cGMP-Elevating Agents Suppress Proliferation of Vascular Smooth Muscle Cells by Inhibiting the Activation of Epidermal Growth Factor Signaling Pathway
Background Abnormal proliferation of vascular smooth muscle cells (VSMC) is a key event in the pathogenesis of atherosclerosis and many vascular diseases. It is known that nitric oxide released from the endothelium participates in the regulation of VSMC proliferation via a cyclic 3′,5′-guanosine monophosphate (cGMP)-mediated mechanism. In a series of experiments, sodium nitroprusside (SNP) and A02131-1 were evaluated for their antiproliferative effect and the mechanism of their cGMP-elevating action.
Methods and Results The effect of SNP and A02131-1 on epidermal growth factor (EGF)-stimulated proliferation of rat aortic smooth muscle cells (VSMC) was examined. Cell proliferation was measured in terms of [3H]thymidine incorporation, flow cytometry, and the cell number. Further, their effect on the EGF-activated signal transduction pathway was assessed by measuring mitogen-activated protein kinases (MAPK), MAPK kinase (MEK), Raf-1 activity, and the formation of active form of Ras. SNP and A02131-1 inhibited EGF-induced DNA synthesis and subsequent proliferation of VSMC. These two increased cGMP but only a little cAMP in VSMC. A similar antiproliferative effect was observed with 8-bromo-cGMP. The antiproliferative effect of the two was reversed by KT5823 but not by dideoxyadenosine nor Rp-cAMPS. SNP and A02131-1 blocked the EGF-inducible cell cycle progression at the G1/S phase. Further experiments indicated that the two cGMP-elevating agents primarily blocked the activation of Raf-1 by EGF-activated Ras.
Conclusions These results demonstrate that cGMP-elevating agents inhibit [3H]thymidine incorporation and thus the growth of VSMC, and this inhibition appears to attenuate EGF-activated signal transduction pathway by preventing Ras-dependent activation of Raf-1.
Extensive VSMC proliferation is considered to be a hallmark of atherosclerosis,1 and the experimental removal of endothelium has been associated with VSMC proliferation.2 Generally, this process is attributed to the release of growth factors (eg, PDGF and EGF) by adherent blood cells such as platelets, leading to VSMC migration and proliferation. A potential mechanism that could modulate VSMC proliferation is the release of NO by the endothelium; the NO induces inhibition of mitogenesis in VSMC via a cGMP-dependent mechanism.3 4 It has been shown that the cGMP and PKG pathway can be activated by a potent activator of guanylate cyclase such as SNP,4 and the activation, in turn, elicits inhibition of VSMC proliferation.5 In most tissues, the intracellular concentration of cGMP is determined by the rate of formation that is regulated by agonist-induced stimulation of guanylate cyclase and hydrolysis of cGMP by a related group of PDE. There are at least seven distinct mammalian PDE families, which differ from each other in biochemical and physical properties, responses to specific effectors, inhibitors, and regulatory control mechanisms.6 Type V PDE, commonly referred to as cGMP-specific PDE, has been isolated from a number of tissues, including human platelets,7 8 trachea,9 and VSMC.10 PDE V is characterized by selectively hydrolyzing only cGMP, independent of Ca2+/calmodulin. 11 Inhibitors of PDE V such as A02131-1 and dipyridamole have vasodilating and antiplatelet aggregating properties,8 12 which may protect the vascular wall against arteriosclerotic changes. Whether or not PDE V inhibitors would suppress proliferation of VSMC and show such effect related to the protection of vascular wall has not been explored.
Recent studies have established that some mitogens stimulate the synthesis of DNA and cell proliferation by activating the phosphorylation cascade of MAPK.13 This signal pathway is initiated by the binding of a mitogen such as EGF to its receptor and subsequent autophosphorylation of tyrosine residues of the receptor, followed by the interactions of “small adaptor proteins” that promote the conversion of membrane-anchored Ras-GDP to the active form, Ras-GTP.13 14 Ras-GTP then promotes the translocation of a protooncogene-coded protein Raf-1 from the cytoplasm to the membrane, resulting in the activation of the protein kinase moiety of Raf-1.15 Raf-1 is the first protein kinase that phosphorylates and activates the downstream protein kinases, MEK, leading ultimately to the activation of the key enzyme of the MAPK cascade.16 This activation triggers DNA synthesis and initiates the cell cycle transition via phosphorylation of nuclear transcription factors of immediate early genes, such as c-fos, c-jun, and c-myc.17
Despite intensive and extensive works in 7 years, the precise mechanism of cGMP-PKG pathway in regulating cell proliferation remains a matter of considerable debate. We report here that both cGMP-elevating agents SNP and A02131-1, do inhibit the DNA synthesis and cell proliferation of VSMC stimulated by EGF, and the inhibition appears to be due to increased amount of cGMP and enhanced activity of PKG that inhibit EGF signal transduction by preventing Ras-dependent activation of Raf-1.
DMEM, FBS, antibiotics, 0.1% trypsin-EDTA solution, and all other tissue culture reagents were obtained from GIBCO. EGF (human, recombinant) was purchased from Collaborative Biomedical Products. 8-Bromo-cGMP, (R)-p-adenosine-3′,5′-cyclic monophosphothioate (Rp-cAMPS) and KT5823 were obtained from Research Biochemical Inc. SNP, dipyridamole, 2′,5′-dideoxyadenosine (DDA), myelin basic protein (MBP), β-glycerophosphate, dithiothreitol, HEPES (N-2-hydroxyethyl piperazine-N-2-ethanesulphonic acid), leupeptin, and lactate dehydrogenase (LDH) kit were all supplied by Sigma Chemical Co. Protein dye reagent, sodium dodecyl sulfate (SDS) and bisacrylamide were from Bio-Rad Laboratories, Inc. A02131-1 [3-(5′-hydroxymethyl-2′-furyl)-1-benzyl thieno (3,2-C)pyrazole] was a gift from Dr S.C. Kuo (Graduate Institute of Pharmaceutical Chemistry, China Medical College, Taiwan). [3H]Thymidine (124 Ci/mmol), [γ−32P]ATP (7000 Ci/mmol), cGMP and cAMP enzyme immunoassay kits were purchased from Amersham Corp. Enzymes including human recombinant MEK, rat MAPK (glutathione-S-transferase conjugated), Raf-1 (glutathione-S-transferase conjugated), and PKG1α (bovine lung) as well as monoclonal antibodies against the two ERK 1 (against p44 mapk) and ERK 2 (against p42 mapk), Ras (Y13-259), Raf-1, and polyclonal antibody against the N-terminal 16-amino acid peptide (PKKKPTPIQLNPNPEG) of MEK, respectively, were obtained from Upstate Biotechnology Inc.
Aortic Smooth Muscle Cell Isolation and Culture
Aortic smooth muscle cells (VSMC) were dissected, harvested from rat aortic strips, prepared, and maintained as previously described.18 Cells from passages 5 through 15 were used for all growth studies. The cells were characterized as smooth muscle cells by morphology and immunostaining with monoclonal antibody specific for smooth muscle α-actin (CGA-7).
Preparation of quiescent cells and measurement of [3H]thymidine incorporation have been described previously.18 Briefly, quiescent cells, incubated in 1 mL DMEM/FBS-free medium with SNP or A02131-1 for 60 minutes, were added with EGF and incubated further for 20 hours. [3H]thymidine (1 μCi/mL) was then added for pulse labeling, and 4 hours later, 10% trichloroacetic acid-insoluble material was prepared. From this acid-insoluble fraction, the DNA was extracted with 0.1N NaOH, and the 3H-counts were measured in a liquid scintillation counter.
The hemocytometer measurement method described earlier was used.18 Triplicate counts were taken for each plate, and quadruplicate plates were used for each determination.
To estimate the proportion of cells at various stages in different phases of the cell cycle, the previously described flow cytometry was used to measure cellular DNA content 24 hours after stimulation of the cells. Flow cytometric determination of DNA content was used as an index of cell proliferation.19
Measurement of cGMP and cAMP
cGMP and cAMP levels in the supernatant of rat VSMC culture were determined as previously described.20 VSMC were grown in 35-mm dishes, and at confluence, monolayer cells were washed twice with PBS and incubated with SNP or A02131-1, for various lengths of time. The reaction was stopped by adding 0.1N HCl, followed immediately by boiling for 5 minutes. Upon cooling to 4°C, the precipitated protein was removed by centrifugation (Eppendorf, model 5415C). The supernatant was assayed for cGMP and cAMP by enzyme immunoassay.
This activity, as a sign of cell viability, was assayed in the medium of cultured VSMC at the end of incubation period by measuring the rate of reduction of pyruvate to lactate using a Sigma Diagnostics kit; LDH activity was then calculated from the change in 340 nm absorbance and expressed in per-unit volume of medium.
Assay of MAPK Activity
Quiescent VSMC were incubated with SNP, A02131-1, or 8-bromo-cGMP for 60 minutes to ensure their penetration into the cells, then added with EGF and incubated further for 5 minutes. The cells were prepared and immunoprecipitated as described earlier.18 The kinase assay was performed using 32P phosphorylation of MBP as a measurement of MAPK activity as described by Morinelli et al.21 After preparing the sample according to the procedure,18 21 the amount of 32P incorporated into MBP was determined in a scintillation counter.
In some experiments, MAPK activity was also assayed using the “in gel” MBP phosphorylation method. Cell extracts were subjected to SDS-PAGE (10%) containing 0.5 mg/mL MBP. After denaturation and renaturation of the separated proteins, the gel was immersed for 1 hour at 25°C in buffer containing (mmol/L) HEPES 20, pH 7.0, dithiothreitol 2, EGTA 0.1, MgCl2 10, ATP 0.05, and 5 μCi [γ−32P] ATP. After incubation, the gel was washed, dried, and exposed to Kodak X-Omat film for autoradiography.
Assay of MEK Activity
MEK activity was assayed using a recombinant rat inactive form MAPK as the substrate as described earlier. Cell lysates were immunoprecipitated with 5 μg of the mouse polyclonal antibody against the N-terminal 16-amino acid peptide (PKKKPTPIQLNPNPEG) of MEK and protein G-Sepharose. The immunoprecipitation and assay for MEK activity were similarly carried out as for the measurement of MAPK activity.
Raf-1 Kinase Assay
As above, quiescent cells were incubated with SNP or A02131-1 for 60 minutes and then EGF for 5 minutes. Protein in the cell lysate was immunoprecipitated with anti–Raf-1 serum. Immunoprecipitated Raf-1 or recombinant Raf-1–GST kinase activity was measured with kinase-negative MEK as the substrate. The activation of the MEK→MAPK→MBP phosphorylation cascade by Raf was tested at 37°C for 30 minutes in the kinase assay mixtures that contained (mmol/L, unless otherwise noted) HEPES 50, pH 7.5, MgCl2 10, dithiothreitol 1, ATP 0.02, 0.5 μg recombinant MEK, 100 ng MAPK, 0.5 μg MBP, and 5 μCi [γ−32P]ATP. The reaction was terminated by spotting 20 μL of the reaction mixture onto p81 paper and the radioactivity counted. In some experiments, EGF-induced phosphorylation of Raf-1 or PKG-induced phosphorylation of inactive form Raf-1-GST was performed at 30°C for 30 minutes in kinase buffer. Reactions were terminated by adding 5 μL sample buffer and the mixture subjected to 10% SDS-PAGE and autoradiography. The phosphorylation bands were visualized and quantitated by using a computing densitometer with ImageQuant software (Molecular Dynamics).
Analysis of Ras-Bound GTP and GDP
Quiescent cells were labeled with 0.2 mCi/mL [32P]-orthophosphate in phosphate-free DMEM for 4 hours, incubated with or without A02131-1 or SNP and stimulated with EGF for 5 minutes. The cells were lysed in Triton X-114 buffer containing 50 mmol/L HEPES, pH 7.4, 1% Triton X-114, 100 mmol/L NaCl, 5 mmol/L MgCl2, 1 mmol/L PMSF, 100 μmol/L GTP, 100 μmol/L GDP, 1 mmol/L ATP, 20 μg/mL aprotinin, 1 mmol/L sodium pyrophosphate, and 1 mmol/L Na3VO4. Phosphatase inhibitors were added in the buffer to improve the recovery of Ras in GTP.22 Membrane-bound Ras was recovered by detergent phase splitting,22 and immunoprecipitated with anti-p21Ras monoclonal antibody (Y13-259) with the aid of protein G-Sepharose. The immune complex was washed with washing buffer containing 50 mmol/L HEPES, pH 7.4, 5 mmol/L MgCl2, 50 mmol/L NaCl, 1 mmol/L sodium pyrophosphate, 1 mmol/L Na3VO4, and 0.1% Triton X-100. Ras-GTP and Ras-GDP were eluted and measured in thin-layer chromatography on a polyethyleneimine-cellulose plate, and GTP fraction [%, GTP/(GDP+GTP)×100] was calculated.
All data are expressed as mean±SEM. One-way ANOVA was used for multiple group comparisons and the Student's t test for others. P<.05 was considered statistically significant.
Inhibition of [3H]Thymidine Incorporation and Cell Proliferation by SNP and A02131-1
Cell proliferation must be preceded by DNA synthesis, ie, inhibition of DNA synthesis would lead to inhibition of cell proliferation. Therefore, the capability of SNP and A02131-1 to inhibit the incorporation of [3H]thymidine into DNA was first examined. Addition of EGF (0.001 to 0.1 μg/mL) into the quiescent VSMC stimulated [3H]thymidine incorporation in a dose-dependent manner. The EC90 was 0.03 μg/mL, and at this amount, EGF stimulated [3H]thymidine incorporation to ≈5.6±0.2-fold that in EGF-free medium. In all subsequent experiments, therefore, 0.03 μg/mL EGF was used as the stimulatory dose. The incorporation of [3H]thymidine thus induced by EGF was inhibited by SNP and A02131-1, also dose-dependently (Fig 1⇓). SNP was more potent than A02131-1, but neither affected the basal incorporation. Since SNP and A02131-1 would cause the elevation of cGMP, their effect was generally thought to be mediated by cGMP as the second messenger. Therefore, we also tested the effect of 8-bromo-cGMP, an analogue of cGMP, and it did inhibit EGF-stimulated [3H]thymidine incorporation, also in a dose-dependent manner (Fig 1⇓). The inhibition of [3H]thymidine incorporation reached >90% for SNP at 30 μmol/L and for A02131-1 at 100 μmol/L. Therefore, unless otherwise mentioned, these concentrations, respectively, were used for most of the subsequent experiments.
Since SNP and A02131-1 inhibited DNA synthesis, these agents should also inhibit cell proliferation. Their effect on the EGF-induced increase of cell number was therefore tested daily for 5 days. As expected, both SNP and A02131-1 significantly suppressed the increase of cell number induced by EGF (Fig 2⇓), and again, SNP was more inhibitory than A02131-1. In addition, these two also inhibited similar events caused by PDGF and angiotensin II (data not shown).
The antiproliferative effect of SNP and A02131-1 was also assessed by flow cytometry. As shown in Fig 3A⇓, in unstimulated quiescent VSMC, 95% of the cells were in the growth-arrested (G0/G1) phase of the cell cycle, whereas after 24-hour stimulation with EGF, 50% of the VSMC entered the S phase of the cell cycle (Fig 3B⇓). When either SNP (Fig 3C⇓) or A02131-1 (Fig 3D⇓) was present, the effect of EGF was almost completely abolished and the cells stayed in the G0/G1 phase.
There was a possibility that the apparent antiproliferative effect of SNP and A02131-1 was the result of the loss of the cells due to their toxicity. A series of experiments was, therefore, performed to check on this possibility. (a) First, the number of cells in the supernatant was determined daily for 5 days, in the presence or absence of SNP or A02131-1. The result showed that <1% of the cells was found in the supernatant and could be stained with trypan blue, indicating little cell inactivation. (b) Next, cells were cultured for 5 days as in (a), and the LDH level in the supernatant was determined daily, since the release of LDH is a sign of cell inactivation. No significant release of LDH was observed during the period. Thus, these agents were not toxic to the cells.
Effect of SNP and A02131-1 on the Amount of cGMP and cAMP in VSMC
The earlier observation that 8-bromo-cGMP could mimic the antiproliferative effect of SNP and A02131-1 suggested that this effect was likely to be mediated by cGMP as the second messenger. Therefore, whether or not the two would increase the amount of cGMP, which in turn would increase the amount of cAMP by inhibiting PDE III,23 was checked. The basal level of cGMP and cAMP in VSMC was determined to be 2.4±0.2 and 4.9±0.5 pmol/mg protein, respectively. SNP and A02131-1 increased cGMP but only a little cAMP in VSMC (Fig 4⇓).
Effect of Inhibitors of cGMP- or cAMP-Dependent Pathway on cGMP-Elevating Agents: Induced Suppression of EGF-Stimulated [3H]Thymidine Incorporation
The increase in cGMP, which appeared to be an effect of SNP and A02131-1, presumably stimulated the activity of PKG. Thus, if the protein kinase activity was inhibited, their effect might be reversed. When KT5823 (10 μmol/L), an inhibitor of PKG,24 was added, the inhibition of EGF-stimulated [3H]thymidine incorporation by SNP and A02131-1 was indeed reversed (Fig 5⇓). On the other hand, neither DDA (200 μmol/L), an adenylate cyclase inhibitor,25 nor Rp-cAMPS (10 μmol/L), a PKA inhibitor,26 antagonized the inhibitory effect of SNP and A02131-1 (Fig 5⇓).
Inhibition of EGF-Stimulated MAPK Activation by cGMP-Elevating Agents
It was known that EGF would activate MAPK, leading eventually to cell proliferation, and EGF was indeed a potent activator of MAPK, as shown in Fig 6A⇓, which also showed that SNP and A02131-1 did inhibit the activation of MAPK, by 99±10% and 98±7%, respectively. Such effect of SNP and A02131-1 was likely achieved via increase of cGMP, since 8-bromo-cGMP inhibited EGF stimulation of MAPK (Fig 6A⇓). In addition, A02131-1 also reduced the effect of PDGF, phorbol 12-myristate 13-acetate (PMA), and okadaic acid on the MAPK activity by 69±7%, 80±11%, and 77±6%, respectively. Furthermore, the anti-MAPK effect of SNP and A02131-1 was reversed by KT5823, whereas little such reversion effect was observed when Rp-cAMPS was present (data not shown).
Activation of MAPK in Extracts of SNP-Treated or A02131-1–Treated Cells by MEK
The generally known steps in the pathway stimulated by EGF leading to the activation of MAPK were activation of tyrosine kinase receptor, which activated p21Ras; the active p21Ras, then activated p74raf-1 which, in turn, activated MEK; and finally, the active MEK phosphorylated and activated MAPK. When any one of these steps was affected, the activation of MAPK would be suppressed. The last step, phosphorylation, thus activation, of MAPK was catalyzed directly by one or more MEK.16 This step might somehow be inhibited by SNP and A02131-1. This was tested in vitro. Quiescent VSMC were added, respectively, with none (control), EGF, and EGF plus SNP (or A02131-1). Cytosolic extracts were obtained, respectively, from these cultures, incubated with purified MEK (50 ng) plus Mg2+/ATP and then analyzed for the activation of MAPK in terms of ERK 1 and ERK 2 activities by measuring the SDS-PAGE “in gel” phosphorylation of MBP. Two MBP-phosphorylating activities corresponding to ERK 1 and ERK 2 were obtained. As shown in Fig 6B⇑, where only the result of SNP on ERK 1 and ERK 2 is shown, by adding MEK, MAPK was activated (first pair of the bar graph) and in the cell extracts from EGF added cells (third pair of the bar graph), the addition of MEK only slightly enhanced the MAPK activation. This was natural, since EGF had already activated majority of the MAPK. In contrast, the MAPK activation could not be seen as expected in the extract derived from the cells treated simultaneously with EGF and SNP, but when MEK was added, the MAPK activation was conspicuously observed with a level equivalent to the control (second pair of the bar), suggesting that SNP and A02131-1 did not directly inhibit MAPK activation.
Inhibition of EGF-Stimulated MEK Activation by cGMP-Elevating Agents
Another possibility of the suppression of MAPK activation was that the MEK were in an inactive form, ie, because of the action of the inhibitors, the MEK failed to be activated by EGF. This was tested using a similar procedure as above for MAPK activation. In this test, however, the ability of the cell lysates (the source of MEK) to phosphorylate recombinant rat MAPK-GST was examined. As shown in Fig 7A⇓, the activity of the MEK in EGF-treated cells was about 7-fold that of control, whereas the activation was greatly inhibited in those cells that were treated with SNP and A02131-1. The inhibition was also seen when 8-bromo-cGMP was added (Fig 7A⇓).
Activation of MEK in Extracts of SNP-Treated or A02131-1–Treated Cells by Raf-1
There were two possibilities for the failure of the MEK to be activated: (1) because of the presence of the inhibitors, the MEK were not in a form for activation; and (2) Raf-1 was not in an active form for activation of the MEK. Thus, to determine these points, cell extracts were prepared as above and from the cell extracts, purified fractions of MEK were obtained. These fractions were then incubated with Mg2+/ATP and activated Raf-1 (50 ng). The result is shown in Fig 7B⇑. The MEK were activated by Raf-1 in the cell extracts derived either from control (first pair of bars in Fig 7B⇑) or cells added with SNP (second pair of bars in Fig 7B⇑). The MEK in the cell extracts derived from cells treated with EGF was already activated, as in the case of MAPK activation described above, and consequently, the addition of activated Raf-1 activated the MEK only a little further (third pair of bars in Fig 7B⇑). When EGF and SNP were added together, the EGF activation of MEK was inhibited, but when Raf-1 was added, it was activated to the same extent as that of EGF-treated extracts (pair of bars with an asterisk in Fig 7B⇑). These results indicated that the MEK themselves were not affected by SNP for activation by Raf-1, and that the second possibility, ie, Raf-1 was in an inactive form, might be the reason for MEK inactivation. This was next examined.
Effects of cGMP-Elevating Agents on EGF-Stimulated Activation of Raf-1 Kinase, Formation of Ras-GTP, and EGF Receptor Autophosphorylation
Raf-1 has been defined biochemically as being a likely target of Ras,15 27 and Ras is critical for activation of Raf-1 by EGF.28 Thus, Raf is likely to be a link between Ras and the MAPK cascade. Therefore, the effect of the inhibitors on this link was investigated by measuring the Raf-1 kinase activity in a coupled reaction as described in “Methods.” Raf-1 kinase was activated by EGF as expected, but this activation was significantly suppressed (P<.001) when the cell was preincubated with SNP and A02131-1 (Fig 8A⇓). However, the inhibitory effect of SNP on Raf-1 activity was reversed by KT5823 but not by Rp-cAMPS. The enzymatic activity of Raf-1 stimulated by EGF required phosphorylation of Raf-1. As shown in Fig 8B⇓, EGF-induced activation of Raf-1 was indeed accompanied by increased Raf-1 phosphorylation, as seen in a characteristic shift in apparent molecular weight. EGF-induced Raf-1 phosphorylation was also suppressed by SNP and A02131-1.
The question was then, how was Raf-1 activity inhibited by PKG, which would have been activated by increasing cGMP? One potential mechanism was that PKG directly phosphorylated Raf-1. This was tested in vitro. Incubation of purified inactive form of Raf-1-GST with PKG (from bovine lung) and [γ−32P]ATP resulted in phosphorylation of a 76-kD band in the Raf-1 preparation. The 76-kD phosphoprotein was identified as Raf-1 by protein immunoblotting. Thus, Raf-1 was an in vitro substrate of PKG (Fig 9A⇓). In addition to substrate phosphorylation, Raf-1 activity was assessed in vitro in a coupled reaction by checking the phosphorylation of catalytically inactive MEK→MAPK→MBP cascade. Preincubation of activated Raf-1 protein with PKG interfered with the induction of the phosphorylation cascade (Fig 9B⇓).
The inhibition of the activation of Raf-1 by the cGMP-elevating agents shown above might be due to insufficient formation of GTP-bound Ras (the active form). This was checked. The addition of EGF naturally increased the amount of Ras GTP complex, and similar increases could still be seen with the addition of the inhibitors (Fig 10⇓, A and B), indicating that these steps in the EGF pathway were probably not involved in the inhibition of Raf-1. SNP also did not affect EGF-activated receptor autophosphorylation (data not shown). These results clearly showed that a target step of the cGMP/PKG action in the EGF pathway was the activation of Raf-1 by activated Ras.
SNP inhibits [3H]thymidine incorporation and growth of rat aortic smooth muscle cells, consistent with the observations of others.2 4 The result presented here, on the other hand, is the first demonstration that PDE V inhibitor, A02131-1, inhibits the proliferation of VSMC. The antiproliferative effect of SNP and A0213l-1 is mimicked by a stable analogue of cGMP, 8-bromo-cGMP. Furthermore, KT5823, an inhibitor of PKG, is effective in reversing the antiproliferative effect of SNP and A02131-1. These results support the notion that the antiproliferative effect of SNP and A02131-1 is likely to be mediated by cGMP as the second messenger. Indeed, the dose of SNP and A02131-1 required to inhibit EGF-stimulated [3H]thymidine incorporation correlates that needed for cGMP formation. They also stimulate a greater increase of cGMP than cAMP. Furthermore, unlike KT5823, DDA, an adenylate cyclase inhibitor, and Rp-cAMPS, a PKA inhibitor, are unable to reverse the antiproliferative effect of SNP and A02131-1, suggesting that the cGMP-PKG, not cAMP-PKA, pathway is involved in the SNP- and A02131-1-induced antiproliferative effect. Such antiproliferative effect observed is unlikely to be due to the toxicity of the drug to the cell.
Despite numerous investigations for 7 years,2 the precise mechanism of cGMP in regulating cell proliferation remains a matter of considerable debate.4 Kinetic analyses indicate that cGMP exerts a cytostatic effect in the G0/G1 phase of the cell cycle.29 30 The flow cytometric result obtained here shows that the SNP-treated and A02131-1–treated cells produce an acute blockage of the cell cycle progression from the G1 into the S phase of the cell cycle. This result is consistent with the previous works29 that show that an elevated cGMP concentration at a later stage of the G1 phase of the cell cycle seems to be necessary to inhibit proliferation.
A converging point for several pathways leading to cell proliferation is MAPK. MAPK have been shown to be activated during the transition of entering the S phase from the G0/G1 phase of the cell cycle.31 The cGMP-elevating agents affect the activation of MAPK in the EGF pathway, of which events involve attachment of EGF to the receptor and subsequent phosphorylation of the EGF receptor, formation of Ras-GTP, activation of Raf-1, MEK, and finally activation of MAPK. The phosphotyrosine content of the EGF receptor is observed to be unaffected by SNP. Moreover, increase in cGMP is associated with marked inhibition of MAPK activation induced by a number of agonists, including PDGF, PMA, and okadaic acid, which presumably act independently of the EGF receptor in the signal transduction pathway. MAPK is phosphorylated and activated by MEK,16 31 which in turn appears to be activated by either of the two kinases, Raf-1 14 15 and MEK kinase.17 Accumulating evidence indicates that Ras mediates the action of growth factors that transfer signal via receptor protein tyrosine kinase,14 and recent findings link Ras to the activation of Raf-1 and MEK.27 28 In the present study, we show that the EGF-stimulated accumulation of the Ras-GTP is unaffected; instead, activation of Raf-1 kinases is inhibited by the cGMP-elevating agents. Thus, cGMP appears to inhibit signal transduction from Ras by preventing Ras-dependent activation of Raf-1. This inhibition presumably decreases the activity of the downstream kinases, including MEK and MAPK. Raf-1 is a 72- to 76-kD protein with intrinsic serine/threonine kinase activity. Although the mechanism of Raf-1 activation is unclear, phosphorylation may alter the conformation of Raf-1, thereby activated the catalytic activity. The molecular mechanism of PKG inhibition of Raf-1 activation is not precisely known, either. However, the result of an in vitro study suggests that PKG suppresses the activity of Raf-1 by directly phosphorylating Raf-1, likely altering the conformation of Raf-1.
In the present study, the cGMP-elevating agents inhibited only about 60% to 70% of Raf-1 activation (Fig 8A⇑). Despite the inhibition of MAPK, DNA synthesis and proliferation are nearly complete (90%) at the respective concentrations used. This suggests that the inhibition likely involves other unknown pathways. Recently, Sugimoto et al32 have demonstrated that atrial natriuretic peptide is able to induce the expression of the MAPK phosphatase, MKP-1, by cGMP-dependent mechanism in concentrations sufficient to inhibit the proliferation of mesangial cells. MKP-1 is a dual specificity phosphatase that selectively dephosphorylates MAPK in vitro and in vivo.33 It remains to be determined whether cGMP-elevating agents increase MKP-1 activity in VSMC.
Several investigators recently found that cAMP is a more potent inhibitor of VSMC proliferation than cGMP,34 and in some cell types the MAPK phosphorylation cascade and stimulation of cell proliferation by a growth factor can be inhibited by a “cross talk” with another signaling system, the cAMP-activated protein kinase (PKA).35 36 PKA blocks activation of Raf-1, MEK, and MAPK in Rat1h ER fibroblasts, accompanied by increase in Raf-1 phosphorylation on serine 43 in the regulatory domain.35 In our case, Rp-cAMPS is unable to reverse the anti-MAPK and anti–Raf-1 effect of SNP and A02131-1, suggesting that it is cGMP-PKG, not cAMP-PKA pathway, that is involved in the SNP and A02131-1-induced effects. This is the first demonstration that the cGMP-PKG could “cross talk” with the MAPK cascade that leads to cell proliferation. All these indicate that increasing cGMP is a counterregulatory mechanism for attenuating the Raf-1 activation pathway. Therefore, given the central role of the MAPK phosphorylation cascade in mediating the actions of growth factors and other mitogens, cGMP-dependent inhibition of the MAPK pathway may be of broad physiological importance.
We show for the first time that cGMP-elevating agents suppress EGF-induced VSMC proliferation via increasing cGMP that acts, in part, to inhibit EGF signal transduction pathway. Similar to SNP, PDE V inhibitor, A02131-1, may exert beneficial effects (including antiplatelet, vasorelaxing, and antiproliferative effects) on vascular structure. Therefore, it may be useful in reducing abnormal growth and proliferation of VSMC observed in hypertension and atherosclerosis.
Selected Abbreviations and Acronyms
|cAMP||=||cyclic 3′,5′-adenosine monophosphate|
|cGMP||=||cyclic 3′,5′-guanosine monophosphate|
|EGF||=||epidermal growth factor|
|ERK||=||extracellular signal-regulated protein kinases|
|MAPK||=||mitogen-activated protein kinases|
|PDE||=||cyclic 3′,5′-nucleotide phosphodiesterases|
|PDGF||=||platelet-derived growth factor|
|PKA||=||cAMP-dependent protein kinases|
|PKG||=||cGMP-dependent protein kinases|
|VSMC||=||vascular smooth muscle cells|
Pei-yi Huang's skillful preparation of the manuscript is gratefully acknowledged. We acknowledge Jonathan T. Ou and Jau-Song Yu for critical review of the manuscript. This work was supported by research grants CMRP 576 from Chang Gung Medical Research Foundation, NSC 84-2331-B182-095 and NSC 85-2331-B182-019 (to J.T.O.) from the National Science Council, and DOH 85-CM-043 from the Department of Health, the Republic of China.
- Received June 13, 1996.
- Revision received October 2, 1996.
- Accepted October 13, 1996.
- Copyright © 1997 by American Heart Association
Garg UC, Hassid A. Nitric oxide-generating vasodilators and 8-bromo-cyclic guanosine monophosphate inhibit mitogenesis and proliferation of cultured rat vascular smooth muscle cells. J Clin Invest.. 1989;83:1774-1777.
Dubey RK. Vasodilator derived nitric oxide inhibits angiotensin II- and fetal calf serum-induced growth of renal arteriolar smooth muscle cells. J Pharmacol Exp Ther.. 1994;269:402-408.
Macphee CH, Harrison SA, Beavo JA. Immunological identification of the major platelet low-km cAMP phosphodiesterase: probable target for antithrombotic agents. Proc Natl Acad Sci U S A.. 1986;83:6660-6668.
Yu SM, Tsai SY, Kuo SC, Ou JT. Inhibition of platelet function by A02131-1: a novel inhibitor of cGMP-specific phosphodiesterase, in vitro and in vivo. Blood.. 1996;87:3758-3767.
Beavo JA, Conti M, Heaslip RJ. Multiple cyclic nucleotide phosphodiesterase. Mol Pharmacol.. 1994;46:399-405.
Weishaar RE, Burrows SD, Kobylarz DC, Quade MM, Evans DB. Multiple molecular forms of cyclic nucleotide phosphodiesterase in cardiac and smooth muscle and in platelets: isolation, characterization, and effects of various reference phosphodiesterase inhibitors and cardiotonic agents. Biochem Pharmacol.. 1986;35:787-800.
Nishida E, Gotoh Y. The MAP kinase cascade is essential for diverse signal transduction pathways. Trends Pharmacol Sci. 1993;18:128-131.
Hall A. Biochemical Function for Ras: at Last. Science.. 1993;264:1413-1414.
Stokoe D, MacDonald SG, Cadwallader K, Symons M, Hancock JF. Activation of Raf as a result of recruitment to the plasma membrane. Science.. 1994;264:1463-1467.
Davis RJ. The mitogen-activated protein kinase signal transduction pathway. J Biol Chem.. 1993;268:14553-14556.
Seth A, Gonzalez FA, Gupta S, Raden DL, Davis RJ. Signal transduction within the nucleus by mitogen-activated protein kinase. J Biol Chem.. 1992;267:24796-24804.
Yu SM, Tsai SY, Guh JH, Ko FN, Teng CM, Ou JT. Mechanism of catecholamine-induced proliferation of vascular smooth muscle cells. Circulation.. 1996;94:547-554.
Ohtsubo M, Roberts JM. Cyclin dependent regulation of G1 in mammalian fibroblasts. Science.. 1993;259:1908-1911.
Morinelli TA, Zhang LM, Newman WH, Mier KE. Thromboxane A2/prostaglandin H2-stimulated mitogenesis of coronary artery smooth muscle cells involves activation of mitogen-activated protein kinase and S6 kinase. J Biol Chem.. 1994;269:5693-5698.
Maurice DH, Haslam RJ. Molecular basis of the synergistic inhibition of platelet function by nitrovasodilators and activators of adenylate cyclase: inhibition of cyclic AMP breakdown by cyclic GMP. Mol Pharmacol.. 1990;37:671-677.
Haslam RJ, Davidson MML, Desjardins JV. Inhibition of adenylate cyclase by adenosine analogues in preparations of broken and intact human platelets: evidence for the unidirectional control of platelet function by cyclic AMP. Biochem J.. 1978;176:83-95.
Rothermel JD, Botello LHP. A mechanistic and kinetic analysis of the interactions of the diastereoisomers of adenosine 3′,5'(cyclic) phosphorothionate with purified cyclic AMP-dependent protein kinsase. Biochem J.. 1988;251:757-762.
Assender JW, Southgate KM, Hallet MB, Newby AC. Inhibition of proliferation, but not of Ca2+ mobilization, by cyclic AMP and GMP in rabbit aortic smooth muscle cells. Biochem J.. 1992;288:527-532.
Yu SM, Cheng ZJ, Guh JH, Lee FY, Kuo SC. Mechanism of anti-proliferation caused by YC-1, an indazole derivative, in cultured rat A10 vascular smooth muscle cells. Biochem J.. 1995;306:787-792.
Blenis J. Signal transduction via the MAP kinases: proceed at your own RSK. Proc Natl Acad Sci U S A.. 1993;90:5889-5892.
Sugimoto T, Haneda M, Togawa M, Isono M, Shikano T, Araki S-I, Nakagawa T, Kashiwagi A, Guan KL, Kikkawa R. Atrial natriuretic peptide induces the expression of MKP-1, a mitogen-activated protein kinase phosphatase, in glomerular mesangial cells. J Biol Chem.. 1996;271:544-547.
Duff JL, Monia BP, Berk BC. Mitogen-activated protein (MAP) kinase is regulated by the MAP kinase phosphatase (MKP-1) in vascular smooth muscle cells. Effect of actinomycin D and antisense oligonucleotides. J Biol Chem.. 1995;270:7161-7166.
Wu J, Dent P, Jelinek T, Woltman A, Weber MJ, Sturgill TW. Inhibition of the EGF-activated MAP kinase signaling pathway by adenosine 3′,5′-monophosphate. Science.. 1993;262:1065-1068.
Cook SJ, McCormick F. Inhibition by cAMP of Ras-dependent activation of Raf. Science.. 1993;262:1069-1072.