but their migration was suppressed by SS, much like fully differentiated SMCs (3, 10).
MATERIALS AND METHODS
Medial SMC and adventitial FB isolation and culture.
Rat medial SMCs and adventitial FBs were isolated from the thoracic aortas of male Sprague-Dawley rats weighing 150–200 g. After the removal of outer adherent fat and connective tissue layers, aortas were incubated at 37°C, 5% CO2-95% air in an enzyme solution composed of 1 mg/ml collagenase II, 1 mg/ml soybean trypsin inhibitor, 0.25 mg/ml elastase (all from Worthington Biochemical), and 100 U/ml penicillin, 100 μg/ml streptomycin (1% P/S, Sigma) in Hank's balanced salt solution (Irvine Scientific). The outer adventitial layer was then removed and placed in a fresh quantity of the enzyme solution described above. The endothelial lining was then gently removed from the remaining tissue pieces with a curved forceps, and the remaining medial layer was placed into a fresh quantity of the enzyme solution described above. After the respective incubations, cells were isolated from medial and adventitial tissue sections and seeded in separate tissue culture flasks with Dulbecco's modified Eagle's medium/F12 (DMEM/F12, Sigma) supplemented with 20% fetal bovine serum (FBS, HyClone) and 1% P/S serving as the culture media. SMCs and FBs from the thoracic aortas of five rats were pooled together and subsequently expanded in primary culture. FBs were maintained at 30–50% confluence and SMCs at 80–90% confluence in culture flasks until reaching passage 3 at which point they were stored in liquid nitrogen. SMCs were thawed and maintained at 80–90% confluence with DMEM/F12 + 10% FBS + 1% P/S serving as the culture media. FBs to be used as subconfluent FBs (SFBs) were thawed and allowed to grow to 30–50% confluence in a culture flask. Once this confluence level was reached, cells were either used in experiments or split 1:6 and allowed to again reach 30–50% confluence. FBs to be used as CFBs were thawed and allowed to grow to 100% confluence in a culture flask. When this confluence level was reached, cells were either used in experiments or split 1:3 and allowed to again reach 100% confluence. SFBs and CFBs were used at the same passage in all companion experiments. This protocol was developed in a preliminary study as a method to allow each FB subpopulation to spend a similar amount of time in culture before use in experiments. DMEM/F12 + 10% FBS + 1% P/S served as the culture media in all cases. SFBs and CFBs were characterized by their expression of two proteins, SM α-actin and SM MHC. Passages 4–9 were used in all experiments. This procedure was approved by the City College/City University of New York Medical School Institutional Animal Care and Use Committee.
Transwell filter supports (24.5 mm diameter) were attached to 75 × 38-mm glass slides (Corning Glass Works) with the Sylgard silicone elastomer kit (Dow Corning), and slides were sterilized via overnight exposure to ultraviolet light. SFBs or CFBs were then plated on the slides and maintained in DMEM/F12 + 10% FBS + 1% P/S until reaching the appropriate level of confluence. The same protocol, outlined elsewhere (2), was followed for evaluating SM α-actin and SM MHC expression in both cell populations. Culture media were removed and cells were rinsed in Dulbecco's PBS (Mediatech) for 5 min. Cells were then fixed in a 4% paraformaldehyde solution (Fisher Scientific) for 10 min, permeabilized with 0.2% Triton X-100 (Sigma) for 30 min, and blocked with 2% bovine serum albumin (Sigma; blocking solution) for 30 min. Cells were next incubated with either a monoclonal mouse SM α-actin antibody (Sigma) diluted in blocking solution to a final concentration of 10 μg/ml or a monoclonal mouse anti-rat MHC antibody (Santa Cruz Biotechnology) diluted 1:100 in blocking solution for 90 min. In all experiments, two slides were exposed to either the SM α-actin or SM MHC antibody as described above (experimental slides), and one slide was exposed to blocking solution containing neither antibody (negative control slide). Cells were then incubated in darkness for 1 h with an Alexa Fluor 488 goat anti-mouse IgG secondary antibody (Molecular Probes; approximate maximum excitation wavelength = 495 nm, approximate maximum emission wavelength = 519 nm) diluted to a final concentration of 10 μg/ml in blocking solution. Cells were rinsed with phosphate-buffered saline (PBS) for 1 min between each step, and all incubations were carried out at room temperature. After the final incubation, cells were rinsed in PBS for 5 min, and a drop of Fluoromount-G mounting media (Southern Biotechnology Associates) was added to slides, followed by a glass coverslip. Slides were stored at 4°C until fluorescent intensity levels, corresponding to the amount of either SM α-actin or SM MHC expression, were quantified. Three sets of experiments, each containing two experimental slides (n = 6 slides total) and one negative control slide (n = 3 total), were conducted for each combination of cell population and phenotype marker.
Fluorescent intensity measurements.
Slides were observed with a Nikon Eclipse TE2000-E inverted microscope. Four random ×10 fields were chosen per slide, and both brightfield and fluorescent images of each field were captured using a Photometrics Cascade 650 camera (Roper Scientific) connected to the microscope. Images were imported into the MetaVue version 6.2r2 imaging software (Universal Imaging) where fluorescent intensity levels were determined. To evaluate fluorescent intensity in slides containing SFBs, the boundary of each cell in the brightfield image was manually traced and copied to the corresponding fluorescent image where the average intensity in that region was determined using the MetaVue software. To evaluate fluorescent intensity in slides containing CFBs, brightfield images were first used to confirm that each field contained negligible cell-free space. The average intensity in the corresponding fluorescent image was then determined using the MetaVue software. Fluorescent intensity values obtained for all experimental conditions were stored in text files, which were later imported into MS Excel for analysis. Measurements were conducted in an identical manner for experimental and negative control slides. For each experiment, the average intensity value obtained from the negative control slide provided a background fluorescence level that was subtracted from values for corresponding experimental slides to determine a net fluorescent intensity value. For each cell population and phenotype marker, fluorescent intensity values obtained from all experiments were averaged to determine an overall net fluorescence intensity.
Western blotting was done to further quantify SM α-actin and SM MHC content in SFBs and CFBs relative to SMCs. For each cell type, cells in T-75 culture flasks were washed twice with ice-cold PBS with Ca2+ and lysed in a SDS extraction buffer (0.2% SDS, 100 mM NaCl, 1% Triton X-100, 0.5% deoxycholic acid, 2 mM EDTA, 10 mM HEPES, 10 mM NaF, 1 mM NaVO4, 1 mM benzamidine, and 1 mM phenylmethylsulfonyl fluoride). Lysate was centrifuged in a microfuge (14,000 rpm for 10 min at 4°C) following which the supernatant was collected, and the remaining pellets were discarded. Protein concentrations in supernatant were evaluated through the use of a total protein assay (Bio-Rad). Equal amounts of protein were loaded onto 10% SDS-polyacrylamide gels. After electrophoresis, proteins were transferred to Immun-blot polyvinylidene difluoride (PVDF) membranes (Bio-Rad) and blocked at room temperature with 2% (wt/vol) enhanced chemiluminescence (ECL) Advance Blocking Agent (Amersham) in Tris-buffered saline containing 0.1% Tween 20. Each membrane was then cut into two sections to separate SM α-actin and SM MHC bands according to the positions of protein markers. Sections containing SM MHC bands were incubated overnight with a 1:100 dilution of a mouse monoclonal SM-MHC antibody raised against SM MHC (Santa Cruz Biotechnology) followed by a 1-h room temperature incubation with an ECL horesradish peroxidase-linked anti-mouse IgG antibody (from sheep; Amersham) diluted 1:1,000. Sections containing SM α-actin were incubated with a 1:10,000 dilution of a mouse monoclonal α-smooth muscle actin antibody (Sigma) for 1 h followed by a 1-h room temperature incubation with an ECL horseradish peroxidase-linked anti-mouse IgG antibody (from sheep; Amersham) diluted 1:10,000. SM α-actin or SM MHC content on the respective membrane was then detected through the use of an ECL advanced Western blotting detection kit (Amersham) and the ChemiDoc XRS system. Content levels were analyzed with the Quantity One software (Bio-Rad). SFB and CFB SM α-actin and SM MHC levels were normalized by the level of the respective protein in SMCs for each set of culture flasks examined.
Cell culture inserts with 8.0-μm pores (Becton Dickinson) housed in six-well companion plates (Becton Dickinson) were coated with 476 μg/ml growth factor reduced Matrigel Matrix (GFR Matrigel, BD Biosciences) diluted in DPBS using a method described previously (3). Rat recombinant platelet-derived growth factor (PDGF)-BB (Sigma) (2.5 ml of either 10 or 100 ng/ml) (3, 11) diluted in DMEM/F12 + 1% P/S was placed in the well of the companion plate to act as the chemoattractant. Either SFBs or CFBs were trypsinized from culture flasks and suspended in DMEM/F12 + 10% FBS + 1% P/S at a concentration of 250,000 cells/ml. Two milliliters of this solution were placed in each insert, and the inserts and companion plates were placed in a 37°C, 5% CO2-95% air incubator for 7 h.
Inserts were then moved to a hood containing a rotating disk apparatus (see Ref. 23 for a thorough description) and were exposed to either 1 or 20 dyn/cm2 shear stress (SS) for 4 h (3). The following equation was used to calculate SS: τ = μωr/h, where τ is the SS, μ is the fluid viscosity (1.2 cP), ω is the rotational velocity of the disk (23.9 or 478 rpm for 1 or 20 dyn/cm2 SS, respectively), r is the disk radius (11.12 mm), and h is the gap distance between the disk and the cell surface (251 μm). Maximal SS (either 1 or 20 dyn/cm2) was applied at the edge of the disk, minimum SS (0 dyn/cm2) was applied at the center of the disk, and the average SS imposed on the cells was calculated as two-thirds of the maximal value. Cells were maintained at 37°C, 5% CO2-95% air during SS exposure, and no significant change in cell morphology was observed due to shear exposure. Control inserts were also housed in the experimental hood with a rotating disk in place but stationary (no shear exposure).
After a shear exposure was completed, inserts were examined under ×100 light microscopy to ensure that the cell monolayer was not compromised by exposure to SS. Cells that had migrated through the GFR Matrigel layer to the underside of the insert were fixed and stained with the DiffQuik staining solution (Dade Behring), whereas cells remaining on top of the Matrigel layer and the layer itself were mechanically removed with a cotton swab. The average number of cells counted in five ×100 fields (one in the center and four around the edges) on the underside of the insert was used to quantify migratory activity. In all experiments, migration levels in an insert subjected to SS were normalized with respect to its companion control insert (no shear exposure).
All data are presented as means ± SE. Data sets were analyzed for statistical significance using a Student's t-test with a two-tailed distribution. P < 0.05 was considered statistically significant.
SFB and CFB expression of SM α-actin and MHC.
Results of immunofluorescence experiments conducted to evaluate SFB and CFB expression of SM α-actin (one of the proteins expressed earliest during SMC differentiation) and SM MHC (a protein found only in fully differentiated, mature SMCs) are summarized in Fig. 1. Average (net) fluorescent intensity levels representing SM α-actin expression were significantly greater (P < 0.0001; Fig. 1) in CFB slides (21,097 ± 1,518 net fluorescent intensity units; n = 24 regions analyzed) than in SFB slides (7,623 ± 951 net fluorescent intensity units; n = 74 regions analyzed). Although net fluorescent intensity levels representing SM MHC expression were significantly greater (P < 0.0001; Fig. 1) in CFB slides (575 ± 101 net fluorescent intensity units; n = 21 regions analyzed) than in SFB slides (205 ± 15 net fluorescent intensity units; n = 95 regions analyzed), SM MHC intensity in both SFBs and CFBs was significantly lower than SM α-actin intensity in the same cell population (P < 0.0001 in both cases), indicating that SM MHC levels were minimal in all slides analyzed.
In a separate set of experiments, SFBs, CFBs, and SMCs from the same primary culture were lysed following exposure to the normal culture conditions described earlier, and SM α-actin and SM MHC content in lysate from all three cultures was evaluated via Western blotting. A representative blot is shown in Fig. 2A. SM α-actin and SM MHC levels for SFBs and CFBs were normalized by SMC levels of the respective protein for each individual experiment. SFB relative SM α-actin content was 0.04 ± 0.02 that of SMC SM α-actin content (normalized to 1), whereas CFB SM α-actin content was 0.85 ± 0.03 that of SMC SM α-actin content (n = 4 experiments; Fig. 2B). SFB SM α-actin levels were significantly lower than those of both CFB and SMC levels (P < 0.001 in each case). CFB SM α-actin levels were also significantly lower than that of SMC levels (P < 0.05). SFB SM MHC content was negligible when normalized by SMC SM MHC content, and CFB SM MHC content was 0.05 ± 0.03 that of SMC content (n = 4 experiments; Fig. 2B). Both SFB and CFB SM MHC levels were significantly lower than those of SMC levels (P < 0.001 in each case), but SFB and CFB levels were not statistically different from each other.
Effects of shear stress on SFB and CFB migration.
The effects of 20 dyn/cm2 SS applied for 4 h on SFB and CFB migration is summarized in Fig. 3A. SFB migration levels in sheared inserts were 2.05 ± 0.32 times greater than controls (normalized to 1) in the presence of 10 ng/ml PDGF-BB (n = 7) and 1.54 ± 0.14 times greater than controls in the presence of 100 ng/ml PDGF-BB (n = 6). Both of these enhanced levels of migration induced by SS were statistically significant (P < 0.05 in each case). In stark contrast to the response observed for SFBs, SS significantly inhibited CFB migration to both 10 ng/ml (P < 0.01) and 100 ng/ml PDGF-BB (P < 0.001 in both cases). Migration levels were 0.62 ± 0.03 in sheared inserts compared with control (normalized to 1) in response to 10 ng/ml PDGF-BB (n = 4) and 0.47 ± 0.05 in sheared inserts relative to control when 100 ng/ml PDGF-BB served as the chemoattractant (n = 11). It is also of note that the baseline migration levels of SFBs and CFBs were quite different. Under control (no shear) conditions, ∼5.25-fold more CFBs (84 ± 44 cells migrated per ×100 field) migrated toward 10 ng/ml PDGF-BB compared with SFBs (16 ± 5 cells/×100 field) and ∼1.5-fold more CFBs (130 ± 19 cells/×100 field) migrated toward 100 ng/ml PDGF-BB compared with SFBs (85 ± 24 cells/×100 field). Neither SFBs nor CFBs migrated in the absence of PDGF-BB on time frames up to 24 h.
The effects of 4 h of 20 dyn/cm2 SS on SFB and CFB migration in both the center (where shear is low) and the edges (where shear is maximal) of the inserts described above were evaluated. A summary of results obtained from examining migration levels in the center of inserts is provided in Fig. 3B, and a summary of results obtained from examining the edges of inserts is indicated in Fig. 3C. SS significantly enhanced SFB migratory activity relative to control levels in both the center (2.10 ± 0.41; n = 7; P < 0.05) and edges (2.01 ± 0.36; n = 7; P < 0.05) of inserts exposed to 10 ng/ml PDGF-BB. SS also significantly elevated SFB migration levels in the center of sheared inserts when 100 ng/ml PDGF-BB served as the chemoattractant (1.63 ± 0.16; n = 6; P < 0.05) but not at the edges of the same sheared inserts (1.32 ± 0.21; n = 6; P > 0.05). SS significantly suppressed CFB migration levels in both the center (0.68 ± 0.09) and edges (0.44 ± 0.13) of sheared 10 ng/ml PDGF-BB inserts (n = 4; P < 0.05 in both cases). SS also significantly inhibited CFB migratory activity in the center (0.45 ± 0.06) and edges (0.55 ± 0.05) of sheared 100 ng/ml PDGF-BB inserts (n = 11; P < 0.001 in both cases).
The effects of 4 h of 1 dyn/cm2 SS on SFB and CFB migratory activity in the presence of 100 ng/ml PDGF-BB were also determined in a manner identical to that described for the 20 dyn/cm2 SS experiments above. Neither SFB- (0.99 ± 0.06; n = 6; P > 0.05) nor CFB (0.78 ± 0.15; n = 6; P > 0.05)-normalized migration levels were significantly altered in the presence of 4 h of 1 dyn/cm2 SS. Analysis of the central and edge areas of SFB and CFB inserts exposed to 4 h of 1 dyn/cm2 SS in the presence of 100 ng/ml PDGF-BB showed that SS exposure significantly inhibited CFB migration in the edges of sheared inserts, where normalized migration levels were 0.70 ± 0.10 (n = 6; P < 0.05) but did not significantly alter migration in any of the other cases (P > 0.05 in all cases; data not shown).
The primary findings of this study indicate that two populations of adventitial FBs, one maintained at subconfluent culture conditions (SFBs) and the other grown to complete confluence in culture (CFBs), display marked differences in their phenotype marker expression as well as their migratory response to fluid SS. SFBs express a modest amount of SM-specific α-actin (Figs. 1 and) and enhance their migratory rate in response to SS (Fig. 3). CFBs, on the other hand, express abundant SM-specific α-actin (Figs. 1 and 2) and reduce their migratory activity in response to SS (Fig. 3). Both cell populations express minimal SM MHC (Figs. 1 and 2). Migratory responses similar to the CFBs have been reported previously in rat aortic SMCs (3, 10). The present work is, to our knowledge, the first to characterize the effects of shear stress on adventitial FB migration and the first to demonstrate FB differentiation induced by state of confluence in culture.
It has been hypothesized that fibroblasts, myofibroblasts, and SMCs all derive from a common progenitor cell and that these different cell populations are distinguishable through their varied expression of phenotype markers, among them SM α-actin and SM MHC. FBs express neither SM α-actin nor SM MHC, myofibroblasts express SM α-actin but not SM MHC, and fully differentiated SMCs express both of these phenotype markers (18). Whereas it is not possible to firmly categorize SFBs as “fibroblasts” and CFBs as “myofibroblasts” according to the above convention, the data presented here (Figs. 1 and 2) suggest that SFBs resemble a traditional fibroblast phenotype based on their modest SM α-actin and minimal SM MHC expression, whereas CFBs are similar to a traditional myofibroblast phenotype based on their significant expression of SM α-actin and minimal expression of SM MHC. Both cell populations used in this study were obtained from the same primary culture. The only difference between them was the degree of confluence at which they were maintained during expansion in culture. A significant increase in SM α-actin expression in CFBs relative to SFBs indicates variation in phenotype between the two cell populations and suggests a novel cell culture technique that allows for examination of two distinct cell populations from the same primary culture. It should be noted that immunostaining and Western blotting experiments presented here were conducted after SFBs, CFBs, and SMCs were expanded under normal culture conditions for an extended time frame (order of days). When SFBs and CFBs were seeded onto cell culture inserts at the same density 7 h before shear exposure, they displayed marked differences in migratory activity in response to SS, indicating a difference in phenotype. This suggests that a time frame on the order of days is required to obtain a variation in phenotype between the two cell populations. These results also suggest a mechanistic link among shear stress, SM α-actin expression, and migration.
Li et al. (7) provided evidence that adventitial FBs respond to mechanical stimulation by showing that conditioned media obtained from FBs exposed to mechanical stretch in vitro induced significant FB migration in a Boyden chamber assay. This, coupled with studies showing that SMC migratory activity is inhibited in response to SS (3, 10), motivated our study of the role of SS on SFB and CFB migration. The new results indicate that SFBs and CFBs have distinct responses to mechanical stimulation (Fig. 3). In experiments following methods similar to those reported previously by Garanich et al. for SMCs (3), the migratory activity of SFBs was enhanced in response to 4 h of 20 dyn/cm2 SS, whereas the same SS levels reduced the migration levels of CFBs (Fig. 3). The inhibition of migration observed for CFBs in this study (∼53%) was not as dramatic as the inhibition of SMC migration in response to 20 dyn/cm2 SS for 4 h reported earlier (∼88%) (3). It appears that although CFBs become “SMC-like” through their expression of SM α-actin, they are not fully differentiated SMCs and therefore do not display responses to stimuli that are entirely characteristic of SMCs.
Four hours of 20 dyn/cm2 SS inhibited CFB migration in both the center of inserts where shear is minimal and at the edges of inserts where shear is maximum. Likewise, 4 h of 20 dyn/cm2 SS enhanced SFB migration in both the center and edge fields of inserts. Whereas neither SFB nor CFB migration was significantly affected by 4 h of 1 dyn/cm2 SS when central and edge fields were evaluated together, CFB migratory activity was significantly inhibited by 4 h of 1 dyn/cm2 SS when edge fields were examined alone. This suggests that FBs are also responsive to lower orders of SS, and when considered with the 20 dyn/cm2 SS data, indicates that both cell populations are releasing soluble factors in response to SS that are communicated among cells and control their migration rates. Studies examining the role of SS on SMC migration have linked the shear-induced inhibition of migration to the downregulation of matrix metalloproteinase (MMP)-2, an endopeptidase with the ability to degrade extracellular matrix molecules (3, 10), and others have reported changes in MMP-2 levels under altered flow conditions (1, 4), so it seems plausible that SS increases MMP-2 expression in SFBs, which leads to elevated migratory activity and conversely attenuates MMP-2 expression in CFBs, leading to decreased migration levels in these cells. This hypothesis remains to be tested.
The primary finding of this study, that SFBs and CFBs respond differently to mechanical stimulation, has potential implications for vascular cell involvement in intimal hyperplasia. FBs are known to participate in intimal lesions (6, 20, 21, 24), and shear-induced FB migration potentially contributes to this phenomenon in the following manner. When a vessel is injured (endothelial damage), transvascular interstitial flow increases [doubles in a medium-sized artery (30); illustrated schematically in Fig. 4 ], which results in increased SS on adventitial FBs, facilitating their migration to the intima. Under the same injury conditions, damage to the internal elastic lamina removes the funneling mechanism responsible for exposing superficial SMCs to elevated shear levels (28) (that inhibits their migration) and results in enhanced SMC migration, contributing to lesion formation.
The distinct SFB and CFB migratory responses to shear stress revealed in the present study also suggest a potential alternative mechanism through which mechanical stresses contribute to the capillary arterialization process in addition to that proposed by others (12–16, 25, 26) (illustrated schematically in Fig. 5). Price and colleagues (12–16, 25, 26) have provided evidence that capillaries connecting two high-pressure arterioles are exposed to elevated luminal pressures and therefore elevated circumferential wall stresses compared with other capillaries in processes such as wound healing and exercise hypertrophy. They hypothesize that elevated wall stresses recruit SM-like cells to the arterializing vessel by upregulating EC growth factor release. A potential alternative view of this process is that elevated luminal pressure drives elevated interstitial flow (shown in red in Fig. 5) through the capillary wall and into the interstitial space surrounding the vessel. FBs located adjacent to the arterioles are thereby exposed to elevated levels of interstitial flow SS, which in conjunction with EC-derived growth factor production, recruits these cells to the arterializing vessel. As they approach the vessel and come into closer contact with neighboring cells, these FBs differentiate to a SM-like phenotype, their migration is suppressed, and they integrate into the vessel wall.
The results presented in the current work as well as the potential alternative mechanisms through which mechanical stresses regulate FB involvement in intimal hyperplasia and capillary arterialization proposed above must be viewed as preliminary for several reasons. First, the current study utilized a rotating disk apparatus to subject SFBs and CFBs to SS while they were migrating. Whereas spatial SS gradients do exist in this system, they are several orders lower than those that have been shown to be physiologically relevant (23). Therefore, it is unlikely that they contributed to the changes in migratory activity observed in response to shear exposure. Also, in the in vivo setting, FBs migrate in a direction opposite that of flow (SS), whereas with the rotating disk system, cells migrate perpendicular to the flow (SS) direction. Finally, the current study and its applicability to FB involvement in intimal hyperplasia and capillary arterialization need to be validated in an in vivo setting to further establish its merit.
In summary, this study showed that two populations of rat aortic adventitial FBs derived from the same primary culture, differing only in their treatment in culture (SFBs and CFBs), exhibited marked differences in their migratory response to fluid SS. These results allow for potential alternative views of FB participation in physiological (arterialization) and pathological (intimal hyperplasia) processes on both the macrocirculatory and microcirculatory levels. This initial two-dimensional study provides the impetus for a more physiological three-dimensional one in which CFBs and SFBs are grown in gels and exposed to interstitial flow SS.
This work was supported by National Heart, Lung, and Blood Institute Grant HL-35549.
The authors recognize Blair Urish for contributions to the migration experiments conducted as a part of this work. We also recognize the original artwork of Rishi A. Mathura presented in Figs. 4 and 5.
The 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.
- Copyright © 2007 by the American Physiological Society