QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.105.088542v1 315/2/729    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Liu, H. Articles by Farley, Sr., J. M. Search for Related Content PubMed PubMed Citation Articles by Liu, H. Articles by Farley, Sr., J. M.

GASTROINTESTINAL, HEPATIC, PULMONARY, AND RENAL

Prostanoids Secreted by Alveolar Macrophages Enhance Ionic Currents in Swine Tracheal Submucosal Gland Cells

Department of Pharmacology and Toxicology, University of Mississippi Medical Center, Jackson, Mississippi

Received April 27, 2005; accepted July 26, 2005.

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
We examined the effect of substances released by swine alveolar macrophages (AMs) on ionic currents in airway submucosal gland cells (SGCs). AMs obtained by lavage were activated by 24-h zymosan exposure (0.1 mg/ml). Supernatant was collected and used to stimulate short-circuit current changes (I_SC) in SGC monolayers in Ussing chambers. Dexamethasone (1 µM) or indomethacin (5 µM) during zymosan exposure of AMs reduced or abolished the supernatant-induced I_SC. Zymosan exposure induced a 5-fold increase in cyclooxygenase (COX)-2 but not COX-1 protein levels in AMs. Prostaglandin E₂ (PGE₂) concentration in the supernatant from zymosan-activated AMs was 550 ± 10 nM (n = 3) compared with 28 ± 3 nM for unstimulated AMs (n = 3). PGE₂, applied serosally, induced I_SC with an EC₅₀ of 15.5 ± 1.3 nM (n = 4) and 3.6 ± 1.8 µM (n = 3) when applied apically. Four types of endoprostanoid receptors (EP_1–4) were detected in SGCs using Western blot. PGE₂-induced I_SC were inhibited by AH6809 (6-isopropoxy-9-oxoxanthene-2-carboxylic acid) but not by SC19220 (8-chloro-dibenzo[b,f][1,4]oxazepine-10(11H)-carboxylic acid, 2-acetylhydrazide), suggesting that endoprostanoid (EP)₂ but not EP₁ receptors were activated by PGE₂. Pretreatment of SGCs with supernatant from zymosan-activated AMs, PGE₂, or forskolin enhanced the sensitivity to acetylcholine (ACh)-induced I_SC. PGE₂-induced I_SC were blocked by charybdotoxin (ChTX), chromanol 293B, or glibenclamide. ACh-induced I_SC were only blocked by ChTX or glibenclamide. None of these blockers altered PGE₂ pretreatment-induced sensitization of ACh-induced I_SC. These results demonstrate that prostanoids released from activated AMs directly increase cystic fibrosis transmembrane conductance regulator and K⁺ channel activity. ACh-induced I_SC are also enhanced due to enhanced activation of Ca²⁺-activated K⁺ channels (K_Ca).

AMs activated by particulates release cytokines, such as interleukin-1, -6, the chemokine macrophage inflammatory protein-1, hematopoietic growth factor, granulocyte-macrophage colony-stimulating factor, and reactive oxygen species (Becker et al., 1996; Dorger and Krombach, 2000; Suwa et al., 2002). Activation of AMs in the lung by particulates has been shown to induce systemic effects via the circulation by increasing leukocytosis in the bone marrow (van Eeden and Hogg, 2002). Media taken from cocultures of macrophages and epithelial cells exposed to particulates and instilled into rabbit lung stimulate bone marrow (Fujii et al., 2002). AMs can also alter the function of respiratory system. For example, AM-released mediators enhance the responsiveness of rat lungs to muscarinic stimulation (Padrid et al., 1993) and particulate matter exposure of naive mice increases airway reactivity (Walters et al., 2001). Little is known however about the effects of AM-derived products in regulating epithelial function and airway fluid and mucus secretion.

We hypothesize that activated AMs release a substance or substances, which directly induce I_SC across SGC monolayers; and the substance(s) secreted modulate ionic current changes induced by other secretagogues and neurotransmitters such as ACh. The latter effect may contribute to hypersecretory changes during airway inflammation. We examined the effect of substances released by zymosan-activated AMs and PGE₂ on SGCs, measured as I_SC across confluent SGC monolayers in Ussing chambers. Our results demonstrate that zymosan induces COX-2 expression and PGE₂ release by AMs. Supernatant from zymosan-activated AMs or PGE₂-induced I_SC mediated via activation of CFTR chloride channels and K⁺ channels as well as enhancing the response to ACh-induced I_SC and K⁺ current in SGCs. These results suggest a role for the AMs activated by particulates in stimulating airway fluid secretion.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Macrophage Culture and Drug Treatment. Male weanling pigs (Yorkshire) from a local vendor weighing 30 kg were sacrificed by exsanguination after isoflurane anesthesia. This method was approved by the local Institutional Animal Care and Use Committee. After exsanguination, macrophages were collected by bronchioalveolar lavage using 300 ml Ca²⁺- and Mg²⁺-free Hanks' balanced salt solution (137 mM NaCl, 4.2 mM NaHCO₃, 0.3 mM Na₂HPO₄, 5.4 mM KCl, 5.5 mM glucose, 0.5 mM EGTA, 100 U/ml penicillin, 100 µg/ml streptomycin, and 25 µg/ml kanamycin, pH 7.4, 4°C). The cells were recovered from the lavage solution by centrifugation, washed in the same lavage solution (100 g x 10 min at 4°C) three times, and then resuspended in DMEM/F-12 medium with 1% heat-inactivated fetal bovine serum and plated at a density of 5 x 10⁵ cells/cm²/ml in six-well culture dishes (4 ml of culture media, 9.6-cm² surface areas). Cells were maintained in a humidified atmosphere containing 5% CO₂ at 37°C for 1 h. Unattached cells were removed by washing with phosphate-buffered saline (PBS) (3x) and culture media (1x). The estimated final cell density was 1.7 x 10⁵ cells/cm²/ml or 1.6 x 10⁶ cells/ml media. Usually at least 1 to approximately 2 x 10⁸ cells were collected by a single lavage. Greater than 98% of the cells were CD14 positive as determined by immunohistochemistry. Macrophages were activated by adding zymosan A in suspension to the culture medium (0.1 mg/ml). Indomethacin and dexamethasone were dissolved in DMSO (1000x stock solution) and added to the culture dish with or without zymosan. After 24 h of incubation, the culture medium was collected and centrifuged at 400g for 6 min at 0°C. The supernatant was frozen immediately and stored at -80°C until use. The osmolarity of macrophage culture media and supernatant was measured using a vapor pressure osmometer (model 552O; Wescor Inc., Logan, UT).

Isolation and Culture of SGCs. Unless specifically noted, isolation and culture of SGCs were conducted according to Chan et al. (1996). After AMs were collected, the trachea was quickly removed and transported to the lab in physiological saline solution containing 140 mM NaCl, 5.5 mM KCl, 1 mM CaCl₂, 5.5 mM glucose, and 10 mM Hepes, pH = 7.4, supplemented with penicillin and streptomycin. The epithelium was stripped off as a single layer and digested in 2 mg/ml protease and 1 mg/ml deoxyribonuclease in 15 ml of physiological saline solution for approximately 60 to 70 min. The isolated cells were then mixed with 5 ml of fetal bovine serum to stop the digestion. The cells were centrifuged through a discontinuous Percoll gradient that consisted of five layers: 10, 20, 30, 40, and 60% Percoll at 500g for 10 min. SGCs, located at the interfaces of the 40 and 60% layers, were collected and washed twice. The SGC pellet was resuspended in BioWhittaker PC-1 medium containing 2 mM Glutamax, serum substitutes, and antibiotics. For Ussing chamber studies, 10⁶ SGCs were plated onto each Millicell-HA insert (12-mm diameter; Millipore Corporation, Billerica, MA) precoated with human placental collagen type IV (20 µg/cm²). Inserts were maintained in PC-1 medium in the cell incubator (37°C, 5% CO₂) for 24 h. The medium inside the inserts was then removed, allowing SGCs to grow at an air interface (Chan et al., 1996). Medium in each culture dish was changed every 48 h, and apical fluid was removed daily. Confluent SGC monolayers formed 3 to 5 days after plating. In some cases, we used 0.1 mg/ml collagenase, 0.1 mg/ml deoxyribonuclease, and 0.5 mg/ml bovine serum albumin in PC-1 media to digest minced fresh epithelium layer for four consecutive 1-h digestion periods at 37°C, with each digestion stopped with fetal bovine serum. Cells were isolated using Percoll gradient as mentioned above and plated in inserts. This method achieved better mucus gland cell survival ratio than that achieved by protease digestion in the final SGC suspension (Yang et al., 1988). Resting transepithelial potentials and resistances for confluent SGC monolayers were measured using a Millicell-ERS voltohmmeter (Millipore Corporation), and currents were calculated according to Ohm's law.

Measurements of I_SC were performed according to previously reported methods from our lab (Chan et al., 1996). Inserts were mounted in Lucite chambers (Corning Life Sciences, Acton, MA). For transmembrane current measurement, normally both sides of chamber were filled with Krebs-Ringer buffer solution (113 mM NaCl, 4.8 mM KCl, 2.5 mM CaCl₂, 18 mM NaHCO₃, 1.2 mM KH₂PO₄, 1.2 mM MgSO₄, 5.5 mM glucose, and 30 mM mannitol, pH adjusted to 7.4) bubbled continuously with 95% O₂/5% CO₂ and circulated by a bubble lift device at 37°C. The transmembrane potential was held at 0 mV by voltage clamp, and transmembrane current changes were measured using VCC-600 amplifiers (Physiologic Instrument, San Diego, CA) and acquired at 50 Hz by a PCI data acquisition card (DAS1602/16; Measurement Computing, Middleboro, MA) using DA-SYLab 6.0 (Dasytec USA, Bedford, NH). I_SC were measured by subtracting baseline I_SC to the peak I_SC response after agonist stimulation. Data were analyzed using Origin 7.0 (OriginLab Corp, Northampton, MA).

Prior to addition of agonist, 10 µM amiloride was added to the apical chamber to inhibit sodium absorption during all experiments. Compounds such as ACh, PGE₂, and forskolin were added to the serosal solution cumulatively from stock solutions (at least 1000 times concentrated). In some experiments, diphenylamine-2-carboxylic acid (DPC; 0.51 mM) or disodium 4,4'-diisothiocyanatostilbene-2,2'-disulfonate (DIDS; 100 µM) was added to the apical side of SGC monolayers to block chloride channel activities.

To measure the serosal membrane K⁺ channel activity, 180 µg/ml nystatin was added to the apical chamber to permeabilize the apical membrane to monovalent ions, such as Cl^-, K⁺, and Na⁺ (Hwang et al., 1996). A high K⁺ gradient across the serosal membrane was established, and Cl^- current was also minimized by replacing NaCl with potassium gluconate in the apical Ussing solution and sodium gluconate in the serosal solution. Therefore, I_SC reflected primarily K⁺ channel activity in the serosal membrane. Ussing chamber solution components for the apical side were 120 mM potassium gluconate (C₆H₁₁O₇K), 25 mM KH₂PO₄, 0.8 mM K₂HPO₄, 1.2 mM MgSO₄, 4 mM CaCl₂, and 10 mM glucose, and the solution for the serosal side was similar to that of apical side except that the potassium gluconate was replaced with equimolar concentration of sodium gluconate (C₆H₁₁NaO₇). Solutions were bicarbonate-free and bubbled continuously with 100% O₂.

Measurement of Prostanoid Concentration. Radioimmunoassay of PGE₂ in the supernatant was performed by Dr. Jay Westcott of National Jewish Medical and Research Center (Denver, CO). DMEM/F-12 medium was used as a blank control, and supernatant from zymosan-activated AMs was diluted before performing the assay.

Protein Purification and Western Blot. To examine the COX-1 and COX-2 protein expression levels, polyclonal antibodies for COX-1 and 2 and their respective secondary antibodies were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). After 24 h of zymosan or vehicle treatment, supernatants from AM culture were collected, and AMs were rinsed once with cold PBS. AMs from six-well dishes treated under the same conditions were combined and lysed in 200 µl of lysis buffer (1x PBS, pH 7.4, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate, 10 µl/ml phenylmethanesulfonyl fluoride, 50 U/ml aprotinin, and 10 µl/ml Na₃VO₄) at 0°C. Cell lysates were centrifuged at 10,000g for 10 min at 4°C to yield whole-cell extracts. Protein concentration in the whole-cell extract was determined using a Coomassie protein assay kit according to manufacturer's instructions (Pierce Chemical, Rockford, IL). Whole-cell extracts (10 µg protein/lane) were then denatured and electrophoresed into a 7.5% SDS-polyacrylamide running gel. Proteins were transferred from the gel to a nitrocellulose membrane (Bio-Rad, Hercules, CA) according to manufacturer's instructions. Membranes were blocked for nonspecific binding by incubating in 5% milk in 1x TBS (10 mM Tris-HCl and 150 mM NaCl, pH 8.0) at 4°C overnight. Membranes were then incubated with primary antibodies diluted with 5% milk in 1x TBST (0.05% Tween 20 and 1x TBS) for 1 h at room temperature. Primary antibody concentrations for COX-2 (sc-1745, goat polyclonal) and COX-1 (sc-7950, rabbit polyclonal) were optimized at 1:2000 and 1:100 dilutions, respectively. After primary antibody incubation, the membranes were washed three times with 1x TBST for 5 min each and then incubated for 45 min at room temperature with horseradish peroxidase-conjugated secondary antibodies. Secondary antibodies concentrations were optimized at 1:5000 for COX-2 (sc-2033, donkey anti-goat IgG) and 1:10,000 for COX-1 (sc-2317, donkey anti-rabbit IgG), respectively. Membranes were washed three times for 5 min each with TBST and once for 5 min with TBS. The protein bands and molecular weight markers were detected using Hyperfilm and ECL plus reagent (GE Healthcare, Little Chalfont, Buckinghamshire, UK). The protein levels and molecular weights were determined using a personal densitometer and the ImageQuant program (GE Healthcare). Band densities were normalized to the signal from AMs not exposed to zymosan and drugs.

For identification of endoprostanoid receptor proteins, polyclonal antibodies against these receptors (EP₁, EP₂, EP₃, and EP₄) were purchased from Cayman Chemical (Ann Arbor, MI). SGCs were collected using discontinuous Percoll gradient (1.5 x 10⁶ 2.5 x 10⁶ cells/sample), and protein was extracted and measured as described above. Whole-cell lysates were precleared by adding 0.25 µg/ml normal rabbit or goat IgG together with 20 µl of Protein A/G-PLUS-Agarose (Santa Cruz Biotechnology, Inc.) and incubating for 30 min at 4°C. Supernatant was collected after centrifugation at 2500 rpm for 5 min. To 1 ml of supernatants, primary antibodies to EP₁, EP₂, EP₃, or EP₄ (2 µg each) were added and incubated for 1 h at 4°C. Protein A/G PLUS-Agarose (20 µl) was then added to each sample and after 1-h incubation, immunoprecipitates were collected by centrifugation at 3000 rpm for 5 min (Mamoon et al., 2004). After washing the pellets four times with PBS, 40 µl of electrophoresis sample buffer was added to each sample and boiled for 90 s. Proteins were separated, transferred to nitrocellulose membranes, and detected as described above.

Materials. Protein purification reagents were purchased from Santa Cruz Biotechnology, Inc. Fetal bovine serum was purchased from Hyclone Laboratories (Logan, UT). PC-1 medium containing 2 mM Glutamax and serum substitutes was purchased from Cambrex Bio Science Walkersville (Walkersville, MD). AH6809 and SC19220 were purchased from Cayman Chemical. Chromanol 293B was purchased from Tocris Cookson Inc. (Ellisville, MO). Amiloride, ChTX, collagenase I, collagen type IV, dexamethasone, DIDS, DMEM, DMSO, DNase, DPC, glibenclamide, indomethacin, PGD₂, PGE₂, PGF₂, Percoll, protease, Zymosan A, and other reagents were purchased from Sigma-Aldrich (St. Louis, MO). DMSO was used to dissolve water-insoluble reagents used in Ussing chamber studies and stored at -20°C. These stocks were further diluted in Krebs-Ringers buffer solution immediately before use. The final dilution of DMSO in the macrophage culture or Ussing chamber solution was equal to or less than 0.1% (v/v). Charybdotoxin was dissolved in water as a 100 µM stock solution. DPC (Sigma-Aldrich) was dissolved in with 0.1 N NaOH as a 1 M stock solution.

Data Analysis. All data are expressed as mean ± S.E.M., and n values reported are the number of animals used for each experiment. To examine the relative sensitivities of SGCs to agonists under different experiment conditions, EC₅₀ values were calculated from individual cumulative concentration-response data using sigmoid fit functions of Origin 7.0 (OriginLab Corp) and then averaged. Data from multiple treatment groups were analyzed by one-way ANOVA or one-way repeated measures ANOVA followed by the Student-Newman-Keuls pair-wise test for multiple comparisons, whenever appropriate (Sigma Stat software; SPSS Inc., Chicago, IL). In some cases, ANOVA on ranks or repeated measures ANOVA on ranks were used instead when the variances of the data were not equal among all treatment groups. Student's t test was used to compare the data from two groups with either paired or unpaired tests as appropriate. A value of p < 0.05 was regarded as significant.

	Results

Top Abstract Materials and Methods Results Discussion References

 
The resting transepithelial potentials, currents, and resistances for confluent SGC monolayers were 5.1 ± 0.5 mV, 5.0 ± 0.7 µA/cm², and 1.8 ± 0.2 K/cm², respectively, at approximately 3 to 5 days after primary culture (n = 11). At the beginning of all Ussing chamber experiments, 10 µM amiloride was added apically to block sodium absorption. This caused an average 1.0 ± 0.1 µA/cm² decrease in I_SC (n = 11). Supernatant taken from 24-h zymosan-exposed AMs had average osmolarities of 320 mOsm/l and pH values of 7.2, close to values of fresh DMEM (320 and 7.4, respectively). The addition of DMEM to the Ussing chambers at up to 10% (v/v) had little effect on the total osmolarity or pH value of the Krebs-Rings buffer solution (310 mOsm/l and pH 7.4, respectively) and did not change I_SC of SGC monolayers.

Effect of Supernatant from Zymosan-Activated AMs in Inducing I_SC. To examine the direct effect of substances released by activated AMs in increasing I_SC, supernatant from 24-h zymosan-exposed AMs (M + zymosan supernatant) was applied cumulatively to the serosal side of the SGC monolayers. The supernatant-induced I_SC is illustrated in Fig. 1A. A small increase in I_SC was induced at 0.3% dilution (I_SC = 1.1 ± 0.3 µA/cm²). I_SC increased with increasing volume added. The I_SC induced by 1 and 10% dilution were 3.1 ± 0.7 and 9.0 ± 0.8 µA/cm², respectively (n = 5). The increase in I_SC was persistent, reaching a plateau after about 5 min of exposure to M + zymosan supernatant. The transmembrane resistance also decreased as shown by the increase in amplitude of the currents induced by voltage pulses (2 mV every 30 s) applied to the epithelium. The I_SC increased by M + zymosan supernatant was abolished by 1 mM DPC but not by 100 µM DIDS applied apically (n = 3). The concentration-response relationship of M + zymosan supernatant-induced I_SC is shown in Fig. 1D (filled squares, n = 5). A maximal response was not reached at the dilutions used (up to 10%) in these experiments. Apical addition of M + zymosan supernatant (up to 10%) did not induce significant increases in I_SC (n = 3). Supernatant applied serosally from 1-, 2-, 3-, 6-, and 12-h zymosan-treated AMs did not cause significant increases in I_SC (data not shown). Supernatant removed from AMs incubated for 24 h without zymosan exposure (M-only supernatant) applied to the serosal side of the SGC epithelium induced I_SC of 1.4 ± 0.8 µA/cm² at 10% dilution (Fig. 4A, middle trace marked with M-only, n = 3), similar in magnitude to that caused by 0.3% M + zymosan supernatant (I_SC = 1.1 ± 0.3 µA/cm², n = 5, p > 0.05).

View larger version (36K): [in this window] [in a new window]   Fig. 1. Supernatant from zymosan-activated AMs increased ISC across the confluent SGC monolayers in the Ussing chamber. A, ISC was recorded under voltage-clamp at 0-mV potential difference across SGC monolayers. Pulses (2 mV; every 30 s) were used to monitor conductance changes. Supernatant from zymosan-activated macrophage or control macrophage (shown in Fig. 4A, trace, M-only) was added cumulatively to the serosal side of monolayers to give final dilutions marked by arrows (from 0.1–10%). The difference of ISC from plateau to baseline at each dilution was measured as the response to supernatant normalized to the area of the SGC monolayers (0.6 cm2) as ISC (microamperes per centimeter squared). DIDS or DPC was added to the apical side to inhibit Cl- channels marked at arrows. B, indomethacin (5 µM) was added to macrophage cultures during 24-h zymosan exposure. Indomethacin treatment abolished the increase in ISC induced by supernatant (trace 3, M + Indo + zymosan supernatant, n = 5). Residual indomethacin (50 nM) in the Ussing chamber originating from supernatant did not affecting ISC (trace 2, M + Indo Supernatant, n = 5). C, PGE2 or PGF2 was applied serosally to the SGC monolayers from 3 x 10-11 to 10-5 M concentrations (cumulative), which induced persistent increases in ISC. D, summary of concentration-response data for M + zymosan supernatant (filled squares, n = 5), PGE2 (open squares, n = 4), and PGF2 (open circles, n = 3) in inducing ISC from A and C. Data shown as mean ± S.E.M., and n is the number of animals used. The dotted line shows that 100% M + zymosan supernatant has an estimated equivalent potency comparable with 10-7 M PGE2.

View larger version (44K): [in this window] [in a new window]   Fig. 4. Pretreatment of SGC monolayers with supernatant from zymosan-activated AMs, PGE2, and forskolin-sensitized ACh-induced ISC. A, dilutions (0.1–10%) of supernatant from AMs without zymosan exposure (trace, M-only) or from AMs with zymosan exposure (trace, M + Zymosan) were applied cumulatively to the serosal side of SGCs followed by cumulative application of ACh (marked at arrows). ACh-induced increases in ISC in SGCs not treated with supernatants (trace, control) were used as control. B, concentration-response relationships for ACh-induced ISC in SGCs pretreated with M + zymosan supernatant (upward triangles, n = 3) and M-only supernatant (squares, n = 3) and in control SGCs (circles, n = 6). C, in the bottom trace (Control), DMSO vehicle applied before ACh treatment had no effect on the responsiveness to ACh. The top two current recordings show that, 5 min prior to ACh addition, 10-7 M PGE2 or 5 x 10-6 M forskolin caused stable increases in ISC. Notice that PGE2 or forskolin pretreatment sensitized SGC monolayers to low ACh concentrations (10-8 to approximately 3 x 10-7 M) to increase ISC. D, effect of forskolin (5 x 10-5 M, squares, n = 4) or PGE2 (10-7 M, upward triangles, n = 10) pretreatment on the concentration responses for ACh-induced ISC (normalized to the maximal ACh-induced ISC response in Control SGC monolayers, circles, n = 6). Data are expressed as mean ± S.E.M., and the number of animals used in each group is indicated in the figure.

Effects of Indomethacin and Dexamethasone in Inhibiting the Production of Active Substance by Zymosan-Activated AMs. There are numerous substances released from macrophages on activation, the mix changing with the stimulus used and the length of time after activation (Lohmann-Matthes et al., 1994). The active substances produced that elicited a change in I_SC were not observed in supernatant until more than 12 h after zymosan exposure, suggesting that release of preformed mediators was not involved. In addition, the substances produced must be fairly stable to accumulate in culture medium for 24 h at 37°C without being totally degraded. Therefore, since it has been shown that zymosan can induce the expression of cyclooxygenase (Vicente et al., 2001) and that some prostanoids, such as PGE₂ and PGF₂ are quite stable in culture medium, we examined the possibility that the active products were prostaglandins.

As shown in Fig. 1B, indomethacin (5 µM), a nonselective COX blocker, was added to the AM culture during 24-h zymosan exposure. Dilution (1%) of supernatant from AMs exposed to both indomethacin and zymosan did not increase I_SC (trace 3, M + Indo + zymosan supernatant), I_SC = 0 ± 0 µA/cm², whereas 1% M + zymosan supernatant induced an increase in I_SC (trace 1, M + zymosan supernatant), I_SC = 2.7 ± 0.3 µA/cm² (n = 5 each, p < 0.05). Supernatant from AMs exposed to indomethacin alone had no effect on I_SC (trace 2, M + Indo supernatant, 0 ± 0 µA/cm², n = 5). Residual indomethacin in the supernatant had no direct effect on the SGCs since application of equivalent dilution of indomethacin (50 nM, n = 5) directly to the Ussing chamber did not change the response of SGCs to M + zymosan supernatant (data not shown).

We also treated AM culture with 1 µM dexamethasone, a known suppressor of macrophage function (Becker and Grasso, 1985). Supernatant from AMs treated with dexamethasone (1%) during 24-h zymosan exposure induced significantly less I_SC compared with I_SC induced by 1% M + zymosan supernatant. Dexamethasone treatment caused a 70.5 ± 2.4% inhibition of I_SC compared with M + zymosan supernatant (n = 3, p < 0.05). There was no direct action of the residual dexamethasone in the supernatant on the SGCs since application of equivalent dilution of dexamethasone (10 nM, n = 3) directly to the Ussing chamber had no effect on the SGC response to M + zymosan supernatant (data not shown). DMSO vehicle (0.1% dilution) in macrophage culture did not influence the supernatant-induced I_SC.

The Comparison of Prostaglandins and Supernatant in Inducing I_SC. The ability of prostaglandins to increase I_SC of SGC was examined In Fig. 1C. PGE₂ or PGF₂ applied cumulatively to the serosal side of SGC monolayers increased I_SC. I_SC rises to a stable plateau within 5 min after the agonist application, similar to the response to M + zymosan supernatant. PGE₂-induced I_SC reach a maximum at approximately 10^-6 M in the case of PGE₂ (Fig. 1D, open squares, maximal I_SC = 15.7 ± 1.4 µA/cm², n = 4). The EC₅₀ for serosal PGE₂-induced I_SC is 10.0 ± 1.8 nM (n = 4). Serosal PGF₂ was less potent than PGE₂ in inducing I_SC, with an estimated EC₅₀ of 1.1 ± 0.1 µM (Fig. 1D, I_SC = 6.4 ± 0.7 µA/cm² at 10^-5 M, open circles, n = 3). We also applied another prostaglandin, PGD₂, to the serosal side of SGC monolayer, which induced I_SC with an EC₅₀ of 19 nM (one experiment, data not shown). As also shown in the Fig. 1D, 100% M + zymosan supernatant is estimated to induce I_SC comparable with that induced by 10^-7 M PGE₂.

PGE₂ Concentration in the Supernatant of Zymosan-Activated AMs. PGE₂ levels in M-only supernatant measured using radioimmunoassay was 10 ± 1 ng/ml (28 ± 3 nM, n = 3). The PGE₂ concentration reached 195 ± 28 ng/ml (550 ± 79 nM, n = 3) in M + zymosan supernatant, an approximate 20-fold increase in PGE₂ release into the culture medium compared with AMs not exposed to zymosan (p < 0.05). PGE₂ was not detectable in fresh culture media.

Cyclooxygenase Expression and Zymosan Exposure of AMs. The delay in production of active substance in inducing I_SC is consistent with increased expression of a protein. We examined the effects of zymosan exposure on COX-1 and COX-2 expression levels in AMs. As shown in Fig. 2, Western blots for both COX-1 and COX-2 yielded single bands with estimated molecular masses of approximately 79 to 90 kDa. COX-1 levels, as estimated using Western blot (Fig. 2, A and inset), were not affected by zymosan exposure for 24 h (98 ± 13% of basal level, n = 3). Indomethacin (5 µM) or dexamethasone (1 µM) present during zymosan exposure did not alter COX-1 expression levels (Fig. 2A, bar graph, 153 ± 17% and 122 ± 29% of basal level, respectively, n = 3 each). COX-1 expression levels were not significantly different among all treatment groups (p > 0.05).

View larger version (17K): [in this window] [in a new window]   Fig. 2. Western blot analysis of COX-1 and COX-2 expression in AMs exposed to zymosan, dexamethasone, and indomethacin. A, effect of dexamethasone (Dex; 1 µM) or indomethacin (Indo; 5 µM) applied during exposure of AMs to zymosan treatment on COX-1 expression levels. The expression levels were normalized to the basal COX-1 level in AMs not treated with zymosan or drugs as shown in lane 1 (n = 3 for each treatment). B, effect of Dex (1 µM) or Indo (5 µM) applied with or without zymosan treatment on COX-2 expression levels in AMs. The COX-2 expression levels were normalized to the basal COX-2 level in AMs not treated with zymosan or drugs as shown in lane 1 (n = 3 for each treatment). *, significantly different from paired macrophage groups not treated with zymosan (empty bars, p < 0.05). , significantly different from zymosan-treated AMs (first filled bar, p < 0.05). Data shown as mean ± S.E.M., and n is the number of animals used.

Serosal and Apical PGE₂ Application in Inducing I_SC. Since AMs are resident on the luminal surface of the airway, would substances, specifically PGE₂, produced luminally cause an effect? As shown in Fig. 3A, trace 2, PGE₂ applied to the apical side of the SGC monolayers increased I_SC with similar characteristics to those observed upon serosal addition of PGE₂ or supernatant. The EC₅₀ for apical PGE₂ was 3.6 ± 1.8 µM (n = 3, Fig. 3B, circles), significantly higher than that for serosal PGE₂ application in parallel experiments (EC₅₀ = 15.5 ± 1.3 nM, n = 4, squares, p < 0.05 using Student's t test).

View larger version (36K): [in this window] [in a new window]   Fig. 3. The difference in the serosal and apical PGE2 sensitivities and the role of EP1 and EP2 receptors in PGE2-induced ISC in SGC monolayers. A, PGE2 was applied cumulatively from 3 x 10-11 to 10-5 M to the serosal side (trace 1) or apical side (trace 2) of SGC monolayers in the Ussing chamber. B, concentration-response relationships (normalized to maximal serosal response) for PGE2-induced ISC are shown. The estimated EC50 for serosal PGE2 was 15.5 ± 1.3 nM (squares, n = 4) and 3.6 ± 1.8 µM (circles, n = 3) for apical PGE2. C, inhibition of PGE2-induced ISC by EP receptor antagonists. PGE2 applied cumulatively from 3 x 10-9 to 10-5 M concentrations to the serosal side of SGC monolayers induced similar response (control trace) to those shown in Figs. 1C and 3A. In parallel inserts, SC19220 (10 µM) and AH6809 (30 µM) were applied serosally prior to PGE2 application. D, PGE2-induced ISC and its inhibition by SC19220 and AH6809. Control SGC has an EC50 of 167 ± 34 nM (circles, n = 4) for PGE2-induced ISC, SC19220-treated SGCs had an EC50 of 103 ± 33 nM (squares, n = 3), and AH6809-treated SGCs had an EC50 of 666 ± 162 nM (upward triangles, n = 3). E, immunoprecipitation and Western blot indicated the presence of all four subtypes of endoprostanoid receptors in SGCs dissociated with either protease or collagenase (EP1–4, approximately three to five animals were used in each group), with estimated molecular masses being 42, 52, 53, and 65 kDa for EP1, EP2, EP3, and EP4 subtypes, respectively. Data points are shown as mean ± S.E.M., and n is the number of animals used in the experiment.

PGE₂ is less potent in inducing I_SC in collagenase-dissociated SGCs than protease-dissociated SGCs (167 ± 34 nM, n = 4 versus 10.0 ± 1.8 nM, n = 4 in Fig. 1D or 15.5 ± 1.3 nM in Fig. 3B, n = 4; p < 0.05). PGE₂ had similar EC₅₀ values in two separate protease-dissociated SGC groups shown in Figs. 1D and 3B (p > 0.05).

Receptors Responsible for PGE₂-Induced I_SC. We examined the types of EP receptor involved in the PGE₂ actions. As shown in Fig. 3C, control trace, PGE₂ increased I_SC with an estimated EC₅₀ of 167 ± 34 nM (Fig. 3D, circles, n = 4). When 10 µM SC19220, an EP₁ antagonist, or 30 µM AH6809, an EP₁ and EP₂ antagonist, was applied to the serosal side of SGC monolayers, AH6809 caused right shift of the concentration-response relationship of PGE₂-induced I_SC (Fig. 3D, upward triangles, EC₅₀ = 666 ± 162 nM, n = 3), but SC19220 did not (Fig. 3D, squares, EC₅₀ = 103 ± 33 nM, n = 3). EC₅₀ for AH6809-treated group was significantly different from the control and SC19220-treated groups (p < 0.05).

Figure 3E shows that all four EP receptor subtypes (EP_1–4) were detected using immunoprecipitation and Western blot methods in SGCs isolated with protease or collagenase -dissociated fresh SGCs. The estimated molecular masses were approximately 42, 52, 53, and 65 kDa, respectively (n = 3–5).

Effect of Supernatant, PGE₂, and Forskolin on ACh-Induced I_SC. We tested the ability of supernatant to modulate ACh-induced I_SC. To estimate the concentration-response relationship for ACh-induced I_SC, we applied ACh cumulatively to the serosal side of SGC monolayers. As shown in Fig. 4A, ACh transiently increased I_SC followed by a slow decay. Changes in I_SC were observed from 10^-7 to a maximum at 10^-5 M ACh (trace marked by control). The estimated EC₅₀ for ACh-induced I_SC in control SGCs was 438 ± 64 nM (Fig. 4B, circles, n = 6). In parallel inserts, prior to ACh treatment, SGC monolayers were treated serosally with M-only supernatant or M + zymosan supernatant as described previously (traces marked by M-only and M + zymosan, respectively). As mentioned before, M-only supernatant at up to 10% (v/v) only induced a small increase in I_SC, whereas M + zymosan supernatant induced a stable increase in I_SC that is similar to that shown in Fig. 1A. Subsequent cumulative application of ACh to the same SGC monolayers induced increases in I_SC similar in characteristics to those in the control trace. ACh-induced I_SC had an EC₅₀ of 335 ± 67 nM for SGCs pretreated with M-only supernatant (Fig. 4B, squares, n = 3), a value not significantly different from the EC₅₀ in control SGCs (p > 0.05). However, the EC₅₀ for ACh-induced I_SC in SGCs treated with M + zymosan supernatant was significantly reduced to 149 ± 40 nM (Fig. 4B, upward triangles, n = 3, p < 0.05 compared with EC₅₀ values in control SGCs or M-only supernatant-treated SGCs). In addition, the pretreatments with M + zymosan supernatant and M-only supernatant significantly increased the maximal increase in ACh-induced I_SC in SGCs, with average values being 140 ± 6% and 122 ± 4% of the ACh-induced maximal I_SC in control SGCs run in parallel (p < 0.05).

We further examined the effect of PGE₂ or forskolin pretreatment on ACh-induced I_SC. SGC monolayers were pretreated with PGE₂ or forskolin (5 x 10^-6 M, to elevating cytosolic cAMP level) serosally for 5 min before ACh application. Both PGE₂ and forskolin induced persistent increases in I_SC in 3 to 5 min (Fig. 4C, middle and top traces, respectively). Pretreatment of SGC monolayers with 10^-7 M PGE₂ or 5 x 10^-6 M forskolin-sensitized SGCs to ACh, ACh-induced increases in I_SC were now observed from 10^-8 M to the maximum at 10^-6 M. This is compared with range of 10^-7 to 10^-5 M ACh concentrations in SGCs not pretreated with PGE₂ or forskolin (control trace, Fig. 4C). PGE₂ (10^-7 M) and 5 x 10^-6 M forskolin pretreatment shifted the EC₅₀ for ACh response to 121 ± 20 nM (n = 10) and 84 ± 19 nM (n = 4) (Fig. 4D, top triangles and squares, respectively), whereas the EC₅₀ for ACh-induced I_SC in controls was 418 ± 28 nM (Fig. 4D, circles, n = 12). The shift in the sensitivity to ACh depended on the concentration of PGE₂. An increase in sensitivity to ACh could be detected at 10^-8 M PGE₂ and reached maximum at 10^-6 M PGE₂. PGE₂ pretreatment shifted the EC₅₀ for ACh-induced I_SC to 292 ± 37 (n = 3), 188 ± 35 (n = 3), and 103 ± 35 (n = 3) nM at 10^-9,10^-8, and 10^-6 M PGE₂ concentrations, respectively. The EC₅₀ values for PGE₂-(from 10^-8 M and up) and forskolin-pretreated groups were significantly different from control (p < 0.01). Pretreatment with 10^-7, 10^-6 PGE₂, or 5 x 10^-6 M forskolin had similar effects in sensitizing ACh-induced I_SC, and the EC₅₀ values for these three groups were not significantly different from each other (p > 0.05).

PGF₂ (10^-5 M) was also applied serosally before ACh application. The EC₅₀ was shifted for the ACh-induced I_SC to 87 ± 8 nM (n = 3), significantly different from that in controls (418 ± 28 nM, n = 12, p < 0.05).

The ACh-induced maximal I_SC in groups treated with 10^-8,10^-7, and 10^-6 M PGE₂ were significantly greater than control monolayers tested in parallel, the average increases being 143 ± 6%, 135 ± 5%, and 131 ± 7% of ACh-induced maximal I_SC in control (p < 0.05), respectively. ACh-induced maximal I_SC in SGCs treated with 10^-9 M PGE₂ or 5 x 10^-6 M forskolin were not significantly different from those of control, the average values being 104 ± 9% and 107 ± 8% of control, respectively (p > 0.05).

K⁺ and Cl^- Channels Involved in PGE₂- or ACh-Induced I_SC. ChTX, a K_Ca blocker, and chromanol 293B, a blocker of K_VLQT1(KCNQ1)/KCNE3 K⁺ channels (Lohrmann et al., 1995), were used to examine the involvement of these K⁺ channels in the PGE₂- and ACh-induced increases in I_SC. Previous reports showed that the K_VLQT1 (KCNQ1) are present in human bronchial epithelial cells and the Calu-3 serous cell line. K_VLQT1 are involved in cAMP-mediated Cl^- secretion (Mall et al., 2000; Cowley and Linsdell, 2002). Chromanol 293B, at 200 µM (supramaximal concentration), was applied serosally to SGC monolayers after 10^-7 M PGE₂ but before ACh applications, which caused an average 63 ± 3% decrease in PGE₂-induced I_SC (Fig. 5A, trace 3, n = 4). However, chromanol 293B did not change the amplitude or potency of ACh-induced increase in I_SC. ACh-induced maximal I_SC in PGE₂ + 293B-treated group was 108 ± 13% of PGE₂ treatment group (Fig. 5B, p > 0.05). Chromanol 293B treatment did not alter the PGE₂-induced reduction in EC₅₀ of ACh-induced I_SC (Fig. 5B, circles, 123 ± 36 nM for PGE₂ treatment group versus 191 ± 82 nM for PGE₂ + 293B treatment group, top triangles, n = 4 each, p > 0.05).

View larger version (29K): [in this window] [in a new window]   Fig. 5. PGE2- and ACh-induced ISC were blocked by K+ and Cl- channel blockers. A, trace 1, control is a similar current recording to Fig. 4C, PGE2 (10-7 M) was applied before ACh treatment. Also, in parallel inserts, ChTX (100 nM, trace 2, n = 3) or chromanol 293B (200 nM, trace 3, n = 4) was applied serosally after PGE2 treatment but before ACh treatment. B, concentration-response relationships for ACh-induced ISC in SGCs pretreated with PGE2 alone (circles, n = 6), PGE2 + ChTX (squares, n = 4), or PGE2 + 293B (upward triangles, n = 4). C, similarly, 2 mM glibenclamide was applied apically after PGE2 treatment, but before ACh treatment (n = 3). D, an enlarged view (greater than 5x) of partial current recording in C, which shows ACh-induced increase in ISC after PGE2 and glibenclamide treatment. The straight line represents the estimated baseline. n represents the numbers of animals used in each group.

ChTX (100 nM) was added serosally after 10^-7 PGE₂ application and blocked 24 ± 4% (n = 4) of 10^-7 M PGE₂-induced I_SC. ChTX also blocked 66 ± 1% of ACh-induced maximal I_SC compared with that in SGCs not treated with ChTX (Fig. 5, A, trace 2, and B, squares, n = 4). EC₅₀ values for ACh-induced I_SC were 171 ± 37 and 109 ± 26 nM for SGC monolayers treated with and without 100 nM ChTX, respectively (n = 4 each, p > 0.05). All these SGC monolayers were pretreated with 10^-7 M PGE₂. The effect of K⁺ channel inhibition on the ACh concentration-response relationships for PGE₂-pretreated SGCs is shown in Fig. 5B. In SGC monolayers not treated with PGE₂, ChTX (100 nM) added before ACh actions also blocked 60 ± 7% of ACh-induced maximal I_SC (n = 4).

To test the possibility that PGE₂ activates CFTR channels and facilitates the ACh response, 2 mM (maximum concentration) glibenclamide was applied apically to the SGC monolayers to block CFTR channel activity. Glibenclamide, a relatively nonspecific CFTR channel blocker, has been used to study the role of CFTR in the epithelial cells (Krouse et al., 2004). Glibenclamide abolished 10^-7 M PGE₂-induced increase in I_SC to below baseline levels (Fig. 5C, n = 3). After PGE₂ and glibenclamide treatment, ACh-induced increases in I_SC were also inhibited (Fig. 5D, a partial enlarged view of 5C, averaged maximal I_SC was 1 µA/cm², n = 3). The EC₅₀ of ACh-induced increases in I_SC after PGE₂ and glibenclamide treatment was 182 ± 28 nM (n = 3), which was not significantly different from control (123 ± 36 nM, n = 4, p > 0.05). These data were consistent with results that supernatant- and PGE₂-induced increases in I_SC were abolished by apically applied 1 mM DPC but less affected by 100 µM DIDS (Fig. 1A).

Effect of PGE₂ on the ACh-Induced Serosal K⁺ Current. To study the modulation of ACh-induced increases in serosal K⁺ current by PGE₂, we used 180 µg/ml nystatin to permeabilize the apical membrane of the SGC monolayers to small monovalent ions, replaced Cl^- in the solution with gluconate ion, and established a high apical to serosal K⁺ ion concentration gradient as described under Materials and Methods. About 10 min after addition of nystatin, basal I_SC increased and then stabilized an indication of successful permeabilization of the apical membrane (Fig. 6A). ACh addition increased serosal K⁺ current with an EC₅₀ of 720 ± 111 nM (Fig. 6B, circles, n = 3). Pretreating permeabilized SGC monolayers with 10^-7 M PGE₂ prior to ACh application increased I_SC (3.6 ± 1.1 µA/cm², n = 3) and significantly sensitized ACh-induced increases in K⁺ current (EC₅₀ = 185 ± 52 nM, Fig. 6B, upward triangles, p < 0.05, n = 3). No enhancement of ACh-induced maximal K⁺ current by PGE₂ pretreatment occurred, and ACh-induced maximal K⁺ current was 82 ± 5% of control (p > 0.05, n = 3).

View larger version (15K): [in this window] [in a new window]   Fig. 6. PGE2-sensitized ACh-induced serosal K+ current in nystatin-permeabilized SGC monolayers. A, recording of K+ current in response to PGE2 and ACh. The apical membrane of SGC monolayers was permeabilized with 180 µg/ml nystatin (at first arrow), causing a stable increase in current. NaCl was replaced with potassium gluconate and sodium gluconate in the apical and serosal solutions, respectively. PGE2 (10-7 M) was added serosally 5 min before ACh treatment (gray trace). The black trace is a control recording without PGE2 pretreatment. Cumulative concentration-response relationships for ACh-induced K+ current were then generated. B, average concentration-response relationships for ACh-induced K+ current (calculated as percentage of ACh-induced maximal response in control). Data are expressed as mean ± S.E.M., and the number of animals used in each group is indicated in the figure.

	Discussion

Top Abstract Materials and Methods Results Discussion References

Unopsonized zymosan activates naive macrophages via the mannose receptor and Toll-like receptors to initiate innate immune responses (Takeuchi and Akira, 2001), leading to phagocytosis, arachidonic acid release, and COX-catalyzed prostanoid productions (Girotti et al., 2004), as well as cytokine production (Lohmann-Matthes et al., 1994). In this study, it took more than 12 h for sufficient amount of active substances to increase I_SC to appear in the supernatant, suggesting that protein synthesis was required. COX-2 expression was increased at 6 to 24 h after zymosan exposure, an effect blunted by glucocorticoid exposure. The ability of zymosan-activated AMs supernatant to increase I_SC was blocked by indomethacin and reduced by dexamethasone, suggesting that the active components of supernatant were COX-related prostanoids. PGE₂ accumulated in the supernatant to a concentration of 5 x 10^-7 M after 1.6 x 10⁶/ml AMs were exposed to 0.1 mg/ml zymosan for 24 h. At this concentration, PGE₂ induces significant I_SC when applied serosally (Fig. 1, C and D). About 2 x 10⁸ AMs were recovered by a single lavage of the lungs, we estimated that if this number of AMs were to be activated in vivo to the same degree as in vitro, 5 x 10^-7 M PGE₂ could be reached in >100 ml of airway fluid.

Products released by AMs in the lumen are secreted onto the apical surface of the epithelium. In Calu-3 cells, a cell line with both serous and mucus gland cell properties (Shen et al., 1994), isoprostane 8-iso-prostaglandin E₂ induced I_SC with apical EC₅₀ higher than serosal EC₅₀ (Cowley, 2003). In this study, PGE₂ applied apically significantly induced I_SC at 10^-7 M, but its EC₅₀ was significantly higher than that for serosal application (3.6 µM versus 15.5 nM, Fig. 3, A and B). Supernatant applied apically (10% dilution) did not induce significant I_SC, but applying undiluted supernatant would induce significant I_SC, comparable with 10^-7 M PGE₂ (Fig. 1D). Although PGE₂ is poorly metabolized in cell culture, it has a half-life of less than 30 s in the circulation (Camus and Jeannin, 1984); thus, PGE₂ is an autocrine or paracrine hormone acting locally to stimulate airway secretion when released into the surface airway liquid. In vivo, products secreted by activated AMs cross the epithelium to induce both local and systemic actions. Lung instillation of products from cocultures of epithelial cells and macrophages exposed to PM10 causes bone marrow stimulation in rabbits (Fujii et al., 2002). Inhaling particulates induces airway hyperreactivity (Walters et al., 2001) and has systemic inflammatory effects (van Eeden and Hogg, 2002). Kitano et al. (1992) showed that intact airway constricted and secreted mucus after intraluminal application of methacholine, although the EC₅₀ was significantly higher than that for serosal application. Relatively lipophilic compounds in supernatant from activated AMs, such as PGE₂, should cross the epithelium to affect underlying tissues by activating serosal receptors, although we cannot rule out specific receptors in the apical cell surface.

PGE₂ binds to four EP receptor types (EP_1–4). All are present in SGCs (Fig. 3E). EP_1,3 receptors are linked to phospholipase C, and EP_2,4 receptors are linked to adenylyl cyclase (Breyer et al., 2001). In SGCs, supernatant-, PGE₂-, or forskolin-induced stable I_SC similar to cAMP-elevating agent-induced I_SC in Calu-3 cells (Cowley, 2003) by activation of CFTR since both supernatant- and PGE₂-induced I_SC were blocked by apically applied DPC or glibenclamide but less by DIDS. In addition, a cAMP-activated K⁺ channel (K_VLQT1) is also involved since PGE₂-induced I_SC were reduced by serosally applied chromanol 293B. The preferential inhibition by the EP₁₊₂ receptor antagonist AH6809, but not by the EP₁ receptor antagonist SC19220, suggests that the EP₂ receptor induces I_SC to PGE₂.

Unlike PGE₂, ACh induced transient I_SC followed by plateaus. ACh belongs to Ca²⁺-elevating neurotransmitters released by parasympathetic nerves (Coulson and Fryer, 2003). ACh activates M₃ receptors to increase I_SC (Liu and Farley, 2005) and to initiate fluid secretion via SGCs (Yang et al., 1988; Ishihara et al., 1992). The increase in I_SC is most likely brought about through production of inositol 1,4,5-trisphosphate and Ca²⁺ release from internal stores. Ca²⁺ not only activates K_Ca, which induces membrane hyperpolarization and net flux of into the lumen (Ballard and Inglis, 2004) but also stimulates mucus secretion (Ishihara et al., 1992). ACh-induced I_SC were significantly blunted by ChTX, suggesting the ACh response depended on the K_Ca.An intermediate conductance K_Ca has been found in Calu-3 cells (Cowley and Linsdell, 2002). The cAMP-activated K_VLQT1 K⁺ channels are not likely to be involved in the ACh-induced I_SC since the latter was not affected by chromanol 293B.

In SGCs, there are two cell types with characteristics consistent with serous and mucus cells. Serous cells and the Calu-3 cells respond to cAMP-elevating agents with increases in both CFTR and K_VLQT1 K⁺ currents, the net effect being exit through CFTR. In Calu-3 cells, muscarinic activation also induces membrane hyperpolarization and net flux of (Ballard and Inglis, 2004). Mucus gland cells have fewer CFTRs, and cAMP-elevating agents do not induce significant ionic current (Tamada et al., 2000). Our previous data showed that SGCs isolated using discontinuous Percoll gradients (Yang et al., 1991; Chan et al., 1996) consist of about 70% mucus cells and 30% serous cells using periodic acid-Schiff and Alcian Blue staining methods. Thus, I_SC in Ussing chamber measurements are combination of ionic currents mediated by serous and mucus gland cells. Farley et al. (1991) reported that the magnitude of ACh-induced I_SC was independent of isoproterenol-induced I_SC in isolated tracheal epithelium preparations at supramaximal drug concentrations (10^-5 M each). They concluded that isoproterenol and ACh responses occurred in different cell types, presumably serous and mucus cells. In this experiment, supernatant from zymosan-activated AMs, PGE₂, or forskolin shifted the concentration-response relationships for the ACh-induced I_SC to the left relative to untreated controls resulting in a greater than 3-fold increase in apparent sensitivity to ACh. ACh-induced maximal I_SC were increased by supernatant or PGE₂ pretreatment. Forskolin pretreatment did not increase ACh-induced maximal I_SC, similar to the effect of isoproterenol on ACh response in isolated tracheal epithelium as reported by Farley et al. (1991). Therefore, PGE₂ or supernatant from zymosan-activated AMs increases cytosolic cAMP concentration to enhance the apparent sensitivity to ACh but may activate another pathway, resulting in an increased maximal response to ACh. The increase in maximal response to ACh may not involve increased maximal activation of serosal K⁺ channels since the amplitude of ACh-induced serosal K⁺ current in nystatin permeabilized SGCs was not influenced by PGE₂ (Fig. 6). PGE₂ is known to activate tyrosine kinase/phosphatidylinositol 3 kinase via the EP₄ receptor, in addition to activation of PKA (Regan, 2003), but whether this pathway is activated in SGCs is not known. EP₄ receptors are present in isolated SGCs (Fig. 3E).

The apparent increase in sensitivity to ACh occurred rapidly after exposure to PGE₂ and is therefore probably not due to an increased expression of receptors. Also, PGE₂ sensitized SGCs to histamine-induced I_SC (H. Liu and J. M. Farley, unpublished data) that are mediated via H₁ receptors (Liu and Farley, 2005). It seems likely that steps common to both ACh and histamine transduction pathways are enhanced by PGE₂. One common event may be the sensitization of ion channels by PKA-mediated phosphorylation. PKA reportedly sensitizes large-conductance Ca²⁺-activated K⁺ channels to Ca²⁺ (Tian et al., 2001). PGE₂ pretreatment apparently sensitized ACh-induced serosal K⁺ current without enhancing its maximal response (Fig. 6, A and B). ChTX reduced the ACh-induced maximal I_SC without changing the apparent sensitization of ACh response by PGE₂ treatment (Fig. 5B). Thus, PGE₂-induced direct sensitization of ChTX-sensitive K_Ca is not solely responsibly for the apparent sensitization of SGCs to ACh. Chromanol 293B-sensitive cAMP-activated K⁺ channels are not involved in such sensitization either because 293B had no effect on the ACh-induced I_SC in both sensitivity and magnitude, although it inhibited the PGE₂-induced I_SC (Fig. 5B).

CFTR channels are reported to be the exclusive Cl^- conductance in Calu-3 cells for cholinergically mediated gland secretions (Moon et al., 1997). Our data show that glibenclamide applied apically reduced both PGE₂- and ACh-induced I_SC. However, Joo et al. (2002) demonstrated that cholinergic-stimulated fluid secretion occurred in tracheal/bronchial epithelium from cystic fibrosis patients, although cAMP-induced secretion did not. It is possible that CFTR loss in serous cells is compensated for partially by Ca²⁺-activated Cl^- channels in mucus cells (Ballard and Inglis, 2004). Glibenclamide may also have nonspecific effects, blocking important ion transporters essential for ACh-induced I_SC (Ballard and Inglis, 2004). Although glibenclamide almost completely blocked PGE₂- and ACh-induced I_SC, the sensitization of the EC₅₀ for ACh-induced I_SC was not affected by glibenclamide (Fig. 5D). These data suggest that the apparent sensitization of SGCs to ACh by PGE₂ is not due to the direct sensitization of ion channels such as CFTR and K_Ca but rather through effects on the signal transduction pathway activating ion channels, presumably by enhancing the elevation of intracellular Ca²⁺.

Overall implications of this study are that exposure to particulates induces the release of PGE₂ from AMs, which have significant effects on the mucosa of the airway. The products released increase ion flux (and therefore secretion of fluid) into the airway and sensitize the SGCs to respond to secretagogues. Inhibition of CFTR and K_Ca changed the magnitude of ACh-induced response without affecting PGE₂-induced sensitization to ACh. If PGE₂-induced sensitization of SGCs to ACh was due to enhanced Ca²⁺ mobilization, we suggest that events activated by Ca²⁺, such as ACh-induced mucus release, would also be enhanced by PGE₂ even if CFTR function was lost. PGE₂ generally has been considered anti-inflammatory in the lung (Vancheri et al., 2004). Our study suggests that PGE₂-induced enhancement of SGC secretory response to ACh may constitute a protective mechanism during acute exposure of airway to particulates by aiding in particulate clearance; however, it also may lead to excessive fluid/mucus secretion and therefore exacerbate pathological conditions found in asthma, cystic fibrosis, or chronic obstructive pulmonary disease.

	Footnotes

 
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.

doi:10.1124/jpet.105.088542.

ABBREVIATIONS: SGC, submucosal gland cell; AM, alveolar macrophage; I_SC, short circuit current; ACh, acetylcholine; PG, prostaglandin; COX, cyclooxygenase; CFTR, cystic fibrosis transmembrane conductance regulator; DMEM, Dulbecco's modified Eagle's medium; PBS, phosphate-buffered saline; DMSO, dimethyl sulfoxide; DPC, diphenylamine-2-carboxylate; DIDS, disodium 4,4'-diisothiocyanatostilbene-2,2'-disulfonate; TBS, Tris-buffered saline; TBST, TBS/Tween 20; EP, endoprostanoid; AH6809, 6-isopropoxy-9-oxoxanthene-2-carboxylic acid; SC19220, 8-chloro-dibenzo[b,f][1,4]oxazepine-10(11H)-carboxylic acid, 2-acetylhydrazide; ChTX, charybdotoxin; ANOVA, analysis of variance; M, macrophage; PKA, cAMP-dependent protein kinase A.

Address correspondence to: Dr. Jerry M. Farley Sr., Department of Pharmacology and Toxicology, University of Mississippi Medical Center, 2500 North State Street, Jackson, MS 39216. E-mail: jfarley{at}pharmacology.umsmed.edu

	References

Top Abstract Materials and Methods Results Discussion References

 

Ballard ST and Inglis SK (2004) Liquid secretion properties of airway submucosal glands. J Physiol 556: 1-10.[Abstract/Free Full Text]

Becker J and Grasso RJ (1985) Suppression of phagocytosis by dexamethasone in macrophage cultures: inability of arachidonic acid, indomethacin and nordihydroguaiaretic acid to reverse the inhibitory response mediated by a steroid-inducible factor. Int J Immunopharmacol 7: 839-847.[Medline]

Becker S, Soukup JM, Gilmour MI, and Devlin RB (1996) Stimulation of human and rat alveolar macrophages by urban air particulates: effects on oxidant radical generation and cytokine production. Toxicol Appl Pharmacol 141: 637-648.[CrossRef][Medline]

Breyer RM, Bagdassarian CK, Myers SA, and Breyer MD (2001) Prostanoid receptors: subtypes and signaling. Annu Rev Pharmacol Toxicol 41: 661-690.[CrossRef][Medline]

Camus P and Jeannin L (1984) The diseased lung and drugs. Arch Toxicol Suppl 7: 66-87.[Medline]

Chan MH, Dwyer TM, and Farley JM (1996) Cocaine inhibits chloride secretion by swine tracheal submucosal gland cells grown in culture. Toxicol Appl Pharmacol 139: 387-393.[Medline]

Coulson FR and Fryer AD (2003) Muscarinic acetylcholine receptors and airway diseases. Pharmacol Ther 98: 59-69.[CrossRef][Medline]

Cowley EA (2003) Isoprostane-mediated secretion from human airway epithelial cells. Mol Pharmacol 64: 298-307.[Abstract/Free Full Text]

Cowley EA and Linsdell P (2002) Characterization of basolateral K⁺ channels underlying anion secretion in the human airway cell line Calu-3. J Physiol 538: 747-757.[Abstract/Free Full Text]

Dorger M and Krombach F (2000) Interaction of alveolar macrophages with inhaled mineral particulates. J Aerosol Med 13: 369-380.[Medline]

Farley JM, Adderholt G, and Dwyer TM (1991) Autonomic stimulation of short circuit current in swine trachea. Life Sci 48: 873-880.[Medline]

Fujii T, Hayashi S, Hogg JC, Mukae H, Suwa T, Goto Y, Vincent R, and van Eeden SF (2002) Interaction of alveolar macrophages and airway epithelial cells following exposure to particulate matter produces mediators that stimulate the bone marrow. Am J Respir Cell Mol Biol 27: 34-41.[Abstract/Free Full Text]

Girotti M, Evans JH, Burke D, and Leslie CC (2004) Cytosolic phospholipase A2 translocates to forming phagosomes during phagocytosis of zymosan in macrophages. J Biol Chem 279: 19113-19121.[Abstract/Free Full Text]

Hwang T, Suh D, Bae H, Lee S, and Jung J (1996) Characterization of K⁺ channels in the basolateral membrane of rat tracheal epithelia. J Membr Biol 154: 251-257.[CrossRef][Medline]

Ishihara H, Shimura S, Satoh M, Masuda T, Nonaka H, Kase H, Sasaki T, Sasaki H, Takishima T, and Tamura K (1992) Muscarinic receptor subtypes in feline tracheal submucosal gland secretion. Am J Physiol 262: L223-L228.

Joo NS, Irokawa T, Wu JV, Robbins RC, Whyte RI, and Wine JJ (2002) Absent secretion to vasoactive intestinal peptide in cystic fibrosis airway glands. J Biol Chem 277: 50710-50715.[Abstract/Free Full Text]

Kitano S, Wells UM, Webber SE, and Widdicombe JG (1992) The effects of intraluminal and extraluminal drug application on secretion and smooth muscle tone in the ferret liquid-filled trachea in vitro. Pulm Pharmacol 5: 167-174.[CrossRef][Medline]

Knowles MR and Boucher RC (2002) Mucus clearance as a primary innate defense mechanism for mammalian airways. J Clin Investig 109: 571-577.[CrossRef][Medline]

Krouse ME, Talbott JF, Lee MM, Joo NS, and Wine