The endothelial cell glycocalyx, a structure coating the luminal surface of the vascular endothelium, and its related mechanotransduction have been studied by many over the last decade. However, the role of vascular smooth muscle cells (SMCs) glycocalyx in cell mechanotransduction has triggered little attention. This study addressed the role of heparan sulfate proteoglycans (HSPGs), a major component of the glycocalyx, in the shear-induced proliferation, migration, and nitric oxide (NO) production of the rat aortic smooth muscle cells (RASMCs). A parallel plate flow chamber and a peristaltic pump were employed to expose RASMC monolayers to a physiological level of shear stress (12 dyn/cm2). Heparinase III (Hep.III) was applied to selectively degrade heparan sulfate on the SMC surface. Cell proliferation, migration, and NO production rates were determined and compared among the following four groups of cells: 1) untreated with no flow, 2) Hep.III treatment with no flow, 3) untreated with flow of 12 dyn/cm2 exposure, and 4) Hep.III treatment with flow of 12 dyn/cm2 exposure. It was observed that flow-induced shear stress significantly suppressed SMC proliferation and migration, whereas cells preferred to aligning along the direction of flow and NO production were enhanced substantially. However, those responses were not found in the cells with Hep.III treatment. Under flow condition, the heparinase III-treated cells remained randomly oriented and proliferated as if there were no flow presence. Disruption of HSPG also enhanced wound closure and inhibited shear-induced NO production significantly. This study suggests that HSPG may play a pivotal role in mechanotransduction of SMCs.
- heparan sulfate proteoglycans
- shear stress
vascular smooth muscle cells (SMCs) normally reside in the tunica media of the arterial wall and are shielded from blood flow by an intact endothelium. However, in the cases of endothelium and internal elastic lamina (IEL) injury, as occur, for example, in angioplasty, near the anastomoses of vascular grafts and in atherosclerotic disease, SMCs are directly exposed to blood flow, and their function may be modulated by changes in the local hemodynamic environment (1, 28). In these situations of endothelial disruption, blood flow-induced shear stress, on the order of 10–20 dyn/cm2, may impose on the superficial SMCs (33). Even in an intact artery, the underneath SMCs are continuously exposed to a shear stress due to interstitial flow driven by the transmural pressure gradient (typically 100 mmHg in an artery) on the order of 1–3 dyn/cm2 (44). In addition, it has been predicted by Wang and Tarbell (45) that the most superficial SMCs lying close to the IEL may be exposed to a higher level of shear stress (the order of magnitude, 10–50 dyn/cm2) via the funneling of flow across the fenestral pores in the IEL.
It has been well established that the pathogenesis of intimal hyperplasia, the major cause of failure of prosthetic bypass grafts and angioplasty procedures, involves SMC proliferation and migration to the intima in response to injury-induced specific biochemical and mechanical environment (1). To assess the role of shear stress in SMC proliferation and migration processes, Sterpetti et al. (38) subjected bovine aortic smooth muscle cells to a shear stress of 6 dyn/cm2. They found that cells aligned along the direction of flow showing a spherical morphology and a decreased proliferation rate relative to the control (No flow) group. In addition, Ueba et al. (43) revealed that shear stress inhibited vascular SMC (VSMC) proliferation in vitro, which was mediated predominantly by transforming growth factor β1 (TGFβ1) in an autocrine manner. Moreover, Palumbo et al. (27) demonstrated that shear stress (12 dyn/cm2) could inhibit bovine aortic SMC migration and invasion via diminished platelet-derived growth factor (PDGF)-Rβ expression, which was associated with decreased matrix metalloproteinase (MMP-2) secretion and membrane-type MMP downregulation. Garanich et al. (14) quantified the migration activity of rat aortic SMCs and suggested that shear stress suppressed SMC migration by directly upregulating the cellular production of nitric oxide (NO), which in turn inhibits MMP-2 activity.
NO is a relatively short-lived molecule that plays a dual role in the arterial system. In physiological conditions, NO is involved in numerous functions including vasorelaxation, reduction of platelet aggregation, and inhibition of adhesion of leukocytes to the vascular wall (13). While in abnormal situations, NO participates in the pathogenesis of many acute inflammatory and autoimmune diseases for its cytostatic and cytotoxic actions on various cell types (32). It has been shown that the exposure of vascular cells [endothelial cells (ECs) or SMCs] to shear may lead to a notable alteration of NO production, which can be mediated by mechanotransduction, or the transformation of blood flow induced shear stress to the cellular biochemical responses (6, 13, 32). Previously, it was proposed that the mechanosensors of endothelial cells included integrins, G proteins, intercellular junction proteins, and ion channels (9). Recent investigations on the glycocalyx of the vascular endothelial cells have verified its pivotal role in cell's mechanotransduction (12).
The glycocalyx of the endothelial cells is a surface layer composed primarily of proteoglycans, with their associated glycosaminoglycan (GAG) that includes heparan sulfate, chondroitin sulfate, and hyaluronic acid,and glycoproteins bearing acidic oligosaccharides with terminal sialic acids (26). The EC glycocalyx has been studied extensively for its role in cellular mechanotransduction over the last decade. However, the mechanisms by which SMCs sense and transform shear stress into cellular biochemical responses have not been conclusively determined. On the SMC surface, ∼50–60% of the GAGs are chondroitin sulfate, and the rest is predominantly heparan sulfate (24). Recently, Ainslie et al. (2) reported that heparan sulfate and chondroitin sulfate components of the SMC glycocalyx played an important role in the mechanotransduction of shear stress into a contractile response. To our best knowledge, this may be the only study that examined the role of the SMC glycocalyx in the mechanotransduction of shear stress.
To seek more evidence of SMC glycocalyx mechanotransduction, we employed a parallel plate flow chamber and a peristaltic pump to expose cultured rat aortic SMC monolayers to a physiological level of shear stress (12 dyn/cm2). Heparinase III (Hep.III ) was applied to selectively degrade heparan sulfate on the SMC surface. Cell proliferation, migration, and NO production rates were assessed and compared among the following four groups of cells: 1) untreated with no flow, 2) Hep.III treatment with no flow, 3) untreated with flow of 12 dyn/cm2 exposure, 4) Hep.III treatment with flow of 12 dyn/cm2 exposure.
MATERIALS AND METHODS
The following chemicals were obtained from Invitrogen (Camarillo, CA): collagenase II, Dulbecco's modified Eagle's medium (DMEM), fetal bovine serum (FBS), and penicillin-streptomycin. Trypsin, paraformaldehyde, glutaraldehyde, BSA, and Hep.III from Flavobacterium heparanum were obtained from Sigma (St. Louis, MO). Type I rat-tail collagen was obtained from Millipore (Bedford, MA). Heparan sulfate primary antibody (HepSS-1) was obtained from US Biological (Swampscott, MA). Alexa Fluor 488-labeled secondary antibody (goat anti-mouse IgM) was obtained from Molecular Probes (Eugene, OR). 2-(4-Amidinophenyl)-6-indolecarbamidine dihydrochloride (DAPI), a cell count kit-8 (CCK-8), and Griess reagent (Nitric Oxide Assay Kit) were obtained from Beyotime Biotech of China (Jiangsu, China).
Rat aortic smooth muscle cells (RASMCs) were isolated from the thoracic aortas of 200- to 250-g male Sprague-Dawley rats by collagenase digestion with the use of established techniques (22). Briefly, rat thoracic aortas were excised and put into Hanks' balanced salt solution under sterile conditions. After being rinsed three times, the adventitia was removed and the aortas were minced and digested in 10 ml of digestion solution (0.125 mg/ml elastase and 10 mg/ml collagenase) at 37°C for 45 min. The cellular digests were filtered through sterile 100-μm nylon mesh, and centrifuged at 1,000 rpm for 10 min. After being rinsed twice, cells were resuspended in DMEM complete culture media. The experiment followed a protocol approved by the institutional committee on animal use, and all animal care was complied with the ‘Principles of Laboratory Animal Care’ and the “Guide for the Care and Use of Laboratory Animals” (NIH Publication No. 86-23, Revised 1985). Cultures were confirmed as RASMCs by staining positively for α-smooth muscle actin and negatively for von Willebrand factor after 2 passages, and the studies here were performed between passages 4 and 10.These cells were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% (vol/vol) fetal bovine serum (FBS), 100 U/ml penicillin, 100 μg/ml streptomycin, at 37°C in 5% CO2. In all experiments (static and flow), cells were seeded on 25.4 × 76.2-mm glass slides precoated with type I rat-tail collagen under sterile conditions just followed the manufacture's instruction.
Figure 1 is the schematic drawing of the flow system used in the present study. A parallel plate flow chamber developed in our own laboratory was used to expose cultured RASMC monolayers to laminar fluid shear stress. Briefly, the flow chamber was made by sandwiching a silicone gasket between the glass slide and a Plexiglas plate. Fluid flow was provided by a peristaltic pump (Longer pump, Hebei, China) and two reservoirs situated one above another. Shear stress was determined by the following equation: τ = 6μQ/Wh2, where τ is shear stress (dyn/cm2), μ is viscosity of the medium (0.0084 poise), Q is the flow rate across the flow chamber (ml/s), h is channel height (254 μm for standard gasket), and W is channel width (1.6 cm for our chamber). During the flow experiments, the flow system (Fig. 1) was kept at 37°C by putting the main reservoir in a temperature-controlled water bath. All experiments were conducted at 12 dyn/cm2. This level of shear stress is in the physiological range found in human major arteries.
Hep.III was used to cleave heparan sulfate on RASMC surface. The enzyme was used at a concentration of 0.2 U/ml in DMEM as described elsewhere (2). Briefly, the conflunent monolayers on glass slides were treated with this enzyme for 30 min in a Petri dish inside a 95% air and 5% CO2, 37°C incubator. Before being mounted on the flow chamber, the glass slides were washed with fresh PBS twice.
Immunofluorescence and confocal microscopy.
To assess the effectiveness of Hep.III in removing heparan sulfate from the glycocalyx selectively, immunofluorescence antibody images for both untreated and pretreated cells were used following the staining procedure described by Florian et al. (12). Furthermore, to verify the recovery extent of heparan sulfate during our experiment time, cells were kept under static condition for another 6 h after Hep.III exposure. Briefly, the confluence monolayer was placed on ice for 5 min and washed twice with ice-cold PBS. The cells were then fixed with 4% paraformaldehyde and 0.5% glutaraldehyde for 20 min at 4°C and blocked with 5% BSA in PBS overnight. After being washed twice with ice-cold PBS, a solution of 2 μl HEPSS-1 was diluted to 1 ml using PBS and applied to the monolayers for 15 min. To gain the fluorescence image, 2 μl Alexa Fluor 488-labeled secondary antibody was diluted to 1 ml with PBS and applied to the monolayers for another 15-min period. DAPI solution (the final concentration, 10 μg/ml) was added to stain the cell nucleus for 10 min. After being washed three times with PBS at 5-min intervals, cells were covered with 50% glycerol in water and scanned under a confocal microscope (×20 objective lens) (Leica TCS SPE, Leica Microsystems, Wetzlar, Germany). Overall fluorescence intensity was calculated for a series of image fields, using Image J (National Institutes of Health). Comparison of the fluorescence intensity of the untreated and treated image fields provided an estimate of the degree of heparan sulfate removal associated with the heparinase treatment.
CCK-8 was applied in this experiment to quantitatively evaluate the cell viability. After RASMCs were exposed to a shear stress of 12 dyn/cm2 or static conditions for 0, 6, 12, 18, and 24 h, the glass slides were translated from the flow chamber to a Petri dish. Meanwhile, 2 ml fresh 10% FBS DMEM and 200 μl CCK-8 were added. It was incubated at 37°C for 1 h to form water-dissoluble formazan. Then 100 μl of the above formazan solution were taken from each sample and added to one well of a 96-well plate; eight parallel replicates were prepared. The absorbances at 450 and 630 nm (calibrated wave) were finally determined using a microplate reader (Multiskan MK3, Thermo Labsystem, Beverly, MA). Ten percent FBS DMEM containing 10% CCK-8 was used as a control.
Cell alignment and migration studies.
Cell alignment before or after 24 h shear treatment was monitored by an invert microscope (Olympus Optical, Japan) equipped with a charge-coupled device camera (TK-C1381, JVC, Tokyo, Japan) and a computer with a frame grabber. It was quantified by measuring the angle of the longest cell-body direction with respect to flow direction. The axis of the cell elongation was manually identified by using Image J for 100 cells in each experiment, and its relative angle to flow direction was then calculated and grouped into the corresponding category, ranging from 0° to 90° for each 10° increment (7).
For cell migration studies, each confluent monolayer was mechanically scratched with a plastic pipette tip to form four to five wounds perpendicular to the long axis of the slide (15). Images of the wound under either static or flow conditions were acquired at 0, 6, 12, 24 h, respectively. RASMC migration during the wound closure process was analyzed using image J software (National Institutes of Health). After the image was acquired, it was converted from pixels to micrometers with the use of a calibration image. For each experiment, percent original wound width was used to evaluate cell migration quantitatively.
The enzymatic production of NO from l-arginine by cultured rat aortic smooth muscle cells was so transient. Usually, it was assayed by measuring nitrite/nitrate, its oxidation product. In this experiment, the Griess method was employed to detect NO, which is based on the chemical diazotization reaction that was originally described by Griess in 1879, which uses sulfanilamide and N-1-napthylethylenediamine dihydrochloride under acidic (phosphoric acid) conditions (31). In all experiments, 50 μl conditioned media from both stationary controls and cultures exposure to flow were taken at 0, 15, 30, 60, 120 min and replaced with an equal volume of fresh media. Because of this dilution effect, cumulative values were computed and normalized to the cell numbers on each glass slide.
Data are presented as means ± SE. Statistical analysis were performed by repeated measures analysis of variance and paired t-test. Differences between means were considered significant if P <0.05.
Verification of heparan sulfate proteoglycans removal and recovery efficiency.
To verify the effectiveness of heparan sulfate proteoglycans (HSPG) degradation by Hep.III and HSPG recovery efficiency after Hep.III treatment, both control and treated cells were exposed to an antibody specific to heparan sulfate and visualized by a fluorescently labeled secondary antibody before and after enzyme treatment. Figure 2 shows a representative immunostaining. As shown in Fig. 2, cells without Hep.III treatment exhibited a uniform HSPG distribution across the cell surface (Fig. 2A), whereas Hep.III pretreatment caused a 39.67 ± 1.3% (P < 0.05, n = 3) reduction in fluorescence intensity relative to the untreated controls (Fig. 3). Six hours after Hep.III treatment, cell surface HSPG regrew by 23.78 ± 0.9% (P < 0.05, n = 3) relative to the fresh Hep.III exposure cells (Fig. 3).
Effect of HSPG disruption on shear stress-induced proliferation.
We subjected both untreated and heparinase III-treated RASMCs to a shear stress of 12 dyn/cm2 and observed the proliferation of cells at the time intervals of 6, 12, 18, 24 h and compared the results with those under no-flow conditions. As evident from Fig. 4, the proliferation of the untreated RASMCs was significantly retarded relatively to the no-flow condition (Fig. 4A), whereas the disruption of heparan sulfate on RASMC surface led to suppressed cell responses to shear stress so that the cells proliferated as if there were no flow presence (Fig. 4B). Moreover, our observation showed that under no-flow condition, Hep.III treatment itself could greatly affect cell proliferation rate (Fig. 4C).
Effect of HSPG disruption on cell alignment with flow.
In this set of experiment, we subjected both untreated and Hep.III-treated RASMCs to a shear stress of 12 dyn/cm2 for 24 h and then monitored the morphology and orientation changes of the cells. The images demonstrated that the cells without enzyme treatment aligned along the flow direction as expected, whereas Hep.III-treated RASMC did not and oriented randomly (Fig. 5A). Cell alignment was quantified by measuring the aligning angle that was the angle between the cell's long axis and the flow direction. Before being subjected to flow, both untreated and treated cells elongated randomly in all possible angles from 0° to 90°(Fig. 5B), whereas after 24 h of flow exposure only the untreated cells aligned in the flow direction (the aligning angles of ∼83% of cells were <30°).
Effect of HSPG disruption on cell migration behavior during the wound closure process.
To identify the role of surface heparan sulfate in the RASMC wound closure process, each of the test confluent cell monolayers were mechanically scratched with a plastic pipette tip to form 4 to 5 wounds perpendicular to the long axis of the slide, and then the monolayers were observed for wound closure under flow condition (12 dyn/cm2) or static condition (Fig. 6). Figure 7 shows that without enzyme treatment, the RASMC wounds at static condition would close up to 67.24 ± 0.38% (means ± SE) in 24 h, whereas the wounds exposed to flow condition only closed to 9.26 ± 1.63%, suggesting that shear stress significantly inhibited wound closure of the untreated SMCs. On the other hand, cells treated with Hep.III showed a different wound closure behavior. Under static condition, the treated RASMC wound closed up to 92.03 ± 0.72% in 24 h, significantly higher than the untreated cells. Under flow condition, although the treated cells showed a similar shear-inhibited wound closure behavior, its wound closure rate in 24 h was much higher than that of the untreated group (23.06 ± 0.94% vs. 9.26 ± 1.63%), indicating that surface heparan sulfate played a crucial role in SMC wound healing.
Effect of HSPG disruption on NO production under flow condition.
Figure 8 shows the effect of Hep.III treatment to cells on NO production. As shown in the figure, NO levels were increased significantly after a laminar shear stress of 12 dyn/cm2 was imposed on the cells. No matter whether the cells were treated with Hep.III or not, the increase in NO level was observed at all time intervals during the 2-h experiment, indicating that flow-induced shear stress stimulated NO production of the cells. However, the comparison of NO level between the treated cells and the untreated ones showed that when subjected to flow, the treated cells produced much less NO than the untreated cells. Moreover, for the treated cells, the earliest time point at which NO level reached significant difference from its static counterpart was 30 min after the flow was imposed. For the untreated cells, it was only 5 min (P < 0.05, n = 8).
Previous experiments have demonstrated that heparinase, an enzyme specific for cell surface heparan sulfate, has the ability to alter the thickness (7, 30) and barrier property (23) of the glycocalyx. It has been testified by Florian et al. (12) that a concentration of 15 mU/ml of Hep.III can cause a 45% reduction in heparan sulfate but a negligible degradation of chondroitin sulfate and undetectable protease activity. The concentration and the duration for heparinase treatment to SMCs in the present study was based on the report by Ainslie et al. (2) in which a 38% reduction in fluorescence due to heparinase treatment was observed on rat aortic SMCs. In our study, the reduction in fluorescence with heparinase treatment was 39.67 ± 1.3% (Fig. 2), which is consistent with the report by Ainslie et al. (2). In line with the theory that the reduction of fluorescence intensity was proportional to the degradation of heparan sulfate proteoglycans on the cell surface, we could verify the effectiveness of our Hep.III treatment for RASMCs.
Yao et al. (46) demonstrated that under static conditions, heparinase treatment alone did not affect endothelial cell proliferation. This seems contradictory to our observation that Hep.III treatment itself inhibited RASMC proliferation significantly (Fig. 4). The discrepancy between their results and ours may be interpreted as follows: first, the cell type used by them (46) was bovine aortic endothelial cells (BAEC), whereas ours was RASMC. The differences in species and cell type cannot be neglected. Second, to degrade the cell-surface glycocalyx, they incubated BAECs in a 60 mU/ml Hep.III solution for 30 min, whereas the Hep.III treatment to RASMCs in our study was conducted at a final concentration of 0.2 U/ml for 30 min, which is the same as that used by Ainslie et al. (2). As far as we know, under static conditions, some growth factors such as fibroblast growth factor-2 (FGF-2) can bind to its high affinity protein kinase receptor (FGFR) and HSPG on the cell membrane to form a ternary complex, which may trigger the downstream signaling to induce SMC proliferation (4, 8). Therefore, it seems rational to speculate that this proliferation inhibition phenomenon was caused directly by HSPG degradation which then affected downstream signaling pathway activation. Furthermore, Nilsson et al. (37) reported the similar phenomenon from cultured RASMCs that heparitinase treatment reduced the rate of entrance into S phase, which supplied extra experimental evidence.
We observed that shear stress could remarkably suppress SMC migration, which is very similar to the observation by Tarbell et al. (34). Our results showed that without flow imposing on the cells, Hep.III treatment alone enhanced SMC migration. This phenomenon has been observed also in endothelial cell by Moon et al. (21). However, our result that the SMC migration was not enhanced by Hep.III treatment under flow condition was far beyond our expectation. The underlying mechanism for this phenomenon has still been under investigation by us. But we speculate that there must be other mechanosensors that can also mediate the response of SMCs to the flow-induced shear stress in terms of cell migration. As we know, mechanical signals such as tensile, compressive, and shear forces are also transmitted by the extracellular matrix (ECM) to cells via integrin receptors that link the external environment to the cytoplasm and the cytoskeleton (36). This suggests that HSPGs or the glycocalyx is only one of the mechanosensors of vascular cells.
The mechanisms by which laminar shear stress regulates SMC proliferation, migration, and NO production remain to be elucidated. Recent studies have provided insight into some of the pathways involved. Regarding the shear stress-regulated SMC proliferation, two major signaling mechanisms have been suggested in the literature: Akt pathway (10, 11) and autocrine or paracrine pathway (3, 43). The former has been suggested to be modulated by the tyrosine phosphorylation of platelet endothelial cell adhesion molecule-1 (PECAM-1) (42). Tzima et al. (42) reported that PECAM-1, vascular endothelial cell cadherin, and vascular endothelial growth factor 2 (VEGFR2) comprised a mechanosensory complex that could detect shear stress and trigger intercellular signals. On the other hand, the latter involves TGF-β1 (43) and Fas-Fas ligand pathway (3), which is independent of Akt pathway. On the aspect of the shear stress regulated SMC migration, Goldman et al. (16) suggested that it might be mediated by extracellular signal-regulated kinase (ERK1/2)-myosin light chain kinase (MLCK) signaling pathway. Regarding the shear stress regulated production of NO, it has been revealed that SMCs secrete NO on exposure to flow by activating a constitutively expressed NOS I isoform (1) or an inducible NO synthase isoform (18). These mechanisms aforementioned are downstream cascades triggered by the upstream mechanotransduction on the surfaces of SMCs.
It has been demonstrated that on the surface of EC, there lies multiple mechanosensors; the glycocalyx on the surface of a cell is believed to be one of them. Numerous theoretical models have been developed to elucidate the structure and the mechanotransduction function of the glycocalyx, including a three-dimensional fibrous meshwork structure model (35), a bumper car model (40), wind in the trees model (39), and a spatial inhomogeneities model (19). These models predict that the flow-induced shear stress is dissipated through the glycocalyx and thus becomes negligible near endothelial cell membranes (19). Furthermore, experimental studies on ECs have demonstrated that some specific components of the EC glycocalyx participate in mechanosensing that mediates NO production in response to shear stress (12, 20, 25). Davies (5) has put forth two possible mechanotransduction mechanisms for the glycocalyx, namely, the decentralized and centralized mechanisms. In terms of the former, because syndecans of the glycocalyx are linked with the cytoskeleton (41), through the syndecans the signal of flow-induced shear stress can be transmitted to the cytoskeleton, spreading to multiple sites within the cell (i.e., nucleus, organelles, focal adhesions, and cell junctions). In terms of the latter mechanism, both syndecans and glypicans of the glycocalyx interact with signaling proteins that can initiate cascades. For example, it has been established that syndecan-4 HSPGs oligomerization that is essential for PKCa activation (41), and glypicans located in caveolae along with many endothelial NO synthase interacting proteins are both involved in shear-induced NO production and endothelial NO synthase activation pathways (17, 29).
In conclusion, the exposure of RASMCs to a flow-induced shear stress of 12 dyn/cm2 enhanced their NO production substantially but suppressed their proliferation and migration behavior, whereas cells preferred to align along the direction of flow. On the other hand, Hep.III treatment to RASMCs diminished the aforementioned responses of the cells to the flow shear. Based on this, we believe that HSPGs or the glycocalyx as a whole may play a crucial role in the mechanotransduction of SMCs.
This work is supported by Grants-in-Aid from the National Natural Science Foundation of China (No. 11072023, 10632010).
No conflicts of interest, financial or otherwise, are declared by the author(s).
- Copyright © 2011 the American Physiological Society