Syndecan-4 (S4) belongs to a family of transmembrane proteoglycans, acts as a coreceptor for growth factor binding as well as cell-matrix and cell-cell interactions, and is induced in neointimal smooth muscle cells (SMCs) after balloon catheter injury. We investigated S4 expression in SMCs in response to several force profiles and the role of MAP kinase signaling pathways in regulating these responses. S4 mRNA expression increased in response to 5% and 10% cyclic strain (4 h: 200 ± 34% and 182 ± 17%, respectively; P < 0.05) before returning to basal levels by 24 h. Notably, the SMC mechanosensor mechanism was reset after an initial 24-h “preconditioning” period, as evident by an increase in S4 gene expression following a change in cyclic stress from 10% to 20% (28 h: 181 ± 1%; P < 0.05). Mechanical stress induced a late decrease in cell-associated S4 protein levels (24 h: 70 ± 6%; P < 0.05), with an associated increase in S4 shedding (24 h: 537 ± 109%; P < 0.05). To examine the role of MAP kinases, cells were treated with U-0126 (ERK1/2 inhibitor), SB-203580 (p38 inhibitor), or JNKI I (JNK/SAPK inhibitor). Late reduction in cell-associated S4 levels was attributed to ERK1/2 and p38 signaling. In contrast, accelerated S4 shedding required both ERK1/2 (5-fold reduction in accelerated shedding; P < 0.05) and JNK/SAPK (4-fold reduction; P < 0.05) signaling. Given the varied functions of S4, stress-induced effects on SMC S4 expression and shedding may represent an additional component of the proinflammatory, growth-stimulating pathways that are activated in response to changes in the mechanical microenvironment of the vascular wall.
- heparan sulfate proteoglycan
syndecan-1 and -4 are members of a family of cell surface heparan sulfate (HS) proteoglycans (HSPGs) and are capable of modulating cell behavior through their capacity to act as cell surface coreceptors for a variety of heparin binding proteins. Recent investigations have emphasized that accelerated syndecan shedding may provide an important mechanism for regulating local host responses to tissue injury (27). Indeed, agents that increase syndecan shedding, including a variety of proteases and growth factors, are released during acute wound repair (8, 16, 26, 35). Moreover, several reports have noted that an absolute reduction in these cell surface receptors, as a consequence of shedding, may alter cell responses to both soluble and insoluble heparin binding proteins. For example, we recently observed (19, 20) that stress-induced loss of syndecan-4 is associated with an increase in fibroblast and vascular smooth muscle cell (SMC) movement on fibronectin.
Although shedding may alter the cell surface concentration of syndecans, it is noteworthy that shed HSPGs may exert a direct biological effect through the presence of HS glycosaminoglycans (GAGs) that remain attached to the shed core protein. Through retention of ligand binding activity, shed syndecans are capable of sequestering heparin binding growth factors such as FGF-2, protecting these factors from heat-, pH-, and protease-related degradation mechanisms present in the extracellular matrix (ECM). After heparanase-induced cleavage of HS chains on the syndecan core protein, heparin binding proteins are then able to interact with cell-bound receptors (17). Alternatively, shed syndecans may potentially act as dominant-negative modulators by competing with cell surface receptors for the same ligand. All told, considerable evidence now supports the notion that shed syndecan ectodomains likely have important physiological roles in morphogenesis, tissue repair, and host defense. In this regard, we have observed (20) persistent upregulated expression of syndecan-4 protein in both the neointima and the adventitia after balloon injury in a rat carotid artery model that appears consistent with the presence of substantial shed protein at the site of local vascular wall injury. Although a direct causal relationship has yet to be proven, we have postulated that locally accelerated syndecan shedding may lead to maladaptive responses that play a critical role in vascular lesion formation.
The restenosis response that often follows balloon angioplasty with or without stent placement may be due, in part, to extreme vessel wall stress that is associated with this mechanical intervention (24). Although balloon angioplasty represents an acute change in local stress, chronic changes in the mechanical microenvironment may also be responsible for the development of neointimal hyperplasia at vascular anastomoses (28). Likewise, systemic arterial hypertension is an independent and potent risk factor for the development of atherosclerosis, and a direct correlation has been noted between regions of increased wall tension and sites that are predisposed to the formation of atherosclerotic lesions (36, 37). Recent investigations have also emphasized that the magnitude of the phasic changes in wall stress may be an important stimulus for atherosclerosis. For example, in a study of 19,083 men, pulse pressure was the strongest predictor of cardiovascular mortality (3). Since our initial investigations suggest that mechanical stress promotes syndecan shedding from vascular SMCs, we have postulated that this response may represent an additional component of the proinflammatory, growth-stimulating pathways that are activated in response to changes in the mechanical microenvironment of the vascular wall (19).
Fitzgerald et al. (8) previously demonstrated that extracellular signal-related kinase (ERK)1/2 activity was required for syndecan shedding induced by epidermal growth factor (EGF) and thrombin receptor activation, whereas p38 mitogen-activated protein (MAP) kinase activity was not involved. These authors also demonstrated that hyperosmolarity and ceramide treatment each trigger shedding via protein kinase C (PKC)-dependent pathways that are likely upstream of c-Jun NH2-terminal kinase (JNK)/stress-activated protein kinase (SAPK) activation. Since a number of reports have demonstrated that MAP kinase pathways are activated by mechanical stress (10, 15, 18, 32), we postulated that ERK1/2 and/or other MAP kinase signaling mechanisms are responsible for mechanical stress-induced upregulation of syndecan-4 expression and shedding. In the present study, we investigated syndecan-4 expression and shedding in response to several distinct force profiles that are representative of varied mechanical microenvironments that may be experienced by vascular SMCs in vivo. We demonstrate that syndecan-4 expression and shedding are uniquely influenced by changes in the phase and magnitude of the local stress field. Moreover, we determined that ERK1/2 and p38 pathways mediate stress-regulated expression of syndecan-4 protein, but accelerated shedding is largely controlled by ERK1/2 and JNK/SAPK signaling pathways.
MATERIALS AND METHODS
Clonetics normal human aortic SMCs, culture medium, and supplements were obtained from Cambrex BioScience (Walkersville, MD). Smooth muscle cell growth medium (SmGM)-2 was prepared according to the supplier's recommendations. Cells were maintained in tissue culture-treated petri dishes at 37°C, 5% CO2, and humidified atmosphere. Growth medium was changed every 2 days, and the cells were passaged 1:4 when the dishes were 80% confluent with 0.05% trypsin-0.53 mM EDTA (Invitrogen Life Technologies, Carlsbad, CA). Experiments were performed on cells between passages 6 and 10 with Quiescence medium, which consisted of basal medium supplemented with 0.5% FBS, 50 μg/ml gentamicin, and 50 mg/ml amphotericin B.
Mechanical strain application.
Although the in vivo complexity of both the biomechanical and associated biochemical events cannot be fully captured in a cell culture system, in vitro model assays can replicate force profiles that are representative of those experienced by vascular SMCs in vivo. In general terms, changes in the biomechanical microenvironment relevant to the development of lesions within the vascular wall may be considered to fall within three distinct force profiles. First, balloon angioplasty induces an acute increase in wall tension that on a cellular level may be recapitulated in vitro by a single mechanical stretch of cells cultured on an elastomeric substrate (12, 23, 38, 39). Second, states that transiently increase blood pressure may lead to an acute increase in cyclic wall tension. As a third regime, sustained or poorly controlled hypertension may be represented biomechanically by a persistent increase in cyclic tension. As illustrated in Fig. 1, SMCs were subjected to one of three regimens of cyclic or static strain.
Strain dishes were assembled by fitting the bases of bottomless, custom-made plastic petri dishes with a silicone membrane (0.005-in. thickness; 40 Durometer, Specialty Manufacturing, Saginaw, MI). In all experiments, 5 μg/ml of human plasma fibronectin (Sigma) in PBS was preadsorbed to the growth area for at least 8 h at 4°C, serving as a substrate for cell attachment. The cells were then seeded onto the membranes, cultured to 80% confluence in SmGM-2, and then growth arrested for 24 h in Quiescence medium. Depending on the experimental conditions, defined static or cyclic tension was applied to the cells with a StrainMaster apparatus (Z-Development, Cambridge, MA) that allowed control over the frequency and amplitude of the radial and circumferential strains of the membrane and enabled the application of homogeneous and biaxially uniform strains over the entire growth area of the membrane. For a given protocol, cells were either cyclically strained at 1 Hz or held static at predefined strain amplitudes and harvested at various time points. In selected experiments, cells were preconditioned at 10% cyclic strain for 24 h, followed by 20% strain thereafter. A two-color live/dead cell assay confirmed that cells remained viable under all mechanical conditions imposed in these studies (see Supplemental Fig. S1; the online version of this article contains supplemental data).
Human aortic SMC RNA analysis.
Real-time quantitative PCR was used to determine syndecan gene expression in human aortic SMCs. Briefly, total RNA was harvested with TRIzol reagent (GIBCO BRL) in which RNA was extracted with chloroform and isopropanol, rinsed with 75% ethanol, and resuspended in RNase-free water. The absence of significant DNA or protein contamination was verified by ultraviolet (UV) absorbance ratio [optical density (OD)260/OD280]. cDNA was generated with SuperScript III reverse transcriptase (Invitrogen), using 5 μg of RNA. Real-time PCR was performed with the TaqMan primer and probes for syndecan-4 (Hs00161617_m1) and 18S rRNA (Hs99999901_s1). Amplification and detection were performed on an ABI Prism 7000 Sequence Detection System using AmpliTaq Gold DNA Polymerase (Applied Biosystems) in a 5′ nuclease reaction. Data generated from syndecan gene amplification were normalized to the simultaneously amplified endogenous 18S rRNA controls. Gene expression in mechanically strained cells was represented relative to that measured in nonstrained controls.
Isolation and quantification of shed syndecan-4.
The following protocol was adapted from Reiland et al. (30) and Rioux et al. (31). Conditioned culture medium was collected and passed through a 0.2-μm membrane filter (Millipore, Billerica, MA) to remove detached cells. Typically, 15 ml of medium was concentrated to 400 μl with a 10,000 molecular weight cutoff Amicon Ultra-15 centrifugal filter unit. The retentate was bound to a Vivapure Mini diethylamine weak anion exchange column (Vivascience, Edgewood, NY) with an equilibration buffer (20 mM NaOAc, pH 5, 4 M urea, 0.5% Triton X-100, 1 mM EDTA, 1 mM Na2SO4, 200 mM NaCl). Proteoglycans were eluted with 100 μl of equilibration buffer containing 1 M NaCl. With a Bio-Spin 6 (Bio-Rad, Hercules, CA) gel filtration column, samples were transferred to a reaction buffer containing 3 mM Ca(OAc)2, 10 mM EDTA, 10 mM HEPES, pH 7.0, 0.1% Triton X-100, 10 mM N-ethylmaleimide, and protease inhibitors (200 μM AEBSF, 160 nM aprotinin, 10 μM bestatin, 3 μM E-64, 4 μM leupeptin, and 2 μM pepstatin A). GAG chains were digested by addition of 2 mU of heparitinase and 8 mU of chondroitinase ABC (Proteus vulgaris, EC 18.104.22.168, Seikagaku America) to 30 μl of sample for a 4-h incubation at 37°C. Total protein was quantified with the bicinchoninic acid method.
Quantification of syndecan was performed according to Elenius et al. (6) and Sneed et al. (34). Shed syndecan-4 was detected with a Bio-Dot SF slot blot apparatus (Bio-Rad) and 0.450-μm Immobilon-P polyvinylidene difluoride (PVDF) membrane (Millipore) on top of three sheets of Bio-Dot SF filter paper (Bio-Rad). Conditioned medium was diluted 1:10 in heparitinase buffer without Triton X-100 and applied directly to the membrane by vacuum filtration. The membrane was blocked for 1 h at room temperature in 5% nonfat dry milk in 0.04% Tween-Tris-buffered saline (TBS) pH 7.4. The immunoblot was labeled overnight at 4°C with anti-syndecan-4 MAb (antibody 8G3; provided by Dr. Guido David, Katholieke Universiteit Leuven, Leuven, Belgium) diluted 1:2,000 in 1% nonfat dry milk in 0.04% Tween-TBS pH 7.4. The membrane was washed in 0.04% Tween-TBS pH 7.4, followed by incubation with horseradish peroxidase anti-mouse IgG diluted 1:5,000 in blocking buffer for 1 h. The immunoblot was developed with enhanced chemiluminescence, and labeled protein was detected with BioMax MR autoradiography film. Levels of protein were quantified by densitometry using ONE-Dscan image analysis software (Scanalytics). To control for inter- and intrasample loading variability, equivalent cell numbers were used for each experiment and protein load was normalized with a total protein assay. All assays were performed in triplicate.
Isolation and quantification of cell-associated syndecan-4.
Cells were washed twice with cold PBS, after which 500 μl of lysis buffer consisting of 20 mM Tris·HCl pH 7.4, 150 mM NaCl, 10 mM EDTA, 1% Triton X-100, and protease inhibitors was added for a 15-min incubation on ice. Cells were mechanically removed and homogenized by passage several times through a 26-gauge needle, and the lysate was centrifuged at 4°C for 15 min at 16,000 g. Urea was added to 350 μl of the resulting supernatant to a final concentration of 4 M, and the lysate was bound to a Vivapure Mini diethylamine weak anion exchange column with an equilibration buffer, as described above. Proteoglycan elution and GAG chain digestion were otherwise identical to those described in the previous protocol. Samples were resolved by SDS-PAGE and transferred to a 0.45-μm Immobilon-P PVDF membrane, where syndecan-4 was detected as detailed previously.
MAP kinase inhibition.
U-0126 (Cell Signaling Technology), SB-203580 (EMD Biosciences, San Diego, CA), and (L)-JNK inhibitor (JNKI) I (EMD Biosciences) were used to inhibit the phosphorylation cascades of the MAP kinase family in selected experiments. U-0126 is a selective inhibitor of activation of MAP kinase kinase-1 and -2, which are upstream activators of ERK1 and ERK2. SB-203580 and (L)-JNKI I act by directly inhibiting the activity of p38 MAP kinase and JNK/SAPK, respectively. U-0126 and SB-203580 were reconstituted in dimethyl sulfoxide (DMSO) and administered to cells at a final concentration of 20 μM. An equal volume of DMSO was used as a vehicle control. (L)-JNKI I was reconstituted in water and administered at 2 μM. All inhibitors were added 30 min before the onset of the experimental study, with a second dose administered at half the initial concentration 4 h before completion of the 24-h test period. All values were first normalized to the nontreated, nonstretched controls and expressed in terms of percent change from control.
Statistics and data analysis.
To analyze changes in syndecan-4 mRNA expression, quantities at given experimental time points were divided by the quantities of the 0 h time points (unstrained controls) for each experiment. Consequently, the 0 h time points always had the value of 1.0 (100% of control) and thus are not presented graphically in the figures. For the experimental data, expression levels presented as >100% of control represent increases and those presented as <100% of control represent decreases. Values are reported as means ± SE. Statistical analysis was performed with the SigmaStat 3.0 software package, and, where applicable, logarithmic transformation of the data was performed. ANOVA was used, with the Holm-Sidak or Tukey tests for multiple pairwise comparisons and the Kruskal-Wallis nonparametric test to assess three or more unpaired groups when normality assumptions did not apply. Interaction effects between two independent variables were examined by two-way ANOVA. A P value <0.05 was considered statistically significant for all tests.
Acute changes in strain induce syndecan-4 expression in vascular SMCs.
Syndecan-4 mRNA expression was initially investigated in response to static strain. When SMCs were subjected to 5% strain, syndecan-4 expression increased to 141 ± 40% of controls at 4 h and then decreased to 79 ± 10% compared with nonstrained cells (Fig. 2A). A similar trend was observed in response to 10% strain (Fig. 2B). In response to 5% and 10% cyclic strain, syndecan-4 gene expression exhibited a biphasic response, with a maximal increase to 200 ± 34% and 182 ± 17% at 4 h, respectively, followed by a return to basal levels (Fig. 2, C and D). In order to have control values for comparison with the preconditioning experiments described below, 10% cyclic strain studies were examined up to 48 h. When SMCs were subjected to 10% cyclic strain for periods of up to 48 h, syndecan-4 expression at late time points was reduced to nearly one-half compared with nonstrained cells (66 ± 12%).
We subsequently examined whether a change in strain amplitude would alter syndecan-4 gene expression when imposed after an initial preconditioning period of cyclic strain (Fig. 3). Specifically, SMCs were subjected to 24 h of 10% cyclic strain, followed by the presence or absence of a step increase of an additional 10% strain (20% total) for up to 24 h. Syndecan-4 mRNA levels 1, 4, and 24 h after the preconditioning period followed a pattern similar to that observed during the initial onset of cyclic strain. That is, expression increased to 181 ± 1% at 4 h, followed by a significant reduction in basal expression to 78 ± 2% at 24 h. In contrast, a significant change in syndecan-4 gene expression was not observed for SMCs that were cultured under continuous 10% cyclic strain.
Cyclic strain regulates shed and cell-associated syndecan-4 protein.
Since syndecan-4 can be released from the cell membrane via ectodomain shedding, characterizing the effect of mechanical strain on syndecan expression required an analysis of both cell-associated and shed protein (Fig. 4). Shed protein levels increased throughout the duration of an imposed 10% cyclic strain, ranging from 237 ± 57% of nonstrained controls at 1 h to 537 ± 109% at 24 h. In contrast, cell-associated syndecan-4 initially decreased to 38 ± 11% of controls at 1 h, followed by an increase to 181 ± 22% at 4 h, with levels at 24 h that were below those in nonstrained controls (70 ± 6%).
Cyclic strain activates MAP kinases.
With the observation that shed and cell-associated syndecan-4 protein were both regulated by cyclic mechanical strain, the signaling pathways that initiate these events were investigated. As anticipated, ERK1/2, p38 MAP kinase, and JNK/SAPK were activated in response to 10% cyclic strain (Fig. 5). While total ERK1/2 remained constant, the ratio of phosphorylated to total ERK1/2 increased from 107% before strain to 246% at 30 min and remained elevated at 60 min. Similarly, the ratio of phosphorylated to total p38 increased from 8% before strain to 284% at 15 min and the ratio of phosphorylated to total JNK/SAPK increased from 6% before strain to 183% at 60 min.
Strain-regulated expression of cell-associated syndecan-4 is differentially controlled by MAP kinase pathways.
To examine the role of MAP kinases in regulating strain-induced expression of syndecan-4 protein, SMCs were pretreated for 30 min with 20 μM U-0126 (ERK1/2 inhibitor), 20 μM SB-203580 (p38 MAP kinase inhibitor), or 2 μM JNKI I (JNK/SAPK inhibitor) before initiation of mechanical stretch. Inhibition of either ERK1/2 or p38 limited the reduction in cell-associated syndecan-4 typically observed after 24 h of cyclic strain (Fig. 6, A and B). Specifically, for cells treated with U-0126, syndecan-4 levels remained at 62 ± 16% of nonstrained, nontreated controls at 24 h vs. 30 ± 9% in nontreated cells. For cells treated with SB-203580, syndecan-4 levels remained at 82 ± 23% vs. 30 ± 9%. In contrast, inhibition of the JNK/SAPK pathway induced a general suppression of cell-associated syndecan-4 expression (Fig. 6C).
Strain-induced shedding of syndecan-4 is controlled by ERK1/2 and JNK/SAPK pathways.
Shed syndecan-4 was assayed in culture medium exposed to SMCs for 24 h. During this period, cells were either incubated under static conditions or subjected to 10% cyclic strain for periods of up to 24 h. Cells treated with an inhibitor of the ERK1/2 pathway displayed significant suppression of syndecan-4 shedding when subjected to either 4 or 24 h of cyclic strain (Fig. 7A). Specifically, when cells were cyclically strained for 4 h, shed levels were reduced to 12 ± 4% of nonstrained controls, compared with 227% ± 42% observed in control samples that were not treated with the ERK1/2 pathway inhibitor. Similarly, after 24 h of cyclic strain, shed syndecan-4 increased to 209 ± 66% of nonstrained controls but remained markedly reduced compared with 1,110 ± 376% observed in control cultures that did not receive the ERK1/2 inhibitor. A similar effect was noted after inhibition of the JNK/SAPK pathway. After 4 h of cyclic strain, shed syndecan-4 levels were reduced to 7 ± 3% of nonstrained controls, compared with 97 ± 28% in the samples not treated with the JNK inhibitor. Likewise, after 24 h, shed levels were 67 ± 23% of nonstrained controls, compared to 230 ± 49% for the nontreated samples (Fig. 7C). However, strain-induced increase in syndecan-4 shedding was unaffected by p38 MAP kinase inhibition (Fig. 7B).
Previous studies suggest that changes in local mechanical stress within the vascular wall influence matrix-driven SMC adhesive and migratory behavior due, at least in part, to alterations in cell surface levels of syndecan-4 and by the shedding of these HSPGs into the pericellular environment (20). For example, because of the involvement of syndecan-4 in the development of focal adhesions and stress fibers, loss of syndecan-4 may lead to a decrease in focal adhesions, with an increase in cell proliferation, migration, and secretion of ECM components. Likewise, the activity of heparin-binding growth factors, cytokines, proteases, and related ligands may be altered by their interactions with shed syndecans in a manner that promotes local tissue repair. We (20), and others (17), have noted that syndecans may promote local sequestration of growth factors and cytokines, thereby inducing cellular chemotaxis, which may be necessary for the effective resolution of tissue damage. In this report, we observed that syndecan-4 gene and protein expression in vascular SMCs is uniquely affected by distinct force profiles. For example, in response to static stress little alteration in syndecan 4-mRNA levels was initially observed, although a reduction in syndecan-4 expression below initial levels was noted at 24 h. In contrast, when SMCs were subjected to cyclic strain syndecan-4 gene expression exhibited a bimodal response with maximum expression at 4 h. Notably, SMC responsiveness to subsequent alterations in the stress field was confirmed by enhanced syndecan-4 gene expression that followed an imposed increase in cyclic stress after an initial 24-h “preconditioning” period. These data imply that the mechanism by which cells sense a mechanical stimulus can be both rapidly activated and reset once a new biomechanical environment is established.
The kinetics of syndecan-4 protein expression was characterized by an initial decrease in total cell syndecan-4 after the onset of an imposed cyclic stress, which paralleled an observed increase in shed syndecan-4. Thus the increase in syndecan-4 gene expression appears to be a compensatory response to a transient increase in syndecan-4 shedding above constitutive levels. The combination of increased gene expression and continued protein synthesis during the first 4 h of strain likely resulted in an initial elevation of cell-associated syndecan-4 protein, after which a decrease in gene expression, coupled with enhanced ectodomain shedding, results in a return of protein expression to basal levels.
Cells regulate expression and shedding of syndecans via complex mechanisms that are not completely understood, including gene expression, posttranslational modification, and protein trafficking (1, 7, 8). Since an increase in protein synthesis does not necessarily result in increased protein shedding, we elected to separately characterize the kinetics of both cell-associated and shed syndecan-4 protein expression in response of cyclic mechanical stress. Moreover, previous studies of syndecan turnover in response to other stimuli and our present understanding of the signaling pathways that are activated in response to mechanical force provided motivation to examine the role of MAP kinase signaling in strain-induced syndecan-4 expression and shedding.
MAP kinases are a family of serine/threonine kinases involved in the conduction and amplification of critical cellular signaling pathways. The family includes ERK1/2, p38 MAP kinase, and JNK/SAPK. When initially discovered, these kinases were thought to be activated by distinct stimuli. For example, ERK1/2 represents a pair of 44- and 42-kDa protein kinases also known as p44/42 MAP kinase and was initially considered to trigger responses to growth and neurotrophic factors (9). p38 MAP kinase, a 38-kDa protein kinase, was presumed to be associated with responses to inflammatory cytokines (22), and JNK/SAPK was first described as a UV-activated kinase (2, 4). More recently, a variety of cellular stresses, such as heat shock, hyperosmolarity, oxidative stress, radiation, and UV light, are all known to activate ERK1/2 signaling pathways.
Our investigations demonstrated that all three MAP kinase pathways were involved in strain-induced regulation of cell-associated syndecan-4. In particular, JNK/SAPK inhibition reduced the observed increase in cell-associated syndecan-4 that was typically noted after 4 h of cyclic strain. Therefore, the JNK/SAPK signaling pathway appears to be required for the acute and transient increase in cell-associated syndecan-4. Conversely, ERK1/2 and p38 MAP kinase inhibition reduced the observed decrease in cell-associated syndecan that occurred after 24 h of cyclic strain. This suggests that ERK1/2 and p38 MAP kinase signaling are responsible for the decrease in cell-associated syndecan-4 that was observed after prolonged periods of cyclic strain. Admittedly, despite additional dosing of MAP kinase inhibitors, the potential for loss of activity during the course of the experiment may have influenced these results.
Syndecan shedding is regulated by a variety of stimuli, including extracellular ligands (7, 35), bacterial virulence factors (25, 26), and stress (13, 21). However, only a select subset of these stimuli have been shown to act via MAP kinase signaling. In particular, ERK1/2 and PKC signaling have been shown to be required for RANTES-induced shedding of syndecan-1 and syndecan-4 in HeLa cells (5) and for EGF- and thrombin receptor-induced shedding in mouse epithelial cells (8). Most other reports on the mechanism of syndecan shedding have focused on the involvement of intracellular protein tyrosine kinase (29) and matrix metalloproteinase (MMP) (1, 11, 21) activity in this process.
In this report, inhibition of ERK1/2 and JNK/SAPK signaling reduced constitutive syndecan shedding and also appeared to limit accelerated shedding that occurred after initiation of cyclic strain. In contrast, signaling via the p38 MAP kinase pathway did not appear to be involved. In epithelial cells hyperosmolarity and ceramide promote syndecan shedding from epithelial cells through a PKC-dependent pathway, whereas EGF and the LasA Pseudomonas virulence factor act via ERK and JNK, respectively (8, 16, 26). While we did not identify the protease responsible for syndecan-4 shedding in SMCs subjected to mechanical stress, prior reports suggest that ectodomain shedding is a MMP-dependent process, in which different metalloproteinases regulate constitutive and accelerated shedding in a species-specific manner. Specifically, investigations in a mouse model suggest that signaling pathways converge to activate the tissue inhibitor of metalloproteinase (TIMP)-3-sensitive zinc metalloproteinase, MMP-7, whereas in human cells the membrane-type MMPs MT1-MMP and MT3-MMP appear to be involved. Given that syndecan-4 is regulated in SMCs in response to diverse mechanical stimuli and that there is a potential for a wide range of functional roles for both cell surface and shed syndecan-4, further studies are warranted to determine the physiological significance of syndecan-4 in the evolution of vascular wall pathology in vivo.
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 the American Physiological Society