Focal Adhesion Protein Zyxin Is a Mechanosensitive Modulator of Gene Expression in Vascular Smooth Muscle Cells
Excessive deformation of vascular smooth muscle cells (SMCs) caused by a prolonged increase in blood pressure (eg, in hypertension) results in an adaptive remodeling of the vessel wall that is characterized by SMC hypertrophy or hyperplasia and contributes to fixation of the increase in blood pressure. The onset of this process is characterized by a unique change in gene expression in the SMC. However, thus far, no transcription factor has been identified that specifically mediates mechanosensitive gene expression in these cells. Therefore, the role of a putative mechanotransducer, the cytoskeletal protein zyxin, was investigated in rat aortic cultured SMCs. Immunofluorescence and Western blot analysis revealed that on exposure to cyclic stretch, but not to osmotic stress or treatment with proinflammatory cytokines, zyxin dissociates from focal adhesions and accumulates in the nucleus. Unlike zyxin, vinculin, another focal adhesion-associated protein, did not translocate. Moreover, antisense oligonucleotide downregulation of zyxin protein abundance suggested that zyxin accumulation in the nucleus is a prerequisite for mechanosensitive gene expression in these cells. Thus, stretch-induced endothelin B receptor expression, for example, was attenuated, whereas that of tenascin-C was augmented after zyxin suppression. The data are consistent with a role for zyxin in transducing mechanical stimuli from the cell membrane to the nucleus in vascular SMCs and in controlling the expression of mechanosensitive genes that have been implicated in hypertension-induced arterial remodeling.
By virtue of the resulting changes in circumferential wall stress and the degree of stretch that the cells are exposed to, blood pressure variations have a major impact on the phenotype of smooth muscle cells (SMCs) in conduit and resistance-size arteries. Although transient increases in blood pressure result in a compensatory increase in SMC tone, as exemplified by the myogenic response, pressure overload triggers a profound change in gene expression in these cells that ultimately leads to an adaptive remodeling of the vessel wall.1 This is characterized by SMC hypertrophy (conduit arteries) or SMC hyperplasia (resistance-size arteries), resulting in an increase in wall thickness that offsets the blood pressure-induced rise in wall stress. However, this structural adaptation causes an increase in vascular resistance and, although aimed at regaining local blood flow control, eventually contributes to fixation of the elevated blood pressure as, eg, in hypertension.2 Arterial hypertension is one of the primary risk factors for atherosclerosis, hence myocardial infarction, stroke, and dilated cardiomyopathy. Therefore, pressure-dependent arterial remodeling is a major contributor to the pathogenesis of these diseases.3
Although the aforementioned interrelation is well established in the clinic, little is known about SMC mechanotransduction, namely the activation of transcription factors that specifically link deformation of the cell to the increase in expression of mechanosensitive genes. This knowledge, however, could pave the way to developing drugs that specifically interfere with pressure-induced arterial remodeling. Therefore, we have analyzed the role of a putative mechanotransducer in rat aortic cultured SMC, the cytoskeletal protein zyxin.
Isolation and culture of rat aortic SMCs were performed as previously described.4 Only batches of SMCs (passage 3) with ≥98% of cells revealing the typical pattern of α-SMC actin staining (according to immunofluorescence analysis) were used. They were cultured on BioFlex collagen I elastomers and exposed to cyclic stretch (15% elongation at 0.5 Hz) in a Flexercell FX-3000 strain unit as previously described.4 Alternatively, cells were treated with the cytokine combination interleukin-1β (IL-1β) (60 U/mL) plus tumor necrosis factor-α (TNF-α) (1000 U/mL) or exposed to osmotic stress (addition of 0.1 mL of 900 mmol/L sodium chloride per mL medium, resulting in a final osmolarity of 454 mosmol/L) for the times indicated.
The antisense oligonucleotide (ODN) directed against zyxin had the sequence 5′-GGGGCCGCCATGGCC-3′ (position −4 to 11 of the coding sequence; GenBank accession number NM_146002). As a control, a scrambled ODN (5′-CGCGCGGCA-CTCGGG-3′) with no homology to zyxin was also designed. The terminal 4 bases of either ODN were linked by phosphorothioate esters for added stability. The ODN were added to the cultured SMCs at a concentration of 5 μmol/L for 48 hours without a transfection reagent before commencing of the experiments. This delivery method previously has been successfully used by us in rat aortic SMCs.5
Reverse-Transcription Polymerase Chain Reaction Analysis
Isolation of RNA and reverse-transcription polymerase chain reaction analysis of relative amounts of cDNA were performed as described previously.4 Primers used had annealing temperatures of 58°C and are listed in References 4 and 5 except those for cyclooxygenase-1 (COX-1), forward primer, 5′-TACTATCCATGCCAGAACCAGG-3′; reversed primer, 5′- TGGTAACTGCTTCTTCCCTTTG-3′; plasminogen activator inhibitor-1 (PAI-1), 5′-CCTCCAAAGACCGAAATGTG-3′, 5′-GAATCTGGCTCTTTCCACCTC-3′; and tenascin-C, (5′-CAGTGCCATAGCAACAACAGC-3′, 5′-TGTATTCCCAGACAC-TGTGCG-3′).
Preparation of Nuclei and Western Blot Analysis
Proteins were prepared essentially as described previously,5 except that the nuclear and cytosolic fractions were subsequently precipitated with 3 volumes of ice-cold ethanol, solubilized with the appropriate volume of loading buffer containing 2% SDS, and subjected to SDS-PAGE.
Western blot analysis for zyxin and β-actin as a loading control followed by densitometry was performed as described previously.5 For detection of zyxin, a primary rabbit polyclonal antibody (provided by Dr M. Beckerle, Department of Biology, University of Utah, Salt Lake City, Utah) or a mouse monoclonal antibody6 was used together with the corresponding secondary HRP-conjugated antibodies (Sigma, Deisenhofen, Germany) and the Super Signal Blaze chemiluminescence reagent (Pierce, St. Augustin, Germany).
Cells were fixed with p-formaldehyde, and immunostaining with antibodies against vinculin (DPC, Bad Nauheim, Germany) and zyxin6 was performed by a standard protocol7 using Alexa Fluor 488-coupled anti-mouse antibodies (Molecular Probes via MoBiTec, Göttingen, Germany).
Unless indicated otherwise, results are expressed as means±SEM of n observations with cells obtained from different animals. One-way analysis of variance followed by Dunnett post-hoc test was used to determine statistically significant differences between the means and/or the means and control, with P<0.05 considered significant (InStat v3.05; GraphPad Software, San Diego, Calif).
Nuclear Translocation of Zyxin in Response to Cyclic Stretch
Immunofluorescence analysis revealed that vinculin and zyxin are exclusively located in focal adhesions in the rat aortic cultured SMCs under static conditions. However, on stretching of the cells (15% elongation at 0.5 Hz), zyxin started to dissociate from the focal adhesions and to accumulate in the nucleus within 10 minutes (not shown). After 30 minutes, most of the zyxin was present in the nucleus (Figure 1a), whereas at 6 hours the majority of the protein had relocated to the focal adhesions (Figure 1b). Vinculin, however, remained in the focal adhesions, indicating that these were not simply disrupted by mechanical strain. Separate Western blot analyses confirmed the nuclear translocation of zyxin in response to cyclic stretch (Figure 1b).
In contrast to stretching of the cells, and irrespective of the time of exposure (ranging from 15 minutes to 12 hours), neither osmotic stress (increase in osmolarity from 290 mosmol/L to 454 mosmol/L) nor simulation of a proinflammatory response (60 U/mL IL-1β plus 1000 U/mL TNF-α) altered the intracellular localization of zyxin (Figure 2). Also, addition of leptomycin B (1 μmol/L),8 a specific inhibitor of the nuclear export of proteins harboring a nuclear export signal such as zyxin, did not result in an accumulation of zyxin in the nucleus of quiescent SMC during a 12- to 24-hour period (Figure 2), suggesting that, normally, the protein is effectively excluded from the nucleus in these cells.
Zyxin and Mechanosensitive Gene Expression
To analyze whether the nuclear translocation of zyxin in SMCs exposed to cyclic stretch affects the expression of mechanosensitive genes, the antisense ODN approach was chosen. Apparently, the turnover of zyxin in quiescent rat aortic cultured SMCs is rather low, because the antisense ODN, but not the corresponding scrambled control ODN, was effective only (78±9% decrease in protein abundance, n=3, P<0.05) (Figure 3a) when the cells were stretched for 30 minutes followed by a 48-hour incubation period with the antisense ODN. This finding and the increase in zyxin mRNA on cyclic stretch (4.4±0.7-fold after 3 hours, n=5, P<0.05 versus static control) suggest that turnover of the protein in SMCs is greatly enhanced in response to mechanical deformation.
As readout, 3 mechanosensitive gene products were selected whose expression in vascular SMCs is upregulated by cyclic stretch and thought to play a role in vascular remodeling, ie, the endothelin B receptor (ETB-R),1,4 the matrix protein tenascin-C,9,10 and plasminogen activator inhibitor-1 (PAI-1).9,11 Downregulation of zyxin protein abundance affected expression of these mechanosensitive gene products, albeit in a different manner (Figure 3b). Thus stretch-induced ETB-R mRNA expression was clearly suppressed, while that of tenascin-C was augmented. Stretch-induced expression of PAI-1 mRNA was insensitive to zyxin depletion. Moreover, stretch-induced expression of other gene products that are not considered to be truly mechanosensitive9,12 was also differentially affected by zyxin antisense ODN treatment (monocyte chemoattractant protein-1 [MCP-1], 110%±46% of scrODN control, n=9; COX-1, 41%±6% of scrODN control, n=9, P<0.01).
Hypertrophy in conduit and hyperplasia in resistance-size arteries is the clinically visible outcome of a supraphysiological pressure-induced deformation of vascular SMCs (eg, in arterial hypertension).1,2 By virtue of fixing vascular resistance at a higher level, this adaptive remodeling of the vessel wall is a major contributor to atherosclerosis, which is ultimately responsible for coronary heart disease, dilated cardiomyopathy, renal insufficiency, or stroke.3 Although phenomenologically characterized in detail, the signal transduction pathways underlying the deformation-induced shift in SMC gene expression are still poorly understood. With the cytoskeletal protein zyxin, we may have identified a specific mechanosensitive signaling protein altering gene expression in vascular SMC.
Mechanotransduction in Smooth Muscle Cells
Focal adhesions mediating cell-matrix interactions have been postulated to serve as a mechanosensor in vascular SMC both in vitro and in vivo.13 However, signaling through these structures appears to be rather complex, involving several protein kinases whose activation has also been implicated in the response of SMCs to other stimuli, such as proinflammatory cytokines or vasoactive peptides. In addition, transcription factors, such as activator protein-1 (AP-1) or CCAAT/enhancer binding protein β that have been shown to be mechanosensitive,14 are also activated by such stimuli. The specific alterations in gene expression in vascular SMCs exposed to cyclic stretch, however, argue for the activation of a signaling molecule that, in addition to these common stress factors, acts as a specific mechanotransducer. This signaling molecule may, but not necessarily must, be associated with focal adhesions in vascular SMCs.
Zyxin, a Zinc-Finger Protein With Dual Function
Zyxin is predominantly located in focal adhesions and stress fibers. Besides the N-terminal domain mediating its association to focal adhesions,15 zyxin has a C-terminal LIM-domain comprising 3 zinc-finger motifs known to mediate protein–protein and/or protein–DNA interactions.16 Originally, zyxin was characterized as a protein coordinating the organization of actin filaments at focal adhesions and cell-to-cell contacts in fibroblasts and epithelial cells. Thus, zyxin is likely to function as a structural protein in vascular SMCs, too.
The second function of zyxin, shown herein, is that of modulating stretch-induced gene expression in vascular SMCs. Thus, within minutes zyxin translocates to the nucleus of SMCs exposed to cyclic stretch and alters the expression of mechanosensitive genes. Osmotic stress or simulating a proinflammatory reaction, however, did not result in a detectable dissociation of zyxin from focal adhesions.
How its effect on mechanosensitive gene expression is brought about remains to be elucidated. The presence of the LIM-domains and the finding that a similar LIM-domain present in the zyxin homolog, lipoma-preferred protein, can directly induce gene expression in an artificial assay system17 suggest that zyxin may act as a transcription factor. Otherwise, zyxin may affect gene expression exclusively through protein–protein interactions, as described for related zinc-finger proteins such as the GATA family of transcription factors.18
Zyxin-Induced Gene Expression in SMCs Exposed to Cyclic Stretch
A role for zyxin in mechanosensitive gene expression was evidenced by antisense ODN-based downregulation of zyxin protein abundance. Although only a small number of genes was analyzed whose expression had been shown to be mechanosensitive in vascular SMCs,4,9,12 the role of zyxin varied depending on the type of gene studied. Stretch-induced PAI-1 and MCP-1 expression appeared to be zyxin-insensitive, ETB-R and COX-1 expression were attenuated, and that of tenascin-C was unblocked.
The lack of effect of zyxin on PAI-1 and MCP-1 expression becomes intelligible when considering that regulation of these genes is normally mediated by common transcription factors, such as AP-1 or CCAAT/enhancer binding protein β, which are mechanosensitive as well.14 Stretch-induced ETB-R expression, however, is only partially mediated by these transcription factors;14 therefore, it may rely on zyxin activation. SMC expression of the receptor along with stretch-induced synthesis of endothelin-1 by the endothelium results in SMC apoptosis,4 which together with the concomitant changes in matrix protein composition may form the basis for pressure-induced vascular remodeling.1–3
The role of COX-1 in this context is difficult to interpret.9 Stretch-induced COX-1 expression may counterbalance or further the remodeling process, depending on the type of prostanoid synthesized. In either case, regulation of the COX-1 gene, like that of the ETB-R, is positively controlled by zyxin, whereas the opposite seems true for the tenascin-C gene. Perhaps, the blockade of tenascin-C expression is a kinetic effect that wears off in conditions of enduring pressure overload. This would make sense insofar that premature upregulation of tenascin-C expression may antagonize SMC hypertrophy and/or hyperplasia.10
When combined, the aforementioned data indicate that zyxin not only contributes to organizing the actin cytoskeleton6,15 but also contributes to a second important feature of SMCs, ie, the changes in gene expression, hence, phenotype that occur as an adaptive response to enduring mechanical strain. The protein may thus serve to stabilize the vessel wall in conditions of a transient, eg, exercise-induced increase in blood pressure and during pressure overload, as in arterial hypertension.
The data presented herein are consistent with a role for zyxin in transducing mechanical stimuli from the cell membrane to the nucleus in vascular SMC. Future work must elucidate the mechanisms of stretch-induced zyxin activation and zyxin-mediated gene expression. This will lead to a better understanding of the principles of mechanotransduction in vascular cells and will also help to define a new therapeutic approach to limit the often-exaggerated pressure-induced remodeling of conduit and/or resistance-size arteries, such as in hypertension.
This work was supported by the Deutsche Forschungsgemeinschaft (SFB/TR2; project B2). The authors are indebted to Drs Mary Beckerle (Department of Biology, University of Utah, Salt Lake City, Utah), Jürgen Wehland (GBF, Braunschweig, Germany), and Ulrich Walter (Institute of Clinical Biochemistry and Pathobiochemistry, University of Würzburg, Germany) for providing the anti-zyxin antibodies, and to Dr Minoru Yoshida (Department of Biotechnology, Graduate School of Agriculture and Life Sciences, University of Tokyo, Japan) for providing leptomycin B. The expert technical assistance of Annette Bennemann, Melanie Böning, and Renate Dohrmann is gratefully acknowledged.
- Received September 24, 2003.
- Revision received October 13, 2003.
- Accepted January 14, 2004.
Intengan HD, Schiffrin EL. Vascular remodeling in hypertension: roles for apoptosis, inflammation and fibrosis. Hypertension. 2001; 38: 581–587.
Mulvany MJ, Small artery remodeling and significance in the development of hypertension. News Physiol Sci. 2002; 17: 105–109.
Cattaruzza M, Dimigen C, Ehrenreich H, Hecker M. Stretch-induced endothelin B receptor-mediated apoptosis in vascular smooth muscle cells. FASEB J. 2000; 14: 991–998.
Cattaruzza M, Schäfer K, Hecker M. Cytokine-induced down-regulation of zfm1/splicing factor 1 promotes smooth muscle cell proliferation. J Biol Chem. 2002; 277: 6582–6589.
Rottner K, Krause M, Gimona M, Small JV, Wehland J. Zyxin is not colocalized with vasodilator-stimulated phosphoprotein (VASP) at lamellipodia tips and exhibits different dynamics to vinculin, paxillin, and VASP in focal adhesions. Mol Biol Cell. 2001; 12: 3103–3113.
Ausubel, F. M. ed. Current protocols in molecular biology. New York: Wiley and Sons; 1997.
Feng Y, Yang JH, Huang H, Kennedy SP, Turi TG, Thompson JF, Libby P, Lee RT. Transcriptional profile of mechanically induced genes in human vascular smooth muscle cells. Circ Res. 1999; 85: 1118–1123.
Ingber DE. Mechanical signaling and the cellular response to extracellular matrix in angiogenesis and cardiovascular physiology. Circ Res. 2002; 91: 877–887.
Cattaruzza M, Eberhardt I, Hecker M. Mechanosensitive transcription factors involved in endothelin B receptor expression. J Biol Chem. 2001; 276: 36999–37003.
Crawford AW, Michelsen JW, Beckerle MC. An interaction between zyxin and alpha-actinin. J Cell Biol. 1992; 116: 1381–1393.
Sadler I, Crawford AW, Michelsen JW, Beckerle MC. Zyxin and cCRP: two interactive LIM domain proteins associated with the cytoskeleton. J Cell Biol. 1992; 119: 1573–1587.
Petit MM, Fradelizi J, Golsteyn RM, Ayoubi TA, Menichi B, Louvard D, Van de Ven WJ, Friederich E. LPP, an actin cytoskeleton protein related to zyxin, harbors a nuclear export signal and transcriptional activation capacity. Mol Biol Cell. 2000; 11: 117–129.
Crossley M, Merika M, Orkin SH. Self-association of the erythroid transcription factor GATA-1 mediated by its zinc finger domains. Mol Cell Biol. 1995; 15: 2448–2456.