Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

There is increasing evidence that nitric oxide (NO) (1), produced by the enzyme nitric oxide synthase (NOS), plays an important role in physiologic and pathologic processes in the airway (1). NO is present in expired gas (4), and studies with animal models of air embolism as well as measurements of bronchiolar NO gas at end-expiration in humans suggest that the principal source of expired NO is the airway rather than the pulmonary vasculature (5, 6). In addition, immunohistochemical studies have revealed the presence of all three of the major isoforms of NOS in the airway. Endothelial NOS (eNOS) has been localized to airway epithelial cells and neutrophils, neuronal NOS is found in airway epithelium and nonadrenergic, noncholinergic inhibitory neurons, and inducible NOS is present in airway epithelial cells, macrophages, fibroblasts, and neutrophils (1, 2). The functions of NO in the airway include smooth-muscle relaxation, neurotransmission, and bacteriostasis, as well as the modulation of ciliary motility, mucin secretion, and plasma exudation (1, 2). Furthermore, expired NO levels are altered in asthma and upper respiratory infections (4, 7, 8), suggesting that NO has a pathologic or compensatory role in certain airway disease states.

Although little is known about the regulation of airway NO production, indirect evidence suggests that there may be hormonal modulation of this process through the effects of estrogen. In studies of menstruating women, exhaled NO concentrations were markedly increased during midcycle, when systemic estrogen levels are highest (9). It has also been noted that asthma symptoms are less significant at mid-menstrual cycle (10). On the basis of these findings, we designed the present experiments to determine the direct effects of estrogen on airway NOS. Studies were focused on airway epithelial eNOS, since there is evidence of estrogen-mediated effects on this isoform of NOS in vascular endothelium (13). We tested the hypothesis that estrogen stimulates eNOS in airway epithelial cells. In addition, we conducted studies to determine whether nuclear effects of estrogen were involved in this stimulation, whether estrogen modifies NOS expression or the level of activation of existing NOS, and whether the process requires the activation of epithelial estrogen receptors (ERs).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Culture

Experiments were performed with NCI-H441 human bronchiolar epithelial cells, which are of Clara cell lineage (10), because we have previously shown that these cells exclusively express the eNOS isoform in a constitutive manner (14). The use of this continuous cell line allows for the specific examination of airway epithelial cell eNOS function without contamination by resident macrophages or endothelial cells, which may be present in primary cell cultures. In addition, NOS activity in the NCI-H441 cell line is not altered after multiple passages in culture. H441 cells have been used previously in numerous studies of airway epithelial cell gene expression and function (15). We propagated the cells in RPMI medium containing 10% fetal bovine serum, 1% L-glutamine, 1% antibiotic-antimycotic mixture, 0.5% ampicillin, 0.15% nystatin, 0.15% gentamicin, and 0.10% tylosin in a humidified incubator with 5% CO₂ in air at 37°C. The cells were studied in 24-well or 150 × 25-mm tissue culture plates (Becton Dickinson, Lincoln Park, NJ) at subconfluence.

NOS Stimulation in Intact Cells

We assessed NOS stimulation in whole H441 cells by measuring the conversion of [³H]L-arginine to [³H]L-citrulline, using methods modified from those of Davda and colleagues (18). This procedure provides a direct assessment of the acute activation of existing eNOS, keeping signal transduction mechanisms intact (18). The H441 cells were placed in L-arginine-deficient, serum-free endothelial- SFM Growth Medium (Life Technologies, Inc., Great Island, NY) containing 1% L-glutamine, 1% antibiotic-antimycotic mixture, 0.5% ampicillin, 0.15% nystatin, 0.15% gentamicin, and 0.10% tylosin for 18 h prior to study. L-arginine-deficient medium was replaced for 15 min at 38°C with phosphate-buffered saline (PBS; 500 µl per well), pH 7.4, containing 120 mM NaCl, 4.2 mM KCl, 2.5 mM CaCl₂, 1.3 mM MgSO₄, 7.5 mM glucose, 10 mM N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (Hepes), 1.2 mM Na₂HPO₄, and 0.37 mM KH₂PO₄. After this preincubation, the assay for NOS activity was initiated by aspirating the PBS and replacing it with 400 µl PBS containing 1.5 µCi/ml [³H]L-arginine (Amersham International, Buckinghamshire, UK). An aliquot of 500 µl of 1 N trichloroacetic acid (TCA) was added to wells designated as blanks. The cells were then incubated at 37°C for 5-60 min, and the NOS reaction was terminated by adding 500 µl of ice-cold 1 N TCA to all wells except blanks. The majority of experiments was performed over a period of 60 min because there was the least amount of variability at this time point. The cells were freeze-fractured twice in liquid nitrogen for 2 min each, and were thawed at 37°C for 8 min. They were scraped from the tissue culture plate with a plastic spatula, and the contents of each well were transferred to a sialonized glass test tube. Ether extraction was performed three times with water-saturated ether. The samples were neutralized with 1.5 ml of 25 mM Hepes, pH 8, applied to Dowex AG50WX-8 (Tris form) columns, and were eluted with 1 ml of 40 mM Hepes buffer, pH 5.5, containing 2 mM disodium ethylenediaminetetraacetic acid (Na₂EDTA) and 2 mM ethyleneglycol-bis-(-aminoethyl ether)-N,N'-tetraacetic acid (EGTA). The eluate was collected in glass scintillation vials, and the [³H]L-citrulline generated was quantitated by liquid scintillation spectroscopy (Model A3000; United Tech Packard, Downers Grove, IL). In individual experiments performed in 24-well plates, four to six wells were used for each treatment group. Basal [³H]L-citrulline generation over 60 min ranged from 4.2 to 11.5 fmol/100,000 cells on different experimental days. Because of this variability in basal NOS activity in separate experiments done in different plates, results are expressed as the percent of basal NOS activity measured in the same 24-well plate. All findings were confirmed in at least three independent studies.

Experimental Design

To determine whether estrogen stimulates NOS in H441 cells we measured the conversion of [³H]L-arginine to [³H]L-citrulline in intact cells under basal conditions, in the presence of 10⁸ M estradiol-17 (E₂), or in the presence of the calcium ionophore A23187 (10⁵ M), a known stimulator of NOS that served as a positive control. The dose-response to E₂ was evaluated in incubations performed with concentrations of E₂ ranging from 10¹² to 10⁶ M. Parallel studies were also performed with estradiol-17. To determine whether the effect of estrogen on NOS is mediated at the level of gene transcription, we assessed NOS stimulation in the presence of actinomycin D (25 µg/ml) after pretreatment with this agent for 2 h. To identify the role of extracellular calcium and to reveal which NOS isoform is affected by estrogen, we conducted the incubations for NOS stimulation in the intact cells in both the presence and absence of extracellular Ca²⁺, and in the presence and absence of LaCl₃ (10⁴ M), an agent that blocks calcium influx (19).

Many of the effects of estrogen on cellular function are mediated by the activation of ERs, which include the classical ER, ER, and the recently discovered ER, ER (13, 20, 21). To determine whether NOS stimulation by E₂ is mediated through ER activation, we measured NOS stimulation in both the presence and absence of the ER antagonist ICI182,780 (10⁵ M) (13). Because recently described ER-mediated effects of estrogen are inhibited by the related antagonist ICI164,384 (20), it is likely that ICI182,780 inhibits both ER- and ER-mediated effects. ICI182,780 was the kind gift of Dr. B. M. Vose (Zeneca Pharmaceuticals, Cheshire, UK).

NOS Abundance

To determine whether estrogen modifies NOS abundance, we measured NOS enzymatic activity in the presence of excess substrates and cofactors in the lysates of cells treated for 60 min with either control medium or medium containing 10⁸ M E₂. The cells were washed with ice-cold buffer containing 100 mM NaCl, 25 mM NaH₂PO₄, and 80 mM Na₂HPO₄, pH 7.5, and were then pelleted and resuspended in ice-cold 50 mM Tris buffer (pH 7.4) containing 1.0 mM EDTA, 5 mM mercaptoethanol, 10 µg/ml pepstatin A, 10 µg/ml leupeptin, 90 µg/ml phenylmethylsulfonyl fluoride, and 1.0 µM tetrahydrobiopterin. The cells were disrupted by freeze-thawing in liquid nitrogen, and NOS activity was determined in the cell lysates by measuring the conversion of [³H]L-arginine to [³H]L-citrulline, as previously described (22). Briefly, 50 µl of cell lysate was added to 50 µl of buffer to yield final concentrations of reagents as follows: 2 mM -nicotinamide adenine dinucleotide phosphate, 2 µM tetrahydrobiopterin, 10 µM flavin adenine dinucleotide, 10 µM flavin mononucleotide, 0.5 mM CaCl₂ in excess of EDTA, 15 nM calmodulin, 2 µM cold L-arginine, and 2 µCi/ml [³H]L-arginine. After incubation at 37°C for 60 min, the reaction was terminated by the addition of 400 µl of 40 mM Hepes buffer, pH 5.5, with 2 mM EDTA and 2 mM EGTA. The terminated reactions were processed further as described previously for the intact-cell experiments. The protein content of the samples was determined by the method of Bradford, using bovine serum albumin as the standard (23). The amount of [³H]L-citrulline generated was linearly related to the duration of incubation and the amount of lysate tested. Under the conditions employed, the limiting factor is the abundance of NOS enzyme. We have previously found this determination to be a sensitive indicator of changes in eNOS abundance (22).

Immunoblot Analysis

To confirm the findings for abundance of eNOS as assessed through measurement of NOS enzymatic activity in cell lysates, we performed immunoblot analysis for eNOS. The methods used generally followed those we have previously reported (14). H441 cells were harvested in ice-cold PBS, pelleted, resuspended in 50 mM Tris buffer (pH 7.4) containing 16 mM 3-[(cholamidopropyl)-dimethylamino]- 1-propanesulfonate, 100 mM NaCl, 0.5 mM EDTA, 0.02 mM EGTA, 0.4 mM -mercaptoethanol, 1.6 mM dithiothreitol, and 2 µg/ml each of soybean trypsin inhibitor, lima bean trypsin inhibitor, antipain, and leupeptin, and were ultrasonically disrupted (Branson Ultrasonics, Chicago, IL). The protein content of the preparation was determined, sodium dodecylsulfate-polyacrylamide gel electrophoresis was performed on 100 µg protein with 10% acrylamide, and the proteins were electrophoretically transferred to poly(vinylidene difluoride) membranes. The membranes were blocked for 1 h in buffer containing 137 mM NaCl and 20 mM Tris (pH 7.5) with 0.5% Tween-20 and 5% dried milk, and were incubated overnight at 4°C with a 1:5,000 dilution of antiserum to eNOS (22). After incubation with primary antiserum, the membranes were washed with the 137 mM NaCl buffer with Tween-20 at 0.2% and dried milk at 0.2%, and were incubated for 1 h with a 1:2,000 dilution of a goat antirabbit immunoglobulin antibody-horseradish peroxidase conjugate (Amersham). The membranes were washed in the 137 mM NaCl buffer with Tween-20, and the band for eNOS was visualized through chemiluminescence (ECL Western Blotting Analysis System; Amersham).

The expression of ER protein in the H441 cells was also evaluated with immunoblot analysis, using an antiserum directed against the estrogen binding domain of ER. Membranes were incubated for 1 h with 2 µg/ml of the mouse monoclonal antibody AER 320, directed against amino acids 495-595 of human ER (Neomarkers, Inc., Fremont, CA). The secondary antibody (1:2,000) was a peroxidase-linked antimouse antibody raised in sheep (Amersham). Because the ligand binding domains of ER and ER are highly homologous (20, 21), the primary antiserum may recognize either receptor subtype.

Reverse Transcription-Polymerase Chain Reaction Assays

To determine whether ER mRNA is expressed in the H441 cells, we performed reverse transcription-polymerase chain reaction (RT-PCR) assays, using primers designed from the complementary DNA (cDNA) sequence of human ER. Total cellular RNA was obtained from the cells by a single extraction with an acid guanidinium thiocyanate-phenol- chloroform mixture (24). RT was done according to methods previously reported, using 5 µg total RNA (25). Briefly, cDNA synthesis was done with 200 U Moloney murine leukemia virus reverse transcriptase, 5 µM oligo(deoxythymidine), 1 mM deoxynucleotide triphosphates (dNTPs), and 3 mM Mg²⁺ in a volume of 20 µl. In selected tubes, the reverse transcriptase was omitted to control for amplification from contaminating cDNA or genomic DNA. The temperature profile consisted of annealing at room temperature for 5 min, extension at 42°C for 60 min, and termination at 99°C for 5 min.

PCR was done on the resulting RT product, using specific oligonucleotide primers designed from the estrogen binding domain of human ER (26). The estrogen binding domain was selected because it is highly specific to the ER compared with other steroid-hormone receptors. The sequence of the sense primer for the ER was 5'-CTGTTTGCTCCTAACTTGCTCTTGGACAGG-3' (exon 6), and that of the antisense primer was 5'-GATGCTCCATGCCTTTGTTACTCATGTTGC-3' (exon 8) (26). The PCR reactions contained 2.0 mM Mg²⁺, 1 µM primers, 200 µM dNTPs, reaction buffer, and 5 µl cDNA in a final volume of 50 µl. To minimize nonspecific amplification, a "hot start" procedure was used in which the PCR reaction tubes were placed in a thermal cycler (Model 9600; Perkin-Elmer, Norwalk, CT) prewarmed to 94°C. After 2 min, each tube was opened sequentially and 2.5 U (in 2 µl) of Taq DNA polymerase was added. The PCR temperature profile consisted of 35 cycles at 94°C for 20 s (denaturation), 51°C for 30 s (annealing), and 72°C for 30 s (extension), followed by a final extension for an additional 5 min at 72°C. The primer location, primer concentration, Mg²⁺ concentration, and annealing temperature were optimized to produce the greatest amount of a single PCR product.

The PCR products were size-fractionated by agarose gel electrophoresis, and product identity was confirmed by transferring the DNA to nylon filters and probing with a ³²P-labeled internal oligonucleotide specific for the ER (5'-CTCCAGGGAGAAGAGTTTGTCTC-3') (26). PCR product identity was also confirmed by direct double-stranded sequencing. RT-PCR was also performed on total RNA from sheep uterus, which served as a positive control. The results were confirmed with three different RNA samples from the H441 cells.

To confirm that eNOS and ER are coexpressed in freshly obtained airway epithelium, we isolated airway epithelial cells from adult sheep immediately after harvesting the lung. The epithelial cell layer of mainstem bronchi was microdissected on ice, using sterile techniques, and total RNA was obtained as described previously. The identity of the cell layer removed from the bronchi was confirmed histologically with hematoxylin and eosin staining (28). RT-PCR assays were done with the same primers as used for the ER assay and with primers specific for sheep eNOS (27). PCR product identity was revealed by Southern blot analysis and by direct double-stranded sequencing. The results were confirmed with three different RNA samples from sheep airway epithelium.

Immunocytochemistry

Immunocytochemistry was also done on preparations of the H441 airway epithelial cells to localize further the ER within these cells. H441 cells cultured in six-well plates (Costar Corp., Cambridge, MA) were washed with PBS (pH 7.4), fixed in 4% paraformaldehyde in PBS for 30 min at room temperature, and washed again with PBS. The cells in each well were encircled with a PAP Pen (Zymed Laboratories, Inc., South San Francisco, CA), and a protein-blocking agent (Lipshaw Immunon, Pittsburgh, PA) was applied for 30 min at room temperature. The cells were blotted and then incubated for 2 h at room temperature with either a mouse monoclonal antibody to human ER (MAB461; Chemicon, Temecula, CA) diluted 1:100 in diluent (Cell Marque, Austin, TX), or with preimmune serum. Following quenching of endogenous peroxidase activity with 3% H₂O₂ in H₂O at room temperature, sequential 10-min incubations were performed with a biotinylated secondary antibody and a streptavidin-labeled conjugate of horseradish peroxidase (DAKO Corp., Carpinteria, CA). The ER was detected indirectly through visualization, using diaminobenzidine (Research Genetics, Huntsville, AL) as the chromagen. The cells were rinsed with H₂O, cleared in xylene, air-dried overnight at room temperature, and coated with crystal mount (Biomeda Corp., Foster City, CA). The immunocytochemical results were visualized and photographed with an Olympus BX50 microscope and SC35 camera (Olympus Optical Co., Tokyo, Japan). Findings were confirmed in two independent experiments.

Statistical Analysis

Analysis of variance (ANOVA) with Neuman-Keuls post hoc testing was used to compare mean values for multiple groups. Nonparametric ANOVA was used when indicated (28). Results are expressed as mean ± SEM. Significance was accepted at P < 0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Estrogen-Stimulated NOS Activity

The effects of E₂ on NOS activity in intact H441 cells are shown in Figure 1. In the presence of 10⁸ M E₂ for 60 min, NOS activity was stimulated to 243% of basal levels (Figure 1A). The stimulation that occurred in response to E₂ was comparable to the 201% increase elicited by the calcium ionophore A23187. Studies of varying concentrations of E₂ revealed that maximal stimulation of 241% was achieved at 10⁸ M E₂ (Figure 1B). The time course of E₂-stimulated NOS activation indicates that the response was evident within 5 min (Figure 1C). Estradiol-17 had no effect on NOS activity, yielding levels that ranged from 85 ± 23% to 91 ± 26% of basal activity at concentrations of estradiol-17 ranging from 10¹² M to 10⁶ M. 

View larger version (17K): [in this window] [in a new window]   Figure 1.   E2-mediated activation of NOS in H441 cells. [3H]L-arginine conversion to [3H]L-citrulline was measured in intact cells. (A) Comparison of effects of E2 (108 M) and A23187 (105 M) over 60 min. (B) Effect of varying concentrations of E2 on NOS activity over 60 min. (C) Time course of the effect of 108 M E2 (mean ± SEM, n = 4, *P < 0.05 versus basal NOS activity). E2 caused a 2.4-fold stimulation of NOS, which was comparable to the response to A23187, apparent at 108 M E2, and discernible within 5 min.

To determine whether E₂-mediated stimulation of NOS involves transcriptional effects of the hormone, we also performed studies following actinomycin D pretreatment (Figure 2). Actinomycin D had no effect on basal NOS activity. In addition, there was no change in E₂-stimulated or A23187-stimulated NOS activity following exposure to actinomycin D. 

View larger version (13K): [in this window] [in a new window]   Figure 2.   Effect of actinomycin D on E2-mediated NOS activity in H441 cells. Cells were treated with actinomycin D (25 µg/ml) for 120 min, and conversion of [3H]L-arginine to [3H]L-citrulline was then measured in nonstimulated cells (basal) and in the presence of E2 (108 M) or A23187 (105 M) over a period of 60 min in the continued presence of actinomycin D (mean ± SEM, n = 4, *P < 0.05 versus basal NOS activity). Actinomycin D had no effect on basal or stimulated NOS activity.

To investigate whether the stimulation of NOS by E₂ is due to a direct upregulation of NOS abundance, we measured NOS enzymatic activity in the presence of excess substrates and cofactors in lysates of control cells and cells exposed to 10⁸ M E₂ for 60 min (Figure 3A). NOS enzymatic activity was similar in control and E₂-treated cells. Correspondingly, immunoblot analysis revealed that eNOS protein abundance was comparable in control and E₂-treated cells (Figure 3B).

View larger version (14K): [in this window] [in a new window]   Figure 3.   Effect of E2 on NOS abundance in H441 cells. After treatment with control medium (Con) or E2 (108 M) for 60 min, NOS activity was determined in H441 cell lysates in the presence of excess cofactors and substrates (A), or eNOS protein expression was evaluated by immunoblot analysis (B). Summary data are shown for n = 3 experiments (mean ± SEM), and a representative immunoblot is shown in the inset. E2 had no effect on abundance of eNOS as evaluated with either approach.

Role of Calcium

The role of calcium in E₂ activation of NOS is shown in Figure 4. The removal of extracellular calcium caused a 60% decrease in basal NOS activity, and also completely prevented the E₂ activation of eNOS (Figure 4A). LaCl₃, which had no discernible effect on basal NOS activity, also caused full inhibition of E₂ stimulation of NOS (Figure 4B).

View larger version (13K): [in this window] [in a new window]   Figure 4.   Role of calcium in E2-stimulated NOS activity in intact H441 airway epithelial cells. (A) [3H]L-arginine conversion to [3H]L-citrulline was measured in the presence (+) or absence () of extracellular calcium (Ca2+), under basal conditions or with 108 M E2 added for 60 min. (B) NOS activity was measured in the absence () or presence (+) of 104 M LaCl3, under basal conditions or with 108 M E2 added for 60 min (mean ± SEM, n = 6; *P < 0.05 versus basal NOS activity; #P < 0.05 versus Ca2+ [+]; P < 0.05 versus LaCl3 []). The removal of extracellular calcium caused a decrease in basal NOS activity, and E2-mediated activation of eNOS was completely prevented. LaCl3 had no discernible effect on basal NOS activity, but caused full inhibition of E2-mediated activation of eNOS.

Role of ER

The effects of the ER antagonist ICI182,780 on E₂-stimulated NOS activity are shown in Figure 5. The basal level of NOS activity was not altered by ER antagonism. However, the ER antagonist completely inhibited E₂-mediated activation of NOS.

View larger version (13K): [in this window] [in a new window]   Figure 5.   Effect of ICI182,780 on E2-stimulated NOS activity in intact H441 airway epithelial cells. [3H]L-arginine conversion to [3H]L-citrulline was measured in the absence () or presence (+) of 105 M ICI182,780, under basal conditions or with 108 M E2 added for 60 min (mean ± SEM, n = 6; *P < 0.05 versus basal NOS activity; #P < 0.05 versus ICI182,780 []). ICI182,780 did not alter basal NOS activity, but completely inhibited E2-mediated activation of eNOS.

Figure 6A shows a typical Southern blot of the RT- PCR products for ER in the H441 cells. A single PCR product, with the predicted size of 367 bp, was obtained with RNA from sheep uterus (positive control) and with RNA from the H441 cells. The identity of the PCR product was confirmed by direct sequencing. PCR product was not obtained when the reverse transcriptase enzyme was omitted from the RT step.

View larger version (16K): [in this window] [in a new window]   Figure 6.   (A) Southern blot analysis of RT-PCR products from ER mRNA in H441 cells and ovine uterus, which served as a positive control (C). (B) Southern blot analysis of RT-PCR products from eNOS and ER mRNA in ovine airway epithelial cells. Assays with epithelial RNA were performed in the presence (+) or absence () of the RT enzyme. The predicted band sizes were 367 bp and 281 bp for ER and eNOS, respectively. PCR product identity was confirmed by direct sequencing. Results are representative of three independent experiments. Single PCR products were obtained for ER in the H441 cells, and for both eNOS and ER in the freshly obtained sheep airway epithelial cells.

To confirm that eNOS and ER are coexpressed in freshly obtained airway epithelium, RT-PCR assays were performed for both mRNAs from ovine bronchial epithelial cells. Figure 6B shows typical Southern blots from these experiments. Single PCR products were obtained for eNOS and for ER, with the predicted sizes of 281 bp and 367 bp, respectively. The identity of the PCR products was confirmed by direct sequencing, and the PCR products were not obtained when the reverse transcriptase was omitted from the RT step. The sequence of the sheep ER cDNA insert obtained with RT-PCR was 91% homologous to that of human ER at the nucleotide level (26).

To determine whether ER protein is expressed in H441 airway epithelial cells, we performed both immunoblot analysis and immunocytochemistry. Immunoblots revealed a signal for ER protein at 67 kD (Figure 7). Similar observations were made in three independent experiments. Immunocytochemistry for ER revealed positive staining for the protein, of variable intensity, in both the nuclei and the cytoplasm of H441 epithelial cells (Figure 8A). Immunostaining was absent when preimmune serum was substituted for the primary antibody (Figure 8B).

View larger version (12K): [in this window] [in a new window]   Figure 7.   Immunoblot analysis for ER protein in H441 airway epithelial cells. Signal for ER protein was evident at 67 kD. Results are representative of three independent experiments.

View larger version (122K): [in this window] [in a new window]   Figure 8.   Immunocytochemistry for ER protein in H441 airway epithelial cells . Magnification: ×380. (A) Staining for ER was evident in both the nucleus and the cytoplasm. (B) Staining was absent when preimmune serum was substituted for the primary antibody. Results are representative of two independent experiments.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In the present study, we demonstrated that estrogen enhances NOS activity in H441 human airway epithelial cells. We showed that this occurs at physiologic concentrations of E₂ (10⁸ M), and that the degree of enhancement is comparable to that produced by the known agonist A23187. We also showed that the less biologically active isomer, estradiol-17, has no effect. In addition, we demonstrated that the NOS response is detectable as early as 5 min after exposure to E₂. To our knowledge, this is the first direct demonstration of NOS stimulation by estrogen in airway epithelial cells.

To determine the mechanism by which estrogen stimulates airway epithelial NOS, we first evaluated whether nuclear effects of estrogen are involved. In experiments done in the presence of actinomycin D, basal activity, A23187-stimulated activity, and E₂-mediated stimulation of NOS were not altered. This finding, as well as the observation that the response to estrogen occurs within 5 min, indicates that the transcription of estrogen-responsive genes does not play a role in this process. Acute, nontranscriptional effects of estrogen have been observed previously in other cell types. For example, E₂ causes changes in the ionic conductance of postsynaptic membranes of neurons within the medial amygdala within minutes of exposure, and which are not blocked by inhibitors of protein synthesis (30). In addition, we have recently demonstrated that E₂ acutely activates eNOS in fetal pulmonary artery endothelial cells via nontranscriptional mechanisms (31).

We also determined whether the estrogen-mediated increase in eNOS activity is due to enhanced eNOS expression. Measuring NOS activity in cell lysates in the presence of excess cofactors and substrates, such that the abundance of NOS enzyme is the limiting factor in substrate conversion, we were unable to demonstrate increased enzymatic activity in cells that had been exposed to estrogen for 60 min as compared with control cells. In addition, immunoblot analysis revealed that the abundance of eNOS protein was unaltered. This indicates that the increased NOS activity in the intact cells involved the activation of preexisting eNOS, and not an upregulation of eNOS expression. We have recently demonstrated that physiologic concentrations of estrogen upregulate eNOS expression in pulmonary endothelial cells, but this effect is not evident before 48 h (22).

To begin to determine the signaling mechanism by which estrogen stimulates eNOS in H441 cells, we evaluated the role of calcium. The removal of extracellular calcium attenuated basal NOS activity and completely prevented the enhancement of NOS activity in response to estrogen. When calcium influx was blocked with LaCl₃ (19), basal NOS activity was not affected, but estrogen-stimulated NOS activity was fully inhibited. These findings indicate that the effect of estrogen on H441 cells depends on the influx of extracellular calcium. The rapid effect of estrogen on eNOS in pulmonary endothelial cells has also been shown to be calcium-dependent (31), and studies with aortic endothelium have revealed that estrogen acutely activates calcium-dependent potassium channels, resulting in a slow increase in intracellular calcium levels (32). In addition, E₂ causes an increase in cyclic adenosine monophosphate in rat pulmonary vascular smooth-muscle cells, which occurs within 5 min, is calcium-mediated, and is not blocked by actinomycin D (33). Collectively, these observations suggest that the acute, nontranscriptional effects of estrogen in a variety of cell types may be mediated through changes in intracellular calcium homeostasis.

We also determined whether ERs are involved in the activation of eNOS by estrogen in H441 cells. The ER antagonist ICI182,780 fully inhibited estrogen-stimulated NOS activity, indicating that ER activation is necessary for this effect. In addition, RT-PCR revealed ER mRNA expression in H441 cells, and both immunoblot analysis and immunocytochemistry demonstrated ER protein expression. There was positive immunostaining of variable intensity for ER protein in both the nuclei and the cytoplasm of H441 cells, in accord with the distribution we have previously observed in ovine fetal pulmonary artery endothelial cells (22). These findings indicate for the first time that functional ER are expressed in human airway epithelial cells, and that they play a role in the acute effects of estrogen on airway epithelial NO production. However, it is not yet clear which ER subtype is involved in these effects. The ER and ER subtypes, which are known to function as transcription factors, are highly homologous, particularly in their DNA-binding domains (95-96%) and C-terminal hormone-binding domains (55- 58%) (20, 21). In addition, both ER and ER are inhibited by ICI compounds (20, 21), and immunoblot analysis and immunostaining may recognize either ER or ER. Thus, the observed effects of estrogen on airway epithelial eNOS may be mediated by either ER, ER, or both receptor subtypes, functioning in a new and unique manner, or by a yet unknown ER subtype. Studies of ER subtype in adult rat lung have revealed the expression of mRNA for both ER and ER, with the latter subtype predominating (34), but the pulmonary cell specificity of ER subtype expression is yet to be determined. Further studies are now indicated to distinguish the roles of ER and ER in acute eNOS activation in airway epithelium.

A degree of caution may be warranted in the direct extrapolation of the findings with a cultured cell line in the present study to processes in the intact lung. However, the use of cultured cells has enabled us to evaluate the direct effects of single factors, such as estrogen, on airway epithelial cell function, avoiding the cardiac and systemic effects of this hormone (35, 36). H441 cells have been used in numerous studies of airway epithelial cell gene expression and function (15). In addition, the use of cultured H441 cells has enabled us to distinguish the effects of estrogen on eNOS in airway epithelium from its effects on pulmonary endothelial eNOS, which would not be possible in the intact lung (31). Furthermore, we have shown that estrogen-mediated activation of airway epithelial eNOS occurs at physiologic concentrations of the hormone. Moreover, we have shown that ER and eNOS are coexpressed in freshly obtained ovine bronchial epithelium.

With these potential limitations in mind, there are important physiologic and pathophysiologic implications of the present findings. Estrogen-mediated effects on airway epithelial cell NO production may underlie hormone-related changes in airway function. It has been repeatedly observed that morbidity in asthma is affected by the menstrual cycle, with fewer symptoms reported at midcycle, when estrogen levels are highest, and worse symptomatology noted during menses (10). In addition, a study of postmenopausal women found lower estradiol levels in asthmatic subjects requiring medication than in nonasthmatic subjects (37). There are also reports of alleviation of asthma symptoms and decreased steroid requirements in women with refractory asthma who are subsequently treated with estrogen (38). Further understanding of the mechanisms by which estrogen modifies airway NO production may lead to new therapeutic interventions for asthma or other airway diseases, such as bronchopulmonary dysplasia, bronchiolitis, and cystic fibrosis. In addition, such understanding will enhance our overall understanding of the rapid, nontranscriptional effects of estrogen in a variety of cell types.