β2-Adrenergic Receptor Agonists and cAMP Arrest Human Cultured Airway Smooth Muscle Cells in the G₁ Phase of the Cell Cycle: Role of Proteasome Degradation of Cyclin D1

Experimental Procedures

Cell Culture.

ASM was dissected from macroscopically normal lung resection specimens obtained from lung transplant recipients and was provided by the Alfred Hospital (Melbourne, Australia). Cultures were prepared as described previously in detail (Tomlinson et al., 1995) and maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 2 mM l-glutamine, 100 μg/ml streptomycin, 100 U/ml penicillin-G, 2 μg/ml amphotericin B, and 0.25% BSA, w/v, which is referred to as serum-free DMEM. Cells were maintained in Falcon culture flasks and incubated (37°C; 5% CO₂) until monolayer confluence was reached. They were harvested weekly by 10-min exposure to 0.5% (w/v) trypsin, 1 mM EDTA in PBS and passaged at a 1:3 split ratio. Cells at passage numbers 3 to 15 were used for experiments.

Immunocytochemistry.

Cells were subcultured into eight-well glass tissue culture chamber slides, grown to 100% confluence in DMEM (10% fetal calf serum), serum-deprived for 4 days, and then fixed and subjected to immunohistochemistry for α-actin, myosin, platelet endothelial cell adhesion molecule 1 (PECAM-1) and cytokeratin, as described previously (Vlahos and Stewart, 1999). The expression of smooth muscle α-actin and myosin was observed in all cultures used in this study. These cultures did not express detectable PECAM-1 staining and less than 5% of the cells were positive for cytokeratin. Paraffin-embedded sections of the airway adjacent to that used for generation of cultures stained positively for smooth muscle α-actin and myosin in bundles of ASM and blood vessels only. The antibody against PECAM-1 stained vascular endothelium, whereas that against cytokeratin stained only the epithelium, confirming the specificity of these antibodies for the target antigens.

DNA Synthesis.

Cells were subcultured into 24-well plates at a 1:3 ratio at a density of approximately 1.5 × 10⁴/cm² and allowed to grow to monolayer confluence over 72 to 96 h in an humidified atmosphere of 5% CO₂ in air at 37°C. The serum-containing media was replaced with serum-free DMEM for 24 h to produce growth arrest. In some experiments, the cells were pretreated with β₂-adrenergic receptor agonists or the cAMP analog 8-bromo-cAMP 30 min before the addition of mitogen, which was added to the appropriate wells in modified serum-free DMEM containing insulin (100 ng/ml), transferrin, (50 ng/ml), and selenium (1.5 pg/ml; specially formulated by CSL Ltd, Parkville, Australia) to provide progression factors that are essential for the mitogen activity. Mitogens and inhibitors were left in contact with cells from the time of addition until the end of the experiment unless indicated otherwise. Cells were incubated for 24 h (37°C; 5% CO₂) before being pulsed with [³H]thymidine (1 μCi/ml for 4 h) to measure incorporation of radiolabeled material into newly synthesized DNA, according to our previous study (Stewart et al., 1997).

Western Analyses of Cell Cycle Regulatory Proteins.

Experiments to determine the effects of β₂-adrenergic receptor agonists or cAMP analogs on cell cycle regulatory proteins were carried out on cells grown to confluence in six-well plates (∼ 9 cm²), using 25- or 75-cm² flasks according to the requirement for lysate protein for Western blotting. These cells were rendered quiescent as described for experiments on DNA synthesis and pretreatment and mitogen exposure were carried out under identical conditions. At the end of the incubation period (usually 20 h), the cells were washed three times in ice-cold PBS (1 ml/10 cm²) and extracted into a lysis buffer (100 mM NaCl; 10 mM Tris; 2 mM EDTA; 0.5% sodium deoxycholate, w/v; 1% triton X-100, v/v; 1 mM PMSF; 10 mM MgCl₂; 100 IU/ml aprotonin; pH 7.5) an aliquot of which was removed for protein assay (Biorad reagent; Biorad, Sydney, Australia). The final concentration of Triton X100 in samples for assay of 0.05% was below the level at which the assay accuracy may be affected (0.1%, according to BioRad instruction booklet). Moreover, the coefficient of variation of duplicate protein assays was 6.8 ± 0.6% (n = 90 duplicate estimates in four different assays). The samples were resolved on polyacrylamide gel electrophoresis and Western-blotted for phospho-ERK, cyclin D1, and pRb according to methods described previously (Fernandes et al., 1999). To visualize the antigen, enhanced chemiluminescence reagent (Amersham, Cardiff, UK) was added for 1 min and the membrane was then apposed to X-ray film (Kodak X-omat AR; Eastman-Kodak, Rochester, NY) for variable periods of time (0.5–4 min) before development. X-ray films were subject to densitometry with a Molecular Dynamics Personal Densitometer (Sunnyvale, CA), and the volumes were normalized to thrombin-induced levels of cyclin D1.

Assay of ERK Activity.

ERK activity was determined by immunoprecipitation by using a specific anti-ERK antibody (goat polyclonal IgG; C-16, Santa Cruz, CA). Cells were seeded onto six-well plates as described previously, grown to confluence, then serum-starved for 24 h. The cells were incubated for 30 min with PD 98059 (30 μM) as indicated, and all cells were incubated with thrombin in serum-free DMEM for 5 min, 2 h, or 20 h. At the end of the stimulation period, the cell lysates were assayed for ERK activity as described previously in detail (Fernandes et al., 1999).

Northern Blot Analyses.

Cells were seeded into 75-cm² flasks at a density of 1.5 × 10⁴ cells/cm², grown to confluence, serum-starved for 24 h as described previously, and stimulated for between 4 and 20 h with thrombin. Total RNA was extracted with 1 ml of Trizol reagent (Gibco BRL, Melbourne, Australia) according to the manufacturer’s instructions. The mRNA was isolated from 75 μg of total RNA by using Dynabeads oligonucleotides (dT)₂₅ (Dynal, Oslo, Norway) according to the manufacturer’s instructions and was separated on a 1.2% formaldehyde denaturing gel and transferred to Immobilon-Ny⁺ nylon membranes (Millipore, Bedford, MA) by using 20× standard saline citrate (SSC). Cyclin D1 mRNA was detected by Northern hybridization to a 440-base-pair human cyclin D1 cDNA probe (Xiong et al., 1991) labeled with [α-³²P]dCTP (Megaprime labeling kit; Amersham). The membranes were hybridized overnight at 65°C, washed twice with 2× SSC and 0.1% SDS at 55°C (30 min), once with 1× SSC and 0.1% SDS at 55°C (30 min), and exposed to autoradiography film (Hyperfilm MP; Amersham). The autoradiographs were quantitated by using a Molecular Dynamics Personal Densitometer. The membranes were also probed for Tubulin by using a 200-base-pair cDNA probe generated by reverse transcription-polymerase chain reaction (with 5′- CCTGGAACCCACAGTTATTGATGAAGAAGTTCG-3′ and 5′-AGAAGCCCTGGAGACCCG- TGCACTGGTCAG-3′ primers) and hybridized as described above. To control for loading differences, cyclin D1 mRNA levels were normalized against the levels of tubulin mRNA.

Preparation of Reagents.

PD 98059, initially dissolved in 100% DMSO (BDH, Dorset, UK) at 50 mM, was diluted 1 in 5 with DMSO to produce a solution of 10 mM. The final concentration of 30 μM PD 98059 in medium resulted in a final concentration of 0.3% DMSO. In experiments with PD 98059, a vehicle control incubation of 0.3% DMSO was used. Growth factors were prepared in BSA (0.25% w/v in PBS).

Statistical Analyses.

When measuring [³H]thymidine incorporation, each treatment in an individual experiment was carried out in quadruplicate. Each experiment was performed in at least three different cultures obtained from three different individual samples. Results are presented as grouped data from multiple cultures and are expressed as mean ± S.E. of n cultures. Fold increments were calculated by dividing the response of treated wells by that of the control wells on the same 24-well plate. The grouped percentage data were normalized by log transformation, before analysis by ANOVA, followed by post hoc tests when differences were detected. Differences were considered to be significant when P < .05. IC₅₀values were calculated from linear regression of log-concentration response data.

Materials.

All chemicals used were of analytical grade or higher. The compounds used and their sources were as follows: Calpain I inhibitor (N-acetyl-Leu-Leu-methioninal),l-glutamine, essentially fatty-acid–free BSA fraction V, albuterol, and thrombin (bovine plasma) were from Sigma Chemical Co. (St Louis, MO); collagenase type CLS 1 and elastase were from Worthington Biochemical (Freehold, NJ); Dulbecco ‘A’ PBS was from Oxoid (Hampshire, UK); trypsin, versene, penicillin-G, streptomycin, and serum-free DMEM were from CSL (Melbourne, Australia); fetal calf serum and amphotericin B (Fungizone) were from Flow Laboratories (Stanmore, Australia); DMEM was from Flow Laboratories (Irving, UK). PD98059 and phospho-specific p42/p44 ERK kinase antibody kit (rabbit polyclonal IgG, 1:1000) were from New England Biolabs (Beverly, MA). MG132 (Carbenzoxyl-l-leucyl-l-leucyl-l-leucinal) was from Calbiochem-Novabiochem GmBH (Usztweg, Bad Seden, Germany). [6-³H]thymidine (185 GBq/mmol, 5 Ci/mmol) and [γ-³²P]ATP (2 mCi/ml) were from Amersham; rabbit polyclonal IgG anti-pRb and goat polyclonal anti-ERK were from Santa Cruz Biotechnology (Santa Cruz, CA); Microscint-O scintillant was from Canberra-Packard (Canberra, Australia). Murine monoclonal anticyclin D1 was from Upstate Biotechnology (Lake Placid, NY). The antibodies used for immunocytochemistry were anti-smooth muscle α-actin (mouse monoclonal) (DAKO M851) and monoclonal mouse anti-PECAM-1 (DAKO-CD31, JC/70A) (DAKO M823) from the DAKO Corporation (Carpinteria, CA); sheep antimouse Ig HRP-conjugated and sheep antirabbit Ig HRP-conjugated were from Silenus (Hawthorn, Australia); murine monoclonal anticytokeratin (CY90) was from Sigma; and anti-smooth muscle myosin (rabbit polyclonal) was provided by Prof. M. Sparrow (University of Western Australia, Perth, Australia).

Results

Inhibition of Thrombin-Stimulated DNA Synthesis by β₂-Adrenergic Receptor Agonists, 8-Bromo-cAMP and PGE₂.

Thrombin (0.3 U/ml) increases incorporation of [³H]thymidine, measured in the last 4 h of a 28-h incubation, and increases cell number when the incubation period is extended beyond 48 h (Tomlinson et al., 1994; Tomlinson et al., 1995; Stewart et al., 1997). Albuterol, 8-bromo-cAMP, or PGE₂ continuously incubated with serum-deprived ASM from 30 min before exposure to thrombin (0.3 U/ml) and then throughout the 28-h incubation, inhibited [³H]thymidine incorporation (Fig.1) in a concentration-dependent manner. Concentrations of albuterol, 8-bromo-cAMP, or PGE₂, (100 nM, 300 μM, and 1 μM, respectively) were chosen for further studies.

Time Course of Action of Albuterol.

In quiescent (G₀) ASM cells, thrombin stimulates DNA synthesis with a delay of 20 to 22 h, defining the duration of G₁. Albuterol inhibits S-phase entry when added as late as 18 h after the addition of thrombin, consistent with an action at or near the restriction point (Stewart et al., 1997), which occurs in mid-to-late G₁ phase. Further evidence of an action at the restriction point of the cell cycle was sought by examining the inhibitory effect of albuterol (100 nM) exposures of varying durations, from 1 h to 8 h, commencing at different times from 4 h before thrombin until up to 26 h later (Fig.2). Different durations of exposure to albuterol were achieved without medium exchange by adding the β₂-adrenergic receptor selective antagonist ICI118551 (1 μM at 1, 2, 4, or 8 h after addition of albuterol). ICI118551 completely blocks the effects of albuterol (100 nM) on [³H]thymidine but had no direct effect on mitogen responses (Tomlinson et al., 1995; Stewart et al., 1997). Significant inhibition of [³H]thymidine incorporation was observed with periods of exposure to albuterol of as short as 1 h when the exposure commenced between 7 and 15 h after the addition of thrombin. Longer periods of exposure to albuterol (2–8 h) had greater inhibitory effects but also showed time-dependence. The greatest activity was observed when addition occurred in a period spanning the presumed timing of the restriction point (5–15 h after thrombin stimulation). These findings directed our further studies toward examination of the levels of cyclin D1 and pRb that control passage of cells through the restriction point of the cell cycle.

Effects of Albuterol on Changes in Levels of Phosphorylated ERK, Cyclin D1, and pRb Levels and Phosphorylation in Response to Thrombin Stimulation.

Cell lysates were prepared after a 20-h incubation in 0.3 U/ml thrombin to determine the effects of albuterol (1–100 nM, added 30 min before thrombin) on the stimulation of phospho-ERK, cyclin D1 levels, and the levels and phosphorylation of pRb (Fig.3). Albuterol opposed the thrombin-induced increases in levels of phospho-ERK and cyclin D1 in a concentration-dependent manner. The potency of albuterol for inhibition of the increase in cyclin D1 levels (pIC₅₀, 8.12 ± 0.19; n = 3) was similar to that for inhibition of [³H]thymidine incorporation (pIC₅₀, 7.83 ± 0.01; n = 3;P > .05, ANOVA). The repressor function of the restriction protein pRb is inhibited by phosphorylation by the active (cyclin D1-complexed) cdk4 (Herwig and Strauss, 1997). The levels of the restriction protein pRb were increased by thrombin (0.3 U/ml) and there was retardation of the pRb on polyacrylamide gel electrophoresis that was indicative of an increase in molecular weight because of phosphorylation (Fig. 3). Increases in pRb levels and phosphorylation induced by thrombin were inhibited in a concentration-dependent manner by albuterol (1–100 nM). The β₂-adrenergic receptor selective antagonist ICI118551 (1 μM) completely prevented the effects of albuterol (100 nM) on the changes in the levels of phospho-ERK, cyclin D1, and pRb (Fig. 3).

Effects of 8-Bromo-cAMP and PGE₂ on Thrombin-Stimulated Changes in Cyclin D1 Levels.

The effects of 8-bromo-cAMP (300 μM) and PGE₂ (1 μM) on thrombin-stimulated cyclin D1 levels were examined to investigate whether the effects of β₂-adrenergic receptor stimulation were observed with other agents that activate protein kinase A and inhibit DNA synthesis. Incubation of ASM cells in 8-bromo-cAMP or PGE₂ from 30 min before addition of thrombin attenuated the increase in cyclin D1 levels observed at 20 h after thrombin stimulation. The inhibitory effects of PGE₂ were significantly greater than those of 8-bromo-cAMP (Table 1).

Effect of the MEK1 Inhibitor PD98059 on ERK Activity, Cyclin D1 Levels, pRb Levels and Phosphorylation, and DNA Synthesis.

Persistent activation of ERK activity seems to be required for cells to progress through G₁ of the cell cycle to S-phase, and ERK activity has been linked to cyclin D1 expression in bovine ASM (Ramakrishnan et al., 1998). Therefore, we investigated the effects of PD98059, a selective inhibitor of MEK1 (Dudley et al., 1995), on phosphorylation of ERK, cyclin D1, and pRb levels (Fig.4). PD98059 (30 μM) inhibited thrombin-induced phosphorylation of ERK, increases in cyclin D1 levels, and increases in levels and phosphorylation of pRb, and in a previous study, has also been shown to inhibit [³H]thymidine incorporation (Fernandes et al., 1999).

Effects of PD98059, Albuterol, and 8-Bromo-cAMP on ERK Activity.

Our evidence (Fig. 4) and that in the literature implicating ERK activity as an upstream regulator of cyclin D1 protein levels (Cocks et al., 1992; Sewen et al., 1993; Ramakrishnan et al., 1998) and as a target for modulation by cAMP (Cook and McCormick, 1993), led to experiments examining the effects of albuterol on ERK activity in thrombin-stimulated cells. ERK activity was measured by an immunoprecipitation kinase assay in serum-deprived cells stimulated with thrombin (0.3 U/ml) for either 5 min, 30 min, or 8 h, at which time points the activity increased by 3.42, 1.45, and 1.77, respectively. Incubation with albuterol (100 nM) or 8-bromo-cAMP attenuated the increase in ERK activity at 30 min and 8 h after the addition of thrombin, but not at 5 min, whereas PD98059 inhibited ERK activity at all evaluated time points (Table2).

Regulation of Cyclin D1 Protein but not mRNA Levels by Albuterol.

Time course experiments were carried out to contrast the effects of albuterol on cyclin D1 mRNA (Fig.5) and protein levels (Fig.6). Cyclin D1 protein was increased above the basal level after 8 h exposure to thrombin and remained elevated until at least 20 h. At each of the time points investigated, albuterol (100 nM) inhibited the thrombin-stimulated increase in cyclin D1 protein level. In contrast, cyclin D1 mRNA levels, which were increased by thrombin as early as 4 h (1.38 ± 0.07;n = 9) and remained elevated at 8 h (1.45 ± 0.10; n = 4) and at 20 h (1.94 ± 0.27;n = 4), were not affected by albuterol or by PD98059 (Fig. 5).

Effects of the Proteasome Inhibitor MG132 and the Calpain I Inhibitors on Cyclin D1 Protein Levels.

The lack of regulation of cyclin D1 mRNA levels by albuterol or PD98059 (Fig. 5) suggested that post-transcriptional mechanisms may underlie the ability of cAMP to reduce the level of cyclin D1 protein. Recent evidence supports a role for proteasome degradation in the control of cyclin D1 in mouse fibroblasts (Diehl et al., 1997). We investigated whether inhibition of the proteasome pathway with MG132 (30 and 100 nM) or inhibition of calpain I with N-acetyl-Leu-Leu-norleucinal (LLN; 3 and 10 μM) (Diehl et al., 1997) would interfere with the regulatory effects of cAMP and albuterol on cyclin D1 protein levels. The reduction by isobutylmethylxanthine (IBMX; 100 μM) or 8-bromo-cAMP (300 μM) of thrombin-stimulated cyclin D1 protein levels (measured 18 h after thrombin stimulation) was reversed by MG132 (100 nM) addition at 15 h after thrombin stimulation (Fig.7). Late addition of MG132 (30 nM) alone had no effect on thrombin-stimulated increases in cyclin D1 protein levels (124 ± 14% of thrombin-stimulated level,n = 4). The decrease in cyclin D1 protein levels induced by albuterol (100 nM) was significantly reduced (P < .05, Dunnett’s test) by the late addition of MG132 (Table 3). In a separate series of experiments examining albuterol responses, incubation with MG132 (30 nM) from 60 min before thrombin and then throughout the 18-h period reduced the inhibition (P < .05, Dunnett’s Test) of thrombin-stimulated increases in cyclin D1 protein levels by albuterol (100 nM) from 48 ± 11% (n = 6) to the nonsignificant level of 18 ± 12% (n = 6). The continuous incubation of the cells with MG132 (30 nM) had no significant effect on the thrombin-stimulated cyclin D1 levels (80 ± 10% of thrombin-stimulated level; n = 5). Similarly, the basal levels of cyclin D1 (20 ± 6% of thrombin-stimulated level) were unchanged (35 ± 9% of thrombin-stimulated level) by continuous incubation with MG132 (30 nM).

Discussion

Elevation of cAMP levels attenuates proliferative responses of cultured airway smooth muscle cells of various species (Tomlinson et al., 1994, 1995; Maruno et al., 1995; Young et al., 1995; Schramm et al., 1996). We now show that albuterol inhibits the passage of cells through the restriction point by decreasing levels of cyclin D1, thereby decreasing cdk-mediated phosphorylation of pRb. Albuterol-mediated reduction in ERK activity was associated with a decrease in the levels of cyclin D1 protein, but not mRNA. This post-transcriptional regulation of cyclin D1 by albuterol was inhibited by the proteasome inhibitor MG132 (Rock et al., 1994) and by the calpain I inhibitor LLN (Diehl et al., 1997), suggesting that the β₂-adrenergic receptor agonist/cAMP pathway may accelerate cyclin D1 protein degradation. However, our findings do not exclude the possibility that cAMP has an additional or alternative action to alter translation of cyclin D1 mRNA.

Albuterol inhibited S-phase entry in ASM when present in the culture medium throughout the period of exposure to thrombin (Tomlinson et al., 1995). The magnitude of this cAMP-dependent G₁arrest was less against receptor tyrosine kinase-type mitogens compared with those activating G protein-coupled receptors (Tomlinson et al., 1995). The inhibitory activity was maintained when albuterol was added up to, but not later than, 18 h after thrombin stimulation (Stewart et al., 1997). Because S-phase commences 20 to 22 h after thrombin addition, this placed the loss of activity of albuterol at a point coinciding with the restriction point in mid-to-late G₁. The restriction point is defined as the time in G₁ when cells continue to progress through the cycle to S-phase in a mitogen-independent manner (Pardee, 1974). This point in the cell cycle is considered to relate to the conversion of pRb from an under-phosphorylated state to a hyperphosphorylated state by active cyclin D1/cdk4 complexes (Herwig and Strauss, 1997). The exposure of ASM to albuterol for between 1 and 8 h commencing at different points of G₁ phase indicated that the inhibitory effect on DNA synthesis was observed even with short pulses of exposure (1 h) when these overlapped with the window of timing of the restriction point. The magnitude of inhibition was dependent on the duration of exposure, but greatest inhibition was consistently obtained between 5 to 15 h after the addition of thrombin. Thus, albuterol targeted discrete biochemical events critical for passage of cells through the restriction point, consistent with its ability to reduce the level and phosphorylation of pRb.

Cyclin D1 is one of two rate-limiting cyclins in the passage of cells through G₁ phase of the cell cycle (the other is cyclin E; Resnitzky and Reed, 1995). Cyclin D1 levels are low in quiescent (G₀) cells, rise throughout G₁, and remain elevated for the remainder of the cell cycle (Matsushime et al., 1991). Cdk4 phosphorylates pRb, leading to derepression of the transcriptional activity of the heterodimeric E2F transcription factor, allowing passage through the restriction point by enabling the expression of proteins that are essential for entry into S-phase. Cyclin E lies downstream of cyclin D1 in the signaling cascade and is considered to play a role in the passage through the G₁/S boundary (Resnitzky and Reed, 1995).

Concentrations of albuterol that inhibited DNA synthesis reduced cyclin D1 levels via activation of a β₂-adrenergic receptor. The membrane-permeant and metabolically stable analog of cAMP, 8-bromo-cAMP, and PGE₂, an established stimulant of cAMP in ASM (Tomlinson et al., 1995), also opposed thrombin-induced increases in the levels of cyclin D1 protein at concentrations that reduced DNA synthesis. These findings are consistent with data showing 8-bromo-cAMP–induced suppression of cyclin D1 protein expression in CCL39 cells (L’Allemain et al., 1991).

The concurrent inhibition by PD98059 (Dudley et al., 1995) of thrombin-stimulated, ERK activity, cyclin D1 levels, pRb phosphorylation, and DNA synthesis is consistent with an important role for ERK in the control of both cyclin D1 levels and ASM cell cycle progression. The reduction in pRb levels and phosphorylation in the presence of PD98059 suggests that this compound prevents ASM passage through the restriction point. The suppression of thrombin-stimulated cyclin D1 levels by PD98059 suggests that MEK1 and ERK activity maintains elevated cyclin D1 levels.

ERK activation persisting through early G₁, is required for mitogenic activity (Meloche et al., 1992), but the duration of this persistence has not been clearly defined. Our findings indicate that the β₂-adrenergic receptor agonists albuterol or 8-bromo-cAMP suppress ERK activity between 30 min and 8 h after addition of thrombin, but do not reduce the peak of ERK activity at 5 min. These observations are consistent with an inhibitory action of cAMP on or upstream of ERK and with those in bovine tracheal smooth muscle cultures, showing that histamine-stimulated increases in the levels of cAMP inhibitraf-dependent ERK activation (Hershenson et al., 1995). In the rat1 fibroblast cell line, cAMP regulates theras/raf/ERK signaling cascade by protein kinase A-mediated phosphorylation of raf, preventing its binding to and activation by ras (Cook and McCormick, 1993). This action of cAMP is evident in fibroblast cell lines at 5 min after exposure to stimulus. The cAMP-insensitive component of the early ERK activation in thrombin-stimulated ASM may be explained by the involvement of alternative ras/raf independent pathways that lead to ERK activation, including protein kinase C. There is evidence for a role of cAMP-insensitive ERK-activation pathways in platelet-derived growth factor-stimulated bovine ASM (Hershenson et al., 1995).

The steps linking ERK activity with the increased levels of cyclin D1 protein have not been fully elucidated, but several studies suggest that cyclin D1 reporter gene activity is increased by an ERK-dependent signaling pathway (Lavoie et al., 1996; Ramakrishnan et al., 1998). In Chinese hamster embryo fibroblasts, sustained activation of ERK is required for ongoing expression of cyclin D1 protein and mRNA (Weber et al., 1997). Superficially, our findings are compatible with ERK being an upstream regulator of cyclin D1 expression, because cyclin D1 protein levels were decreased by the MEK1 inhibitor PD98059 and there was parallel inhibition of ERK and cyclin D1 protein by cAMP elevation. However, a time-course study examining cyclin D1 expression as early as 4 h after exposure to thrombin did not find any evidence for inhibition of the elevated cyclin D1 mRNA levels with either PD98059 or the β₂-adrenergic receptor agonist. We have examined the effects of a number of agents on thrombin-stimulated cyclin D1 expression and found that of PD98059, rapamycin, albuterol, IBMX, 8-bromo-cAMP, and dexamethasone, only dexamethasone reduces cyclin D1 mRNA levels (Fernandes et al., 1999). A recent investigation points to a key role for nuclear factor-κB in the transcriptional regulation of cyclin D1 (Hinz et al., 1999). As glucocorticoids are well established and powerful regulators of the activation of nuclear factor-κB, examination of the role of this transcription factor in human ASM cell cycle is warranted.

Although reduced translation of cyclin D1 mRNA could potentially explain the cAMP-associated decrease in cyclin D1 protein levels, the recently reported involvement of the proteasome degradation pathway in cyclin D1 stability (Diehl et al., 1997, 1998) was considered more likely to be involved. Addition of the proteolysis inhibitors, MG132 or LLN to ASM late in G₁ at 15 h, attenuated the reduction in cyclin D1 protein levels (measured at 18 h) caused by continuous incubation in albuterol, the phosphodiesterase inhibitor, IMBX or 8-bromo-cAMP. These observations suggest that ongoing elevation of cAMP acutely regulates the level of cyclin D1 with a measurable effect within 3 h of interruption of this control mechanism. In some experiments, MG132 and LLN addition was delayed until 3 h before the harvest of cell lysates to reduce the impact of confounding influences of changes in the rate of degradation of other proteins by the proteasome pathway, such as the regulatory unit of protein kinase A. The latter studies also provided evidence that the albuterol/cAMP-induced decline in cyclin D1 levels, but not those under basal conditions or after incubation with thrombin alone, was opposed by MG132. Dibutyryl cAMP-induced increases in proteasomal degradation of the transcription factor GATA-6 (Nakagawa et al., 1997) support the inference that proteasome degradation is cAMP-sensitive. Furthermore, our studies with Clontech Atlas Array membranes (Palo Alto, CA) have identified that albuterol up-regulates ubiquitin ligase mRNA (E.G. and A.G.S., unpublished observations).

In NIH3T3 fibroblasts, phosphorylation of cyclin D1 by glycogen synthase kinase-3β (GSK-3β) triggers its rapid proteasome-dependent proteolysis (Diehl et al., 1998). The influence of cAMP on activation of GSK-3β in ASM is not known. However, in 3T3 fibroblasts, it seems that a ras-dependent PI3K pathway rather than theras/raf/ERK pathway inhibits the activation of GSK-3β (Diehl et al., 1998). Interestingly, GSK-3β activity is inhibited by isoproterenol acting through β₃-adrenergic receptors independently of cAMP in L6 myotubes (Moule et al., 1997). However, β₃-adrenergic receptors are not implicated in the effects of albuterol on the ASM DNA synthesis or cyclin D1, because these are blocked by ICI118551 and propranolol (Tomlinson et al., 1994, 1995).

Although ERK activity is required for S-phase entry in ASM (Ramakrishnan et al., 1998), it is unlikely to be sufficient (Malarkey et al., 1995). In bovine tracheal smooth muscle, the phosphoinositol-3-kinase (PI3K)/p70 ribosomal S6 kinase pathway is also required for S-phase entry and is inhibited by elevated levels of cAMP (Scott et al., 1996; Walker et al., 1998). Inhibitors of PI3K (wortmannin) and p70 ribosomal S6 kinase activation (rapamycin) reduce platelet-derived growth factor-induced DNA synthesis (Scott et al., 1996) and also reduce thrombin-induced DNA synthesis in cultured human ASM (A.G.S. and T.H., unpublished observations). The PI3K pathway therefore represents an additional site for the inhibitory actions of cAMP in ASM, because PI3K is linked to inhibition of the activation of glycogen synthase kinase-3β and acceleration of cyclin D1 protein proteasome-dependent degradation (Diehl et al., 1998).

Our observations indicate that β₂-adrenergic receptor stimulation and cAMP cause a G₁ arrest in human cultured ASM cells by opposing mitogen-induced elevation of cyclin D1 protein levels through a post-transcriptional action that seems to regulate the proteasome-dependent degradation of cyclin D1. Further elucidation of the targets for regulation of cell cycle progression by β₂-adrenergic receptor agonists and glucocorticoids that do reduce cyclin D1 mRNA levels (Fernandes et al., 1999) may indicate how these agents could be used optimally in combination to modulate airway-wall remodeling in asthma.