American Journal of Respiratory Cell and Molecular Biology

Cultured lung epithelial cells release antibacterial activity upon contact with Pseudomonas aeruginosa (PA), which is impaired in cystic fibrosis (CF). In order to identify the factors responsible for killing PA by a biochemical approach, we purified antimicrobial activity from supernatants of the A549 lung epithelial cell line, previously stimulated with PA bacteria, by subsequent high performance liquid chromatography. NH2-terminal sequencing of a major bactericidal compound revealed it to be identical with human beta-defensin-2 (hBD-2). A mucoid phenotype of PA, but not two nonmucoid PA strains, high concentrations ( > 10 μ g/ml) of PA lipopolysaccharide, tumor necrosis factor α , and interleukin (IL)-1 β , but not IL-6, dose-dependently induced hBD-2 messenger RNA in cultured normal bronchial, tracheal, as well as normal and CF-derived nasal epithelial cells. Genomic analysis of hBD-2 revealed a promoter region containing several putative transcription factor binding sites, including nuclear factor (NF) κ B, activator protein (AP)-1, AP-2, and NF-IL-6, known to be involved in the regulation of inflammatory responses. Thus, hBD-2 represents a major inducible antimicrobial factor released by airway epithelial cells either on contact with mucoid PA or by endogenously produced primary cytokines. Therefore, it might be important in lung infections caused by mucoid PA, including those seen in patients with CF.

Certain studies suggest that invertebrate epithelial cells can recognize microorganisms and mount a fast defense response through the production of various inducible antibiotic peptides. This leads to a characteristic broad spectrum of antimicrobial activity against bacteria and fungi (1, 2). The recent discovery of two closely related bovine airway epithelia–derived antibiotic peptides, tracheal antimicrobial peptide (TAP) and lingual antimicrobial peptide (LAP), gave rise to the speculation that these vertebrate peptide antibiotics, which are induced by contact with bacteria, are important compounds of the respiratory innate immune system that helps to keep cattle airways sterile (3, 4).

It was previously reported that normal human airway epithelium has the ability to release antimicrobial activity against Pseudomonas aeruginosa, whereas airway epithelial cells from patients with cystic fibrosis (CF) failed to display such activity (5, 6). Other studies revealed that apart from skin, normal human airway epithelia express messenger RNA (mRNA) for the inducible human beta-defensin (hBD) 2 (7, 8), which is strongly induced in the respiratory epithelia of patients with inflammatory lung diseases and CF (9).

To determine whether human respiratory tract epithelial cells release antimicrobial peptide(s) upon direct contact with gram-negative bacteria, we sought to identify and characterize such peptides and the conditions of their production, and to investigate whether the epithelia of patients with CF showed a different behavior.

Culture and Stimulation of Airway Epithelial Cells

Airway epithelial cells were isolated from nasal, tracheal, and bronchial epithelial mucosal tissue, which was obtained from patients (including those with CF) undergoing surgery for therapeutic reasons. Briefly, mucosae were dissected into small strips and then incubated at 37°C for 90 min in keratinocyte growth medium (KGM) (Clonetics, San Diego, CA) containing 0.1% trypsin. After filtering the cell suspension, epithelial cells were grown to confluence in KGM.

Second passage cultures of primary nasal, tracheal, and bronchial cells cultured for 24 h in KGM lacking bovine pituitary extract were stimulated with heat-killed P. aeruginosa bacteria (clinical isolate obtained from a patient suffering from leg ulcers at the Department of Dermatology, University of Kiel, Kiel, Germany; clone PA-O, mucoid phenotype), two nonmucoid forms (no. 15442 and 27853; American Type Culture Collection [ATCC], Manassas, VA), various recombinant cytokines (Pepro-Tech Inc., London, UK), or P. aeruginosa–derived lipopolysaccharide (LPS) (serotype 10; Sigma, Munich, Germany). The human A549 lung epithelial cells used for purification of antimicrobial activity were cultured in minimal essential medium (MEM) (BioConcept, Freiburg, Germany) containing 10% fetal calf serum (FCS). Before stimulation, cells were grown for 24 h in FCS-free medium and then were incubated with heat-killed (65°C, 30 min) mucoid P. aeruginosa (clone PA-O) in FCS-free MEM (20 ml/175 cm2 flask; bacteria density: 108/ml; bacteria-to-cell ratio: 200:1). Stimulation was carried out for 48 h, and 1.5 liters of each supernatant (derived from approximately 1 × 109 A549 cells) were used for the purification of the antimicrobial peptides.

Purification of Antimicrobial Peptides from Cell Culture Supernatant

Supernatants (1.5 liters) from A549 lung epithelial cells treated for 48 h with heat-inactivated mucoid P. aeruginosa were centrifuged (6,000 × g, 30 min), concentrated using Amicon YM3 filters (Amicon Corp., Danvers, MA), diafiltered against sodium phosphate buffer (10 mM, pH 7.4), and then applied to a P. aeru-ginosa affinity column to selectively trap antibiotic peptides acting against P. aeruginosa.

The P. aeruginosa affinity column was prepared using an N-hydroxy-succinimide (NHS)–activated sepharose column (HiTrap, 5 ml; Pharmacia, Uppsala, Sweden), which had been washed three times with 10 ml ice-cold (1 mM) HCl. Subsequently, 5 ml of mucoid P. aeruginosa suspension (109 bacteria/ml) (clone PA-O) were added to this preactivated column with a flow rate of 1 ml/ min. Application of the bacteria to the column was performed five times to increase the coupling yield of the bacteria. Coupling efficiency was measured by determination of the absorbance of the coupling solution before and after application to the column. After 30 min incubation at room temperature, the column was deactivated by washing three times with 0.5 M aqueous ethanolamine containing 0.5 M NaCl, pH 8.3, followed by 0.1 M sodium acetate containing 0.5 M NaCl, pH 4. Subsequently, the cell culture supernatant was applied to the affinity column that had been previously equilibrated with 10 mM phosphate buffer, pH 7.4. Bound material was eluted using 5 ml of 0.1 M glycine buffer, pH 3.0, containing 1 M NaCl at a flow rate of 1 ml/min. After equilibration of the column with 10 mM phosphate buffer, pH 7.4, the effluent was applied to the column and bound material was eluted as described previously. This step was performed four times to increase the efficacy of the column to bind peptides. The eluate was concentrated using Amicon filters as described previously and diafiltered against 0.1% trifluoroacetic acid (TFA) for subsequent reversed-phase (RP) high performance liquid chromatography (HPLC).

P. aeruginosa affinity column bound material was purified on a preparative wide-pore, reversed-phase (RP-8) HPLC column (300 × 7 mm, C8 Nucleosil, 250 × 12.6 mm; Macherey and Nagel, Düren, Germany) that had been previously equilibrated with 0.1% (vol/vol) TFA in HPLC-grade water containing 20% (vol/ vol) acetonitrile. Proteins were eluted with a gradient of increasing concentrations of acetonitrile containing 0.1% (vol/vol) TFA (flow: 2 ml/min). Aliquots (10 to 30 μl) of each fraction were lyophilized, dissolved in 5 μl 0.1% (vol/vol) aqueous acetic acid, and tested for antimicrobial activity against P. aeruginosa or Escherichia coli. Fractions containing antimicrobial activity against P. aeruginosa were lyophilized, dissolved in 50 mmol/liter ammonium formate, pH 4.0, containing 25% (vol/vol) acetonitrile, and then applied to a micro-Mono S HPLC column (Mono S PC 1.65, 2.1 × 100 mm, 5-μm particle size; Pharmacia), attached to a Smart-Micro-HPLC apparatus (Pharmacia), and equilibrated with the same buffer. Proteins were eluted with a gradient of increasing concentrations of NaCl (maximum: 1 M NaCl) in equilibration buffer using a flow rate of 100 μl/min. Aliquots (1 μl) of each HPLC fraction were diluted with 4 μl 0.1% (vol/vol) acetic acid and tested for antimicrobial activity. Fractions containing antimicrobial activity (which eluted with 0.43 M NaCl) were directly applied to a wide-pore micro RP-18 HPLC column (Sephasil C18, 2.1 × 100 mm, 5-μm particle size; Pharmacia) that had been previously equilibrated with 0.1% (vol/vol) aqueous TFA. Samples were eluted with an increased gradient of acetonitrile similar to the procedures used for purification of eotaxin (10).

Molecular mass estimation was performed using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) in the presence of 8 M urea and tricine (11) as described for chemokines (12) under nonreducing conditions (11). Proteins were visualized by silver staining (12). N-terminal amino acid sequence analysis was done by automated Edman degradation and HPLC determination of the thiohydantoin derivates was done using a 776 automated sequencer (Applied Biosystems, Weiterstadt, Germany). Peptide mapping using endopeptidase Lys C (Boehringer Mannheim, Mannheim, Germany) was performed as described for eotaxin (10).

Antimicrobial Assays

Antimicrobial activity was either determined using the radial diffusion agarose assay system or the microdilution assay system (13). Briefly hBD-2 (purified natural, 41 residues containing material, obtained in part from lung epithelial cells as well as keratinocytes) was incubated in 100 μl 10 mM sodium phosphate buffer (pH 7.4) with 0.03% (wt/vol) trypticase soy broth (TSB) containing approximately 1 × 106 bacteria/ml log phase microorganisms. Incubation was performed for 3 h at 37°C in 96-well microtiter plates (Falcon, Nuembrecht, Germany) with gentle shaking (150 rpm). The antibiotic activity of hBD-2 was analyzed by plating serial dilutions of the incubation mixture and determination of the colony-forming units (CFU) the following day.

Semiquantitative Reverse Transcriptase/Polymerase Chain Reaction

Epithelial cells were washed twice with phosphate-buffered saline and total RNA was isolated using the TRIzol reagent (GIBCO BRL, Eggenstein, Germany). Total RNA (2 μg) was reverse transcribed using standard reagents (GIBCO BRL). The complementary DNA (cDNA) corresponding to 50 ng RNA served as a template in a duplex polymerase chain reaction (PCR) containing 0.08 μM of hBD-2 specific intron spanning primers (forward primer: 5′-CCAGCCATCAGCCATGAGGGT-3′; reverse primer: 5′-GGAGGCCTTTCTGAATCCGCA-3′) with an internal control for equal amounts of cDNA 0.1 μM of a glyceraldehyde-3-phosphate dehydrogenase (GAPDH)–specific intron spanning primer pair (5′-CCAGCCGAGCCACATCGCTC-3′; 5′-ATGAGCCCCAGCCTTCTCCAT-3′), which yielded a 360-bp amplified product. Amplification was performed using 25 to 33 cycles with denaturation at 94°C for 1 min, primer annealing at 60°C for 30 s, and extension at 72°C for 2 min. PCR products were subjected to electrophoresis on a 2% agarose gel and visualized by ethidium bromide staining. Specificity of hBD-2 encoding PCR products was verified by sequencing.

Cloning of hBD-2 cDNA from Airway Epithelial Cells

Total RNA obtained from P. aeruginosa–stimulated primary human tracheal or bronchial epithelial cells from patients with and without CF was reverse transcribed using standard reagents (GIBCO BRL). Specific primers (5′-ACCTTCTAGGGCAAAAGACTG-3′ and 5′-GGAGCCATATGTCATCCAGTC-3′) were designed from our previously obtained human skin hBD-2 cDNA sequence (accession number Z71389, Genbank/EMBL database) to isolate the airway epithelial hBD-2 cDNA using an inverse PCR protocol (14, 15) and the 5′ rapid amplification of cDNA ends (RACE) system (GIBCO BRL), according to the manufacturer's protocol.

PCR products were cloned into a pUAg Vector (R&D Systems, Wiesbaden, Germany) and subsequently sequenced.

Analyses of hBD-2 Genomic DNA

Genomic DNA was amplified as previously described (16) using genomic human lung fibroblast DNA from a DNA library cloned into a Lamda Fix II Vector (Stratagene, Amsterdam, The Netherlands) as template, modified T7/T3 primers (T3-mod: 5′-AATTAACCTCACTAAAGAG-3′; T7-mod: 5′-CTCTAATCAGACTCACTATAGGGC-3′) and gene-specific primers based on the hBD-2 cDNA sequence (UR-1: 5′-ACCTTCTAGGGCAAAAGACTG-3′ for the 5′ upstream region; DR-1: 5′-GGAGCCATATGTCATCCAGTC-3′ for the 3′ downstream region). Amplification products that were identified to be positive by nested PCR were cloned into pUAg Vector (R&D Systems) and subsequently sequenced. Internal sequences were determined using oligonucleotides based on initial sequence data. Six clones were analyzed and a complete genomic sequence of 3851 bp was reconstructed (the complete hBD-2 genomic nucleotide sequence is available from the Genbank/EMBL database, accession number AJ000152). The 5′ promotor region upstream of the hBD-2 coding region was analyzed and compared for potential transcription factors with EMBL sequence database entries using the HUSAR/GCG sequence analysis programs (http://genome. as well as the Transfac transcription factor database (17).

Airway Epithelial Cells Release a Potent Antibiotic Peptide upon Contact with Mucoid P. aeruginosa

To investigate whether epithelial cells secrete antimicrobial compounds that kill P. aeruginosa on contact, we analyzed supernatants of cultured nasal epithelial cells as well as A549 lung epithelial cells for the presence of such compounds. Initial experiments revealed that a mucoid form of P. aeruginosa induced the release of proteinaceous P. aeruginosa, killing activity in both cell types.

To identify the P. aeruginosa killing compounds, we used a P. aeruginosa–coated affinity column to trap the antimicrobial peptides. As a result, we found that the majority of P. aeruginosa killing material, produced by both cell types, bound to the column. Further, RP-8 HPLC of A549 epithelial cell–derived material revealed one major peak containing the highest activity that eluted in an acetonitrile gradient at 42% acetonitrile (Figure 1A).

Final purification by microcation exchange (Mono S)- HPLC (data not shown) and micro-reversed phase (RP-18) HPLC (Figure 1B) revealed a 4-kD polypeptide when electrophoretic analysis was performed in the presence of urea and tricine (Figure 1B, inset). Amino acid sequencing of 25 residues revealed a strong similarity to that of hBD-2.

When A549 epithelial cell–derived hBD-2 was compared with authentic, natural skin–derived (41 residues) hBD-2, no difference in the mobility on tricine/SDS-PAGE analyses was seen (data not shown). Approximately 10 μg of pure hBD-2 was isolated from 1.5-liter culture supernatants corresponding to 109 A549 cells.

Molecular Cloning of the cDNA Encoding hBD-2 from Human Airway Epithelial Cells

Because multiple β-defensins differing only in a few amino acid residues exist in leukocytes of cattle, chickens, and turkeys (13, 18, 19), we cloned the hBD-2 cDNA from human tracheal (TEC) and bronchial epithelial cells (BEC). Analyses of 11 clones (six TEC and five BEC) revealed that all had sequences identical to the hBD-2 cDNA sequence that we had previously isolated from keratinocytes (Genbank/EMBL database accession number Z71389). No clones were obtained having similar sequences, indicating that the number of closely related sequences to hBD-2 (with sufficient nucleotide homology for PCR amplification with the selected primers) seems to be rather limited in the lung. To analyze whether there might exist a defective hBD-2 gene in CF, we have also cloned the hBD-2 gene from cultured CF nasal epithelial cells and found identical sequences (data not shown).

hBD-2 Gene Expression in Airway Epithelia Is Induced by a Mucoid Phenotype of P. aeruginosa and Proinflammatory Cytokines

To confirm our finding that A549 airway epithelial cells produce hBD-2 protein upon contact with bacteria on a transcriptional level, we have investigated A549 airway epithelial cells for hBD-2 mRNA expression upon contact with different strains of P. aeruginosa. Only the mucoid strain that led to hBD-2 protein release induced hBD-2 mRNA, whereas two nonmucoid strains (ATCC No. 15442 and 27853) failed to do so (data not shown).

When primary respiratory epithelial cells were investigated, we again found that the mucoid strain of P. aeruginosa induced hBD-2 mRNA expression in cultured nasal epithelial primary cells (NEC), TEC, as well as BEC in a dose-dependent fashion. The strongest effects were seen at bacterial densities greater than 105 bacteria/ml for TEC and BEC (Figure 2A). Interestingly, NEC were induced to produce hBD-2 mRNA at bacteria densities below 104/ml, indicating different sensitivities of the respiratory epithelia. Time course studies in TEC revealed half maximum hBD-2 induction within 1 to 2 h of contact with bacteria (Figure 2B).

To elucidate whether bacterial LPS is responsible for this effect, we stimulated TEC with P. aeruginosa LPS. As shown in Figure 2C, only a concentration greater than 10 μg/ ml led to a strong hBD-2 mRNA expression when cells were stimulated for 6 h. When NEC cultured from the polyps of patients with CF were treated with the heat-killed mucoid form of P. aeruginosa, a similar induction of hBD-2 mRNA expression as shown in Figure 2A was seen (data not shown).

Because LPS is known to elicit many of its effects via induction of proinflammatory cytokines in mononuclear phagocytes, we have investigated whether cytokines like tumor necrosis factor (TNF)-α, interleukin (IL)-1β, or IL-6 influence hBD-2 mRNA expression in airway epithelial cells. We found a dose-dependent stimulation of hBD-2 mRNA expression in cultured TEC by TNF-α (mean effective dose [ED50]: ≈ 0.5 ng/ml) and IL-1β (ED50: ≈ 0.5 ng/ml), but not, however, by IL-6 (Figure 3).

hBD-2 mRNA Is Differentially Expressed in Airway Respiratory Tissues

Because different respiratory epithelia show differences in the hBD-2 gene induction, we were interested in identifying the topographic basal expression of hBD-2 mRNA in the human airway system. Analyses of hBD-2 mRNA obtained from lymphoid and nonlymphoid respiratory tissues by semiquantitative duplex reverse transcriptase (RT)- PCR revealed that distal parenchyma of the lung, trachea, and tonsils strongly express hBD-2 mRNA, whereas pharynx and adenoid showed the hBD-2 message to a far less extent. In contrast, nasal polyps, tongue, and larynx tissues expressed little hBD-2 message (Figure 4).

Genomic Organization of hBD-2 DNA

To analyze the genomic structure and the promoter region of hBD-2, we generated an hBD-2–specific genomic PCR amplification product of 3,581 bp (Genbank/EMBL databases accession number AJ000152). The hBD-2 gene is composed of two exons interrupted by a 1.69- kb intron. Exon 1 (94 bp) codes for the 5′-UTS and the prepro-coding region, whereas exon 2 (242 bp) includes the mature peptide coding region and the 3′-untranslated sequence (UTS) (Figure 5A). Analysis of the promoter region (1,284 bp) upstream of the hBD-2 coding region revealed the presence of a TATA box at nucleotide position −38 and a CAAT box at nucleotide −47. Several other conserved motifs for putative transcription factor binding sites in this region of the hBD-2 gene could be detected (Figure 5B and Table 1).

Table 1. Putative transcription factor binding sites in the 5 flanking region of the hBD-2 gene

Transcription FactorBinding Site Consensus Sequence* Location (relative to transcription start)References
MEP1TGCRCNC−83; −120032, 33
NF-κB–likeGGGRNTTTCC−19532, 33
MAFGRRGSAAGK−212; −668; −101733, 34
PEA3AGGAAR−209; −373; −668; −1015; −123532, 33
HSF 1/2RGAANRTTCN−235; −60332, 33
GATAWGATAR−24332, 33
NF-IL6TKNNGNAAK−250; −852; −123732, 33
CF1ANATGG−253; −422; −620; −965; −117833, 35
CTCF (transcriptional repressor)CCCTC−281; −517; −108632, 33
AP-2CCCMNSSS−287; −518; −733; −87533, 36
Myc/MaxNNACCACGTGGTNN−304; −92433, 38
E2ARCAGNTG−34432, 33
MZF1NGNGGGA−628; −71433, 39
AP-3TGTGGWWW−85232, 33
IUF1CATYAS−937; −118633, 40
PPARAGGTCA−1243; −125332, 33

Definition of abbreviations: activator protein, AP; common factor 1, CF1; heat shock transcription factor 1/2, HSF 1/2; insulin upstream factor 1, IUF1; mammary activating factor, MAF; metal element protein 1, MEP1; myeloid zinc finger protein 1, MZF1; polyomavirus enhancer A binding protein 3, PEA3; peroxisome proliferator-activated receptor, PPAR; sterol regulatory element binding protein 1, SREB-1; upstream stimulating factor, USF; vitamin D receptor, VDR.

* Base ambiguity code: K = G or T; M = A or C; N = A or C or G or T; R = A or G; S = G or C; W = A or T; Y = C or T.

Antimicrobial Properties of Natural hBD-2 and Its Sensitivity to Salt

The natural 41 residue form of hBD-2 exerted potent antimicrobial activity at low salt concentrations against bacteria such as a mucoid form of P. aeruginosa (PA-O) and Escherichia coli (ATCC no. 11303), as well as yeast such as Candida albicans (clinical isolate), whereas only bacteriostatic activity was seen against Staphylococcus aureus (clinical isolate) at concentrations greater than 100 μg/ml (Figure 6). The calculated LD50 of hBD-2 (dose that achieves 50% reduction of colony-forming units) for mucoid P. aeruginosa was 5 μg/ml, for E. coli, 6 μg/ml, and for C. albicans, 18 μg/ml. Furthermore, antimicrobial activity of hBD-2 against mucoid P. aruginosa was dose-dependently reduced when salt concentrations greater than 25 mM were used (inhibitory dose killing 50% [ID50]: 50 mM) (data not shown).

Despite evidence that cultured human airway epithelial cells have the ability to release antimicrobial activity (5), it has been proved difficult to identify the factor(s) responsible for such activity. Very recent studies (8, 9) revealed that hBD-2, which is highly active against gram-negative bacteria (7, 9), may play a role in lung inflammation, in particular in infected lungs of patients with CF where the mucoid form of P. aeruginosa is important for the pulmonary outcome of the disease (20).

To determine whether human airway epithelia produce antibiotic peptides upon contact with bacteria, in particular mucoid P. aeruginosa, we addressed this question by a direct approach using a P. aeruginosa affinity column to purify antimicrobial peptides directed against these bacteria, which are released from human airway epithelial cells. Analysis of HPLC fractions for mucoid P. aeruginosa killing activities revealed hBD-2, an antimicrobial 4-kD peptide that we recently isolated from lesional psoriatic scales and cloned from keratinocytes (7), as a major P. aeruginosa killing peptide. Natural hBD-2 was found to effectively kill nonmucoid and mucoid P. aeruginosa at low concentrations (LD50: 5 μg/ml; Figure 6) (7-9). Insect cell– derived recombinant hBD-2 was reported to be either less (8) or more active against nonmucoid P. aeruginosa (9), indicating hBD-2 to be a more powerful antibiotic against P. aeruginosa than hBD-1 (9).

Cloning the hBD-2 cDNA from both TEC and BEC revealed identity with the cDNA we have recently cloned from skin keratinocytes (7), the exact same cDNA sequence we also found in clones obtained from patients with CF, indicating that in CF there is no defect in the primary structure of hBD-2, a finding that is supported by a recent detection of immunoreactive hBD-2 in CF bronchoalveolar lavage fluid (9).

In cattle, one characteristic of epithelial β-defensins is that the genes are widely expressed throughout the airway epithelia as well as some other epithelia, including inflamed skin, where they showed strongest expression (21). mRNAs for TAP, and to a lesser extent for LAP, are mainly expressed in the nasal and conducting airway epithelium of cattle with little expression in the gas exchange regions (21).

The present study demonstrates that hBD-2 mRNA is expressed in the entire respiratory airway tract, including nasal, tracheal, and bronchial epithelia. Moreover, investigation of the tissue distribution of hBD-2 mRNA expression by using RT-PCR techniques revealed that it displays the highest basal hBD-2 expression in lung, trachea, and tonsils. Thus, hBD-2 may have a prominent role in the innate immune response of the respiratory tract. The absence of hBD-2 mRNA in the tongue is unexpected, especially because of the known continued colonization by various microorganisms in this organ. Whether the human tongue may express a different β-defensin, an equivalent to the bovine LAP, needs to be investigated. The finding that the hBD-2 gene is highly expressed in the lung parenchyma indicates that this defensin may be also produced in the gas exchange regions of the lung. This suggestion is supported by recent in situ hybridization studies in which hBD-2 mRNA was diffusely expressed at a low level in normal lung (8), although like TAP, it is highly upregulated in inflamed lung epithelia (9).

One of the most striking findings was the observation that only a mucoid phenotype of P. aeruginosa rapidly stimulated hBD-2 mRNA expression in the different re-spiratory tract epithelial cells and that these cells responded with different sensitivities. Its early induction (within 1 h) indicates that it is most likely not mediated via proinflammatory cytokine release and subsequent induction, which would require a longer time period.

Interestingly, induction of hBD-2 mRNA failed with two defined, nonmucoid P. aeruginosa strains (data not shown). Similar results as shown in Figure 2 were seen with respiratory epithelial cells obtained from patients with CF, which suggests no defect in the induction of the hBD-2 gene expression by a mucoid P. aeruginosa strain in CF. It remains to be determined whether quantities of hBD-2 peptide released by respiratory epithelial cells are identical with that released by non-CF epithelia. Nevertheless, a recent semiquantitative estimation of immunoreactive hBD-2 revealed similar results in CF and non-CF bronchoalveolar lavage fluid (9).

To elucidate the mechanism of bacterial hBD-2 gene induction, respiratory epithelial cells were stimulated with a P. aeruginosa LPS preparation. Surprisingly hBD-2 mRNA expression was found only at an extremely high concentration (> 10 μg LPS/ml; Figure 2C), which makes LPS unlikely to be the major hBD-2 inducing compound of mucoid P. aeruginosa and indicates that a mucoid form of P. aeruginosa contains and/or releases unique signaling molecules that stimulate respiratory epithelial cells for hBD-2 production.

Based on our recent observation that human keratinocytes strongly express hBD-2 mRNA when stimulated with TNF-α (7), we were interested to get an insight into the conditions that regulate hBD-2 gene expression in airway epithelial cells. Both proinflammatory cytokines, TNF-α and IL-1β, known to be involved in respiratory tract inflammation, were found to upregulate hBD-2 mRNA expression in a dose-dependent manner and induce hBD-2 release at physiologically relevant concentrations, supporting another recent study, which indicates that the induction of hBD-2 mRNA by IL-1β occurs in cultured lung epithelial cells at a concentration of 100 ng/ml (9).

To our surprise, we found that IL-6 was not able to induce hBD-2, despite the presence of three nuclear factor (NF)–IL-6–like consensus sequences in the promotor region of the hBD-2 gene. IL-6 is often induced together with TNF-α in many inflammatory conditions. However, whether IL-6 plays a pro- or anti-inflammatory role in local inflammation is not clear. Although it is commonly believed that IL-6 acts as an inducer of inflammatory genes, a recent report about IL-6 (−/−) mice indicates a crucial anti-inflammatory role by controlling the level of proinflammatory cytokines (22). Therefore, it is not unexpected that IL-6 does not act as an inducer of hBD-2.

To study the genetic control and regulation of the hBD-2 gene expression, we cloned and sequenced the hBD-2 genomic DNA and demonstrated that the hBD-2 gene consists of two exons and one intron (Figure 3A). This gene organization is highly conserved in some mammal α- and β-defensins as well as in plant defensins. Recently, we and others localized the hBD-2 gene on chromosome 8p (23, 24), where other known human defensin genes are clustered (25). Analyses of genomic sequences have demonstrated that TAP is the closest related gene to hBD-2, which corresponds well with our gene expression studies. Thus the similarities between this bovine airway β-defensin and hBD-2 support the hypothesis that these genes have evolved from a common ancestral gene.

Analysis of the 5′ flanking region of the hBD-2 gene revealed the presence of several conserved, potential transcription factor consensus sequences known to be involved in the regulation of inflammatory responses, which may explain the observed gene induction of hBD-2. For example, the hBD-2 gene contains a NF-κB-like consensus sequence for NF-κB binding (24). Proinflammatory cyto-kines such as TNF-α and IL-1β, which we and others have seen induce hBD-2 gene expression in epithelial cells, are known to mediate their effects on gene induction on genes involved in inflammatory reactions through NF-κB regulatory elements (26). Furthermore, it has been reported that optimal induction of the antimicrobial peptide TAP and several cytokine- and LPS-responsive genes requires the cooperative action of the transcription factors NF-κB and AP-1 (27, 28), both having putative binding sites in the hBD-2 gene. For insects it has been shown that NF-κB– related binding motifs are involved in mediating the induction of antimicrobial peptides in response to injury and bacterial infection (1, 29). The presence of NF-κB–like binding sites in inducible antimicrobial peptide genes of insects (1), cattle (30), and also humans suggests a common evolutionary link for the mediation of this innate host defense system (31).

In contrast to several putative regulatory promotor elements found in the hBD-2 gene, the previously reported genomic hBD-1 sequence does not contain similar putative transcription regulatory elements. Thus, our findings confirm hBD-2 to be the first human epithelia-derived peptide antibiotic that is regulated in response to both bacterial agents and inflammatory mediators, supporting the hypothesis that hBD-2 contributes to a dynamic host defense system of human respiratory epithelia.

This work was supported by a grant from Deutsche Mukoviszidose e.V. The authors are grateful to Prof. Enno Christophers for support, to Jutta Quitzau and Marlies Brand for technical assistance, and to Gabriele Tams and Clair Watts for editorial help. They thank Dr. Carsten Schlüter for providing them with the GAPDH-specific primer sequence. The Alexander von Humboldt Foundation is acknowledged for financial support of Luis Teran.

1. Hultmark D.Immune reactions in drosophila and other insects: a model for innate immunity. Trends Genet.91993178183
2. Boman H. G.Peptide antibiotics and their role in innate immunity. Annu. Rev. Immunol.1319956192
3. Diamond G., Zasloff M., Eck H., Brasseur M., Maloy W. L., Bevins C. L.Tracheal antimicrobial peptide, a cysteine-rich peptide from mammalian tracheal mucosa: peptide isolation and cloning of a cDNA. Proc. Natl. Acad. Sci. USA88199139523956
4. Schonwetter B. S., Stolzenberg E. D., Zasloff M. A.Epithelial antibiotics induced at sites of inflammation. Science267199516451648
5. Smith J. J., Travis S. M., Greenberg E. P., Welsh M. J.Cystic fibrosis airway epithelia fail to kill bacteria because of abnormal airway surface fluid. Cell851996229236
6. Goldman M. J., Anderson G. M., Stolzenberg E. D., Kari U. P., Zasloff M., Wilson J. M.Human beta-defensin-1 is a salt-sensitive antibiotic in lung that is inactivated in cystic fibrosis. Cell881997553560
7. Harder J., Bartels J., Christophers E., Schröder J.-M.A peptide antibiotic from human skin. Nature3871997861
8. Bals R., Wang X., Wu Z., Freeman T., Bafna V., Zasloff M., Wilson J. M.Human beta-defensin 2 is a salt-sensitive peptide antibiotic expressed in human lung. J. Clin. Invest.1021998874880
9. Singh P. K., Jia H. P., Wiles K., Hesselberth J., Liu L., Conway B. A. D., Greenberg E. P., Valore E. V., Welsh M. J., Ganz T., Tack B. F., McCray P. B.Production of beta-defensins by human airway epithelia. Proc. Natl. Acad. Sci. USA9519981496114966
10. Mochizuki M., Bartels J., Mallet A. I., Christophers E., Schröder J.-M.IL-4 induces eotaxin: a possible mechanism of selective eosinophil recruitment in helminth infection and atopy. J. Immunol.16019986068
11. Schägger H., von Jagow G.Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa. Anal. Biochem.1661987368379
12. Schröder J.-M.Identification and structural characterization of chemokines in lesional skin material of patients with inflammatory skin disease. Method Enzymol.2881997166197
13. Selsted M. E., Tang Y. Q., Morris W. L., McGuire P. A., Novotny M. J., Smith W., Henschen A. H., Cullor J. S.Purification, primary structures, and antibacterial activities of beta-defensins: a new family of antimicrobial peptides from bovine neutrophils. J. Biol. Chem.268199366416648
14. Triglia T., Peterson M. G., Kamp D. J.A procedure for in vitro amplification of DNA segments that lie outside the boundaries of known sequences. Nucleic Acids Res.1619888186
15. Bartels J., Schlüter C., Richter E., Noso N., Kulke R., Christophers E., Schröder J.-M.Human dermal fibroblasts express eotaxin: molecular cloning, mRNA expression, and identification of eotaxin sequence variants. Biochem. Biophys. Res. Commun.225199610451051
16. Hein H., Schlüter C., Kulke R., Christophers E., Schröder J.-M., Bartels J.Genomic organization, sequence, and transcriptional regulation of the human eotaxin gene. Biochem. Biophys. Res. Commun.2371997537542
17. Wingender E., Kel A. E., Kel O. V., Karas H., Heinemeyer T., Dietze D., Knuppel R., Romaschenka A. G., Kolchanow N. A.TRANSFAC, TRRD and COMPEL: towards a federated database system on transcriptional regulation. Nucleic Acids Res.251997265268
18. Harwig S. S. L., Swiderek K. M., Kokryakow V. N., Tan L., Lee T. D., Panyutich E. A., Aleshina G. M., Shamov O. V., Lehrer R. I.Gallinacins: cysteine-rich antimicrobial peptides of chicken leukocytes. FEBS Lett.3421994285
19. Evans E. W., Beach G. G., Wunderlich J., Harmong B. G.Isolation of antimicrobial peptides from avian heterophils. J. Leukoc. Biol.561994661665
20. Parad R. B., Gerard C. J., Zurakowski D., Nichols D. P., Pier G. B.Pulmonary outcome in cystic fibrosis is influenced primarily by mucoid Pseudomonas aeruginosa infection and immune status and only modestly by genotype. Infect. Immunol.67199947444750
21. Stolzenberg E. D., Anderson G. M., Ackermann M. R., Whitlock R. H., Zasloff M.Epithelial antibiotic induced in states of disease. Proc. Natl. Acad. Sci. USA.94199786868690
22. Xing Z., Gauldie J., Cox G., Baumann H., Jordana M., Lei X. F., Achong M. K.IL-6 is an antiinflammatory cytokine required for controlling local or systemic acute inflammatory responses. J. Clin. Invest.1011998311320
23. Harder J., Siebert R., Zhang Y., Matthiesen P., Christophers E., Schlegelberger B., Schröder J.-M.Mapping of the gene encoding human β-defensin-2 (DEFB2) to chromosome region 8p22–p23.1. Genomics461997472475
24. Liu L., Wang L., Jia H. P., Zhao C., Heng H. H. Q., Schutte B. C., McCray P. B., Ganz T.The human beta-defensin-1 and alpha-defensins are encoded by adjacent genes: two peptide families with differing disulfide topology share a common ancestry. Gene2221998237244
25. Liu L. D., Zhao C. Q., Heng H. H. Q., Ganz T.Structure and mapping of the human beta-defensin HBD-2 gene and its expression at sites of inflammation. Genomics431997316320
26. Baeuerle P. A., Henkel T.Function and activation of NF-kappa B in the immune system. Annu. Rev. Immunol.121994141179
27. Martin T., Cardarelli P. M., Parry G. C., Kelts K. A., Cobb R. R.Cytokine induction of monocyte chemoattractant protein-1 gene expression in human endothelial cells depends on the cooperative action of NF-kappa B and AP-1. J. Immunol.27199710911097
28. Mackman N., Brand K., Edgington T. S.Lipopolysaccharide-mediated transcriptional activation of the human tissue factor gene in THP-1 monocytic cells requires both activator protein 1 and nuclear factor kappa B binding sites. J. Exp. Med.174199115171526
29. Ip Y. T., Reach M., Engstrom Y., Kadalayil L., Cai H., Gonzalez-Crespo S., Tatei K., Levine M.Dif, a dorsal-related gene that mediates an immune response in Drosophila. Cell751993753763
30. Diamond G., Jones D. E., Bevins C. L.Airway epithelial cells are the site of expression of a mammalian antimicrobial peptide gene. Proc. Natl. Acad. Sci. USA90199345964600
31. Hultmark D.Insect immunology: ancient relationships. Nature3671994116117
32. Faisst S., Meyer S.Compilation of vertebrate-encoded transcription factors. Nucleic Acids Res.201992326
33. Akiyama, Y. 1995. TFSEARCH: Searching Transcription Factor Binding Sites.
34. Mink S., Haertig E., Jennewein P., Doppler W., Cato A. C. B.A mammary cell-specific enhancer in mouse mammary tumor virus DNA is composed of multiple regulatory elements including binding sites for CTF/ NFI and a novel transcription factor, mammary cell-activating factor. Mol. Cell. Biol.12199249064918
35. Riggs K. J., Merrell K. T., Wilson G., Calame L.Common factor 1 is a transcriptional activator which binds in the c-myc promoter, the sceletal alpha-actin promoter, and the immunoglobulin heavy-chain enhancer. Mol. Cell. Biol.11199117651769
36. Courtois S. J., Lafontaine D. A., Lemaigre F. P., Durviaux S. M., Rousseau G. G.Nuclear factor-I and activator protein-2 bind in a mutually exclusive way to overlapping promoter sequences and trans-activate the human growth hormone gene. Nucleic Acids Res.1819905764
37. Kim J. B., Spotts G. D., Halvorser Y.-D., Shih H.-M.Dual DNA binding specificity of ADD1/SREBP1 controlled by a single amino acid in the basic helix-loop-helix domain. Mol. Cell. Biol.15199525822588
38. Fisher D. E., Parent L. A., Sharp P. A.Myc/max and other helix-loop-helix/leucine zipper proteins bend DNA toward the minor groove. Proc. Natl. Acad. Sci. USA8919921177911783
39. Morris J. F., Hromas R., Rauscher F. J.Characterization of the DNA-binding properties of the myeloid zinc finger protein MZF1: two independent DNA-binding domains recognize two DNA consensus sequences with a common G-rich core. Mol. Cell. Biol.14199417861795
40. Scott V., Clark A. R., Hutton J. C., Docherty K.Two proteins act as the IUF1 insulin gene enhancer binding factor. FEBS Lett.29019912730
41. Freedman L. P., Luisi B. F.On the mechanism of DNA binding by nuclear hormone receptors: a structural and functional perspective. J. Cell. Biochem.511993140150
Address correspondence to: Dr. J.-M. Schröder, Department of Derm-atology, Schittenhelmstr. 7, 24105 Kiel, Germany. E-mail:

The complete genomic nucleotide sequence of hBD-2 has been deposited in the Genbank/EMBL database with the accession number AJ 000152.

Abbreviations: activator protein, AP; bronchial epithelial cell, BEC; cystic fibrosis, CF; colony-forming unit, CFU; complementary DNA, cDNA; fetal calf serum, FCS; glyceraldehyde-3-phosphate dehydrogenase, GAPDH; human beta-defensin, hBD; high performance liquid chromatography, HPLC; interleukin, IL; keratinocyte growth medium, KGM; lingual antimicrobial peptide, LAP; lipopolysaccharide, LPS; messenger RNA, mRNA; nasal epithelial cell, NEC; nuclear factor, NF; polymerase chain reaction, PCR; reverse phase, RP; reverse transcriptase, RT; sodium dodecyl sulfate polyacrylamide gel electrophoresis, SDS-PAGE; tracheal antimicrobial peptide, TAP; tracheal epithelial cell, TEC; trifluoroacetic acid, TFA; tumor necrosis factor, TNF.


No related items
American Journal of Respiratory Cell and Molecular Biology

Click to see any corrections or updates and to confirm this is the authentic version of record