Neutral Endopeptidase of a Human Airway Epithelial Cell Line Recovers after Hypochlorous Acid Exposure: Dexamethasone Accelerates This by Stimulating Neutral Endopeptidase mRNA Synthesis Zhihui Lang and Christopher G. Murlas Lung Cell Biology Laboratories and Department of Medicine, Rush University, Chicago, Illinois

Hypochlorous acid (HOCI) exposure of Calu-l cells in situ leads to a relatively rapid and substantial decrease in whole cell neutral endopeptidase (NEP) activity that may result from the internalization of NEP from plasma membrane surfaces. To confirm this, and to assess the time course of changes in cell NEP after oxidant exposure and the potential influence of corticosteroid treatment on these, we evaluated Calu-l NEP activity by high performance liquid chromatography and NEP-specific mRNA over the ensuing 48 h after HOCI in the presence or absence of 1 t-tM dexamethasone. Cells, grown to confluency in Dulbecco's modified Eagle's medium with 10% fetal bovine serum, were exposed for 5 min to 100 t-tM HOCI and then maintained in culture for 48 h thereafter. Before exposure, some cell plates were cooled to 4 0 C and/or incubated for 5 min in 1 mM sodium azide. In some experiments, post-HOCI changes in NEP-specific mRNA in the presence or absence of dexamethasone were also evaluated using Northern blot analysis. We found that sodium azide at 4 0 C totally blocked the effect of HOCI on Calu-1 NEP (n = 6). In the absence of sodium azide, NEP activity spontaneously recovered to preexposure levels within 24 h. This recovery occurred 6 h earlier in the presence of 1 t-tM dexamethasone. Furthermore, dexamethasone increased NEP activity at 24 and 48 h after HOCI. Northern blot analysis indicated that NEP-specific mRNA did not change during spontaneous recovery, but was increased by dexamethasone 24 h after HOCI. We conclude that HOCI exposure decreases Calu-1 plasma membrane NEP by inducing its internalization. Exocytosis of internalized enzyme may be responsible for the spontaneous recovery of cell surface NEP after HOCI. This recovery is substantially accelerated by dexamethasone treatment, which stimulates NEP-specific mRNA synthesis by these cells.

Neutral endopeptidase 24.11 (NEP, also known as enkephalinase, common acute lymphoblastic leukemia antigen [CALLA], and CDlO) is a membrane-bound, glycosylated zinc peptidase with a molecular weight of 94 to 97 kD. It is a typical type II membrane protein with a small intracellular domain (27 amino acids), a transmembrane region (22 amino acids), and a large extracellular catalytic domain (1). The NEP gene has been cloned in several species. Analyses of its eDNA and deduced amino acid sequences suggest that NEP is a highly conserved enzyme. In humans, its gene is

(Received in original form December 5, 1991 and in final form April 1, 1992) Address correspondence to: Christopher Murlas, M.D., Department of Medicine (Pulmonary and Critical Care Medicine), Rush University, 1653 W. Congress Parkway, Chicago, IL 60612. Abbreviations: ala-p-nitroanilide, ANA; common acute lymphoblastic leukemia antigen, CALLA; Dulbecco's modified Eagle's medium, DMEM; hypochlorous acid, HOCI; high performance liquid chromatography, HPLC; p-nitroanilide, NA; neutral endopeptidase 24.11, NEP; phosphate-buffered saline, PBS; N-Succinyl-Ala-Ala-Ala-p-nitroanilide, SA3NA; trichloroacetic acid, TeA. Am. J. Respir. Cell Mol. BioI. Vol. 7. pp. 3110-306, 1992

located on chromosome 3. NEP cleaves peptide bonds at the amino side of hydrophobic amino acids and inactivates many biologically active peptides (1), including substance P, the neurokinins, so-called bombesin-like peptides (2), and atrial natriuretic factor. Although it was discovered in and purified from kidney, NEP is widely distributed in the body, including trachea, lung, intestine, kidney, and brain. In the respiratory system, NEP is present in lung and tracheobronchial epithelial cells where it may subserve a wide variety of functions (1-4), including the modulation of smooth muscle contraction, mucus secretion, and lung cell growth. It has previously been shown that luminal exposure of airway segments in vitroto hypochlorous acid (HOC!) produces airway smooth muscle hyperresponsiveness to substance P and a decrease in NEP activity of tissue segment homogenates, suggesting that HOCI may decrease airway epithelial cell NEP activity (3). A number of other biochemical targets of HOCI have been shown (5-8). To confirm that this effect occurs in humans, we have recently assessed the effects of HOCI exposure on the human airway epithelial cell line Calu-1 (4). As in normal respiratory epithelial cells from other species in our experience, we have found that the plasma membranes of the Calu-l cell line are rich in NEP

Lang and Murlas: Postoxidant Epithelial NEP Reconstitution Stimulated by Corticoids

activity. We have also found that HOCI exposure of whole Calu-l cells in situ leads to a relatively rapid and substantial decrease in whole-cell NEP activity although cell viability is not affected. This decrease in activity of the membrane fraction caused by HOCI is accompanied by a commensurate increase in the cytoplasmic fraction, suggesting that HOCI . induces internalization of NEP from plasma membrane surfaces. It has previously been demonstrated that NEP on human neutrophils is rapidly internalized after exposure to either phorbol myristate acetate or a human NEP-specific monoclonal antibody (9). To further evaluate this, and the potential influence of glucocorticoid treatment, we measured Calu-l NEP activity and NEP-specific mRNA 24 h after HOCI exposure in the presence or absence of 1 IlM dexamethasone. Before HOCI exposure, some cell plates were cooled to 4 0 C and incubated for 5 min in I mM sodium azide, a procedure known to inhibit NEP endocytosis in other cell types (9, 10).

Materials and Methods Reagents Dulbecco's modified Eagle's medium (DMEM), phosphatebuffered saline (PBS) solution, penicillin, streptomycin, and amphotericin B were all obtained from GmCO (Grand Island, NY). N-Succinyl-Ala-Ala-Ala-p-nitroanilide (SA3NA), ala-p-nitroanilide (ANA), p-nitroanilide (NA), fetal bovine serum, trichloroacetic acid (TeA), sodium azide, and dexamethasone were obtained from Sigma Chemical Co. (St. Louis, MO). Bio-Spin 30 chromatography columns were purchased from Bio-Rad Laboratories (Richmond, CA). The human NEP cDNA was a generous gift from Dr. Michelle Letarte of University of Toronto (Toronto, Canada). Mouse -y-actin cDNA was provided by Dr. Xiaoqiang Li of the University of Chicago (Chicago, IL). [a-32P]dCTP was purchased from Amersham (Arlington Heights, IL). All other general chemicals were purchased from Sigma. Dexamethasone was initially dissolved in 100 % ethanol to make a 10 mM stock solution. It was subsequently diluted in DMEM for the experiments.

301

and adjusted to pH 7.0 using phosphoric acid. HOCl/OCIconcentration was quantified as previously described (11) and assuming an extinction coefficient of Em = 142 M-1·cm- 1 for HOCl/OCI-. This HOCI solution diluted in PBS, pH 7.0,was immediately diluted to the final concentration of 100 IlM for use. Absorbance at 291 om was measured. For each experiment, the solution of HOCI in PBS, pH 7.0, was prepared fresh every day and used within 30 min after preparation. Exposure of Whole Cells to HOCI and Sodium Azide After Calu-l cells reached confluency, medium was removed and cells were rinsed twice with 2 ml of PBS, pH 7.0, at room temperature. Two milliliters of 100 IlM HOCI in PBS, pH 7.0, were carefully added to each dish (containing approximately 1 million cells) so as not to disturb the cell layer in situ. The dishes were then incubated at 37 0 C for 5 min. When cells in larger dishes were exposed, the volume of HOCI solution added per million cells was proportional. After exposure, cells were rinsed immediately with PBS, pH 7.4, and assayed for NEP activity. Pretreatment with sodium azide was accomplished by incubating the cells in 1 mM sodium azide at 4 0 C for 5 min after rinsing them with PBS, pH 7.0. After this time, HOCl was added (final concentration, 100 IlM), and the cells were incubated at 4 0 C for another 5 min. Cells were then rinsed with PBS, pH 7.4, at room temperature twice and assayed for NEP activity. In other experiments, cells were exposed to 100 IlM HOCI first and then incubated with sodium azide at 4 0 C for 5 min to assess the direct effect of sodium azide on cell NEP post-HOCl. For experiments concerning recovery post-HOCl, cells were incubated in medium with or without 0.01 % ethanol (vehicle) or 1 IlM dexamethasone in 0.01 % ethanol after being exposed to HOCl. The same was done for experiments in which Northern blot analyses were done. Vehicle alone affected neither Calu-l cell NEP activity nor NEP-specific mRNAs.

Cell Culture Calu-l cells (obtained from the American Type Cell Collection, Rockville, MD [ATCC catalog number HTB-54] and derived from a human lung epidermoid carcinoma) were grown in DMEM supplemented with 10% fetal bovine serum, 2 mM L-glutamine, antibiotics (100 U/ml penicillin and 100 Ilg/ml streptomycin), and antimycotic (0.25 Ilg/ml amphotericin B) in 35 x 10 mm tissue culture dishes at 37 0 C under humidified 5% CO 2 • Medium was replaced twice weekly. Passages 31 through 42 were used for the study. Cells were used only after reaching confluency. When larger amounts of cells were needed for preparation of total RNA, cells were grown in 100 x 20 mm dishes. All experiments were performed at 370 C unless noted otherwise in the text.

Assay of Calu-l Cell NEP Activity In Situ NEP activity was measured as previously described (8). Briefly, Calu-I cells were rinsed with PBS, pH 7.4, after the prescribed incubation period. A volume of 0.5 ml of 0.05 mM SA3NA (a synthetic substrate) in PBS, pH 7.4, containing 10 IlM amastatin was added to each dish. The dishes were incubated at 37 0 C for 30 min with occasional swirling. Thereafter, 0.2-ml aliquots were withdrawn and 50% cold TeA was added to the final concentration of 10% to stop the reaction. This reaction mixture was incubated on ice for 20 min and spun in an Eppendorf (5415C) microcentrifuge at 15,000 rpm for 20 min at 4 0 C. Next, 50 IIIof the clear supernatant was injected onto a Waters high performance liquid chromatography (HPLC) column (3.9 x 150 mm) for analysis. Cells were trypsinized and counted using a hemocytometer. Enzyme activity determined by HPLC (see below) was expressed as pmoles of ANA produced per minute per million cells.

Preparation of HOCI NaOCI (Sigma) was initially diluted 10 times in ice-cold 0.1 N KOH, and this solution was kept at 4 0 C as stock solution for no more than 2 wk, during which time the NaOCI solution was stable. This was further diluted 20 times in PBS

NEP Cleavage Product Analysis by HPLC Enzyme cleavage products were analyzed on a IlBondapak C18 column (3.9 x 150 mm) by a Waters HPLC system (model 600 E; Waters, Milford, MA) using a 30-min, twostep gradient at flow rate of 1 ml/min. Initial conditions con-

302

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 7 1992

sisted of 10% solvent A (methanol) and 90% solvent B (10 mM NaH 2PO., pH 2.5). Solvent A was increased to 30% within 10 min and up to 60% in the next 0.5 min. Solvent A was held at 60 % for 8 min and then brought back to initial conditions. The eluent was monitored at 314 nm. Each peak of products was integrated and quantitated by comparing with authentic standards. NEP activity was identified by production of the substrate cleavage product, ANA, which was phosphoramidon inhibitable and unaffected by the presence of amastatin, an antagonist of aminopeptidase activity. Recovery of Calu-l NEP Activity after HOCI Calu-l cells grown to confluency in 10 x 35 mm dishes were rinsed twice with PBS, pH 7.4, and exposed to 100 /-tM HOCI as described above. After removal of HOCl, cells were incubated in HOCl-free medium for up to 48 h until being assayed for NEP activity as described above. For experiments evaluating the effects of dexamethasone on recovery, cells were incubated in HOCl-free medium with I/-tM dexamethasone in 0.01 % ethanol or vehicle only for up to 48 h until being assayed for NEP activity. Northern Blot Analysis of NEP-specific mRNA NEP-specific mRNA was assessed using previously described methods (12). The cDNA probe for human NEP encodes nucleotides 627 to 2208 and covers the majority of the open reading frame (exons 8 to 24) (13, 14). This 1.6-kb cDNA was inserted into an ampicillin-resistant Bluescript plasmid. Plasmid DNA was isolated and purified using polyethylene glycol precipitation. The insert was released by digestion with the restriction enzyme EcoR! and isolated from agarose gel by phenol-chloroform extraction. It was then labeled with [a- 32P]dCTP by random priming. The unincorporated isotope was removed by using Bio-Spin columns, yielding a probe with a specific activity> 109 cpm/ /-tg DNA. Full-length mouse ')'-actin cDNA was also labeled using the same methodology. Total RNAs from Calu-l cells were extracted using acid guanidinium thiocyanate-phenol-chloroform extraction (15). The final RNA pellet was washed with 70% ethanol (at -20° C) and 100% ethanol (at room temperature) and dissolved in ix TE (10 mM Tris, 1 mM EDTA) buffer, pH 8.0. Concentrations of RNA were determined by absorbance at 260 nm. Aliquots of total RNA were stored at -70° C until used. Electrophoresis of total RNA was performed using 1.2% agarose gel containing formaldehyde in ix MOPS buffer (20 mM 2-[N-morpholino]propanesulfonic acid, pH 7.0; 8 mM sodium acetate; and 1 mM EDTA, pH 8.0). A separate lane of RNA sample was run in parallel and cut after electrophoresis for staining with ethidium bromide. The 28S and 18S rRNAs were used for determination of molecular size. RNAs were transferred to GeneScreen Plus nylon membranes (NEN Research Products, Boston, MA) by capillary blotting with 20x SSC (IX SSC = 0.15 M sodium chloride and 0.015 M sodium citrate) overnight. The membranes were prehybridized at 42° C in 50% formamide, 10% dextran sulfate, and 1% sodium dodecyl sulfate. Denatured, radiolabeled NEP cDNA probe (> 109 cpm//-tg cDNA) and sheared salmon sperm DNA (100 /-tg/ml) were added, and hybridization was allowed to proceed for 40 h at 42° C. The mem-

branes were washed at high stringency (0.2X SSC at 68° C for 30 min). Autoradiographs were exposed overnight at -70° C using an intensifying screen (DuPont, Wilmington, DE). The same membrane was again probed using radiolabeled actin cDNA to correct for variations between samples in the amounts of RNA loaded. Quantitations of both NEPspecific mRNAs and actin-specific mRNA were determined by laser densitometry (Pharmacia LKB Biotechnology, Pleasant Hill, CA). Much shorter exposure autoradiograph, different from the one shown in Figure 4, was used for quantitation of actin. The density of NEP-specific mRNAs was normalized in relation to that of actin for each sample. Slight variations in densities of the actin bands were found in the autoradiographs despite equal amounts of RNA having been loaded per lane. Gamma-actin mRN A was unaffected by dexamethasone treatment. Statistical Analysis Each experiment was done with an identical number of control samples. Data for each time point was collected in triplicate. Mean values ± SEM for like subgroups (both experimental and control) were calculated using independent t tests. Differences between experimental and control groups were considered significant for P values < 0.05.

Results Spontaneous Recovery of Calu-l NEP Activity after HOCI Injury After a 5-min exposure to HOCI, Calu-l cell NEP activity decreased to approximately 70% of preexposure levels (Figure 1). Expressed in terms of activity per minute per million cells, it was 49.2 ± 1.1 at that time. After removal of HOCI

P

' 5If)-

._

100

u= «8 0.. c: 0> 0 U

120

.2: !!2

ell U (LtD

Z~ _ c:

303

c« O>Z

E _a.

40 20

~«

0

0> -

o

12

24

36

48

(L

NaN 3

and incubation in growth medium, this activity decreased another 5% over the next 6 h. Over the ensuing 18 h, NEP activity gradually recovered to baseline levels. A decrease in Calu-l NEP activity of dexamethasoneuntreated cells (control group) was observed at 48 h (Figure 2). This was linked to cell division and a net increase of cell number at that time. There was no increase in cell number during the initial 24 h post-HOCl. It is unlikely that the increase in NEP activity caused by dexamethasone was indirectly related to the relative decrease in the control group that was observed at 48 h. The percentage increase in activity of the dexamethasone-treated cells (165.5%) was much greater than was the decrease seen in controls (27 %) at that time. Effect of Sodium Azide and Low Temperature on HOCl-induced Decrease in Calu-l NEP Activity Data concerning the influence of sodium azide on the decrease in Calu-l NEP activity caused by HOCl are shown in Figure 3. In the presence of 1 mM NaN), HOCI did not decrease Calu-l NEP activity. To exclude the possibility that NaN) caused this result by reacting directly with either Calu-l NEP or HOCl, we also tested the effect of NaN) alone and the order of exposure of cells to HOCI and NaN) on Calu-l NEP activity. As shown in Figure 3, 1 mM NaN3 alone had

+

Conditions

Figure 2. Effects of dexamethasone on the recovery of Calu-l cell NEP activity from HOCI injury. Calu-l cells grown to confluency in 10 X 35 mm culture dishes were exposed to 100 /LM HOCI at 37 0 C for 5 min. HOCI was removed immediately after exposure by washing the cells in PBS, pH 7.4. Whole-cell NEP activities were assayed before and after HOCI exposure in separate but identical dishes of cells. Cells were then incubated in HOCl-free medium containing either 10- 6 M dexamethasone (open circles) or 0.01 % ethanol (closed circles) for up to 48 h. During incubation, wholecell NEP activity was assayed at various time points as indicated. The zero time point represents control activity, i.e., activity of unexposed cells. The next point represents values of NEP activity 5 min after HOCI exposure. Calu-l cell NEP activities are expressed as percent of control activity at individual time points. Values represent means ± SEM (n = 6). Some standard errors are smaller than symbol size. * and ** indicate experimental values significantly different from controls (P < 0.05 and P < om, respectively).

+

HOCI

Time (h)

+

+

Figure 3. Effects of sodium azide and low temperature on HOCIinduced decrease in Calu-l cell NEP activity. Confluent layers of Calu-l cells were preincubated with 1 mM sodium azide or PBS, pH 7.4, at 4 C for 5 min. After this, HOCI or PBS, pH 7.4, was added to the dishes. Cells were then incubated at 4 0 C for another 5 min in the presence of PBS only, 1 mM sodium azide only, 100 /LM HOCI, or 1 mM sodium azide plus 100 /LM HOCI as indicated. At the end of incubation, cells were washed and assayed for wholecell NEP activity as described in MATERIALS AND METHODS. Calu-l cell activities were expressed as percent of control activity (PBS, pH 7.4, only). Values represent means ± SEM (n = 6). Asterisks indicate values significantly different from controls (* P < 0.05; ** P < 0.01). 0

no effect. Furthermore, treatment with NaN3 after HOCI exposure did not block HOCl's effect as did NaN) pretreatment (data not shown). Low temperature alone did not have any effect. Calu-l NEP activity post-HOCI at 4 0 C was 51.9 ± 4.0 pmol ANA/ min/million cells compared with 54.3 ± 3.1 at 37° C. Northern Blot Analysis of Calu-l NEP-specific mRNAs after HOCI Exposure Figure 4 shows the results of Northern blot analysis of NEPspecific mRNA from Calu-l cells. Two NEP-specific mRNA species were identified. Their sizes were 3.5 and 6.5 kb compared with the known molecular sizes of 28S and 18S ribosomal RNA. Gamma-actin mRNA was also probed to normalize densitometric results in relation to actin mRNA. There was no detectable change in the density of either of the two bands 24 h after HOCI exposure (Figure 4, lanes 1 and 3; Figure 5). Although equal amounts of total RNA were loaded in each lane, slight variations in the density of actin bands were still detected. Final comparisons between densitometry values ofNEP-specific mRNA bands were made using values normalized by comparison with lane actin band densities, which were quantitated using a shorter film exposure. Effects of Dexamethasone on Calu-l NEP Activity during Recovery from HOCI Injury Results of dexamethasone treatment of Calu-l cells after HOCI are shown in Figure 2. HOCl-exposed cells were in-

304

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 7 1992

-6.5kb

28S-

-3.5kb

18S-

-actin

HOC!

Dex

+

+

+ +

Conditions Figure 4. Autoradiograph of Northern blot analysis concerning NEP-specific mRNAs after HOCI exposure. Thirty micrograms of total cellular RNA from unexposed (lanes 1 and 2) or HOC1exposed (lanes 3 and 4) cultures were loaded into each lane. Within the two groups of cultures were ethanol-treated (0.01%) and dexamethasone-treated (1 J.'M) samples. Two NEP-specific bands at 3.5 and 6.5 kb and an actin at 1.8 kb are present. The locations of ribosomal RNA bands (28S and I8S) are also shown.

z'00

300

c: Q)

o

ec:

o -

200

o

'0

c:

100 I- ,..--'rT7..,.

Q)

~

Q)

a.

o

+

HOCI Dex

+

+ +

Conditions Figure 5. Quantitation of dexamethasone effect on Calu-l NEPspecific mRNAs before and after HOCI exposure. Densities of the 3.5-kb (open bars) and 6.5-kb (cross-hatched bars) NEP-specific mRNAs of unexposed cells (in absence of dexamethasone) represent control values. Densities of these two NEP-specific mRNA bands in each subgroup (lane) have been normalized to the actin density of that lane (see MATERIALS AND METHODS). Values shown are means of two experiments. In other experiments, we have found that a 12-h treatment with dexamethasone caused 0 and 47.7% increases in densities of the 6.5 and 3.5 kb species, respectively.

cubated in medium containing either 1 J.tM dexamethasone in 0.01 % ethanol or vehicle alone for up to 48 h thereafter. In the presence of 1 J.tM dexamethasone, Calu-l cell NEP activity at 18 h was 116.9 ± 12.1% of pre exposure levels. Compared with controls at 24 and 48 h, dexamethasone increased NEP activity by 70.1 ± 14.1 and 267.4 ± 28.8%, respectively. Effects of Dexamethasone on Calu-l NEP-specific mRNA after HOCI Exposure The effects of dexamethasone on NEP-specific mRNAs 24 h after HOCI exposure are shown in Figures 4 and 5. Although there was slight variation in the density of actin bands, dexamethasone did not affect the levels of ')'-actin mRNA. In different cell systems, this observation has been reported by others as well (16). In the presence of dexamethasone, cells unexposed to HOCI showed 120 and 270% increases in the densities of 3.5- and 6.5-kb NEP-specific mRNAs (Figure 4, lanes 1 and 2), respectively. By comparison, the HOClexposed cells treated with dexamethasone exhibited 44 and 50% increases, respectively, in these two species of NEPspecific mRNAs (Figure 4, lanes 3 and 4).

Discussion As in normal respiratory epithelial cells from other species in our experience (1, 4), plasma membranes of the Calu-l cell line are rich in NEP activity. HOCI exposure of these cells in situ leads to a substantial decrease in their NEP activity. This effect occurs within 5 min after oxidant exposure although cell viability within 30 min is not affected (8). In the present study, we found that this decrease in NEP activity caused by HOCI can be prevented by pretreating the cells with sodium azide at low temperature (4 0 C). Low temperature alone had no such effect. We also found that over the ensuing 24 h after HOCI exposure, there was a spontaneous recovery in NEP activity of Calu-l cells. Our data indicate that this recovery is not associated with an increase in synthesis of NEP-specific mRNA. We also found that recovery after acid exposure could be substantially accelerated by dexamethasone treatment of the cells. Dexamethasone stimulation of the spontaneous recovery in NEP activity postHOCI was associated with an increase in Calu-l NEP mRNA synthesis. This study confirms and extends our past report concerning the acute effects of HOCI exposure in situ onNEP activity of this respiratory epithelial cell line (8). Studying cell sonicates as well as whole cells, we have previously observed that the decrease in cell surface NEP activity post-HOCI is accompanied by a commensurate increase in cytosolic NEP within 60 min after exposure. Several possibilities may explain this HOCI-induced decrease in cell surface NEP activity. HOCI could directly inactivate NEP, cleave it off the cell membrane, or cause its internalization by exposed cells. Because we have previously shown that up to 400 J.tM HOCI does not inactivate isolated NEP, and because there is no detectable NEP in the culture medium during and after HOCI exposure (8), we speculated that HOCI decreases Calu-l cell membrane NEP by prompting internalization of the enzyme. The present studies of this HOCI effect done in the presence of sodium azide at low temperature support this speculation.

Lang and Murlas: Postoxidant Epithelial NEP Reconstitution Stimulated by Corticoids

Incubation in sodium azide at low temperature (4 0 C) has previously been used by other investigators to block endocytosis of NEP in both human neutrophils (9) and a human lymphoblastic leukemia cell line (10). As we have found in Calu-l cells, low temperature incubation alone does not affect endocytosis of NEP in human neutrophils or leukemic cells. In this study, we also found that if Calu-l cells are removed from HOCI after exposure and grown in fresh culture medium, they can reconstitute their NEP activity spontaneously over the ensuing 24 h. Several possibilities may explain this phenomenon. Perhaps internalized NEP is degraded by lysosomal proteolytic enzymes as has been shown in phorbol myristate acetate-activated neutrophils (9). It appears that cells can sense this loss through an as yet unknown signal and begin to synthesize NEP de novo. On the other hand, internalized NEP may not be degraded by cells, and it may be exocytosed after the oxidant exposure. Alternatively, it may be that both sets of processes occur and are in dynamic equilibrium. This appears to be the case in human non-T, non-B leukemic (NALM-6) cells in which the synthesis of CALLA (which has since been shown to be identical to NEP [14]) continues despite modulation of its expression on the cell surface by specific antibody (17). In these leukemic cells, Pesando and colleagues (17, 18) demonstrated that antibodymodulated cells rapidly reexpress CALLA in antibody-free media. In this setting, the antigenic determinant recognized by anti-CALLA monoclonal antibody remains intact during modulation, although the internalized enzyme is slowly degraded. By comparison, we have found in Calu-l cells that NEP activity fully recovers after HOCI exposure. During this spontaneous recovery, we could not detect changes in NEP-specific mRNA. In other words, stimulation for transcription of new NEP by Calu-l cells did not appear to be involved in this process. This suggests to us, therefore, that NEP internalized by Calu-l cells post-HOCI is not degraded and that it is reconstituted by exocytosis to the cell surface during spontaneous recovery. We have previously reported that dexamethasone increases both NEP activity and its steady-state mRNA in Calu-l cells in a time- and dose-dependent manner (19). Such an effect of dexamethasone has been demonstrated in numerous other cell types (20-23). In this study, we found that dexamethasone stimulates the spontaneous recovery of all NEP activity post-HOC!. In the presence of dexamethasone, NEP activity recovered to preexposure levels 6 h earlier than in its absence. As indicated by Northern blot analysis, this acceleration of recovery was linked temporally to an increase in transcription of the NEP gene. Such a response indicates that Calu-l cells after HOCI exposure are capable of transcribing new NEP as well as surviving (8) and reconstituting plasma membrane (NEP). In P388D1 cells, HOCI damage is more severe (5). These cells swell and lose intracellular K+ after exposure to 10 ~M HOCl. By comparison, Calu-l cells appear to be relatively resistant to HOCl. Nonetheless, the amount ofmRNA produced by Calu-l cells in response to glucocorticoids post-HOCI is less than that seen in unexposed cells, and this may be of considerable consequence. The long-term impact of this HOCI effect remains to be fully evaluated. As other investigators have found in rat kidney epithelial

305

cells (24) and human fibroblasts (14), we detected a 6.5-kb mRNA in Calu-l cells along with a 3.5-kb species. It is curious that in a recently reported SV-40transformed human airway epithelial cell line, this 6.5-kb species is either scant or not present normally (23). This SV-40 transformed cell line is also different in that it requires 6 days of I ~M dexamethasone to increase its NEP activity to the same degree that Calu-l cells generate in 48 h. Furthermore, SV-40 transformed cells respond to 6 days of glucocorticoids by generating large amounts of 6.5-kb mRNA (23). At least the latter differences between Calu-l and SV-40 transformed epithelial cell lines may represent cell-specific characteristics and/or differences in quantitating Northern blot results. By normalizing our results to ,,-actin mRNA, which dexamethasone does not affect in our experience, we corrected for variations in amounts of RNA loaded per lane that normally can occur when performing Northern blot analyses. It is also possible, however, that the transformation of human airway epithelial cells by SV-40 virus may alter NEP gene expression itself. In COS-I cells, for example, it has been shown that transfection with plasmids containing the SV-40 promoter and NEP eDNA results in expression of NEP driven by that promoter (25). Our findings lead us to speculate that HOCI exposure causes changes in the structure of NEP residing on the surfaces of airway epithelial cells. These cells appear to sense that damage as do NALM-6 cells (10, 17, 18,26). Oxidation of certain enzyme determinants, such as the sulfhydryl groups of cysteine and methionine (5), may cause such changes. These changes may not inactivate the enzyme, but do stimulate cells to initially endocytose it and thereafter reconstitute it on the plasma membrane post-oxidation. Analogous processes in cells exposed to an immune rather than oxidant stimulus have been reported (10, 17, 18, 26). Although this recovery does not appear to involve transcription, it can be accelerated by glucocorticoids that clearly do increase transcription of the NEP gene in airway epithelial cells. Because of their potential importance to an improved understanding of the pathogenesis and treatment of various airway diseases (27), the regulation and ultimate fate ofNEP oxidized on the respiratory muscosal surface are subjects meriting further investigation. Acknowledgments: This work was supported by Grants HL-01965, HL-3422804, and OH-00060 from the National Institutes of Health and by Grant 1599 from the Council for Tobacco Research USA, Inc. C. Murlas was supported in part by a Research Career Development Awardfrom the National Heart, Lung and Blood Institute.

References 1. Erdos, E. G., and R. A. Skidgel. 1989. Neutral endopeptidase 24.11 (enkephalinase) and related regulators of peptide hormones. FASEB J. 3:145-151. 2. Shipp, M. A., G. E. Tarr, C.-Y. Chenetal. 1991. CDIO/neutralendopeptidase 24.11 hydrolyzes bombesin-like peptides and regulates the growth of small cell carcinomas of the lung. Proc. Natl. Acad. Sci. USA 88:10662-10666. 3. Murlas, C. G., T. P. Murphy, and Z. Lang. 1990. HOCI causes airway substance P hyperresponsiveness and neutral endopeptidase hypoactivity. Am. J. Physiol. 258:L361-L368. 4. Lang, Z., and C. G. Murlas. 1991. HOCI exposure ofahuman airway epithelial cell line decreases its plasma membrane neutral endopeptidase. Lung 169:311-324. 5. Drozdz, R., N. J. Naskalski, andJ. Sznajd. 1988. Oxidation of amino acids and peptides in reaction with myeloperoxidase, chloride and hydrogen peroxide. Biochim. Biophys. Acta 957:47-52.

306

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 7 1992

6. Grisham, M. B., M. M. Jefferson, D. F. Melton, and E. L. Thomas. 1984. Chlorination of endogenous amines by isolated neutrophils. J. Bioi. Chem. 259:10404-10413. 7. Lang, Z., N. Soske1, G. Williams, and C. G. Murlas. 1989. Identification of metalloendopeptidase activity in bovine airway mucosal cells and its inhibition by hypochlorous acid. Am. Rev. Respir. Dis. 139:A236. (Abstr.) 8. Schraufstatter, 1. U., K. Browne, A. Harris et al. 1990. Mechanisms of hypochlorite injury of target cells. J. Clin. Invest. 85:554-562. 9. Erdos, E. G., B. Wagner, C. B. Harbury, R. G. Painter, R. A. Skidgel, and X.-G. Fa. 1989. Down-regulation and inactivation of neutral endopeptidase 24.11. (enkephalinase) in human neutrophils. J. Bioi. Chem. 264: 14519-14523. 10. Pulcuzynski, S., A. M. Boesen, O. M. Jensen, and P. Hokland. 1989. Immunoelectron microscopy studies on the modulation of common acute lymphoblastic leukemia antigen (CALLA) on Nalm-6 cells: delineation of intracellular transport. Leukemia 3:137-144. 11. Thomas, E. L., M. B. Grisham, and M. M. Jefferson. 1986. Preparation and characterization of chloramines. Methods Enzymol. 132:569-585. 12. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1989. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. 13. Tran-Paterson, R., H. F. Willard, and M. Letarte. 1989. The common acute lymphoblastic leukemia antigen (neutral endopeptidase-3.4.24.11) gene is located on human chromosome 3. Cancer Genet. Cytogenet. 42: 129-134. 14. Letarte, M., S. Vera, R. Tran et al. 1988. Common acute lymphocytic leukemia antigen is identical to neutral endopeptidase. J. Exp. Med. 168:1247-1253. 15. Chomczynski, P., and N. Sacchi. 1987. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162:156-159. 16. O'Reilly, M. A., J. C. Clark, andJ. A. Whitsett. 1991. Glucocorticoid enhances pulmonary surfactant protein B gene transcription. Am. J. Physiol. 260(Lung Cell. Mol. Physiol. 4):L37-L43.

17. Pesando, J. M., K. J. Tomaselli, H. Lazarus, and S. F. Schlossman. 1983. Distribution and modulation of a human leukemia-associated antigen (CALLA). J. Immunol. 131:2038-2045. 18. Pesando, J. M., 1. Ritz, H. Lazarus, K. J. Tomaselli, and S. F. Schlossman. 1981. Fate of a common acute lymphoblastic leukemia antigen during modulation by monoclonal antibody. J. Immunol. 126:540-544. 19. Lang, Z., and C. G. Murlas. 1992. Dexamethasone induces neutral endopeptidase expression in the human airway epithelial cell line Calu-1. Am. J. Physiol. In press. 20. Collins, D., M. G. Caron, and R. J. Lefkowitz. 1988.I3-adrenergic receptors in hamster smooth muscle cells are transcriptionally regulated by glucocorticoids. J. BioI. Chem. 263:9067-9070. 21. Schmid, E., W. Shmid, M. Jantzen, D. Mazer, B. Jastroff, and G. Schutz. 1987. Transcriptional activation of the tyrosine aminotransferase gene by glucocorticoids and cAMP in primary hepatocytes. Eur. J. Biochem. 165:499-506. 22. Nakabzyashi, J., K. Watanabe, A. Saito, A. Otsuru, K. Sawadaishi, and T. Tamaoki. 1989. Transcriptional regulation of o-fetoprotein expression by dexamethasone in human hepatoma cells. J. Bioi. Chem. 264:266-271. 23. Borson, D. B., and D. C. Gruenert. 1991. Glucocorticoids induce neutral endopeptidase in transformed human tracheal epithelial cells. Am. J. Physiol. 260(Lung Cell. Mol. Physiol. 4):L83-L89. 24. Malfroy, B., P. R. Schofield, w.-J. Kuang, P. H. Seeburg, A. J. Mason, and W. J. Henzel. 1987. Molecular cloning and amino acid sequence of rat enkephalinase. Biochem. Biophys. Res. Commun. 144:59-66. 25. Devault, A., C. Nault, M. Zollinger et al. 1988. Expression of neutral endopeptidase (enkephalinase) in heterologous COS-1 cells. J. Bioi. Chem. 263:4033-4040. 26. Ritz, J., J. M. Pesando, J. Notis-McConarty, and S. F. Schlossman. 1980. Modulation of common acute lymphoblastic leukemia antigen induced by monoclonal antibody in vitro. J. Immunol. 125:1506-1514. 27. Murlas, C. G. 1991. Environmental airway injury: mucosal changes and airway hyperreactivity. In The Airway Epithelium: Structure and Function in Health and Disease. S. Farmer, D. Hay, and C. Lenfant, editors. Marcel Dekker, New York. 213-254.