Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Respiratory aspiration of gastric contents occurs in patients with a variety of neurologic disorders, in patients under general anesthesia, and in elderly patients (1). Aspiration pneumonia is reported to develop in two pathophysiologically distinct conditions: acid aspiration-induced lung injury originally reported by Mendelson (5) and bacterial infection of the lungs after aspiration of oropharyngeal or gastric secretions (6). The main pathogenic factors that distinguish these two diseases are the acidity and the volume of aspiration. Massive aspiration with severe acidity (pH < 2.5) causes not only ciliated and nonciliated airway damage and airway hyperpermeability but also neutrophil accumulation and activation in the lung associated with proinflammatory cytokine production (7). On the other hand, silent aspiration is another cause that develops into bacterial infection of the lungs in elderly subjects (2). Swallowing and cough reflexes, important protective mechanisms of the lung against aspiration, are depressed in patients with stroke, and a high incidence of silent aspiration during sleep occurs in such patients (8, 9). Recurrent aspiration of small amounts of gastric contents plays a role in the pathogenesis of a chronic inflammation of bronchioles in the elderly population (10). These findings suggest that microaspiration of gastric and oropharyngeal secretions may affect defense mechanisms in the lung. However, in contrast to massive aspiration of gastric contents, the acidity of occult aspiration of gastric acid is rapidly neutralized after instillation (11), and a number of studies have demonstrated that acid aspiration with moderate acidity (pH 3.0-5.0) does not cause permeability edema associated with chemical injury of airway and lung (1, 12, 13). Furthermore, the effect of acid aspiration with moderate acidity (pH 3.0-5.0) on the innate defense mechanisms of the airway has not been well studied.

Defensins, one of the most intensively studied classes of antimicrobial peptides, are identified in a wide distribution of animals, including birds, rodents, ruminants, and humans (14). The main function of defensins is suggested to kill bacteria and fungi either on the surfaces of the epithelial cells or within phagolysosomes of phagocytes. Defensins are small cationic peptides containing arginine-rich 29-47 amino acids with three disulfide bands, which can be divided into the - and -defensin subfamilies in human subjects (14). The -defensins are produced by neutrophils and intestinal Paneth cells, whereas -defensins are produced mainly by epithelial cells. Of the -defensins, airway epithelial cells produce human -defensin (HBD)-1, HBD-2, and HBD-3 (14). mRNA for HBD-1 is constitutively expressed, whereas mRNA expression for HBD-2 and HBD-3 is induced by proinflammatory cytokines such as tumor necrosis factor (TNF)- and interleukin (IL)-1 and by gram-negative bacteria. Recent studies have demonstrated that human airway epithelial cells produce sodium-sensitive antimicrobial peptides into the apical side of surface liquid and that the bactericidal activity of HBDs decreases in cystic fibrosis, suggesting a major role of HBDs in the host defense against bacterial infections (15, 16). However, the effects of acid exposure on the bactericidal factors secreted by the human airway epithelium have not been well studied.

We therefore examined the bactericidal activity of airway surface liquid (ASL) and HBD productions in cultured human tracheal epithelial cells. Furthermore, we examined the effects of acid exposure on the productions of lactoferrin (LTF) and lysozyme (Lyz), because abundant antimicrobials including LTF and Lyz are suggested to be important for innate immunity in the ASL (20).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Media Components

Reagents for cell culture media were obtained as follows: Dulbecco's modified Eagle's medium (DMEM), Ham's F-12 medium, and fetal calf serum (FCS) were from GIBCO BRL (Life Technologies, Palo Alto, CA); dithiothreitol, Sigma type XIV protease, penicillin, streptomycin, gentamicin, and amphotericin B were from Sigma (St. Louis, MO); vitrogen solution was from Collagen (Palo Alto, CA); and Ultroser G serum substitute (USG) was from BioSepra (Marlborough, MA).

Human Tracheal Epithelial Cell Culture

The tracheae were obtained 3 to 6 h after death from 20 patients without overt pulmonary disease (mean age, 65 ± 4 yr; range, 52- 78 yr) under a protocol passed by the Tohoku University Ethics Committee. Isolation and culture of human tracheal surface epithelial cells were performed as previously described (21). In brief, the surface epithelium was scored into longitudinal strips and pulled off the submucosa. The tracheal surface epithelial cells were isolated by digestion with protease (0.4 mg/ml; Sigma type XIV) dissolved in phosphate-buffered saline (PBS) at 4°C overnight. The cells were pelleted (200 × g for 10 min) and suspended in a mixture of DMEM-Ham's F-12 medium containing 5% FCS (50/50, vol/vol). Cell counts were made with a hemocytometer, and estimates of viability were done with trypan blue. The cells were plated at 10⁶ viable cells per cm² onto Millicell CM inserts (0.45-µm pore size and 0.6-cm² area; Millipore Products Division; Bedford, MA). This medium was replaced by DMEM- Ham's F-12 medium containing 2% USG on the first day after the cells were plated. Cells were grown with an air interface (i.e., no medium added to the apical surface). Cell culture medium was supplemented with penicillin (10⁵ U/liter), streptomycin (100 mg/ liter), gentamicin (50 mg/liter), and amphotericin B (2.5 mg/liter). Millicell inserts were coated with vitrogen gels. To make the vitrogen gels, 10-fold minimum essential medium, 0.1 N NaOH, and vitrogen solution (Collagen) were mixed at 4°C (10/10/80, vol/vol/vol). After mixing, 0.4 ml/cm² of this solution was added to the Millicell inserts. Vitrogen gels were formed by incubation at 37°C for 1 h and were used within 2 h of manufacture.

Preparation of ASL and Acid Exposure of the Apical Surface of Human Tracheal Epithelial Cells

Preparation of ASL from cultured human tracheal epithelial cells was performed as described elsewhere (15). On Days 6-8 of culture, the DMEM-Ham's F-12 medium containing 2% USG and antibiotics in the basolateral side was replaced by antibiotic-free medium five times over 48 h, and the apical surface was washed three times over 48 h with antibiotic-free PBS. Thereafter, the antibiotic-free DMEM-Ham's F-12 medium containing 2% USG was changed in the basolateral side at 2- to 3-d intervals. On Days 8-12 of culture, the crude ASL was collected by washing the apical surface of the cultured human tracheal epithelial cells with 350 µl of water. The collected solution was spun down at 200 × g for 10 min, and the supernatant was recovered as ASL. The volume, electrolyte concentration, and acidity of the recovered ASL were measured. Electrolyte concentration was measured using ion-selective electrodes (EA05U automated electrolyte analyzer; A and T, Morioka, Japan), and the acidity of ASL was measured with a pH meter (ISFET pH Meter KS723; Shindengen, Tokyo, Japan). In the preliminary experiments, we found that the recovered volume ranged from 240-260 µl, the sodium concentration ranged from 14-18 mEq/liter, the potassium concentration ranged from 0.6-0.8 mEq/liter, and the pH ranged from 7.6-7.8. Because ASL is reported to be salt sensitive (15), the crude ASL was 1.5 times diluted and adjusted at 40 mEq/liter of sodium concentration with saline and water to estimate the bactericidal activity of the ASL.

To study the effects of acid exposure on the bactericidal activity of ASL of human tracheal epithelial cells, the pH of PBS was adjusted to 3.0, 4.0, or 5.0 with HCl. The apical surface of the cultured human tracheal epithelial cells was exposed to 150 µl of acidic PBS for 10 min. The acidic PBS was then sucked off, and the apical surfaces of the epithelial cells were then rinsed once with fresh antibiotic-free DMEM-Ham's F-12 medium containing 2% USG and once with PBS. Culture medium in the basolateral side was also replaced with fresh antibiotic-free DMEM- Ham's F-12 medium containing 2% USG. The epithelial cells were further cultured for 24 h with 5% CO₂ at 37°C. After the 24-h incubation, ASL was prepared as described above. To examine the effect of acid exposure with a shorter duration, 5 min incubation in the acid solution were also performed.

In the preliminary experiments, we measured the content of gentamicin in ASL with fluorescence polarization immunoassay (22) to examine if antibiotics were left on the cultured cells. We found that ASL did not contain a significant amount of gentamicin (data not shown).

In the preliminary experiments, we found that ASL had no bactericidal activity when it was collected immediately after rinsing with fresh medium and PBS, suggesting that the rinsing could wash out bactericidal molecules on the apical surfaces. Therefore, observed bactericidal activities might be newly produced and secreted on the apical surface during the 24-h incubation after acid exposure and rinsing.

Bacterial Stocks and Quantification of Bacteria

Bacteria were prepared as previously described (16). Single colonies of Escherichia coli (HB101) were inoculated into LB broth and cultured overnight at 37°C. An aliquot of this culture was transferred into fresh LB broth and incubated for an additional 2-3 h at 37°C to obtain a mid-logarithmic phase of the bacteria. The E. coli were washed once in 10 mM PBS and twice in water. The concentrations of E. coli were quantified by measuring the absorbance at 620 nm with a spectrophotometer (U-2000; Hitachi, Tokyo, Japan). The concentrations of E. coli were also measured by counting the colonies in the LB agar plate (15). The suspension of E. coli was seeded on LB agar plates after serial dilution of the bacteria. The bacteria on agar plates were cultured for 16 h at 37°C, and the number of colonies was counted. The concentrations of E. coli were expressed as colony forming units (CFU) per ml. We used an average value of triplicate cultures from the same bacteria suspension. E. coli were then diluted in water to an appropriate density.

Ussing Chamber Study

To determine the effects of acid exposure on the function of tight junctions of cultured human tracheal epithelial cells, electrical resistance and short-circuit current of the epithelial cells were measured using Ussing chamber methods as described previously (21). For studies in the Ussing chamber, Millicell CM inserts with their attached cells without edge damage were mounted in a modified Ussing chamber (21). Experiments were performed in Krebs-Henseleit solution with the following composition (mM): 118 NaCl, 5.9 KCl, 2.5 CaCl₂, 1.2 MgSO₄, 1.2 NaH₂PO₄, 25.5 NaHCO₃, and 5.6 glucose. The solution was maintained at 37°C and aerated continuously by bubbling with a mixture of 95% O₂-5% CO₂ (pH 7.4). Transepithelial resistance (R), electrical conductance (G), and short-circuit current (Isc) were determined from the current produced by the fixed transepithelial potential difference (pulse width, 200 ms; intensity, 0.5 mV; frequency, 0.05 Hz).

Bactericidal Activity Assay of ASL

The bactericidal activity of ASL was examined as previously described (15). Suspension of 500 CFU of E. coli in 2 µl of water was mixed with 30 µl of ASL four times diluted in NaCl solution with 40 mEq/l of sodium concentration in 96-well microplates. After this solution containing E. coli and ASL was incubated for 2 h at 37°C and seeded on LB agar plates, the agar plates were incubated overnight at 37°C and the number of colonies on the plates was counted.

The bactericidal activity of ASL was examined with microdilution assays as described elsewhere (23, 24) with some modification. In brief, 30 µl of ASL was 2-fold diluted with NaCl solution with 40 mEq/liter of sodium concentration in 96-well plates. E. coli suspension (2 µl; 500 CFU) was added to the diluted ASL in 96-well plates and incubated for 2.5 h at 37°C. To stop the bactericidal activities of ASL by increasing the sodium concentration to 150 mEq/liter, 165 µl of LB broth was then added to each well of the 96-well plates and incubated for 3 h at 37°C. The bacterial growth was determined by observation of bacterial turbidity under inverted microscopy and by measuring the absorbance at 620 nm with a microplate reader (Labsystems Multiskan BICHROMATIC; Labsystems, Helsinki, Finland). The bactericidal activity of ASL was expressed as the maximum dilution ratio to inhibit E. coli growth.

To examine the concentration-response effects of sodium on the bactericidal activity of ASL, ASL was 6-fold diluted with NaCl solution and water with sodium concentrations ranging from 14-110 mEq/liter. After 2 µl of E. coli suspension (500 CFU) growing in the log phase was mixed with 30 µl of the diluted ASL with various sodium concentrations in 96-well plates, the solution was incubated for 2.5 h at 37°C. LB broth (150 µl) was then added to the solution in 96-well plates to inhibit bactericidal activity of ASL by increasing the sodium concentration up to 150 mEq/liter. The growth of E. coli was determined after the mixture was cultured for 3 h at 37°C.

To examine the effects of the duration of acid exposure on the bactericidal activity of ASL, the human tracheal epithelial cells were exposed to acid (pH 3.0) for either 5 or 10 min. To examine the effects of acidity of mildly and moderately acidic PBS on the bactericidal activities of ASL, the human tracheal epithelial cells were exposed to acidic PBS with pH 3.0, pH 4.0, or pH 5.0 for 10 min.

Detection of HBDs mRNA with Reverse Transcriptase- Polymerase Chain Reaction

Human tracheal epithelial cells cultured on Millicell CM inserts were lysed by the addition of RNAzol (0.2 ml/10⁶ cells; Biotecx, Houston, TX) at various times (0, 8, 24, or 48 h) after acid challenge and were transferred to 1.5 ml Eppendorf tubes. The cell homogenates were mixed with a 10% volume of chloroform, shaken vigorously for 15 s, placed on ice for 15 min, and centrifuged at 12,000 × g for 15 min at 4°C. The upper aqueous phase containing RNA was collected and mixed with an equal volume of isopropanol. Pellets of RNA were obtained by centrifugation at 12,000 × g for 15 min at 4°C, dissolved in water, and stored at 80°C before use.

The mRNA expressions of HBD-1, HBD-2, and -actin (as a housekeeping gene) were examined by reverse transcriptase-polymerase chain reaction (RT-PCR) as previously described (25, 26). Briefly, 2 µg of RNA from each aliquot of human tracheal epithelial cells were dissolved in a 100 µl buffer containing the reagents for RT reaction with the following composition: 50 mM tris (hydromethyl) aminomethane (Tris) · HCl (pH 8.3), 75 mM KCl, 3 mM MgCl₂, 10 mM dithiothreitol, 5 U/µl Moloney murine leukemia virus RT (GIBCO-BRL Life Technologies), 0.5 mM deoxynucleoside 5'-triphosphate (dNTP; Takara, Ohtsu, Japan), 1 U/µl RNase inhibitor (Promega, Madison, WI), and 5 µM random hexamers (Pharmacia Biotech, Uppsala, Sweden). The RT reaction was performed for 60 min at 37°C, followed by 95°C for 10 min. The resulting cDNA was frozen at 80°C until use in the PCR. For each sample, 5 µl of RT mixture was added to a 45-µl PCR mixture consisting of 10 mM dNTP and 1.25 U Taq polymerase (Takara). Primer pairs for HBD-1, HBD-2, or -actin were present at 2 ng/µl. The oligonucleotide sequences for HBD-1, HBD-2, and -actin were designed as follows: sense primer D1s (5'-CCTGAAATCCTGAGTGTTGC-3') and antisense primer D1a (5'-GCGTCATTTCTTCTGGTCAC-3') for HBD-1; sense primer D2s (5'-CCAGCCATCAGCCATGAGGGT-3') and antisense primer D2a (5'-GGAGCCCTTTCTGAATCCGCA-3') for HBD-2 (26); sense primer ACT1 (5'-CCTTCCTGGGCATGG AGTCCTGT-3') and antisense primer ACT2 (5'-GGAGCAA TGATCTTGATCTTCA-3') for -actin (27). The PCR was performed in an automated thermal cycler (MJ Research, Watertown, MA), and 10 µl of the reaction was removed at 30 cycles for each sample. Samples were separated on a 2% agarose gel (FMC BioProducts, Rockland, ME) and were stained for 30 min in 1 µg/ml ethidium bromide. The DNA bands were visualized on a UV illuminator and were photographed with type 667 positive/negative film (Polaroid, Cambridge, MA). Expressions of HBDs mRNA in human tracheal epithelial cells were examined before and at 8, 24, and 48 h after acid exposure (pH 3.0, 10 min).

Real-Time Quantitative RT-PCR for HBD-1

To quantify the mRNA of HBD-1 and -actin expression in the human tracheal epithelial cells after acid exposure, real-time quantitative RT-PCR, using the Taqman technology (Roche Molecular Diagnostic Systems), was performed as previously described (28). Taqman technology exploits the 5' 3' nucleolytic activity of AmpliTaq DNA polymerase (29, 30). In principle, the method uses a dual-labeled fluorogenic hybridization probe (a Taqman probe) that specifically anneals the template between the PCR primers. The probe contains a fluorescent reporter (6-carboxyfluorescein [FAM]) at the 5' end and a fluorescent quencher (6-carboxytetramethylrhodamine [TAMRA]) at the 3' end. According to the progression of PCR, the Taqman probe is degraded and releases the reporter, resulting in an increase in fluorescence emission. The use of a sequence detector (ABI PRISM 7700; Applied Biosystems, Foster, CA) allows measurement of the amplified product in a direct proportion to the increase in fluorescence emission continuously during PCR. In the present experiment, the reaction mixture for RT-PCR was prepared in a single buffer system without the addition of reagents or changing tubes between the RT reaction and PCR (28). Briefly, 100 ng of RNA dissolved in 10 µl of water from each aliquot of human tracheal epithelial cells were denatured at 90°C for 90 s. Each RNA sample (100 ng/10 µl of water) was mixed in 40 µl of buffer containing the following reagents for the one-step RT-PCR reaction: 50 mM KCl, 10 mM Tris-HCl (pH 8.3), 0.01 mM EDTA, 60 nM Passive Reference 1 (Applied Biosystems), 5 mM MgCl₂, 100 nM sense primer D1s, 100 nM antisense primer D1a, 0.3 mM deoxynucleoside triphosphate (Boehringer), 0.4 U/µl RNase inhibitor (Promega), 0.4 U/µl Moloney murine leukemia virus RT (Perkin Elmer), 0.0025 U/µl Taq Gold Polymerase (Perkin Elmer), and 100 nM Taqman probe D1T (5'-[FAM]CCAGTCGCCATGAGAACTTCCTACCT [TAMRA]-3'). Sequences of the PCR primer pair D1s and D1a used in these experiments were described in above. We used the program PrimerExpress (Applied Biosystems) to design the probe and primers according to the guidelines for the best performance of the PCR. The fragment of mRNA for HBD-1 was reverse transcribed into cDNA (30 min at 48°C) and amplified by PCR for 40 cycles (15 s at 95°C and 1 min at 60°C). Whole reactions of the RT-PCR and detection of the fluorescence emission signal for every PCR cycle were performed at the same time in a single tube in a sequence detector (ABI 7700). The minimum PCR cycle to detect the fluorescent signal was defined as the cycle threshold (C), which is predictive of the quantity of an input target fragment (30). The standard curve was obtained between the fluorescence emission signals and C by means of duplicated serial dilutions of the total RNA from human tracheal epithelial cells cultured in medium alone. Real-time quantitative RT-PCR for -actin was also performed using the same PCR products. Expression of HBD-1 mRNA was normalized to a constitutive expression of -actin mRNA.

Western Blot Analysis of HBDs

Western blot analysis of HBDs was performed as previously described (31). To perform Western blot analysis, 10 samples of ASL in each condition of acid exposure (pH 3.0, 4.0, or 5.0 for 10 min) were pooled. Cationic peptides from pooled ASL were extracted using weak cation exchange matrix MacroPrep CM (carboxymethyl) support (Bio-Rad Laboratories, Hercules, CA). Cationic peptides of CM extracts were separated by acid-urea-polyacrylamide gel electrophoresis (AU-PAGE) and then transferred to Immobilon-P polyvinylidene difluoride membranes (Millipore, Bedford, MA). The blots were subjected to immunoblot analysis according to the procedure for immunoblotting recommended by Santa Cruz Biotechnology (Santa Cruz, CA), which commercially provided the primary and secondary antibody for HBD-1 and HBD-2. The blots were blocked in 3% gelatin in TBS for 30 min and then incubated in 1:100 polyclonal goat anti-HBD-1 or HBD-2 (sc-10849 or sc-10854, respectively; Santa Cruz Biotechnology) as the primary antibody for 3 h. The blots were washed in TBS with 0.05% Tween 20 three times for 10 min each and then incubated in a 1:1,000 dilution of horseradish peroxidase conjugated donkey anti-goat IgG (sc-2056; Santa Cruz Biotechnology) as the secondary antibody for 3 h. Antibody detection was performed by standard ECL techniques as recommended by the manufacturer (Amersham Pharmacia Biotech, Buckinghamshire, England). Peptides of rHBD-1 and rHBD-2 were used as standards (Peptide Institute, Osaka, Japan).

Enzyme-Linked Immunosorbent Assays for Lactoferrin and Enzyme Activity Assay for Lysozyme

To examine the effects of acid exposure on the productions of antimicrobials LTF Lyz by cultured human tracheal epithelial cells, we performed specific enzyme-linked immunosorbent assays (ELISA) for LTF and enzyme activity assay for Lyz using ASL. The sensitivity of the assay for the LTF ELISA kit (Calbiochem-Novabiochem, Darmstadt, Germany) was 1.6 ng/ml. The enzyme activity for Lyz, which induces the lysis of Micrococcus lysodeikticus, is commonly measured in human body fluids by turbidimetric techniques (32). The sensitivity of the enzyme activity assay for Lyz was 1.0 µg/ml.

Statistical Analysis

Results are expressed as means ± SE. Statistical analysis was performed using a one-way ANOVA, and multiple comparisons were made using Bonferroni's method. Significance was accepted at P < 0.05; n is the number of donors from whom cultured epithelial cells were used.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effects of Acid Exposure on G

Baseline R and Isc of the cultured human tracheal epithelial cells were 155 ± 31  · cm² and 48 ± 3 µA/cm², respectively (n = 10). G was significantly increased by 10 min of acid exposure to the apical side of the epithelial cells at pH 1.8 (Figure 1). In contrast, 10 min of acid exposure failed to increase G at pH 2.8 and pH 4.0 (Figure 1).

View larger version (10K): [in this window] [in a new window]   Figure 1.   Transepithe-lial membrane conductance (G) of cultured epithelial cells 24 h after treatment with acidic PBS solution for 10 min at pH 1.8, 2.8, or 4.0. Results are reported as means ± SE from seven samples. A significant difference from treatment with control PBS (Control) is indicated by ***P < 0.001

Effects of Acid Exposure on the Bactericidal Activity of ASL

The number of E. coli, estimated by counting the colony on the LB plate after mixing with ASL, was 43 ± 8 CFU in human tracheal epithelial cells exposed to control PBS. Acid exposure to the apical side of the epithelial cells for either 5 or 10 min at pH 3.0 significantly increased the number of E. coli surviving after mixing with ASL (P < 0.05) (Figure 2A).

View larger version (12K): [in this window] [in a new window]   View larger version (18K): [in this window] [in a new window]   View larger version (12K): [in this window] [in a new window]   Figure 2.   (A) Growth of E. coli measured by counting the number of colonies on the LB plate in the ASL from the cultured human tracheal epithelial cells exposed to acidic PBS at pH 3.0 for either 5 or 10 min. Results are reported as means ± SE from three representative samples. Significant differences from treatment with control PBS (Control) are indicated by *P < 0.05. (B) Growth of E. coli measured by the absorbance of E. coli suspension at OD 620 nm with microdilution assays in the ASL from the cultured human tracheal epithelial cells exposed to acidic PBS at pH 3.0 for either 5 min (open triangles) or 10 min (open squares). Results are reported as means ± SE from three representative samples. Significant differences from treatment with control PBS (open circles) are indicated by *P < 0.05. (C) Growth of E. coli measured by counting the number of colonies on LB plate in the ASL from the cultured human tracheal epithelial cells exposed to acidic PBS at pH 3.0, 4.0, or 5.0 for 10 min. Results are reported as means ± SE from three representative samples. Significant differences from treatment with control PBS (Control) are indicated by *P < 0.05.

The minimum inhibitory dilution ratio of ASL with microdilution assays by measuring the absorbance at 620 nm was 1/24 in control epithelial cells. In contrast, acid exposure significantly inhibited the bactericidal activity of ASL. The inhibitory effects of acid on the bactericidal activity of ASL were dependent on duration of the exposure. The E. coli growth could be detected at ASL dilution ratios of 1/5 and 1/6 after 10 and 5 min of acid exposure at pH 3.0 compared with detection at control dilution ratio of 1/36 (Figure 2B). The minimum inhibitory dilution ratio of ASL increased to 1/3 after acid exposure at pH 3.0 for 10 min and more than 1/5 after acid exposure at pH 3.0 for 5 min, respectively (P < 0.05). Likewise, the inhibitory effects of acid on the bactericidal activity of ASL were dependent on the pH of acidic PBS, and the maximum effect was obtained at pH 3.0 when the epithelial cells were exposed to mild acidic PBS with pH 3.0, 4.0, or 5.0 (Figure 2C).

Salt-Sensitivity of Bactericidal Activity and Sodium Concentrations of ASL

The inhibitory effects of ASL on the growth of E. coli were dependent on the sodium concentrations as described previously (15) (Figure 3). At a 20 mEq/liter sodium concentration, ASL inhibited the growth of E. coli measured with the absorbance at 620 nm in all of human tracheal epithelial cells exposed to acidic PBS at pH 3.0 for either 5 or 10 min or exposed to control PBS (Figure 3). However, ASL with sodium concentrations of 80 mEq/liter or more failed to inhibit E. coli growth in all the epithelial cells exposed to acidic or control PBS.

View larger version (17K): [in this window] [in a new window]   Figure 3.   Growth of the E. coli suspension at OD 620 nm in the ASL at various sodium concentrations from the cultured human tracheal epithelial cells exposed to acidic PBS at pH 3.0 for either 5 min (open triangles) or 10 min (open squares) or to normal PBS (open circles). Results are reported as means ± SE from three representative samples.

Acid exposure did not change the sodium concentrations in ASL in the human tracheal epithelial cells. The sodium concentrations in ASL collected by washing with 350 µl of water were 16 ± 1 mEq/liter in the epithelial cells exposed to acid at pH 3.0 for 5 min, 16 ± 1 mEq/liter in the epithelial cells exposed to acid at pH 3.0 for 10 min, and 15 ± 1 mEq/liter in the control cells, when the epithelial cells made confluent cell sheets in Millicell-CM inserts (0.6-cm²) (P > 0.50; n = 7).

Effects of Acid Exposure on mRNA Expression of HBDs

The baseline expression of HBDs mRNA was constant. Neither smoking habits nor cause of death influenced the baseline expression of HBDs mRNA. Furthermore, the expression of HBD-1 mRNA was more abundant than that of HBD-2 in the epithelial cells cultured in medium alone (Figure 4A). HBD-1 mRNA expression was decreased 24 h after exposure of the cells to acid (pH 3.0, 10 min) and recovered to control levels at 48 h. The inhibitory effects of acid exposure on HBD-1 mRNA expression were dependent on pH, and strong inhibitory effects were observed at pH 3.0 and pH 4.0, whereas a product band was detectable at pH 5.0 (Figure 4B). In contrast, acid exposure did not change the expression of mRNA of HBD-2 and -actin at any time after acid exposure (Figure 4A) and at any pH of acidic PBS (Figure 4B).

View larger version (28K): [in this window] [in a new window]   Figure 4.   (A) Time course of mRNA expression of HBD-1 and HBD-2 in human tracheal epithelial cells after acid exposure (pH 3.0, 10 min) as detected by RT-PCR. -actin was used as a housekeeping gene. Data are representative of three different experiments. (B) mRNA expression of HBD-1 and HBD-2 in human tracheal epithelial cells 24 h after acid exposure at pH 3.0, 4.0, or 5.0 for 10 min as detected by RT-PCR. -actin was used as a housekeeping gene. Data are representative of three different experiments.

Real-Time Quantitative RT-PCR for HBD-1

Exposure of acidic PBS (pH 3.0, 10 min) consistently inhibited the amount of HBD-1 mRNA expression in the cultured human tracheal epithelial cells. mRNA extracted from the epithelial cells at 8 h after acid exposure revealed significant decreases in the amount of HBD-1 mRNA expression. The amount of HBD-1 mRNA expression decreased progressively until 24 h after acid exposure, and significant inhibition of HBD-1 mRNA expression was still observed at 48 h after acid exposure (Figure 5A). Likewise, the inhibitory effects of acid exposure on the amount of HBD-1 mRNA expression was dependent on the pH of acidic PBS, and maximum inhibition was observed at pH 3.0 (Figure 5B). Acid exposure did not change the amount of -actin mRNA expression (data not shown).

View larger version (13K): [in this window] [in a new window]   View larger version (13K): [in this window] [in a new window]   Figure 5.   (A) Time course of mRNA expression of HBD-1 in human tracheal epithelial cells 24 h after acid exposure (pH 3.0, 10 min) as detected by real-time quantitative RT-PCR. Results are expressed as relative amounts of mRNA expression (%) compared with the control values and are reported as means ± SE from three samples. Significant differences from treatment with control PBS (0, time 0) are indicated by ***P < 0.001. (B) HBD-1 mRNA expression in human tracheal epithelial cells 24 h after exposure to acidic PBS at pH 3.0, 4.0, or 5.0 for 10 min as detected by real-time quantitative RT-PCR. Results are expressed as relative amounts of mRNA expression (%) compared with the control values and are reported as means ± SE from three samples. Significant differences from exposure to control PBS (C, control) are indicated by *P < 0.05 and ***P < 0.001.

Effects of Acid Exposure on the Protein Level of HBD in ASL

The protein level of HBD-1 was more abundant than that of HBD-2 in ASL from the epithelial cells cultured in medium alone (Figures 6A and 6B). More than one band immunoreactive with the HBD-1 polyclonal anti-HBD-1 antibody was detected in lengths larger than 36 amino acids. HBD-1 protein level in ASL was decreased 24 h after acid exposure. The inhibitory effects of acid exposure on HBD-1 protein production were dependent on pH, and strong inhibitory effects were observed at pH 3.0 and 4.0 (Figure 6A). In contrast, the band of HBD-2 was very faint, and acid exposure did not change the protein expression of HBD-2 at any pH of acidic PBS (Figure 6B).

View larger version (33K): [in this window] [in a new window]   Figure 6.   (A) Western blot analysis demonstrating HBD-1 protein levels in ASL from human tracheal epithelial cells 24 h after exposure to acidic PBS at pH 3.0, 4.0, or 5.0 for 10 min. Recombinant HBD-1 peptide (15 ng of 36-amino-acid form) was loaded as a standard (Std). Data are representative of three different experiments. (B) Western blot analysis demonstrating HBD-2 protein levels in ASL from human tracheal epithelial cells 24 h after exposure to acidic PBS at pH 3.0, 4.0, or 5.0 for 10 min. Recombinant HBD-2 peptide (15 ng of 41-amino-acid form) was loaded as a standard (Std). Data are representative of three different experiments.

Effects of Acid Exposure on LTF Production and Lyz Activity in ASL

LTF protein level in ASL 24 h after acid exposure was significantly decreased. The inhibitory effects of acid exposure on LTF protein production were dependent on pH, and strong inhibitory effects were observed at pH 3.0 and 4.0 (Figure 7A). In contrast, acid exposure did not change the Lyz activity at any pH of acidic PBS (Figure 7B).

View larger version (13K): [in this window] [in a new window]   View larger version (12K): [in this window] [in a new window]   Figure 7.   (A) LTF protein level in ASL from human tracheal epithelial cells 24 h after exposure to acidic PBS at pH 3.0, 4.0, or 5.0 for 10 min as detected by specific ELISA. Results are expressed as means ± SE from five samples. Significant differences from exposure to control PBS (C, control) are indicated by **P < 0.01 and ***P < 0.001. (B) Lyz activity in ASL from human tracheal epithelial cells 24 h after exposure to acidic PBS at pH 3.0, 4.0, or 5.0 for 10 min as detected by enzyme activity assay. Results are expressed as means ± SE from five samples.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The present study suggests that acid exposure inhibits bactericidal activity of ASL from primary cultures of human tracheal epithelial cells. These conclusions are made on the basis of the observation that the bactericidal activity of ASL for E. coli decreased in cultured human tracheal epithelial cells after exposure to mildly to moderately acidic PBS, which did not alter electrical conductance. The inhibitory effects of acid on bactericidal activity were dependent on the pH of acidic PBS exposed to the epithelium and on the exposure periods. Furthermore, bactericidal activity of ASL from human tracheal epithelial cells was observed at low concentrations (20-40 mEq/liter) of sodium but not at higher concentrations of sodium ( 80 mEq/l), being consistent with sodium sensitivity of HBD (15, 16). The acid exposure decreased the expression of HBD-1 mRNA in human tracheal epithelial cells and the protein level of HBD-1 in ASL from the cells. In contrast, acid exposure did not affect the production of mRNA and protein of HBD-2 and mRNA expression of -actin. Furthermore, ASL did not contain significant content of antibiotics supplemented in the culture medium. These findings suggest that acid exposure may inhibit the production of bactericidal molecules including defensins in airway epithelial cells, thereby reducing the bactericidal activity of ASL.

Massive gastric contents with low pH damage the ciliated and nonciliated airway epithelial cells (1, 33). Furthermore, acid exposure with low pH (less than 3.0) induces conditions of airway inflammation such as hyperpermeability (1, 12, 33), airway edema with neutrophil activation (34), and IL-8 production (36). Therefore, impairment of mucociliary transport by massive aspired gastric contents, as observed in Mendelson's syndrome, is suggested to cause the colonization of microorganisms in airways (37). In contrast, gastric content with pH higher than 3.0 does not induce chemical injury of airways and lung parenchyma (1, 12), and hyperpermeability of human tracheal epithelial cells (33). However, microaspiration of gastric and oropharyngeal secretions also induces pneumonia (2, 5). Because the pH of microaspiration contents containing gastric and oropharyngeal secretions may be high (pH > 3.0) because of the dilution with oropharyngeal secretions or the buffering by the environment of local tissue (11), this may not cause mucociliary damage (1, 12, 33). However, acidic PBS with pH more than 3.0 affected the production of HBD-1 and reduced the ASL bactericidal activity in human tracheal epithelial cells in the present study. Therefore, the impaired bactericidal activity associated with a reduced production of HBD-1 may relate to the pneumonia induced by microaspiration of gastric and oropharyngeal secretions.

To study the effect of acid stimulation on the antimicrobial activity in ASL, we used a primary culture of human tracheal epithelial cells with an air interface and exposed acidic PBS adjusted to pH 3.0-5.0. Under this culture condition, the human tracheal epithelial cells retain the ultrastructures and ion transport properties of the original tissue (21). The cultured epithelial cells were demonstrated not to be contaminated with submucosal gland cells, fibroblasts, or macrophages (21), and the electrical conductance did not change after exposure to an acidic pH (higher than 3.0) as shown by Ohrui and colleagues (33). Likewise, the sodium concentration in ASL after exposure to acidic PBS did not differ from the concentration in ASL after exposure to control PBS. Furthermore, the DMEM-Ham's F-12 medium containing 2% USG did not have bactericidal activity. Therefore, ASL secreted from human tracheal epithelial cells may be a reason for the bactericidal activity of epithelial cells.

The present study suggests that acid exposure may have an inhibitory effect on antimicrobial activity in ASL derived from cultured airway epithelial cells. This antimicrobial activity was sodium sensitive and consistent with a previous report (15) using human tracheal epithelial cells cultured with the same methods (21). The productions of mRNA and protein of HBD-1 were reduced by acid exposure. In contrast, acid exposure did not affect the production of mRNA and protein of HBD-2 and -actin mRNA. In the present study, acidic PBS was sucked off after acid exposure, and the apical surface of cultured human tracheal epithelial cells was rinsed with fresh medium and PBS. Furthermore, ASL was collected during the 24-h incubation after acid exposure and rinsing. Therefore, ASL might not contain acidic PBS, which directly inactivates antimicrobial proteins and peptides. In the preliminary experiments, we found that ASL had no bactericidal activity when collected immediately after rinsing with fresh medium and PBS, suggesting that the rinsing could wash out bactericidal molecules on the apical surfaces. Therefore, observed bactericidal activities might be newly produced and secreted on the apical surface during the 24-h incubation after acid exposure and rinsing.

Airway epithelial cells secrete various bactericidal factors other than defensins such as Lyz, immunoglobulins, and LTF (20). Lyz has antimicrobial activity against some gram-positive bacteria as well as fungi (38), and LTF has antibacterial effects against gram-negative bacteria by destabilizing their outer membrane (39). Although acid exposure significantly reduced the production of LTF without changes in the Lyz activity, the levels of LTF in the ASL were 1-1.5 ng/ml in the present study and much lower than those in pulmonary secretions (40). Furthermore, antimicrobial activity of the ASL was sodium sensitive and consistent with that of HBDs (15, 16). Therefore, HBDs, rather than LTF, may be responsible for antimicrobial activity of the ASL observed in the present study.

The RT-PCR analysis on the mRNA expression for HBDs revealed that the transcript level for HBD-1, but not HBD-2, is transiently reduced by acid treatment. We confirmed the time course of HBD-1 mRNA expression after acid exposure by real-time quantitative RT-PCR for HBD-1. In contrast, acid exposure did not change the mRNA expression of HBD-2. These effects of acid exposure on the expression of HBDs were also confirmed by the protein analysis of HBDs. Because the genomic HBD-1 sequence does not contain transcription factor regulatory elements for NF-B and AP-1 (41), HBD-1 is suggested to be constitutively produced and is not transcriptionally regulated by inflammatory agents (42). On the other hand, the genomic HBD-2 sequence has NF-B and NF IL-6 consensus binding sites in its 5' flanking region (43), and the HBD-2 gene expression is inducible by various proinflammatory agents such as TNF-, IL-1, and microorganisms, including bacteria (44, 45). Furthermore, Singh and colleagues (46) have demonstrated that HBD-1 was found in bronchoalveolar lavage fluid (BALF) from normal subjects as well as from patients with cystic fibrosis (CF) and inflammatory lung disease, whereas HDB-2 was detected in BALF from patients with CF or inflammatory lung disease but not in normal subjects. These findings suggest that HBD-1 may be constitutively expressed and serve as a defense in the lung, whereas HBD-2 expression is induced by stimulation such as airway inflammation. Although HBD-2 mRNA was expressed in the human tracheal epithelial cells, the levels of mRNA and protein of HBD-2 were low compared with those of HBD-1. Therefore, the production of HBD-2 might not be sufficient to have bactericidal activity in the present study.

Western blot analysis in the present study showed more than one band immunoreactive with the HBD-1 polyclonal anti-HBD-1 antibody in lengths larger than 36 amino acids. Previous studies revealed three forms of HBD-1 in human airway epithelia (46) and four forms of HBD-1 in the urogenital tract (31), ranging in length from 36 to 47 amino acids. Therefore, the results in the present study were consistent with those of the previous studies and suggest that human tracheal epithelial cells may produce the higher molecular weight forms of HBD-1 than 36 amino acids.

There are several reports on the regulation of gene expression by environmental pH. Acidic conditions lead to transient increases in c-fos, c-jun, junB, and egr-1 mRNA followed by an increase in Na/H antiporter mRNA in MCT cells, an SV40-transformed mouse proximal tubule cell line (47, 48). Furthermore, acid incubation of LLC-PK1-F+ cells, a porcine proximal tubule cell line, increases the expression of ammoniagenic enzymes mRNA (49). The promoter of the ammoniagenic enzyme mRNA has the hepatic nuclear factor-1 (HNF-1) recognition motif required for the acid response (50). In contrast, mRNA expressions of egr-1 and type 1 collagen in murine osteoblasts are inhibited by acidosis and increased by alkalosis in a physiological range (pH 6.8-7.6), although mRNA expression of c-fos, c-jun, junB, and junD are not modulated by ambient pH (51). Thus, the effects of acid exposure on cell functions may differ among cell species. Further studies are needed to clarify the mechanisms.

In conclusion, the results of the present study may give a new insight into the development of bacterial infection of lungs associated with repeated silent aspiration in elderly persons. Because a small amount of gastric acid aspiration can be buffered in the environment of local tissue (11), it may not cause chemical pneumonitis with permeability edema. However, acidic conditions, with a relatively high pH, inhibited bactericidal activity of ASL from human tracheal epithelial cells by reducing HBD-1 production in the present study. Therefore, frequent exposure to acid, even with moderate acidity, may inhibit the innate immunity of the airway epithelial cells and may increase the susceptibility to bacterial infection, thereby resulting in the development of pneumonia.