INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

VASCULAR SMOOTH MUSCLE cell motility is essential for developmental growth and the progression of vascular disease (42). However, a complete understanding of the mechanisms underlying these processes is lacking. Nitric oxide (NO) is an autacoid the levels of which are increased in inflammatory conditions, including those that occur in vascular injury or disease (52). Moreover, NO inhibits cell proliferation in tissue culture (16) and attenuates the formation of neointimal masses in injured blood vessels (50). NO also decreases the motility of highly subcultured vascular smooth muscle cells (11, 41), and an analogous effect in vivo could contribute to the attenuation of neointima formation.

Cell adhesion is an important determinant of motility (36). There is evidence that NO decreases adhesion, spreading, and integrin-linked signal transduction in mesangial cells, melanoma cells, and chondrocytes (5, 25, 53). However, the mechanisms contributing to NO-induced altered adhesion and spreading and the relationship of cytoskeletal rearrangements and altered adhesion to cell motility in vascular smooth muscle cells have not been determined.

Cell-matrix interactions are mediated by integrin receptors. Interactions of extracellular matrix molecules with integrins occur in structures termed focal adhesions, constituting a specific domain in cell membranes and linking the actin cytoskeleton to the extracellular matrix. Focal adhesions perform both structural and signaling functions by mediating communication between the extracellular matrix and the intracellular domain (6). One of the most important events after the binding of integrins with their cognate ligands is rapid tyrosine phosphorylation of focal adhesion-associated proteins, including paxillin, focal adhesion kinase (FAK), p130^cas, vinculin, and tensin. Moreover, the levels of tyrosine phosphorylation of FAK and paxillin correlate with the assembly of focal adhesion complexes and actin stress fibers (8, 39).

Tyrosine phosphorylation of focal adhesion proteins is controlled by kinases as well as phosphatases. Although the role of tyrosine kinases, such as FAK, in the regulation of the cytoskeleton and signaling via focal adhesions is relatively well documented (21), the role of tyrosine phosphatases is only beginning to be addressed. Recent studies suggest that phosphatases play a major role in the regulation of phosphotyrosine levels in focal adhesion proteins (38, 40).

Our laboratory has recently reported that NO and cGMP decrease phosphotyrosine levels of focal adhesion proteins in primary cultures of aortic smooth muscle cells isolated from newborn rats via activation of one or more protein tyrosine phosphatases (PTPs) (10, 27). The purpose of the current study was to identify a NO-stimulated PTP whose activation could explain the alteration of cell shape, adhesion, and motility induced by NO in aortic smooth muscle cells. We show for the first time that NO increases PTP-1B activity and that it decreases cell adhesion and alters cell shape in aortic smooth muscle cells. Moreover, we report that PTP-1B is necessary for decreased cell adhesion and altered cell shape induced by NO and that it mediates NO-induced inhibition of cell motility.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Materials. Lactating Sprague-Dawley rats and their pups were purchased from Charles River Laboratories (Wilmington, MA), or pups of the same strain were bred in the University of Tennessee vivarium. Primaria tissue culture plates were from Falcon-Becton Dickinson (Oxnard, CA). Type I collagenase, soybean trypsin inhibitor, fetal bovine serum (FBS), bovine serum albumin (fraction V), 8-bromoguanosine 3',5'-cyclic monophosphate (8-BrcGMP), 8-bromoadenosine 3',5'-cyclic monophosphate (8-BrcAMP), S-nitroso-N-acetyl-penicillamine (SNAP), and rabbit anti-mouse IgG antibodies conjugated to horseradish peroxidase were from Sigma (St. Louis, MO). DMEM-Ham's F-12 (1:1) medium was from GIBCO (Grand Island, NY). Porcine pancreatic elastase, insulin, transferrin, and selenous acid were from Collaborative Research (Lexington, MA). 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one (ODQ) was purchased from Alexis (San Diego, CA). Antibodies against phosphotyrosine, paxillin, FAK, and PTP-1B were purchased from Transduction Laboratories (Lexington, KY).

Cell culture. Smooth muscle cells were obtained from the thoracic aortas of newborn Sprague-Dawley rats (6-9 days old), as previously described (10, 27). The cells were seeded into Primaria culture plates at a density of 1.8-2.3 × 10⁴ cells/cm² and cultured for the first 2 days in serum-free DMEM-Ham's F-12 (1:1) medium supplemented with insulin (5 µg/ml), transferrin (5 µg/ml), and selenous acid (5 ng/ml), plus 50 U/ml penicillin and 50 µg/ml streptomycin, in a humidified atmosphere of 5% CO₂-95% air. Most cells (~95%) attached to the culture surface within the first few hours after seeding and were observed to spread after overnight incubation. After the initial 2-day culture in serum-free medium, FBS was added to a final concentration of 10%. Cells were cultured for an additional 3-5 days in serum-containing medium whereupon they reached confluence. Confluent cultures were used as monolayers. It should be noted that the extracellular matrix to which cells adhered in confluent culture was presumably generated by the cells themselves, because rat aortic cells in culture are known to secrete relatively large amounts of matrix proteins consisting mostly of collagen and elastin (45). Cells were identified as smooth muscle in origin by immunostaining for -smooth muscle actin (results not shown). All experiments in this study were performed using cells in primary culture or up to passage 2. Each individual experiment represents results from one such cell isolate, generally obtained from two rat litters.

Measurement of cell motility. Motility of primary aortic smooth muscle cells of newborn rats (6-9 days old) was measured by using a Transwell culture chamber (Costar, Cambridge, MA) in which the upper and a lower chambers are separated by a collagen-coated polycarbonate filter having 8-µm pores, as described in detail in a recent publication from our laboratory (23).

Measurement of smooth muscle cell adhesion. Adhesion of smooth muscle cells was measured via the use of a modified trypsin-based assay (19) as follows. Smooth muscle cells in 24-well plates were treated with 2 µCi/well of [³H]thymidine in DMEM-F-12 (1:1) containing 10% FBS for 24 h to provide a label that is directly related to cell number. Cells were then rinsed with Hanks' balanced salt solution three times and incubated at 37°C in 1 ml of serum-free DMEM-F-12 supplemented with experimental reagents. Adhesion was determined by rinsing cells twice with Dulbecco's PBS, followed by incubation with 1.0 ml of 0.2% trypsin in the same buffer for 10-15 min at 37°C on a rotating platform at 80 rpm. The released cells were gently rinsed away. The cells remaining in the plate were dissolved with 0.1% SDS-0.1 N NaOH, and the radioactivity of an aliquot was measured by scintillation spectrophotometry. The percentage of unreleased cells was calculated based on the total radioactivity and the radioactivity of unreleased cells after treatment with trypsin.

Measurement of protein phosphotyrosine levels via Western blotting. Cells maintained in serum-free medium were treated with SNAP at 37°C for the indicated time periods. After incubation, cells were rinsed twice with ice-cold PBS and subsequently lysed directly in the culture plates using RIPA lysis buffer (150 mM NaCl, 1% sodium desoxycholate, 0.1% SDS, 1% Triton X-100, and 50 mM Tris, pH 7.2) supplemented with 1 mM sodium vanadate, 1 mM phenylmethylsulfonyl fluoride (PMSF), 5 µg/ml aprotinin, and 5 µg/ml leupeptin. Lysates were centrifuged at 14,000 rpm for 15 min, and protein concentrations were determined using the bicinchoninic acid protein assay. Equivalent amounts of protein were separated on 7.5% SDS-PAGE and electrophoretically transferred to polyvinylidene difluoride (PVDF) membranes. Membranes were blocked using PBS, supplemented with 0.1% Tween 20 plus 3% bovine serum albumin, and probed for phosphotyrosine levels using recombinant antiphosphotyrosine antibodies conjugated to horseradish peroxidase, diluted 1:2,500 in blocking buffer. Immunoreactive bands were visualized using Renaissance chemiluminescence reagents (DuPont-NEN).

Immunoprecipitation. Cells were incubated with experimental agents, followed by lysis at 4°C for 30 min using RIPA buffer. Lysates containing equivalent amounts of protein were precleared by incubating with protein G-conjugated agarose beads for 1 h at 4°C. After centrifugation, lysates were incubated with anti-paxillin (1:500) or anti-FAK (1:200) antisera overnight at 4°C, followed by addition of protein G-Sepharose beads (Pharmacia) and further incubation for 1.5 h or overnight. Sepharose beads were then rinsed twice with RIPA buffer and once with 3 M NaCl containing 1 mM sodium vanadate and 1 mM PMSF, followed by boiling of beads in Laemmli sample buffer and loading of supernatant onto gels for SDS-PAGE. Proteins were transferred onto PVDF membranes, followed by probing with antiphosphotyrosine antibodies, and quantitation with the enhanced chemiluminescence (ECL) kit (NEN). The immunoblots were then treated with stripping solution (62.5 mM Tris buffer, pH 6.7, containing 2% SDS and 100 mM -mercaptoethanol) for 30 min at 60°C and reprobed with corresponding primary antibodies against paxillin (1:5,000) or FAK (1:1,000), followed by horseradish peroxidase-coupled secondary antibodies with appropriate specificity followed by ECL.

Immunophosphatase assay for PTP-1B. After treatment with SNAP, cells were lysed in ice-cold RIPA buffer containing 1 mM PMSF, 5 µg/ml aprotinin, and 5 µg/ml leupeptin. Immunoprecipitation was carried out by incubating lysates (500 mg) with 2 µg of PTP-1B antibody for 3 h at 4°C and further with protein G-Sepharose beads for 1 h. Immunocomplexes were washed twice with RIPA buffer and twice with PTP assay buffer (50 mM sodium acetate, pH 5.5, containing 1 mM EDTA). Immunoprecipitated PTP-1B beads were incubated with 10 mM p-nitrophenyl phosphate as substrate in the above PTP assay buffer for 30 min at 22°C. The reaction was stopped with the addition of 1 M NaOH. The enzyme activity was determined by measuring the increase of absorbance at 405 nm. The values in the absence of enzyme were taken as background and were always subtracted.

Treatment of cells with oligodeoxynucleotides. Smooth muscle cells were seeded into 24-well culture plates and cultured until about 80% confluence in 10% FBS-DMEMF-12. Cells were then rinsed with serum-free medium and cultured for 24 h in serum-free medium, without N-[1-(2,3-dioleoyloxy)propyl]-N,N,N-trimethylammonium methylsulfate (DOTAP) only or with addition of antisense (5'-CCATTTCCATGGCGG-3'), three-base mismatched antisense (5'-ACATTTCAATGGCGT-3'), sense (5'-CCGCCATGGAAATGG-3'), or scrambled antisense (5'-CTAGGCATCGTGCTC-3') oligodeoxynucleotides (ODNs), premixed with transfection reagent DOTAP. Smooth muscle cells were then treated without or with SNAP, and motility, cell shape, or adhesion was measured.

Immunocytochemistry. Immunocytochemical staining for focal contact components or smooth muscle -actin was performed as described previously (53).

Statistical analysis. Statistical analyses were performed via unpaired, two-tailed Student's t-test or by analysis of variance followed by Fisher's test. Data are presented as means ± SD. P < 0.05 was considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

SNAP activates PTP-1B via cGMP pathway. PTP-1B is a ubiquitously expressed enzyme that modulates phosphotyrosine levels in focal adhesion proteins such as paxillin and p130^cas, as shown by us and others (23, 29). From our previous results indicating that NO causes dephosphorylation of focal adhesion proteins (10, 27), we hypothesized that this effect might be attributable to NO-elicited stimulation of PTP-1B activity. To determine whether NO increased PTP-1B activity, primary aortic smooth muscle cells from newborn rats were treated with or without NO donor SNAP. PTP-1B from untreated or treated cells was then immunoprecipitated with anti-PTP-1B antibody, and the activity of immunoprecipitated phosphatase was measured using p-nitrophenyl phosphate as substrate. As shown in Fig. 1, treatment of cells with SNAP for 30 min increased PTP-1B activity in a concentration-dependent fashion. Similar results were obtained via the use of ³²P(poly-tyr-glu) as substrate for PTPase activity (not shown). Moreover, we verified that PTP-1B constituted >80% of immunoprecipitated phosphatase activity as determined via an in-gel phosphatase assay (not shown) and that treatment with SNAP did not alter the amount of immunoprecipitated PTP-1B, as determined by Western blotting (not shown). Finally, we verified that vanadate (10 mM), a selective inhibitor of phosphatase activity, completely blocked the activity of PTP-1B from control or SNAP-treated cells, demonstrating that hydrolysis of p-nitrophenyl phosphate was specifically attributable to PTP activity (data not shown).

View larger version (11K): [in this window] [in a new window]   Fig. 1.   S-nitroso-N-acetyl-penicillamine (SNAP) increases protein tyrosine phosphatase 1B (PTP-1B) activity in concentration-dependent fashion. PTP-1B enzyme activity in immunoprecipitate of lysates was measured as a function of SNAP concentration. Incubation time with substrate was 30 min, and enzyme activity was shown to be linear with time for at least 120 min (not shown). Results are means ± SD from 4 independent experiments. * P < 0.05 via analysis of variance followed by Fisher's protected least-significant difference test.

View larger version (11K): [in this window] [in a new window]   Fig. 2.   Guanyl cyclase inhibitor, 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one (ODQ), blocks SNAP-elicited activation of PTP-1B. Cells were preincubated with 2.5 µM ODQ for 30 min, before treatment with 50 µM SNAP and measurement of PTP-1B activity as described in legend for Fig. 1. Results are means ± SD of 3 experiments. * P < 0.05 relative to control. Effect of ODQ alone was not statistically significant.

Antisense ODN against PTP-1B reverses inhibitory effect of NO on cell motility. NO decreases rat smooth muscle cell motility in subcultured cells (11, 41). The purpose of these experiments was to determine the effect of decreasing PTP-1B levels on inhibition of motility by NO in aortic smooth muscle cells isolated from newborn rats.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Antisense oligodeoxynucleotide (ODN), but not sense, scrambled or three-base mismatch ODN, decreases nitric oxide (NO)-inhibited cell motility. Cells were preincubated with 10 µM ODN for 24 h, followed by treatment with 50 µM SNAP. Results are means ± SD of 3 experiments.

NO alters shape and adhesion of smooth muscle cells. Several groups have reported that NO induces a change in cell shape and adhesion in mesangial cells, melanoma cells, or chondrocytes (5, 25, 53). To determine whether this effect could be demonstrated in vascular smooth muscle cells, we measured the effect of SNAP on cell shape. As shown in Fig. 4, SNAP altered the morphology of cells from that of a well-spread and spindle-shaped configuration to one of less spread, round cells. The shape change was evident after 20-30 min of exposure to SNAP and was most pronounced after 1-3 h, with cell shape gradually returning to normal after 16 h. That the return to normal morphology was not due to NO donor degradation was shown by the refractoriness of cells to a second treatment with NO donor (not shown). The refractoriness was presumably related to the well-known desensitization to NO induced by chronic treatment with NO donors. The effect of SNAP on cell shape was concentration dependent (0.5-100 µM) and was already evident at 0.5 µM (data not shown).

View larger version (139K): [in this window] [in a new window]   Fig. 4.   SNAP alters morphology of aortic smooth muscle cells in time-dependent fashion. Confluent cultures of smooth muscle cells were treated with 50 µM SNAP in serum-free DMEM-F-12 medium for indicated time periods. Parallel cultures incubated in absence of SNAP for same time periods showed no change in morphology. Similar results were obtained in 6 additional independent experiments.

View larger version (179K): [in this window] [in a new window]   Fig. 5.   Effect of cyclic nucleotides or ODQ on cell shape. Confluent cultures of smooth muscle cells in 24 wells were treated for 1.5 h with reagents indicated on figure. Similar results were obtained in 2 additional independent experiments.

View larger version (25K): [in this window] [in a new window]   Fig. 6.   A: SNAP elicits reduction of cell adhesion in aortic smooth muscle cells. Subconfluent cultured smooth muscle cells in 24-well culture plates were labeled with [3H]thymidine, as described in MATERIALS AND METHODS. Cells were incubated with SNAP for 1.5 h in serum-free medium. Percentage of cells remaining attached after treatment with trypsin was calculated by measuring cell-associated radioactivity and is shown on ordinate. Results represent means ± SD from 3 independent experiments. * P < 0.05 relative to control. B: guanyl cyclase inhibitor ODQ blocks SNAP-induced reduction of cell adhesion. Tritium-labeled smooth muscle cells were preincubated with or without 10 µM ODQ for 30 min before exposure to 50 µM SNAP for 1.5 h. Cell adhesion was calculated as previously described. Results are means ± SD from 3 independent experiments. * P < 0.05 vs. control or ODQ plus SNAP category.

Pretreatment of smooth muscle cells with phosphatase inhibitor phenylarsine oxide (PAO) or antisense ODN against PTP-1B blocks effect of SNAP on cell shape and adhesion. If SNAP-induced smooth muscle cell-shape change and reduced adhesion were causally linked to increased phosphatase activity, we would expect that inhibition of PTP activity would block these effects. As expected, pretreatment of smooth muscle cells with the PTP inhibitor PAO blocked SNAP-induced shape change (Fig. 7). Similar results were obtained via the use of a second selective PTP inhibitor, pervanadate (not shown).

View larger version (135K): [in this window] [in a new window]   Fig. 7.   Blockade of SNAP-induced shape change by PTP inhibitor phenylarsine oxide (PAO). Smooth muscle cells were seeded into 24-well plates and treated without or with SNAP in presence or absence of phosphatase inhibitor PAO. Original magnification was ×400. Similar results were obtained in 2 additional independent experiments.

View larger version (181K): [in this window] [in a new window]   Fig. 8.   Blockade of SNAP-induced smooth muscle cell-shape change by antisense ODN against PTP-1B. Smooth muscle cells were treated without or with 10 µM ODN (antisense, sense, scrambled sense, and three-base mismatched antisense ODN against PTP-1B) for 24 h before exposure to 50 µM SNAP for 1.5 h. Original magnification was ×200. Scr, scrambled antisense; SS, sense; M-AS, 3 base pair mutant antisense; AS, antisense.

View larger version (16K): [in this window] [in a new window]   Fig. 9.   Blockade of SNAP-induced decrease of adhesion by pretreatment of smooth muscle cells with antisense ODN against PTP-1B. Smooth muscle cells were prelabeled with [3H]thymidine (2 µCi/ml) for 24 h. After cells were rinsed 3 times with serum-free medium, they were treated without or with 10 µM of various ODNs (antisense, sense, scrambled sense, and three-base mismatched antisense ODN against PTP-1B) for 24 h before exposure to 50 µM SNAP for 1.5 h. Cell adhesion was then measured using a trypsin-sensitive adhesion assay as described in MATERIALS AND METHODS. Results are means ± SD from 3 independent experiments. * P < 0.05 relative to control.

NO induces cytoskeletal disorganization and disassembly of focal adhesions. Maintenance of normal cell shape and adhesion requires the presence of an intact cytoskeleton and specific cell-matrix interactions. The change in smooth muscle cell morphology, as well as decreased cell adhesion, prompted us to examine possible alterations of cytoskeletal organization and focal adhesions that could explain these results. As demonstrated in Fig. 10, after incubation of cells with 100 µM SNAP for 30 min, a more rounded morphology manifesting fewer and less well-organized stress fibers emerged compared with untreated cells. SNAP also induced disassembly of focal contacts as manifested by diffuse staining for paxillin, in contrast to focal staining in untreated cells (Fig. 10). Both of these effects were blocked by ODQ (not shown), further supporting the involvement of cGMP in cytoskeletal disorganization.

View larger version (148K): [in this window] [in a new window]   Fig. 10.   SNAP elicits disorganization of smooth muscle cell actin microfilaments and paxillin. Smooth muscle cells were cultured until semiconfluent and then treated with indicated reagents, followed by immunostaining with anti--smooth muscle actin or anti-paxillin. Original magnification was ×400. Similar results were obtained in 2 additional independent experiments.

SNAP decreases phosphotyrosine levels of paxillin via a cGMP-mediated event that temporally precedes shape change. Phosphorylation of focal adhesion proteins is correlated with increased cell adhesion and stress fiber formation (38). We have previously reported that NO and cGMP induce dephosphorylation of phosphotyrosine in several cytoskeletal proteins, including paxillin (27). To explore the mechanisms involved in NO-induced cell-shape change and decreased adhesion, we determined the time course of NO-induced loss of phosphotyrosine levels in paxillin. As shown in Fig. 11, SNAP caused a significant time-dependent decrease in the phosphotyrosine level of paxillin, an effect that could be detected as early as 2.5 min after treatment. This effect was mimicked by 8-BrcGMP and blocked by the soluble guanylate cyclase inhibitor ODQ (data not shown).

View larger version (16K): [in this window] [in a new window]   Fig. 11.   SNAP decreases phosphotyrosine levels in paxillin in time-dependent fashion. Smooth muscle cells were treated with or without 50 µM SNAP for indicated time periods, protein was solubilized, and paxillin was immunoprecipitated as described in MATERIALS AND METHODS. After SDS-PAGE and transfer to polyvinylidene difluoride (PVDF), blots were probed with either antiphosphotyrosine antibody (left) or stripped and reprobed with antipaxillin antibody (right). Densitometric analysis yielded the following values (time 0 set to 100). Left: 0 min, 100; 2.5 min, 42.9; 5 min, 34.8; 15 min, 9.8; 30 min, 11.7. Right: 0 min, 100; 2.5 min, 115; 5 min, 129; 15 min, 98.4; 30 min, 124. Similar results were obtained in 4 additional independent experiments.

SNAP decreases phosphotyrosine levels of FAK. The decrease of paxillin phosphorylation led us to examine the phosphorylation state of another focal adhesion protein, FAK. FAK appears to be colocalized with paxillin in focal adhesions and in most cases is tyrosine phosphorylated in parallel with paxillin. Moreover, FAK plays a central role in integrin-linked signal transduction and paxillin is thought to be a substrate for FAK (21). As shown in Fig. 12, SNAP decreased phosphotyrosine levels of FAK in a time-dependent fashion, with a time course consistent with the potential involvement of decreased FAK tyrosine phosphorylation, and by inference decreased FAK activity in SNAP-induced cytoskeletal reorganization.

View larger version (17K): [in this window] [in a new window]   Fig. 12.   SNAP decreases phosphotyrosine levels in focal adhesion kinase (FAK) in time-dependent fashion. Cells were treated without or with 50 µM SNAP for indicated time intervals, protein was solubilized, and FAK was immunoprecipitated as described in MATERIALS AND METHODS. After SDS-PAGE and transfer to PVDF, blots were probed with either anti-phosphotyrosine antibody (left) or anti-FAK antibody (right). Densitometric analysis yielded the following values (with time 0 set to 100). Left: 0 min, 100; 5 min, 41.3; 30 min, 60.7; 60 min, 35.4. Right: 0 min, 100; 5 min, 110; 30 min, 193; 60 min, 103. Similar results were obtained in 2 additional independent experiments.

Antisense but not control ODN against PTP-1B reverses decrease of paxillin phosphotyrosine levels induced by SNAP. We next measured the effect of reducing the levels of PTP-1B via antisense ODN on the reduction of paxillin phosphotyrosine levels induced by SNAP. As shown in Fig. 13, antisense ODN, but not several control ODNs, including sense, scrambled antisense, or three-base mismatch antisense, blocked SNAP-induced reduction of phosphotyrosine in paxillin, compared with phosphotyrosine levels in cells not treated with either SNAP or antisense.

View larger version (23K): [in this window] [in a new window]   Fig. 13.   Reversal of NO-induced paxillin dephosphorylation by antisense, but not control ODN, against PTP-1B. Cells were pretreated with various ODN (10 µM) for 24 h before treatment without or with 50 µM SNAP and determination of phosphotyrosine levels in paxillin by direct Western blotting. We separately demonstrated that >90% of phosphotyrosine-containing protein in the range of 70-80 kDa represents paxillin, as shown by quantitative immunoprecipitation, therefore making direct Western blotting of paxillin feasible. Densitometric analysis gave the following results (with control in absence of SNAP set to 100). Without SNAP: C, 100; SS, 99.7; Scr, 136; M-AS, 179; AS, 247. With SNAP: C, 58.3; SS, 43.9; Scr, 61.7; M-AS, 56.31; AS, 144.6. Similar results were obtained in 2 additional independent experiments.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

We have previously reported that NO decreases the levels of phosphotyrosine in several focal adhesion proteins in aortic smooth muscle cells and that this effect can be attributed to increased activity of one (or more) PTPs (10, 27). In the current study we report that NO increases the activity of the ubiquitous phosphatase PTP-1B in rat aortic smooth muscle cells. Moreover, we show that NO-induced inhibition of cell motility is attributable to elevated PTP-1B activity, as well as altered shape change and decreased tyrosine phosphorylation of focal adhesion proteins.

Currently, pharmacological agents that decrease PTP activity in enzyme-specific fashion are not available. We have therefore utilized a specific antisense ODN to decrease PTP-1B protein levels. Antisense ODN blocked the motility-inhibitory effect of NO, whereas three-base mismatch antisense, sense, or scrambled antisense ODN had no effect. The use of multiple control ODNs, including that of a stringent three-base mismatch antisense sequence, increases the probability that the effect of antisense ODN on motility is specifically related to its capacity to decrease the levels of PTP-1B, as reported in a previous publication (23). Moreover, in agreement with the previous study from our laboratory (23), antisense ODN increased cell motility, indicating that constitutively expressed PTP-1B plays a role in regulation of motility. Although antisense ODN restored NO-decreased motility to control levels, it did not prevent the inhibitory effect of NO, at least compared with the elevated level of motility elicited by antisense ODN. On first interpretation, these data seem to indicate that increased PTP-1B may be sufficient, but not necessary, for NO-elicited inhibition of cell motility. However, because antisense ODN decreased but did not eliminate the expression of PTP-1B protein as shown in an earlier publication (23), an unambiguous interpretation is difficult because residual levels of PTP-1B protein might be sufficient to mediate the inhibitory effect of NO. An alternative, although not exclusive, mechanistic possibility is that aortic smooth muscle cells express redundant PTPs involved in NO-elicited effects on cell motility. At least one other PTP, PTP-PEST (proline, glutamate, serine, and threonine), elicits dephosphorylation of some of the same focal adhesion proteins as PTP-1B (44) and overexpression of PTP-PEST in fibroblasts inhibits cell motility (18). We are currently testing the hypothesis that the mechanism of NO-induced inhibition of cell motility may also involve PTP-PEST.

The mechanism by which NO increases PTP-1B activity appears to involve cGMP as second messenger because the potent cGMP agonist 8-pCPT-cGMP mimics the effect of the NO donor and the selective inhibitor of guanyl cyclase ODQ blocks it. We have recently reported that NO and cGMP agonists induce dephosphorylation of focal adhesion proteins via reduction of intracellular calcium (27), whereas others have reported that increase of intracellular calcium is associated with the opposite effect, namely, reduction of PTP activity (12, 35). Thus it is reasonable to suggest that cGMP signaling causes PTP activation by reducing cytoplasmic calcium.

The fundamental importance of adhesion and spreading for cell survival, motility, growth, and morphogenesis is now well appreciated (4). Various agents such as growth factors and chemoattractants, and progression through the cell cycle lead to changes in cell shape. These alterations in cell morphology are linked to modified organization of the actin cytoskeleton and cell adhesion (24). NO exerts diverse effects on migration and proliferation of many cell types, including smooth muscle cells (11, 16, 22). Recent observations indicate that NO modulates cell-matrix interactions, and a possible link between NO-induced alteration of cell-matrix interactions and cell function has been suggested (5, 14, 53). However, little is known about the molecular signaling mechanisms underlying these effects. In this study we provide, for the first time, evidence that NO alters the shape and adhesion of aortic smooth muscle cells and that these effects require PTP-1B activity because both the nonspecific PTP-1B inhibitor PAO and the specific inhibitor antisense ODN block these effects.

The effects of SNAP on smooth muscle cell shape and adhesion occur through the cGMP signaling pathway because the cGMP analog 8-BrcGMP mimicked the capacity of SNAP to induce cell shape and dephosphorylation of paxillin. Moreover, the effects of NO on cell-shape change and reduced adhesion as well as dephosphorylation of paxillin were attenuated by the soluble guanyl cyclase inhibitor ODQ. This conclusion is in agreement with previous observations that NO elicits vasodilatory, antimitogenic, and antimigratory effects in smooth muscle cells via cGMP-dependent mechanisms (11, 16, 37). Furthermore, a recent report indicates that cGMP-dependent protein kinase may be required for thrombospondin- or tenascin-dependent focal adhesion disassembly and cytoskeletal reorganization (33). However, cGMP-independent mechanisms cannot be completely excluded from contributing to NO-induced smooth muscle cell-shape change and reduced adhesion. Indeed, protein kinase C, an important participant in processes regulating cell attachment, spreading, and FAK phosphorylation (51), is known to be inactivated by NO (48). cAMP has also been reported to contribute to cell-shape change, decreased cell adhesion, focal adhesion formation, and paxillin phosphorylation (19, 20). Although some of the effects of NO on cell proliferation may be potentially attributable to increased cAMP levels (7), the lack of effect of 8-BrcAMP on cell shape in smooth muscle cells, as demonstrated by our experiments, makes this unlikely.

Recent studies have emphasized the role of phosphatases in the regulation of tyrosine phosphorylation of focal adhesion proteins (38, 40). In principle, the phosphorylation of proteins would be regulated by kinases as well as phosphatases. The focal adhesion-specific kinase FAK has been extensively investigated in this regard (21) and is a major enzyme involved in tyrosine phosphorylation of paxillin (49). FAK and paxillin are both localized to focal adhesions and are phosphorylated in parallel on tyrosine residues. Moreover, FAK has the capacity to phosphorylate paxillin in vitro (49) and is involved in focal adhesion and stress fiber formation (21). Inhibition of FAK phosphorylation with selective tyrosine kinase inhibitors can interfere with formation of stress fibers and focal adhesions. Moreover, knockdown of FAK using antisense oligonucleotides disrupts cellular attachment and induces cell rounding (15, 39). Our current results indicate that NO elicits dephosphorylation of FAK in a pattern similar to that of paxillin, suggesting that dephosphorylation of paxillin may be at least partially attributable to decreased phosphorylation and hence decreased activity of FAK. This notion is also consistent with the findings of a recent publication indicating that overexpression of NO synthase attenuates platelet-derived growth factor-induced paxillin and FAK phosphorylation in vascular smooth muscle cells (13). Thus dephosphorylation of FAK and paxillin could explain NO-induced cell-shape change and decreased adhesion and motility. Because there is strong evidence of the involvement of specific integrins in vascular smooth muscle cell motility (46), we also speculate that the biochemical changes described here are potentially related to NO-elicited alteration of integrin function.

At least three tyrosine phosphatases are thought to regulate phosphorylation in focal adhesions, including LAR, PTP-1B, and PTP-PEST. LAR is a transmembrane PTP expressed on the cell surface as two noncovalently associated subunits derived by cellular processing of a precursor protein (47). LAR and an associated protein LIP.1 are localized in focal adhesions (43). PTP-1B and PTP-PEST are implicated in focal adhesion signaling as both associate with and elicit the dephosphorylation of p130^cas, a tyrosine-phosphorylated focal adhesion-associated protein (17, 29). Moreover, overexpression of PTP-1B leads to decreased phosphotyrosine levels in p130^cas (31) and interferes with cell-adhesion-stimulated signaling pathways (30). Because antisense ODN against PTP-1B reversed the effect of NO on paxillin phosphorylation, we infer that the reduction of phosphotyrosine induced by NO is attributable to the increase of PTP-1B elicited by NO.

Integrin-mediated linkage of the extracellular matrix to the cytoskeleton occurs at focal adhesion domains. Focal adhesions consist of clustered integrin molecules linked to stress fibers via proteins such as vinculin, paxillin, or talin. These cytoskeletal structures participate in the process of adhesion by stabilizing integrin-ligand interactions and supporting cell shape. Shape change, reduced adhesion, and motility induced by NO reported in this study are associated with disorganization of the cytoskeleton and disassembly of focal adhesions, as shown by immunochemical staining of the cytoskeletal protein actin and the focal adhesion protein paxillin. It is well established that stress fiber formation and assembly of focal adhesions are regulated by tyrosine phosphorylation of focal adhesion proteins, as demonstrated by the use of protein tyrosine kinase inhibitors herbimycin and genistein, which inhibit focal adhesion formation and actin stress fiber assembly (8). Conversely, treatment of cells with PTP inhibitors leads to increased integrin-dependent organization of focal adhesions and actin stress fibers (9, 38). It is therefore reasonable to suggest that the dephosphorylation of focal adhesion proteins preceding NO-induced cell-shape change and reduced adhesion is an early downstream event involved in NO-induced shape change and reduced adhesion. Finally, it is also reasonable to speculate that the phosphorylation level of myosin, which is known to be reduced by NO (32), is likely to play a role in the shape change because there is evidence of cross-talk between the levels of phosphotyrosine and the myosin light chain kinase pathway (26). Moreover, there is also evidence of myosin phosphatase and kinase regulation by rho and rho kinase (1, 28), two proteins of established significance in modulation of stress fibers.

What then are the potential pathophysiological implications of these observations? Cell-matrix interactions profoundly influence smooth muscle cells function, such as migration, contraction, proliferation, and apoptosis. It seems likely that the disruption of focal contacts, change of cell shape, and reduced cell adhesion are mechanistically linked with NO-induced changes of smooth muscle cell proliferation and migration. It is interesting to note that the concentrations of SNAP that induce smooth muscle cell-shape change and reduced adhesion overlap with the concentrations reported by us previously for inhibition of smooth muscle cells growth (10). Recent studies suggest that anchorage-dependent cell growth requires both growth factors and an appropriate extracellular matrix context for the transit of cells through the G1 phase and entry into the S phase (2). Moreover, cell shape also appears to be an independent factor in the modulation of cell proliferation (4). Cell-shape change and reduced adhesion induced by NO might be one of the mechanisms responsible for the reported antimitogenic effect of NO. It should also be noted that we have described recently that NO-induced antimitogenesis was correlated with the dephosphorylation of specific proteins of primary aortic smooth muscle cells, perhaps via activation of one of more PTP (10). These data indicate that cell-shape change and reduced adhesion induced by NO, most likely via dephosphorylation of focal adhesion proteins, might be an important step in NO-induced smooth muscle cell growth inhibition. However, it is also important to note that the effect of NO on both cell proliferation and motility differs in different models of vascular smooth muscle cells, in a manner that correlates with the phenotype expressed by the smooth muscle cells. Thus, whereas NO inhibits both proliferation and motility in relatively dedifferentiated cells (16), it stimulates proliferation and motility in more differentiated cells (3, 22). The underlying mechanistic reasons for this difference are not known.

In summary, we have shown that NO regulates smooth muscle cell shape and adhesion, most likely via a mechanism involving increased PTP-1B activity and dephosphorylation of phosphotyrosine in focal adhesion proteins. These findings provide new insight into the mechanisms linked to the effects of NO on proliferation and migration of smooth muscle cells.