Ethanol increases phosphodiesterase 4 activity in bovine bronchial epithelial cells

Alcohol 31 (2003) 31–38

Ethanol increases phosphodiesterase 4 activity in bovine bronchial epithelial cells Mary A. Forge`tb, Joseph H. Sissonb, John R. Spurzema,b, Todd A. Wyatta,b,* b

a Research Service, Department of Veterans Affairs Medical Center, 4101 Woolworth Avenue, Omaha, NE 68105, USA Pulmonary and Critical Care Medicine Section, Department of Internal Medicine, University of Nebraska Medical Center, 985300 Nebraska Medical Center, Omaha, NE, 68198-5300, USA Received 13 March 2003; received in revised form 14 June 2003; accepted 20 June 2003

Abstract Ethanol exposure in airway epithelium increases cyclic AMP (cAMP)–dependent protein kinase (PKA) activity. Activation of PKA and cyclic guanosine monophosphate (cGMP)–dependent protein kinase (PKG) has been shown to increase ciliary beat frequency (CBF) in bovine bronchial epithelial cells (BBECs). We have shown that biologically relevant concentrations of ethanol stimulate increases in CBF in a nitric oxide–dependent manner, mediated through elevated cAMP levels and subsequent PKA activation. This ethanol-driven rapid and transient increase in CBF occurs 15 to 30 min after exposure to 100 mM ethanol. However, after prolonged exposure to 100 mM ethanol (ⱖ6 h), CBF and the catalytic activity of PKA return to baseline levels. We hypothesize that cyclic nucleotide–dependent phosphodiesterase (PDE) activity attenuates the duration of ethanol-stimulated ciliary motility. The effect of ethanol on the PDE activity in BBECs was determined through direct assay of catalytic activity. When BBECs were incubated with 100 mM ethanol, significant increases in cAMP levels occurred within 1 h, with corresponding increases in PKA activity. Treatment of BBECs with 100 mM ethanol increased cAMP–PDE activity significantly by 4 h. 3-Isobutyl-1-methylxanthine, Ro 20-1724, and rolipram inhibited ethanolstimulated cAMP–PDE activity. These agents inhibited ethanol-stimulated cAMP–PDE activity and increased the magnitude of ethanolstimulated PKA activity observed under the same conditions. These findings support the idea that acute exposure (⬍6 h) to ethanol increases cAMP levels, and the associated increase in PKA activation is regulated by cAMP-dependent PDE, specifically PDE4. Other compensatory mechanisms however, may be responsible for the down-regulation of PKA, which occurs after chronic epithelial exposure (ⱖ6 h) to ethanol. 쑖 2003 Elsevier Inc. All rights reserved. Keywords: Lung; Airway epithelium; cAMP; PKA; PDE4; Phosphodiesterase; Ethanol

1. Introduction The results of numerous studies indicate that alcoholism contributes to lung disease (Murray et al., 1998; Tabak et al., 2001). For example, heavy drinkers (individuals who consume ⬎6.5 oz of alcohol per week) have an increased incidence of pneumonia, bronchitis, and lung disease (Lebowitz, 1981). The findings obtained from these studies support the suggestion that alcoholism contributes to the development of lung disease through a variety of compromised physiologic processes. Mucociliary clearance—the

* Corresponding author. Pulmonary and Critical Care Medicine Section, Department of Internal Medicine, University of Nebraska Medical Center, 985300 Nebraska Medical Center, Omaha, NE 68198-5300, USA. Tel.: ⫹1-402-559-3817; fax: ⫹1-402-559-8210. E-mail address: [email protected] (T.A. Wyatt). Editor: T.R. Jerrells 0741-8329/03/$ – see front matter 쑖 2003 Elsevier Inc. All rights reserved. doi: 10.1016/j.alcohol.2003.06.005

production of mucus and the synchronized beating of airway cilia—is one important mechanism through which the body removes microorganisms, dust, and debris. In this way, mucociliary clearance serves as a primary host defense mechanism. The increased incidence of lung disease observed in heavy drinkers may be due to a decrease in effective mucociliary clearance. Like many host defense mechanisms, mucociliary clearance can be regulated in response to external stimuli. During periods of normal function, baseline ciliary beat frequency (CBF) maintains resting airway clearance. During periods of stress, ciliary motility increases and enhances clearance. One of the mechanisms responsible for this stimulation of ciliary motility is an increase in intracellular cyclic nucleotides (Lansley et al., 1992; Wyatt et al., 1998b). Such increases in cyclic nucleotides result in activation of the cyclic AMP (cAMP)– and cyclic guanosine monophosphate (cGMP)–dependent protein kinases (PKA and PKG, respectively). When triggered by beta-agonists, these increases in

32

M.A. Forge`t et al. / Alcohol 31 (2003) 31–38

CBF are nitric oxide (NO) dependent, because (1) increases in CBF parallel NO production in these cells and (2) inhibition of nitric oxide synthase (NOS) blocks CBF stimulation (Jain et al., 1993). Because of its clinical significance, the effect of ethanol on this regulatable host defense has been studied. Results of earlier work have shown that acute exposure (0.5–6 h) to ethanol increases cilia motility in an in vitro bovine model (Sisson, 1995). As with beta-agonists, stimulation of CBF by ethanol is mediated through a NO-dependent pathway and coincides with increases in cAMP levels that result in an increase in PKA activity (Sisson et al., 1999). Increased PKA activity results from the binding of cAMP, which allows dissociation of the catalytic and regulatory subunits and results in an active catalytic site (Francis et al., 2001). The exact mechanisms involved between PKA activation and CBF increases are not fully understood. The PKA activity can return to baseline levels through the action of phosphodiesterases (PDEs), enzymes that catalyze the hydrolysis of the ester bond, which allow for the cyclic structure of the nucleotides, resulting in a biologically inactive 5′-nucleotide monophosphate (Francis et al., 2001). At least 11 families of PDE isoenzymes have been categorized on the basis of their particular structure and function (Beavo et al., 1994; Soderling & Beavo, 2000). A predominant isoenzyme found in lung epithelial cells is type 4 (PDE4), a cAMP-dependent PDE (cAMP–PDE) (Barnes, 1995). However, other PDE isoforms, such as PDE1, PDE5, and PDE7, have also been localized in the human and bovine airways (Fuhrmann et al., 1999). Phosphodiesterase activity is known to provide negative feedback regulation of PKA by means of the hydrolysis of cAMP in many cell types. Therefore, given that acute exposure (⬍6 h) to ethanol increases cilia motility in epithelial cells by means of cAMP activating PKA and that this stimulated activity returns to baseline by 6 h of ethanol treatment, we hypothesize that acute exposure to ethanol activates cAMP–PDE catalytic activity, allowing for the hydrolysis of cAMP, reducing ethanol-stimulated PKA activity and the return of acute ethanol exposure–stimulated CBF to baseline levels in bovine bronchial epithelial cells (BBECs). 2. Materials and methods 2.1. Cell preparation To obtain primary cell cultures, bovine lungs were obtained at a local slaughterhouse, and bronchi were extracted. After the removal of excess adipose and connective tissue, each bronchus was washed two times in phosphate-buffered saline (PBS) containing a mixture of streptomycin (50 U/ ml), penicillin (50 U/ml), and amphotericin B (2 µg/ml). Bronchi were then placed in Media-199 (GIBCO-Invitrogen, Carlsbad, CA) containing 0.01% protease and incubated overnight at 4ºC. To harvest the epithelial cells, each bronchial lumen was washed repeatedly with Media-199

containing 10% fetal bovine serum. This medium was then collected, filtered through a 75-µm mesh to remove debris, and centrifuged at 2,500g for 10 min. The resulting cell pellet was resuspended in Media-199 containing antibiotics and plated at a concentration of 1 × 104 cells/cm2 in 60-mm polystyrene tissue culture dishes coated with 1% type I collagen. Cells were grown and all incubations were performed at 37ºC and 5% CO2. At confluence, each plate contained approximately 2 mg of total cellular protein. For all experiments, cells had greater than 95% viability, as tested by lactic dehydrogenase (LDH) assay (Sigma, St. Louis, MO). 2.2. Phosphodiesterase activity assay The PDE assay measures catalytic activity through the breakdown of [3H]-cAMP to the nucleotide monophosphate. The procedure we used was a modification of methods previously described (Wyatt et al., 1998a). The PDE activity was determined in crude whole cell fractions. Confluent monolayers of primary BBECs were stimulated in the presence or absence of ethanol and PDE inhibitors. Media were removed, and cells were flash frozen in liquid nitrogen after the addition of 250 µl of cell lysis buffer, consisting of 35 mM TRIS-HCl (pH 7.4), 0.4 mM ethyleneglycolbis(β-aminoethyl ester)-N,N,N′N′-tetraacetic acid (EGTA), 10 mM MgCl2, 0.1% Triton X-100, 0.2 mM phenylmethylsulfonyl fluoride (PMSF), leupeptin (1 µg/ml), and aprotinin (100 µg/ml). The dishes were stored at –70ºC until assayed. Monolayers were scraped, placed in chilled microcentrifuge tubes, homogenized by sonication, and centrifuged at 10,000g for 30 min at 4ºC to separate the cytosolic and membrane portions. To initiate the reaction, 100 µl of the cytosolic fraction was added to 150 µl of the PDE reaction mix, consisting of 40 mM 4-morpholinepropanesulfonic acid (MOPS) buffer at pH 7.5, 0.5 mM EGTA, 15 mM magnesium acetate, bovine serum albumin [(BSA); 0.17 mg/ml), 20 µM cAMP, and [3H]-cAMP. Each sample was incubated for 20 min at 30ºC. Samples were subsequently boiled 3 min to halt the first reaction and then chilled on ice for 3 min in preparation for the second reaction. This second reaction was initiated by the addition of 10 µl of snake venom (10 mg/ ml) and incubated for 10 min at 30ºC. Addition of 250 µl of low salt buffer stopped this reaction. Each sample was subsequently loaded on minichromatography columns, and anion-exchange chromatography with A-25 Sephadex resin was performed to separate incorporated [3H]-cAMP. The eluate, which contained the incorporated tritium, was collected and counted in aqueous scintillation solution. Enzyme activity was expressed as picomoles per minute per milligram of protein and analyzed by using a one-way analysis of variance (ANOVA). 2.3. Cyclic nucleotide–dependent kinase activity assay The PKA activity was determined in crude whole cell fractions of primary BBECs. Confluent monolayers of

M.A. Forge`t et al. / Alcohol 31 (2003) 31–38

BBECs were treated, media were removed, and remaining material was lysed and flash frozen as described above. To separate crude fractions, monolayers were scraped and homogenized by sonication and centrifuged at 10,000g for 30 min at 4ºC. The supernatant portion was assayed by using a modification of the methods previously described (Jiang et al., 1992) with the use of 130 µM PKA heptapeptide substrate (LRRASLG) (Peninsula, Belmont, CA), in a buffer containing 20 mM TRIS-HCl (pH 7.5), 100 µM 3-isobutyl1-methylxanthine (IBMX), 20 mM magnesium acetate, and 200 µM [γ-32P] ATP. Samples (20 µl) were added to 50 µl of the reaction mix described above and incubated for 15 min at 30ºC. The reaction was halted by spotting the sample onto P-81 phosphocellulose papers (Whatman, Hillsboro, OR), and papers were then washed five times in 75 mM phosphoric acid for 5 min each and washed once in ethanol. Dry papers were subsequently counted in nonaqueous scintillation solution, and enzyme activity was expressed as picomoles per minute per milligram of protein. Significance was determined by using a one-way ANOVA.

33

as a virtual instrument to display a time versus amplitude waveform of the ROI as a virtual strip chart recorder); (2) to perform a power spectrum analysis (with the use of fastFourier transformation or FFT) and display the number of readings versus frequencies over a 0–15 Hz range present in the ROI; and (3) to determine the dominant frequency among the frequencies displayed across the range of frequencies. All frequencies represent the mean ⫾ 1 S.E.M. from six separate cell groups or fields. 2.5. Materials The Media-199, streptomycin–penicillin, and fungizone were purchased from GIBCO-Invitrogen (Carlsbad, CA). The [γ-32P]-ATP was purchased from ICN (Irvine, CA); the [8-3H]adenosine 3′,5′-cyclic phosphate, from Amersham Biosciences (Piscataway, NJ); the phosphocellulose P-81 paper, from Whatman (Clifton, NJ); and the peptides for kinase assays, from Peninsula Laboratories (Belmont, CA). All other reagents were purchased from Sigma Chemical (St. Louis, MO).

2.4. Cilia beat frequency measurements The CBF measurements were made by first examining ciliated epithelial cells and next quantifying the motion of their cilia. Phase-contrast microscopy, video analysis, and computerized frequency spectrum analysis were used to analyze CBF. To maintain a consistent temperature during experimental periods (24ºC ⫾ 0.5ºC), cultures of ciliated cells were placed on a thermostat-controlled heated stage. Visual images of actively beating cilia were made by using an inverted phase-contrast microscope with a ×20 objective lens with a ×1.5 tube multiplier (IMT-2; Olympus, Melville, NY). These images were recorded with a video camera (WVD5000; Panasonic, Secaucus, NJ), and analog video recordings were made by using an sVHS video cassette recorder (AG-1980; Panasonic). The analog video output sampled the image at 30 frames per second (fps). Quantification of CBF was later performed by analyzing video-taped experiments. A video signal processor (University of Nebraska Biomedical Instrumentation Department, Omaha, NE) was used to split and modify the analog video signal for display and analysis. This processor performed three functions: (1) An unmodified video image was routed to a television monitor for viewing and selecting regions of interest (ROIs); (2) a movable and sizable window was superimposed on the video image to indicate the ROIs to be analyzed; and (3) the light intensity of the video signal contained within the ROI was averaged into an analog output light intensity signal that was routed into the computer for frequency analysis. A computer (Macintosh IIci; Apple Computer, Cupertino, CA) received the analog signal where an analog-todigital converter board [(A/D); National Instruments, Austin, TX] created a digital light intensity signal. Customized software written in Lab-View 2.1 (National Instruments) was used (1) to analyze the digital signal (the software functioned

3. Results 3.1. Ethanol transiently stimulates cyclic AMP levels in airway epithelial cells Because PDE activity may provide negative feedback regulation of ethanol-stimulated PKA activity, we determined the time course of ethanol-stimulated cAMP generation in airway epithelial cells. Cyclic AMP levels were measured in crude homogenates of BBECs that were boiled and assayed for the concentration (picomoles per milligram of protein) of cyclic nucleotide (see Materials and Methods section). Exposure to 100 mM ethanol for 1 h stimulated a fivefold increase in cAMP levels over those detected in BBECs treated with media alone (Fig. 1). The ethanol-stimulated increases in cAMP levels were transient. By 6 h, cAMP levels returned to baseline, where they remained through 24 h. 3.2. Ethanol stimulates cAMP–PDE catalytic activity in airway epithelium Having observed that acute exposure to ethanol transiently increased cAMP levels in BBECs, we hypothesized that the drop in cAMP levels after 1 h of ethanol stimulation was due to activation of PDEs. To test this hypothesis, direct PDE catalytic assays were performed to determine PDE activity changes after exposure to ethanol. To assay PDE activity, monolayers of primary BBECs were treated for 0 to 18 h with 100 mM ethanol in multiple 60-mm tissue culture dishes. The catalytic activity of homogenized cell lysates was determined by direct cAMP–PDE activity assay (see Materials and Methods section). The catalytic activity of cAMP–PDE in BBECs increased with acute exposure (⬍6

34

M.A. Forge`t et al. / Alcohol 31 (2003) 31–38

had returned to baseline levels, where it remained throughout the 18 h after ethanol exposure. This change in PDE activity lagged behind, but roughly paralleled, observed cyclic nucleotide changes, supporting the suggestion that the return of cAMP levels to baseline values with prolonged ethanol exposure may be regulated through the action of cAMP–PDE. 3.3. Varying the treatment concentration does not affect the duration of ethanol’s effect on cAMP–PDE activity

Fig. 1. Ethanol increases the concentration of cyclic AMP (cAMP) in bovine bronchial epithelial cells (BBECs). Primary BBECs were cultured from a bovine bronchus, grown to confluence, and assayed for cAMP concentration. The BBECs were incubated in Media-199 in the presence of 100 mM ethanol (EtOH) for 0 to 24 h and flash frozen in cell lysis buffer (see Materials and Methods section). Cell homogenates were assayed for the concentration (picomoles per milligram of protein) of cAMP as described (see Materials and Methods section). A significant increase in the cAMP concentration occurred at 1 h of ethanol incubation. No significant increase in cAMP production occurred at 6 to 24 h of incubation with 100 mM ethanol. Each data point represents triplicate measurements (n ⫽ 9) for that given time point. A P value of ⬍.05 was considered significant, as indicated by (*). Small bars represent ⫾ S.E.M.

To determine the effect of ethanol concentration on cAMP–PDE activity, BBECs were treated for 2, 4, or 6 h with concentrations of ethanol ranging from 0.1 to 100 mM. Measurable increases in PDE activity occurred after 2 and 4 h of exposure to ethanol at concentrations ranging from 0.1 to 100 mM (Fig. 3). These activity levels showed a concentration-dependent change in activity over a biologically relevant range of concentrations. The highest overall elevations in PDE activity occurred at 100 mM ethanol after 4 h of incubation. The catalytic activity present at 4 h was significantly greater than that observed with 6 h of ethanol treatment. The cAMP–PDE activity was no longer stimulated by ethanol (0.1–100 mM) after 6 h of incubation. These findings support the idea that PDE activity in ethanol-treated BBECs is elevated by 2 h, continues to increase by 4 h,

h) to 100 mM ethanol (Fig. 2). By 2 h, a measurable, but transient, increase was observed, whereas peak cAMP–PDE activity (1,400 pmol/min/mg) occurred after 4 h of ethanol treatment. The ethanol-stimulated PDE activity rapidly decreased after 4 h of ethanol exposure. By 5 h, the activity

Fig. 2. Ethanol stimulates increases in the catalytic activity of cyclic AMP (cAMP)–dependent phosphodiesterase (cAMP–PDE) during acute incubations. Bovine bronchial epithelial cells (BBECs) were treated with 100 mM ethanol (EtOH) for 0 to 18 h, and the cAMP–PDE activity was measured. Ethanol stimulated significant increases in cAMP–PDE activity after 4 h of incubation. Although no significant increases in cAMP–PDE activity occurred at 1 to 3 h of ethanol treatment, an increasing trend in activity was observed. No significant increase in cAMP–PDE occurred with 5 to 18 h of ethanol incubation, as cAMP–PDE activity levels returned to baseline. The cAMP–PDE catalytic activity was measured as described (see Materials and Methods section). Each data point represents triplicate measurements (n ⫽ 9) for a given sample. A P value of ⬍.05 was considered significant, as indicated by (*). Small vertical bars represent ⫾ S.E.M.

Fig. 3. Acute ethanol treatment results in increased cyclic AMP (cAMP)– dependent phosphodiesterase (cAMP–PDE) catalytic activity. Confluent monolayers of primary bovine bronchial epithelial cells (BBECs) were incubated with ethanol (0.1–100 mM) for 2, 4, or 6 h. Cell media were removed, and cells were flash frozen (see Materials and Methods section). Cell lysates were assayed for cAMP–PDE activity. Measurable increases in cAMP–PDE activity occurred at 2 h of ethanol (EtOH) incubation, for all concentrations measured. The greatest increase after 2 h of incubation with ethanol occurred at a concentration of 100 mM. Ethanol incubation for 4 h resulted in a measurable increase in cAMP–PDE activity at 10 mM ethanol and asignificant increase at 100 mM ethanol. However, after 6 h of ethanol incubation, cAMP–PDE activity returned to unstimulated levels, and no significant change occurred with ethanol treatment. Vertical bars represent ⫾ S.E.M., and each experimental data point resulted from triplicate samples (n ⫽ 9). Significance occurred with a P value of ⬍.05, as indicated by (*).

M.A. Forge`t et al. / Alcohol 31 (2003) 31–38

35

and returns to baseline by 6 h of ethanol stimulation, regardless of the ethanol concentration. 3.4. Ethanol-increased PDE activity is blocked by a nonspecific phosphodiesterase inhibitor, IBMX Having established, through the use of direct catalytic–activity assays, that cAMP–PDE levels in BBECs increase in response to stimulation with 100 mM ethanol, we examined the effect of PDE inhibition. Bovine bronchial epithelial cells were initially preincubated for 1 h with 200 µM IBMX, a nonspecific cAMP–PDE inhibitor, and subsequently stimulated with ethanol for 1 to 18 h. The IBMX blocked ethanol stimulation of cAMP–PDE activity (Fig. 4). 3.5. Inhibition of PDE 4, but not of PDE5, decreases cAMP–PDE activity in response to ethanol To determine the specific cAMP–PDE isoform activated by ethanol, cells were pretreated with 200 µM IBMX, 10 µM zaprinast, or 20 µM Ro 20-1724, followed by exposure to 100 mM ethanol (Fig. 5). Catalytic activity of cAMP– PDE in BBECs increased after 4 h of incubation with 100 mM ethanol. These activities were blunted by pretreatment with a nonspecific PDE inhibitor, IBMX. Pretreatment with Ro 20-1724 profoundly blocked the ability of ethanol to increase cAMP–PDE activity. In contrast, zaprinast, a PDE5 inhibitor, had no effect on cAMP–PDE activity in mediaor ethanol-treated cells. These findings indicate that ethanolactivated PDE activity is likely the PDE isoform type 4.

Fig. 4. A nonspecific phosphodiesterase (PDE) inhibitor, 3-isobutyl-1methylxanthine (IBMX), lowers ethanol-stimulated cyclic AMP (cAMP)– PDE activity in bovine bronchial epithelial cells (BBECs). Cells were preincubated with or without 200 µM IBMX for 1 h and subsequently treated with 100 mM ethanol (EtOH) for 0 to 18 h. Cell lysates were assayed for cAMP–PDE catalytic activity. At all time points, preincubation with IBMX resulted in no measurable increases in cAMP–PDE activity after cells were treated with ethanol (0.5–18 h). Significant increases in cAMP–PDE activity occurred only when BBECs were incubated with ethanol without IBMX pretreatment. Bars represent ⫾ S.E.M. for each data point. Samples were repeated in triplicate (n = 9), and significance is indicated by (*) when a P value of ⬍.05 occurred.

Fig. 5. Preincubation of bovine bronchial epithelial cells (BBECs) with phosphodiesterase type 4 (PDE4) inhibitors, but not with PDE5 inhibitors, decreases 100 mM ethanol–stimulated cyclic AMP (cAMP)–dependent phosphodiesterase (PDE) activity. Confluent monolayers of primary BBECs were pretreated with media only, 200 µM 3-isobutyl-1-methylxanthine (IBMX), 20 µM Ro 20-1724, or 10 µM zaprinast for 1 h and subsequently incubated with (gray bars) or without (white bars) 100 mM ethanol (EtOH) for 4 h. Cell lysates were assayed for cAMP–PDE activity. Cells that were pretreated with media only or with zaprinast had a significant increase in the cAMP–PDE activity when incubated with 100 mM ethanol for 4 h. Ethanol-stimulated cAMP–PDE activity was blocked by pretreatment with IBMX and Ro 20-1724. The cAMP–PDE activity was expressed as fold increase over baseline values for media control cells. Samples were assayed in triplicate (n = 9), and significance occurred with a P value of ⬍.05, as indicated by (*). For each data point the ⫾ S.E.M. is indicated by small vertical bars.

3.6. Magnitude of PKA activity stimulated by ethanol is extended with cAMP–PDE activity inhibition The catalytic activity of PKA was assayed in ethanol (100 mM)-treated BBECs with and without cAMP–PDE inhibitors. The effect of PDE4 inhibition on the ability of ethanol to increase PKA activity was studied by using a direct PKA-activity assay. Significant increases in PKA activity occurred after 1 h of incubation with 100 mM ethanol (Fig. 6). This corresponds to observations that acute exposure (⬍6 h) to ethanol significantly increased PKA activity in BBECs. However, inhibition of cAMP–PDE activity by preincubating cells for 1 h with Ro 20-1724 (20 µM) or rolipram (8 µM) before ethanol treatment resulted in significantly greater PKA activation compared with findings for ethanol exposure alone. No significant change in PKA activity was observed when cells were treated for 6 or 24 h with ethanol. Preincubation of BBECs with PDE inhibitors for 1 h and subsequent treatment of cells for 6 or 24 h resulted in no significant increase in PKA activity. No significant change in PKA activity was observed when cells were treated with PDE inhibitors in the absence of ethanol (data not shown). 3.7. Effect of PDE4 inhibitors on cilia beat frequency The effect of PDE4 inhibitors on ciliary motion, both with and without ethanol, was examined. To accomplish this,

36

M.A. Forge`t et al. / Alcohol 31 (2003) 31–38

Fig. 6. Effect of phosphodiesterase type 4 (PDE4) inhibition on ethanolstimulated cyclic AMP–dependent protein kinase (PKA) activity in bovine bronchial epithelial cells (BBECs). Confluent monolayers of BBECs were pretreated for 1 h with Ro 20-1724 (20 µM), rolipram (8 µM), or media and subsequently treated with 100 mM ethanol (EtOH). The cytosolic fraction of cell homogenates was assayed for PKA activity, as expressed in picomoles per minute per milligram of protein. The PKA activity was significantly increased over baseline (media only) values in response to 1 h of ethanol treatment. In comparison with findings for cells treated only with ethanol, the PKA activity for BBECs that were preincubated with rolipram or Ro 20-1724 and subsequently incubated with ethanol for 1 h was significantly greater. No significant change in PKA activity was observed when BBECs were treated with ethanol for 6 or 24 h in the presence or absence of PDE inhibitors. Each data point represents triplicate measurements (n = 9) for that given time point. A P value of ⬍.05 was considered significant. For media versus ethanol treatment, significance is indicated by (*), and for treatment with Ro 20-1724 or rolipram and ethanol versus ethanol alone, significance is indicated by (**). Small bars represent ⫾ S.E.M.

ciliated cells were preincubated with rolipram, Ro 20-1724, or media alone, followed by treatment with media or ethanol. Cells exposed to media during the preincubation and treatment periods maintained baseline beating frequency (Fig. 7). Cells treated with ethanol had an increase in CBF whether or not they were preincubated with PDE4 inhibitors. The presence of PDE4 inhibitors did not extend ethanol-stimulated CBF from 6 to 24 h. These findings support the idea that chronic ethanol desensitization of ciliary beating is not mediated by activation of PDE.

4. Discussion Findings obtained from many recent studies support that moderate alcohol consumption is associated with protective

Fig. 7. Effect of phosphodiesterase type 4 (PDE4) inhibition on ethanol (EtOH)-stimulated ciliary beat frequency (CBF) in ciliated bovine bronchial epithelial cells (BBECs). Confluent monolayers of ciliated BBECs were preincubated for 1 h with Ro 20-1724 (20 µM), rolipram (8 µM), or media and subsequently treated with 100 mM ethanol. The motion of actively beating cilia was quantified and expressed as CBF in Hertz (Hz). The CBF was significantly increased over baseline (media only) values in response to ethanol treatment. The CBF for ciliated BBECs that were preincubated with rolipram or Ro 20-1724 and subsequently incubated with ethanol was significantly greater than that observed in BBECs treated only with media. Each data point represents triplicate measurements (n = 9) for that given time point. A P value of ⬍.05 was considered significant, as indicated by (*). Small vertical bars indicate ⫾ S.E.M.

health effects (Belleville, 2002; Redmond et al., 2000). In contrast, chronic, heavy alcohol consumption is associated with many diseases, including those of the lung (Guidot & Roman, 2002). It has been shown that exposure to a biologically relevant range of ethanol concentrations increases bronchial epithelial cell ciliary motility (Sisson, 1995). This ethanol-stimulated increase is regulated by increased airway epithelial cell PKA activity (Sisson et al., 1999). On the basis of this observation alone, one could infer that alcohol consumption would be beneficial for mucociliary clearance. However, these ethanol-stimulated increases are only transient (⬍6 h), as both CBF and PKA activity return to baseline levels with chronic exposure (ⱖ6 h) to ethanol. After this chronic exposure period, no additional ethanol stimulation of the cilia is observed. Further, the ciliated cells become desensitized to additional stimulation of CBF, even by cilia-stimulatory agents other than ethanol such as betaagonists (Wyatt & Sisson, 2001). Thus, our in vitro model reveals a striking difference between the effects of acute versus those of chronic ethanol exposure on ciliary beating. In the current study, we found that exposure of BBECs to 100 mM ethanol stimulated an increase in cAMP levels. This increase was transient, as cAMP levels returned to baseline values by 6 h. This return to baseline values of ethanol-stimulated increases in cAMP supports the idea of a compensatory mechanism of cyclic nucleotide degradation

M.A. Forge`t et al. / Alcohol 31 (2003) 31–38

whereby cAMP homeostasis occurs. Using direct PDE catalytic–activity assays, we found that ethanol stimulates PDE4 in BBECs. Such selective PDE activation in response to acute ethanol exposure supports a mechanism for the observed reduction and return to baseline cAMP levels. This reduction in cAMP levels temporally precedes the lowering of acute ethanol exposure–stimulated PKA activity and the return of CBF to baseline levels. Therefore, PDE activation represents one mechanism by which the acute effects of ethanol exposure might transition to the observed chronic exposure effects in relation to cAMP. Phosphodiesterases are ubiquitous enzymes that function to maintain the cellular concentration of cyclic nucleotides through nucleotide hydrolysis. Several of the 11 families of PDEs have been isolated in lung epithelial cells (Barnes, 1995; Fuhrmann et al., 1999). Fuhrmann et al. (1999) have identified PDE1, PDE4, PDE5, and PDE7 in human and bovine airway epithelium, with PDE4 being the major cAMP-hydrolyzing enzyme and PDE5 being the major PDE for cGMP. Our findings support this observation because PDE4-specific inhibitors blocked ethanol-stimulated cAMP–PDE activity. Using direct catalytic–activity assays of cGMP–PDE, we have also observed a zaprinast-sensitive PDE5 activity in BBECs (data not shown). Because much of the machinery of cyclic nucleotide–regulated CBF is localized to the ciliary axoneme, including PKA (Kultgen et al., 2002; Sisson et al., 2000) and PKG (Li et al., 2000), it will be important to determine the subcellular localization of these PDEs in ciliated cells. The site of cAMP production may co-localize with the site of cAMP degradation, facilitating the orchestration of the acute versus chronic effects of ethanol exposure on CBF. The 100 mM concentration of ethanol was chosen because this concentration has been shown to increase acutely the epithelial cell PKA activity and it is biologically relevant. However, chronic exposure (ⱖ6 h) of BBECs to 100 mM ethanol results in a desensitization of PKA (Wyatt & Sisson, 2001). In this case, desensitization occurs when PKA no longer responds to ethanol exposure with increased activity as seen during acute exposure. This return to baseline activity is not due to a metabolism of, nor to a decrease in, the ethanol concentration in the media of ethanol-treated cells (Wyatt & Sisson, 2001). This finding supports that additional regulatory mechanisms such as PDE activity account for the return of cAMP to baseline levels after chronic exposure to ethanol. However, such specific cAMP– PDE activity does not account for the desensitization of ciliary beating after chronic exposure to ethanol, as we have observed that cyclic nucleotide analogs resistant to PDE hydrolysis fail to stimulate CBF in those cells exposed chronically to ethanol (Wyatt & Sisson, 2001). This observation indicates that, although cAMP–PDE activity functions as the mechanism to return ethanol-stimulated CBF to baseline levels, the maintenance of the prolonged desensitization response is the result of another, as yet unidentified, effect of chronic exposure to ethanol on the ciliated cells.

37

The relation between ethanol exposure and cAMP–PDE activity has been explored outside the lung. Dibutyryl–cyclic AMP increases alcohol dehydrogenase (ADH) activity and mRNA in cultured primary rat hepatocytes (Potter et al., 1995). In addition, theophylline also increased rat hepatocyte ADH activity in that study. This observation supports that the cAMP levels stimulated by ethanol could degrade ethanol through the stimulation of ADH. However, the ethanol-stimulated activation of cAMP–PDE would counteract such a pathway over time. Findings obtained from other studies reveal that ethanol directly inhibits PDE. Alcohol has been shown to augment beta-agonist stimulation of cAMP production in rat parotid cells through inhibition of PDE (Harper & Brooker, 1980). Phorbol-ester stimulation of both cAMP– PDE and cGMP–PDE activity was shown to be inhibited by 1% ethanol in CD3⫹ peripheral blood monocytes (ZakaroffGirard et al., 1999). At high concentrations (5%–10%), ethanol competitively inhibited the activity of cAMP–PDE in human gastric mucosa (Karppanen et al., 1976). Other reported results have shown that ethanol is associated with elevated PDE levels. Cyclic AMP–PDE activity is elevated in the CNS of developing rats with fetal alcohol syndrome (Suzuki et al., 1981). In summary, our findings reveal, for the first time, the stimulation of cAMP–PDE activity by ethanol in BBECs. This PDE activation results from cell exposure to intoxicating, but physiologically relevant, concentrations of ethanol. The ethanol-mediated activation of cAMP–PDE correlates temporally with the observed decrease in ethanol-stimulated cAMP levels and PKA activity. These findings support our hypothesis that PDE activation is the mechanism responsible for the return of acute ethanol exposure–stimulated CBF to baseline levels in bronchial epithelium. Such a mechanism likely represents an important transition pathway to chronic ethanol desensitization of ciliary beating.

Acknowledgments An NIH Grant # 5 RO1 AA08769-07 to J.H.S. and a Department of Veterans Affairs Merit Review Grant to T.A.W. supported this study. Todd A. Wyatt is an American Lung Association Career Investigator.

References Barnes, P. J. (1995). Cyclic nucleotides and phosphodiesterases and airway function. Eur Respir J 8, 457–462. Beavo, J. A., Conti, M., & Heaslip, R. J. (1994). Multiple cyclic nucleotide phosphodiesterases. Mol Pharmacol 46, 399–405. Belleville, J. (2002). The French paradox: possible involvement of ethanol in the protective effect against cardiovascular diseases. Nutrition 18, 173–177. Francis, S. H., Turko, I. V., & Corbin, J. D. (2001). Cyclic nucleotide phosphodiesterases: relating structure and function. Prog Nucleic Acid Res Mol Biol 65, 1–52.

38

M.A. Forge`t et al. / Alcohol 31 (2003) 31–38

Fuhrmann, M., Jahn, H.-U., Seybold, J., Neurohr, C., Barnes, P. J., Hippenstiel, S., Kraemer, H. J., & Suttorp, N. (1999). Identification and function of cyclic nucleotide phosphodiesterase isoenzymes in airway epithelial cells. Am J Respir Cell Mol Biol 20, 292–302. Guidot, D. M., & Roman, J. (2002). Chronic ethanol ingestion increases susceptibility to acute lung injury: role of oxidative stress and tissue remodeling. Chest 122(6 Suppl), 309S–314S. Harper, J. F., & Brooker, G. (1980). Alcohol potentiation of isoproterenolstimulated cyclic AMP accumulation in rat parotid. J Cyclic Nucleotide Res 6, 51–62. Jain, B., Rubinstein, I., Robbins, R. A., Leise, K. L., & Sisson, J. H. (1993). Modulation of airway epithelial cell ciliary beat frequency by nitric oxide. Biochem Biophys Res Commun 191, 83–88. Jiang, H., Colbran, J. L., Francis, S. H., & Corbin, J. D. (1992). Direct evidence for cross-activation of cGMP-dependent protein kinase by cAMP in pig coronary arteries. J Biol Chem 267, 1015–1019. Karppanen, H., Puurunen, J., Kairaluoma, M., & Larmi, T. (1976). Effects of ethyl alcohol on the adenosine 3′,5′-monophosphate system of the human gastric mucosa. Scand J Gastroenterol 11, 603–607. Kultgen, P. L., Byrd, S. K., Ostrowski, L. E., & Milgram, S. L. (2002). Characterization of an A-kinase anchoring protein in human ciliary axonemes. Mol Biol Cell 13, 4156–4166. Lansley, A. B., Sanderson, M. J., & Dirksen, E. R. (1992). Control of the beat cycle of respiratory tract cilia by Ca2⫹ and cAMP. Am J Physiol 263(2 Pt 1), L232–L242. Lebowitz, M. D. (1981). Respiratory symptoms and disease related to alcohol consumption. Am Rev Resp Dis 123, 16–19. Li, D., Shirakami, G., Zhan, X., & Johns, R. A. (2000). Regulation of ciliary beat frequency by the nitric oxide–cyclic guanosine monophosphate signaling pathway in rat airway epithelial cells. Am J Respir Cell Mol Biol 23, 175–181. Murray, R. P., Istvan, J. A., Daniels, K., & Beaudoin, C. M. (1998). Alcohol and morbidity in the Lung Health Study. J Stud Alcohol 59, 250–257. Potter, J. J., MacDougald, O. A., & Mezey, E. (1995). Regulation of rat alcohol dehydrogenase by cyclic AMP in primary hepatocyte culture. Arch Biochem Biophys 321, 329–335.

Redmond, E. M., Sitzmann, J. V., & Cahill, P. A. (2000). Potential mechanisms for cardiovascular protective effect of ethanol. Acta Pharmacol Sin 21, 385–390. Sisson, J. H. (1995). Ethanol stimulates apparent nitric oxide–dependent ciliary beat frequency in bovine airway epithelial cells. Am J Physiol 268(4 Pt 1), L596–L600. Sisson, J. H., May, K., & Wyatt, T. A. (1999). Nitric oxide–dependent ethanol stimulation of ciliary motility is linked to cAMP-dependent protein kinase (PKA) activation in bovine bronchial epithelium. Alcohol Clin Exp Res 23, 1528–1533. Sisson, J. H., Mommsen, J., Spurzem, J. R., & Wyatt, T. A. (2000). Localization of PKA and PKG in bovine bronchial epithelial cells and axonemes. Am J Respir Crit Care Med 161, 449A. (Abstract). Soderling, S. H., & Beavo, J. A. (2000). Regulation of cAMP and cGMP signaling: new phosphodiesterases and new functions. Curr Opin Cell Biol 12, 174–179. Suzuki, N., Tanaka, H., & Arima, M. (1981). [Activity of 2′,3′-cyclic nucleotide 3′-phosphohydrolase in the central nervous system of rat fetal alcohol syndrome during development (author’s transl)]. No To Shinkei 33, 1207–1212. Tabak, C., Smit, H. A., Rasanen, L., Fidanza, F., Menotti, A., Nissinen, A., Feskens, E. J., Heederik, D., & Kromhout, D. (2001). Alcohol consumption in relation to 20-year COPD mortality and pulmonary function in middle-aged men from three European countries. Epidemiology 12, 239–245. Wyatt, T. A., Naftilan, A. J., Francis, S. H., & Corbin, J. D. (1998a). ANF elicits phosphorylation of the cGMP phosphodiesterase in vascular smooth muscle cells. Am J Physiol 274(2 Pt 2), H448–H455. Wyatt, T. A., & Sisson, J. H. (2001). Chronic ethanol downregulates PKA activation and ciliary beating in bovine bronchial epithelial cells. Am J Physiol Lung Cell Mol Physiol 281, L575–L581. Wyatt, T. A., Spurzem, J. R., May, K., & Sisson, J. H. (1998b). Regulation of ciliary beat frequency by both PKA and PKG in bovine airway epithelial cells. Am J Physiol 275(4 Pt 1), L827–L835. Zakaroff-Girard, A., El Bawab, S., Nemoz, G., Lagarde, M., & Prigent, A. F. (1999). Relationships between phosphatidic acid and cyclic nucleotide phosphodiesterases in activated human blood mononuclear cells. J Leukoc Biol 65, 381–390.