Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Nitric oxide (NO) is antiproliferative for vascular smooth muscle cells (VSMCs) in vitro and in vivo. Chemical NO donors reversibly inhibit the proliferation of cultured VSMCs (Garg and Hassid, 1989; Kariya et al., 1989; Assender et al., 1992; Estrada et al., 1997; Ishida et al., 1997; Yu et al., 1997). Intimal VSMC hyperplasia after balloon injury is inhibited by supplementation with L-arginine, the metabolic precursor of NO (McNamara et al., 1993; Tarry and Makhoul, 1994), and by introduction of endothelial NO synthase (eNOS) gene (von der Leyen et al., 1995). Conversely, mice with a targeted disruption of eNOS display an increase in arterial wall thickness accompanied by cellular hyperplasia, suggesting that endogenous NO is a negative regulator of VSMC proliferation (Rudic et al., 1998).

In an investigation of the effects of S-nitroso-N-acetylpenicillamine (SNAP), a NO-releasing agent, on the cell cycle of VSMCs, we found that the G₁ inhibition induced by NO may result from the induction of p21^{Waf1/Cip1/Sdi1}, a cyclin-dependent kinase inhibitor (Ishida et al., 1997). However, the mechanism underlying NO-mediated p21 induction remains to be determined.

p21 was discovered from genes induced by the tumor suppressor p53 (El-Deiry et al., 1993). p53 accumulates after cells are exposed to DNA-damaging stimuli, such as irradiation and alkylating agents, and induces several genes, including p21, by functioning as a transcription factor (Agarwal et al., 1998). Cytotoxic levels of NO released from high concentrations of NO donors or generated by inducible NO synthase (iNOS) have been reported to increase the level of p53 (Meßmer et al., 1994; Forrester et al., 1996; Ho et al., 1996; Ambs et al., 1997). However, it is unclear whether lower concentrations of NO, which inhibit cell proliferation but do not cause cytotoxicity or apoptosis, activate p53. Therefore, we investigated whether p53 is involved in the up-regulation of p21 and the inhibition of cell proliferation induced by NO.

It is established that NO-induced smooth muscle relaxation is mediated by the soluble guanylate cyclase (sGC)-cGMP pathway (Murad, 1986; Walter, 1989). However, it is a matter of some controversy whether the antiproliferative effect of NO also is mediated by this pathway. Indeed, some studies showed that cGMP analogs and phosphodiesterase inhibitors inhibited VSMC proliferation (Abell et al., 1989; Garg and Hassid, 1989; Kariya et al., 1989; Furuya et al., 1991; Assender et al., 1992; Yu et al., 1997). However, others have implied cGMP-independent mechanisms. NO donors were able to inhibit the proliferation of BALB/c3T3 fibroblasts in spite of the lack of sGC activity (Garg and Hassid, 1990). VSMC proliferation was not inhibited by cGMP analogs and cGMP-specific phosphodiesterase inhibitors (Garg and Hassid, 1990; Estrada et al., 1997). Therefore, we also examined whether the sGC-cGMP pathway is involved in the p21 expression and antiproliferative effect induced by NO.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals. SNAP was purchased from Research Biochemicals Inc. (Natick, MA). Cycloheximide (CHX), actinomycin D, 8-bromo-cGMP, and 3-isobutyl-1-methylxanthine (IBMX) were purchased from Sigma Chemical Co. (St. Louis, MO). Rp-8-bromoguanosine-3',5'-monophosphorothioate (Rp-GMPS) and Rp-8-bromoadenosine-3',5'-monophosphorothioate (Rp-AMPS) were purchased from Biolog Life Sciences Institute (Bremen, Germany). 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one (ODQ) was purchased from Wako Pure Chemicals Industries (Osaka, Japan). Other common chemicals were of reagent grade.

Cell Culture. VSMCs obtained from human umbilical arteries as described in Ishida et al. (1997) were maintained in Dulbecco's modified Eagle's medium (DMEM) containing 20% (v/v) fetal bovine serum (FBS) (Life Technologies, Rockville, MD), 5 ng/ml human recombinant basic fibroblast growth factor (Amersham Pharmacia Biotech, Uppsala, Sweden), 100 U/ml penicillin, 100 µg/ml streptomycin, and 1 µg/ml amphotericin B (growth medium). Cells synchronized in a quiescent state (G₀) by serum starvation for 48 h were stimulated with growth medium to reenter the cell cycle. Cell numbers were determined with a Coulter counter (Z1; Coulter Electronics, Luton, Beds, UK). VSMC line P53LMACO1 cells obtained from thoracic aorta of p53-knock out mice (Ohmi et al., 1997) were kindly provided by Yoshiaki Nonomura, Teikyo University, Tokyo, Japan. P53LMACO1 cells and VSMCs from wild-type mice (C57BL/6J strain) were cultured in DMEM containing 10% FBS on dishes coated with collagen type I-C (Iwaki Glass, Chiba, Japan). A431 cells obtained from Riken Cell Bank (Saitama, Japan) also were maintained in DMEM supplemented with 10% FBS.

DNA Synthesis Assay. DNA synthesis was assessed by the level of [³H]thymidine ([³H]TdR) incorporation as described in Ishida et al. (1997).

Immunoprecipitation and Western Blotting. For detection of p21, cell lysates were immunoprecipitated and analyzed by Western blotting as described in Ishida et al. (1997). For p53, cells were lysed in 10 mM Tris-HCl (pH 7.4) containing 150 mM NaCl, 1 mM EDTA, 1% (v/v) Nonidet P-40 (Nacalai Tesque, Kyoto, Japan), 0.1% (w/v) sodium deoxycholate, 0.1% (w/v) SDS, 10 mM NaF, 200 µM Na₃VO₄, 20 µg/ml phenylmethylsulfonyl fluoride (PMSF), and 20 µg/ml leupeptin (RIPA buffer). Lysates were sonicated for 10 s and rocked on ice for 1 h, followed by a centrifuge at 16,000g for 20 min to remove insoluble pellets. The protein concentrations were determined by the modified Lowry's method with Dc protein assay kit (Bio-Rad Laboratories, Hercules, CA). Proteins (20 µg/lane) were fractionated by SDS-polyacrylamide gel electrophoresis (PAGE) (10%), electroblotted onto a polyvinylidene difluoride membrane, and immunoblotted with an antihuman p53 monoclonal antibody (2 µg/ml, DO-7; PharMingen, San Diego, CA). To normalize the amounts of proteins applied to SDS-PAGE, the membranes were reprobed with anti--tubulin monoclonal antibody (2 µg/ml, DM1A; Oncogene Research Products, Cambridge, MA).

Northern Blotting. cDNA of p21 was prepared as described in Ishida et al. (1997). Total cellular RNAs (10 µg/lane) were electrophoresed and analyzed by Northern blotting as described in Ishida et al. (1997).

Reverse Transcription-Polymerase Chain Reaction (RT-PCR). RT-PCR was performed with Ready-To-Go RT-PCR beads (Amersham Pharmacia Biotech). Total cellular RNAs (1 µg) were used for the RT reaction and the products were amplified by 25 thermal cycles with DNA Thermal Cycler 480 (Perkin-Elmer Cetus Instruments, Norwalk, CT). PCR primers for p53 were synthesized based on the GenBank database (5'-ATGGAGGAGCCGCAGTCAGA-3' and 5'-CACTCGGATAAGATGCTGAG-3'). PCR products were electrophoresed on 2% agarose gel and visualized by staining with ethidium bromide. Amplified DNAs were identified by sequencing.

Transfection and Luciferase Assay. The 2.4-kilobase human genomic p21 promoter, which contained the transcriptional initiation site at its 3' end, was excised with HindIII from wild-type waf1 promoter (WWP)-Luc (El-Deiry et al., 1993) and subsequently cloned into the HindIII site of PGV-B2, a firefly luciferase reporter vector (Toyo Ink Mfg. Co., Tokyo, Japan). To construct a plasmid containing a deletion mutant of the promoter lacking p53 consensus sequences, a 1.3-kilobase DraI/HindIII fragment excised from WWP-Luc was cloned between the SmaI/HindIII sites of PGV-B2. These plasmids were designated PGV-WWP and PGV-deletion mutant (DM), respectively. Cells seeded in 12-well plates (5 × 10⁵ cells/well) were cultured in growth medium for 24 h. After three washes with antibiotic-free DMEM containing 0.1% bovine serum albumin, the cells were incubated in the same medium for 45 h for synchronization in G₀. Plasmid DNAs (600 ng/well) were transiently transfected into the cells with LipofectAMINE PLUS reagent (Life Technologies) as described in the manufacturer's protocol. Simultaneously, pRL-simian virus 40 (SV40) (60 ng/well, Toyo Ink. Mfg. Co.), which contained Renilla luciferase gene with an SV40 promoter, was cotransfected as a control for transfection efficiency. Transfected cells were incubated in growth medium in the absence or presence of SNAP (100 µM) for 18 h. After two washes with phosphate-buffered saline, the firefly and Renilla luciferase activities were measured sequentially with a double luciferase assay system (Toyo Ink Mfg. Co.) and a luminometer (Dia-Iatron, Tokyo, Japan). Firefly luciferase activities were normalized to those of Renilla luciferase.

Electrophoretic Mobility Shift Assay. To prepare nuclear extracts, cells were suspended in the hypotonic buffer [10 mM HEPES/KOH (pH 7.9), 10 mM KCl, 100 µM EDTA, 0.1% Nonidet P-40 (v/v), 1 mM dithiothreitol, 500 µM PMSF, 2 µg/ml aprotinin, 2 µg/ml leupeptin, 1 mM Na₃VO₄] and incubated for 10 min on ice. After centrifugation at 3000g for 1 min, the pelleted nuclei were resuspended in the extraction buffer [50 mM HEPES/KOH (pH 7.9), 420 mM KCl, 5 mM MgCl₂, 100 µM EDTA, 1 mM dithiothreitol, 500 µM PMSF, 2 µg/ml aprotinin, 2 µg/ml leupeptin, 1 mM Na₃VO₄, and 20% glycerol] and incubated for 30 min on ice. The lysates were centrifuged at 16,000g for 15 min and resultant supernatants were used as nuclear extracts. Double-stranded oligonucleotides containing a p53 consensus sequence 5'-TCAGGAACATGTCCCAACATGTTGAGC-3' were labeled at 5'-ends with ³²P by T4 polynucleotide kinase, and purified with Sephadex G-50 (ProbeQuant G-50 Micro Columns; Amersham Pharmacia Biotech). For the DNA binding reactions, nuclear extracts (10 µg of protein) were incubated with the labeled DNA probe (~5 × 10⁴ cpm) and poly(dI-dC)·poly(dI-dC) (2 µg; Amersham Pharmacia Biotech) in the binding buffer [12.5 mM HEPES/KOH (pH 7.9), 105 mM KCl, 1.25 mM MgCl₂, 250 µM EDTA, 25 µM dithiothreitol, 12.5 µM PMSF, 500 ng/ml aprotinin, 500 ng/ml leupeptin, 250 µM Na₃VO₄, and 5% glycerol] for 1 h on ice. For the competition experiments, a 100-fold molar excess of unlabeled oligomers was added before the addition of labeled probe. To identify the protein bound to DNA, nuclear extracts were preincubated with an antihuman p53 monoclonal antibody (5 µg of PAb421; Oncogene Research Products) on ice for 1 h before adding the labeled probe. Protein-DNA complexes were electrophoresed on 6% native polyacrylamide gel in 0.5× Tris, boric acid, and EDTA buffer [1× Tris, boric acid, and EDTA: 89 mM Tris, 89 mM boric acid, and 2 mM EDTA (pH 8.0)] at 4°C. Dried gels were analyzed for radioactivity with a bioimage analyzer BAS-2500 (Fuji Photo Film Co., Tokyo, Japan).

cGMP Assay. To extract cGMP, cells (~1.5 × 10⁵) were frozen at 80°C after the culture medium was replaced with 1 ml of ice-cold pure ethanol. Cells were centrifuged at 2000g for 5 min at 4°C, and the supernatant was transferred to a new tube. A 200-µl aliquot was completely evaporated with a vacuum concentrator (Speed Vac Plus SC210A; Savant, Holbrook, NY), and dissolved in water. cGMP concentrations were determined with a radioimmunoassay kit (Yamasa Shoyu Co., Chiba, Japan).

Statistics. Results are expressed as means ± S.D. of the number of observations. Statistical significance was assessed by Student's t test for paired or unpaired values.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

p53 Is Involved in NO-Induced p21 Expression and Growth Arrest. SNAP stimulates the induction of p21 expression in VSMCs (Ishida et al., 1997). To examine whether NO stabilizes p21 protein and mRNA, we measured the rates of their degradation with CHX, a protein synthesis inhibitor, and actinomycin D, an RNA synthesis inhibitor, respectively. The protein and mRNA, which had been accumulated by incubation with growth medium for 3 h, were degraded as cells were incubated with CHX and actinomycin D, respectively (Fig. 1). SNAP (100 µM) had no significant effect on their rates of degradation. Therefore, it is unlikely that this compound induces p21 expression by stabilizing the protein and mRNA.

View larger version (30K): [in this window] [in a new window]   Fig. 1.   Effects of SNAP on the rates of degradation of p21 protein and mRNA. a, G0 cells were stimulated with growth medium in the absence or presence of SNAP (100 µM). CHX (1 µM) was added 3 h after stimulation. Cell lysates prepared at the indicated times were immunoprecipitated with a polyclonal antibody to p21 (1 µg/ml; Santa Cruz Biotechnology, Santa Cruz, CA). Precipitated proteins were fractionated by SDS-PAGE (12.5%) and immunoblotted with a monoclonal antibody to p21 (1 µg/ml 6B6; PharMingen). b, degrees of intensity of blots obtained in (a) were quantified by a bioimage analyzer. Values were standardized to those obtained at 3 h and are shown as means ± S.D. (n = 3). , vehicle-treated cells; , SNAP-treated cells. c, G0 cells were stimulated with growth medium in the absence or presence of SNAP (100 µM). Actinomycin D (300 nM) was added 3 h after stimulation. Total cellular RNAs were extracted at the indicated times. Equal amounts of RNA (10 µg/lane) were fractionated by electrophoresis and hybridized with 32P labeled cDNA probes for p21 and -actin. d, data obtained in (c) were quantified. The levels of p21 normalized to those of -actin were standardized to the values obtained at 3 h and are shown as means ± S.D. (n = 3). , no addition; , SNAP; , actinomycin D; and , actinomycin D + SNAP.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   p21 Promoter assay. G0 cells cotransfected with PGV-WWP or PGV-DM and pRL-SV40 were stimulated with growth medium in the absence or presence of SNAP (100 µM) for 18 h. Luciferase activities were determined as described in Materials and Methods. Firefly luciferase activities were normalized to those of Renilla to eliminate differences in transfection efficiencies. The activities are expressed as fold increases compared with the activity in the cells transfected with PGV-WWP and incubated without SNAP. Data represent means ± S.D. (n = 3). *P < .05.

View larger version (42K): [in this window] [in a new window]   Fig. 3.   Effect of protein synthesis inhibitor on expression of p21 mRNA induced by SNAP. a, G0 cells were stimulated with growth medium for 18 h. SNAP (100 µM) and CHX (1 µM) were added at 3 h after mitogenic stimulation. Total cellular RNAs extracted at 0 and 18 h were analyzed by Northern blotting for p21 and -actin. b, data obtained in (a) were quantified. The levels of p21 normalized to those of -actin were standardized to the values obtained at time zero and are shown as means ± S.D. (n = 3). *P < .05.

View larger version (37K): [in this window] [in a new window]   Fig. 4.   Effect of SNAP on the expression of p53. a, G0 cells were stimulated with growth medium in the absence or presence of SNAP (100 µM). Cells were lysed in RIPA buffer at the indicated times. Proteins (20 µg/lane) were fractionated by SDS-PAGE (10%) and blotted with an anti-p53 or an anti--tubulin antibody. *, an A431 cell extract (10 µg) was applied as a positive control. b, expression levels of p53 normalized to those of -tubulin were plotted as percentages of the values obtained at time zero. Data represent means ± S.D. (n = 4). *P < .05, **P < .01. , vehicle-treated cells; , SNAP-treated cells. c, G0 cells were stimulated with growth medium in the absence or presence of SNAP (100 µM). Total cellular RNAs extracted at the indicated times were analyzed for p53 by RT-PCR as described in Materials and Methods.

View larger version (92K): [in this window] [in a new window]   Fig. 5.   Electrophoretic mobility shift assay for p53. Nuclear extracts (10 µg) prepared from VSMCs stimulated with growth medium for 18 h in the absence or presence of SNAP (100 µM) were incubated with 32P labeled double-stranded oligonucleotides containing a p53 binding site of human p21 promoter as described in Materials and Methods. For competition experiments, a 100-fold molar excess of unlabeled oligonucleotides (Competitor) was added before the addition of radiolabeled probes. To identify the protein associating with DNA, some of the nuclear extracts were preincubated with an anti-p53 monoclonal antibody (5 µg) or a nonimmunized mouse IgG (5 µg).

View larger version (20K): [in this window] [in a new window]   Fig. 6.   Effects of SNAP on p21 expression and DNA synthesis in p53-deficient and mutated cells. a, after serum starvation for 48 h, VSMCs from wild-type mice (C57BL/6J strain) and P53LMACO1 cells were stimulated with DMEM supplemented with 10% FBS in the absence or presence of SNAP (100 µM) and harvested at 24 h. Immunoprecipitation and Western blotting for p21 were performed using an anti-p21 polyclonal antibody. b, quiescent A431 cells were stimulated with DMEM supplemented with 10% FBS for 24 h in the presence of various concentrations of SNAP. Immunoprecipitation and Western blotting were performed as described in the legend for Fig. 1a. c, after serum starvation for 48 h, wild-type VSMCs and P53LMACO1 cells were incubated with [3H]TdR in DMEM supplemented with 10% FBS in the absence or presence of SNAP (100 µM). Incorporated radioactivities were measured at 30 h. Data represent means ± S.D. (n = 3). *P < .05, **P < .01. Open columns, wild-type VSMCs, and filled columns, P53LMACO1 cells. d, Quiescent A431 cells were incubated with [3H]TdR in DMEM supplemented with 10% FBS in the presence of various concentrations of SNAP. Radioactivities were measured at 24 h and are shown as percentages of values obtained in cells incubated without SNAP. Data represent means ± S.D. (n = 3). **P < .01.

NO-Induced p21 Expression and Growth Arrest Are Independent of cGMP. Because NO activates sGC to generate cGMP by binding to the heme moiety of the enzyme (Murad, 1986; Walter, 1989), we investigated whether the sGC-cGMP pathway is involved in NO-induced p21 expression.

View larger version (21K): [in this window] [in a new window]   Fig. 7.   Effect of an sGC inhibitor on SNAP-induced p21 expression. a, G0 cells were stimulated with growth medium containing SNAP (100 µM) for indicated times. IBMX (100 µM) was added for 30 min before harvest. Cells represented by filled columns were pretreated with ODQ (10 µM) for 30 min before mitogenic stimulation, and the same concentration of ODQ was added to the growth medium. Intracellular cGMP concentrations were determined by radioimmunoassay as described in Materials and Methods. Data represent means ± S.D. (n = 3). **P < .01. Open columns, SNAP and filled columns, ODQ + SNAP. b, G0 cells were stimulated with growth medium for 18 h in the absence or presence of SNAP (100 µM) and analyzed for p21 expression by Western blotting. ODQ was added as described in (a). c, degrees of intensity of the blots obtained in (b) were quantified and are shown as means ± S.D. (n = 3). *P < .05.

View larger version (30K): [in this window] [in a new window]   Fig. 8.   Relation between SNAP-induced p21 expression and cGMP pathway. a, G0 cells were stimulated with growth medium in the absence or presence of SNAP (100 µM), and intracellular cGMP concentrations were determined by radioimmunoassay at the indicated times. Cells represented by hatched columns were treated with IBMX (100 µM) for 30 min before extraction. Data from one representative experiment are shown. Similar results were obtained from three independent experiments. *P < .05, **P < .01. Open columns, control (vehicle-treated); filled columns, SNAP; and hatched columns, SNAP + IBMX. b, G0 cells were stimulated with growth medium in the absence or presence of SNAP (100 µM), 8-bromo-cGMP (1 mM), and IBMX (100 µM), which were added alone or in combination. Cell lysates prepared at 18 h were analyzed for p21 expression by Western blotting. c, degrees of intensity of the blots obtained in (b) were quantified and are presented as means ± S.D. (n = 3). **P < .01.

View larger version (25K): [in this window] [in a new window]   Fig. 9.   Relation between antiproliferative effect of SNAP and cGMP pathway. a, G0 cells were incubated with [3H]TdR in growth medium in the presence of various concentrations of SNAP () or 8-bromo-cGMP (). Incorporated radioactivities were measured at 24 h and are shown as percentages of values in cells incubated in the absence of the compounds. Data represent means ± S.D. (n = 3). *P < .05, **P < .01 versus the values obtained in the absence of the compounds. b, G0 cells were labeled with [3H]TdR in growth medium in the absence or presence of SNAP (100 µM), Rp-GMPS (25 µM), and Rp-AMPS (50 µM). Incorporated radioactivities were measured at 24 h. Data represent means ± S.D. (n = 3).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

p21 is induced by p53-dependent or independent mechanisms (Gartel et al., 1996). DNA-damaging stimuli induce p53 to promote transcription of a number of proteins involved in regulation of the cell cycle, DNA repair, and apoptosis, such as p21, Gadd45, and Bax (Agarwal et al., 1998). Indeed, p53 is induced by cytotoxic levels of NO, which are released from high concentrations of NO donors or generated by iNOS (Meßmer et al., 1994; Forrester et al., 1996; Ho et al., 1996; Ambs et al., 1997). But p53-dependent induction of p21 is not limited to cytotoxic stimuli; for instance, the induction of p21 by depleting ribonucleotides is dependent on p53 but not accompanied by cytotoxicity (Linke et al., 1996).

Our study suggested that p53 is involved in the expression of p21 induced by NO because SNAP up-regulated p53 probably by post-transcriptional mechanisms, activated p53-specific DNA binding, failed to stimulate a deletion mutant of p21 promoter that lacked p53 consensus sequences, and did not induce p21 in VSMCs obtained from p53-knock out mice or in A431 cells carrying mutated p53. An involvement of p53 also was suggested in the inhibition of cell proliferation induced by NO because the antiproliferative effect of SNAP was attenuated in cells lacking p53 or containing mutated p53. The reason that NO partially retained the ability to inhibit proliferation in cells without functioning p53 may be that p21 induction is not the only way for NO to inhibit the cell cycle. In fact, NO has been suggested to inhibit cell proliferation by inhibiting the activity of ribonucleotide reductase, an essential enzyme for DNA synthesis (Roy et al., 1995), consistent with our finding that SNAP interrupted the smooth muscle cell cycle at at least two points located in the G₁ and S phases (Ishida et al., 1997).

SNAP did not appear to change the level of p53 in our previous study (Ishida et al., 1997), in which we lysed cells with a relatively mild detergent (0.5% Nonidet P-40) to avoid interference with subsequently performed immunoprecipitation. In the present study, we used a lysis buffer containing strong detergents to extract nuclear proteins more efficiently and immunoblotted whole-cell lysates instead of immunoprecipitates. The discrepancy between our results may have resulted from the difference in the efficiencies to extract p53 bound to DNA.

In other cell species, p53 has been suggested to be involved in NO-mediated p21 induction. Among several human cancer cell lines, NO gas induced p21 expression in cells containing wild-type p53 but not in cells lacking p53 or containing mutant p53 (Ho et al., 1996). In PC12 cells, the activation of p21 promoter by nerve growth factor was mediated by NO and partially but not totally depended on up-regulated p53 because responsiveness remained in a deletion mutant of the p21 promoter, in which a p53 consensus sequence was removed (Poluha et al., 1997). However, their mutant still contained the proximal p53 binding site that we deleted in the mutant promoter used in the present study, which may explain the incomplete loss of responsiveness.

It is still unclear whether the up-regulation of p21 can be explained simply by the increase in the expression level of p53. SNAP prevented the decrease in the amount of p53 after mitogenic stimulation, resulting in the retention of p53 at levels 2- to 3-fold higher than control cells. However, the SNAP-induced elevation of DNA binding activity of p53 was more prominent. Therefore, the activity of p53 may be regulated not only by the amount of protein but also by other post-transcriptional mechanisms. Recent evidence suggests that the function of p53 is regulated by several modifications such as phosphorylation, glycosylation, acetylation, changes in redox states, and the association with regulatory proteins such as Mdm2 (Agarwal et al., 1998).

NO modifies the function of a variety of proteins by nitration, nitrosation, and nitrosylation. Such proteins so far reported include some enzymes such as ribonucleotide reductase and epidermal growth factor receptor tyrosine kinase (Roy et al., 1995; Estrada et al., 1997) and transcription factors such as activator protein-1, cAMP response element-binding protein, and nuclear factor-B (Lander, 1997). In particular, NO modifies the thiol groups of cysteins contained in these transcription factors, which are important for DNA binding. The DNA-binding domain of p53 also contains several cystein residues that play important roles in binding to DNA (Cho et al., 1994). Recently, NO was found to modify the conformation and function of p53 (Rainwater et al., 1995; Calmels et al., 1997). High concentrations of SNAP (2-5 mM) increased the level of p53 in nuclei but significantly decreased DNA binding activity, whereas lower concentrations of SNAP (250-500 µM) stimulated DNA binding (Calmels et al., 1997). Further studies are needed to determine whether structural changes of p53 contribute to the up-regulation of p21 induced by NO.

Recent in vivo study suggested that p53 plays a significant role in the regulation of cell proliferation during the formation of atherosclerotic lesions (Guevara et al., 1999). The development of aortic atheromas in response to a high-fat diet was accelerated in mice deficient in p53 and apo-E compared with mice lacking apo-E only, being accompanied by a significant increase in the rate of cell proliferation. On the other hand, iNOS has been shown to be expressed in macrophages, smooth muscle cells, and T lymphocytes in human and rabbit atherosclerotic lesions (Buttery et al., 1996; Esaki et al., 1997; Luoma et al., 1998). Because we revealed that an NO donor activates p53 to induce the expression of p21 in VSMCs in vitro, NO generated in vivo by iNOS also may activate the p53-p21 pathway to inhibit excessive VSMC proliferation and thereby prevent development of atherosclerotic lesions. In fact, in vivo iNOS gene transfer has been reported to inhibit intimal hyperplasia in injured arteries of rats and pigs (Shears et al., 1998).

Our results also suggested that NO-induced p21 expression is independent of the sGC-cGMP system. Pretreatment with the sGC inhibitor ODQ did not inhibit p21 expression and 8-bromo-cGMP, IBMX, and their combination did not induce p21. Furthermore, it may be particularly important that SNAP had no significant effect on intracellular cGMP concentration during the G₁ phase, although cGMP production, determined by measuring cGMP levels in the presence of IBMX, was certainly accelerated by SNAP. This implied that SNAP not only stimulated cGMP production but also accelerated hydrolysis of cGMP by phosphodiesterase. Considering that the SNAP-induced up-regulation of p21 is sustained for at least 30 h after mitogenic stimulation (Ishida et al., 1997), it is difficult to explain the up-regulation of p21 simply by the elevation of cGMP.

Smooth muscle relaxation is mediated by the sGC-cGMP pathway (Murad, 1986; Walter, 1989). NO stimulates sGC to convert intracellular GTP to cGMP by forming a nitrosyl-heme at the active center of this enzyme. However, it is unlikely that the sGC-cGMP system plays a predominant role in inhibition of VSMC proliferation induced by NO. The role of this system seemed to be small in the inhibition of G₁/S transition induced by NO because 8-bromo-cGMP resulted in only a slight inhibition of [³H]TdR incorporation and Rp-GMPS did not attenuate the antiproliferative effect of SNAP.

Our results agree with previous reports that have demonstrated that 8-bromo-cGMP and zaprinast, a cGMP-specific phosphodiesterase inhibitor, are not able to inhibit cell proliferation (Garg and Hassid, 1990; Estrada et al., 1997). Interestingly, according to a recent article (Chiche et al., 1998), S-nitrosoglutathione, an NO donor, and 8-bromo-cGMP did not inhibit the proliferation of VSMCs, in which the expression of PKG had decreased after the cells were passaged in culture, but they significantly inhibited the proliferation in cells infected with adenovirus encoding PKG I to restore the kinase activity. Chiche et al. (1998) proposed that an abundant expression of PKG in VSMCs in intact blood vessels increases cellular sensitivity to the antiproliferative effect of NO and cGMP.

In addition, PKA has been suggested to be partially responsible for the antiproliferative effect of NO in VSMCs (Cornwell et al., 1994). In our cells, however, an involvement of PKA is also unlikely because Rp-AMPS did not attenuate the antiproliferative effect of SNAP. Moreover, the antiproliferative effect of S-nitrosoglutathione on VSMCs infected with adenovirus encoding PKG was not blocked by KT5720, a PKA-selective inhibitor (Chiche et al., 1998).

Collectively, our results and those of others suggest that NO inhibits VSMC proliferation through two distinct mechanisms. At low concentrations such as are produced by eNOS, NO inhibits proliferation by the sGC-cGMP-PKG pathway, whereas relatively high concentrations of NO released from chemical NO donors (10⁵-10⁴ M), two to three orders of magnitude higher than that required for smooth muscle relaxation, can inhibit proliferation by activating p53 to induce p21 expression independently of cGMP.