In response to infection, epithelia mount an innate immune response that includes the production of antimicrobial peptides. However, the pathways that connect infection and inflammation with the induction of antimicrobial peptides in epithelia are not understood. We analyzed the molecular links between infection and the expression of three antimicrobial peptides of the β-defensin family, human β-defensin (hBD)-1, hBD-2, and hBD-3 in the human epidermis. After exposure to microbe-derived molecules, both monocytes and lymphocytes stimulated the epidermal expression of hBD-1, hBD-2, and hBD-3. The induced expression of hBD-3 was mediated by transactivation of the epidermal growth factor receptor. The mechanisms of induction of hBD-1 and hBD-3 were distinct from each other and from the IL-1-dependent induction of hBD-2 expression. Thus during inflammation, epidermal expression of β-defensins is mediated by at least three different mechanisms.

Epithelia not only serve as physical barriers against infection but also mount an innate immune response by producing antimicrobial peptides to neutralize invading microbes (1). Antimicrobial peptides are abundant and widely distributed effector molecules of the innate immune response from insects to human (2) and are active against a broad spectrum of Gram-positive and Gram-negative bacteria, as well as some fungi and enveloped viruses.

The key role of antimicrobial peptides in mammalian host defense is made evident by mice with disrupted antimicrobial peptide genes that are prone to infection in the affected epithelial organs, including the skin (3, 4, 5). In humans, the role of epithelial antimicrobial peptides in host defense is suggested by studies of two major human skin diseases with a defective skin barrier, atopic dermatitis and psoriasis. Patients with atopic dermatitis are much more prone to have skin infections than are patients with psoriasis, and this difference has now been linked to the much higher expression of antimicrobial peptides in psoriasis than in atopic dermatitis (6, 7). Indeed some of the major human antimicrobial peptides of human skin were originally isolated from the scales of patients with psoriasis (8, 9). Although inflamed epithelia are a particularly rich source of antimicrobial peptides, the molecular events that lead to increased expression of the different antimicrobial peptides in epithelia in response to infection and inflammation are incompletely understood.

The human β-defensin (hBD)3 constitutes the major class of antimicrobial peptides found in human epithelia (10), and three β-defensins, hBD-1, hBD-2, and hBD-3 are expressed in the skin (8, 9, 11). During inflammation, the expression of hBD-2 in inflamed skin is induced by IL-1 from resident monocyte-derived cells, which establishes a clear link between infection/inflammation and hBD-2 expression (12). Although inflamed skin and other epithelia greatly increase the production of hBD-3 (9, 13), it is not known how the expression of hBD-3 is regulated. The expression of hBD-1 has been deemed constitutive (14), but there is evidence that it too may be somewhat increased by inflammation (15, 16). When we analyzed biopsy specimens of highly inflamed squamous epithelia by immunohistochemistry we found separate areas with intense staining for each hBD-1, hBD-2, and hBD-3. This suggested a differential regulation of the expression of these β-defensins and prompted us to further examine the mechanism for induction of these antimicrobial peptides.

In this study we analyzed and compared pathways that link infection and inflammation with the expression of three β-defensins (hBD-1, hBD-2, and hBD-3) in human epidermis. Remarkably, the mechanisms of induction of hBD-1 and hBD-3 were distinct from each other and from the IL-1-dependent induction of hBD-2 expression.

The hBD-1 (44 aa) was purified from urine, and hBD-2 was recombinantly produced as previously described (17, 18). The anti-hBD-1 and anti-hBD-2 Abs were previously described (17, 18). Anti-hBD-3 Ab for immunohistochemistry was purchased from Orbigen. Alkaline phosphatase-conjugated goat anti-rabbit Ab was obtained from Pierce. SpeB, a cysteine proteinase from Streptococcus pyogenes, was purchased from Toxin Technology. TGF-α and epidermal growth factor (EGF) were obtained from PeproTech. Escherichia coli O55:B5 LPS, peptidoglycan (PGN) from Staphylococcus aureus, heparin-binding EGF (HB-EGF), and amphiregulin were purchased from Sigma-Aldrich. Control Abs, IL-1R antagonist, neutralizing Abs to TGF-α, HB-EGF, amphiregulin, IFN-α, IFN-β, IFN-γ, and IL-6 were purchased from R&D Systems. Neutralizing Abs toward EGFR were obtained from Oncogene Research Products.

Isolation of mononuclear cells.

Blood was collected from healthy volunteer donors according to a protocol approved by the University of California Los Angeles Human Subjects Protection Committee.

The mononuclear cells were collected in the supernatant following centrifugation at 200 × g for 10 min at room temperature on Ficoll-Paque (Amersham Biosciences). The cells were washed twice in sterile PBS and once in IMDM (Invitrogen Life Technologies) and resuspended in the same medium with 10% autologous serum and gentamicin at a concentration of 6 × 106 cells/ml.

Isolation of monocytes and lymphocytes.

The mononuclear cells were washed three times in PBS and layered on top of 46% isotonic Percoll (Sigma-Aldrich) and centrifuged at 550 × g for 30 min at room temperature. After centrifugation the monocytes were collected from the interface, and the lymphocytes were collected as the cells outside the interface. Both monocytes and lymphocytes were washed in IMDM and resuspended in the same medium with 10% autologous serum and gentamicin at a concentration of 6 × 106 cells/ml.

The cells were incubated at 37°C after the addition of 50 ng/ml E. coli LPS, 10 μg/ml peptidoglycan from Staphylococcus aureus, or 1 μg/ml SpeB. After 4 days of stimulation, the cells were pelleted, and the supernatant was collected and stored at −20°C until additional experiments.

Primary epidermal cultures EPI-200-3S (MatTek) containing human epidermal keratinocytes were grown on a collagen-coated Millicel CM membranes. The cultures were placed in 12-well plates with defined media supplied by the manufacturer (which contains no bovine pituitary extract). On day 4, the epidermal cultures were lifted to the air-liquid interface and then cultured in air-liquid interface for another 4 days according to the instructions by the manufacturer. On day 2 after airlifting the cultures, the medium was changed to medium without EGF. On day 4 after airlifting, the cultures were stimulated with the supernatant from mononuclear cells, or monocytes or lymphocytes in the presence or absence of various blocking Abs and inhibitors.

All blocking Abs and the metalloproteinase inhibitor galardin were used in a final concentration of 7.5 μg/ml. The concentration of IL-1R antagonist was 200 ng/ml.

Polyclonal Ab to hBD-3 was generated in our laboratory by E. Valore (Host Defense Laboratory, University of California, Los Angeles, CA) from hBD-3 synthesized chemically by W. Lu (University of Maryland, Baltimore, MD). One milligram of HPLC-purified synthetic hBD-3 was coupled with 2 mg of mariculture keyhole limpet hemocyanin using the cross-linking reagent EDC (1-ethyl-3-(3-dimethylaminopropyl)carbodiimide) according to the manufacturer’s instructions (Pierce). The conjugate was dialyzed against PBS to remove contaminants. Approximately one-fourth of the conjugate was emulsified with Freund’s complete adjuvant (BD Biosciences) and Hunters Titermax adjuvant (Sigma-Aldrich) and used to immunize each rabbit by s.c. injection. Boosts using Freund’s incomplete adjuvant and Hunters Titermax were conducted at 30-day intervals. Serum was collected 10 days after boost. These Abs were subsequently used to detect hBD-3 in Western blot analysis.

The cells from the epidermal keratinocyte cultures were homogenized and sonicated after addition of 1 M HCl. The samples were centrifuged for 10 min at 10,000 rpm, and the supernatants were collected. The pellets were re-extracted once in 5% acetic acid. The acetic acid extract was lyophilized and resuspended in 0.01% acetic acid and pooled with the HCl extract. The combined extracts were diluted 20-fold in distilled H2O, and the pH was adjusted to 7.

The β-defensins from both the media and the extract from the cells were extracted for 3–4 h at room temperature with MacroPrep CM Support bead (Bio-Rad) equilibrated in 25 mM ammonium acetate (pH 6.8). The beads were subsequently washed, and the bound material was eluted with 5% acetic acid. The eluate was lyophilized and resuspended in 0.01% and dialyzed in 0.1% acetic acid. The dialyzed material was concentrated by centrifugation in Microcon filter (Millipore) with molecular cutoff at 3 kDa. The retentate was resuspended in 0.01% acetic acid and concentrated once again by centrifugation.

Acid urea-PAGE and immunoblotting were performed according to the instructions from the manufacturer (Hoeffer). After transfer of proteins from the 12.5% polyacrylamide gels, the polyvinylidene difluoride (PVDF)-membrane was fixed for 30 min in TBS with 0.05% glutaraldehyde (Sigma-Aldrich) and blocked with Superblock Blocking buffer (Pierce). For visualization of the β-defensins, the PVDF membranes were incubated overnight with primary Abs. The following day, the membranes were incubated for 2 h with HRP-conjugated secondary Abs (Pierce) and visualized by Immun-Star HRP luminal/enhancer and Immun-Star peroxide buffer (Bio-Rad). The PVDF membrane was stripped for 20 min in 0.2 M glycine (pH 2.5), 1% SDS, washed twice with TBS with 0.05% Tween 20, and finally blocked before incubating overnight with a different Ab.

Primary keratinocytes were obtained from Cascade Biologics and grown in serum-free medium (KGM2 Bullet kit) obtained from Cambrex. Cells were stimulated beginning 24 h after complete confluence was reached. Cells were stimulated with TGF-α (50 ng/ml), EGF (100 ng/ml), HB-EGF (100 ng/ml), and amphiregulin (100 ng/ml).

Under protocols approved by the Institutional Review Board, skin from healthy donors was obtained from patients undergoing breast or abdominal reduction surgery. The skin was sliced into 1 × 10 mm pieces and incubated in keratinocyte medium (KGM2 Bullet kit; Cambrex) with 5% human serum and then stimulated with 10 μg/ml PGN, 100 ng/ml LPS, or 1 μg/ml SpeB in the presence or absence of 200 ng/ml IL-1R antagonist, anti-EGFR Ab (10 μg/ml), irrelevant control Ab (10 μg/ml), or galardin (7.5 μg/ml). The skin was stimulated for 48 h for RNA purification and 60 h for immunohistochemistry.

Total RNA was isolated with TRIzol (Invitrogen Life Technologies) according to recommendations of the manufacturer. The RNA was precipitated with ethanol and resuspended in diethyl pyrocarbonate H2O. The RNA from the skin was repurified once to obtain RNA of adequate purity. The RNA concentration was determined by spectrophotometry, and the integrity of the RNA was assessed by electrophoresis on an agarose gel.

For Northern blotting, 5 μg of RNA were analyzed by size on a 1% agarose gel with 6% formaldehyde dissolved in 1× MOPS (20 mM 3-(N-morpholino)-propanesulfonic acid, 5 mM sodium acetate, 1 mM EDTA, pH 7.0). The RNA was transferred to a Hybond-N membrane (Amersham Biosciences) by capillary blotting and fixed by UV irradiation. The filters were prehybridized for a minimum of 30 min at 42°C in 10 ml of ULTRAhyb (Ambion) and hybridized overnight at 42°C after addition of a further 5 ml of ULTRAhyb containing the 32P-labeled probe. The membranes were washed twice for 5 min at 42°C in 2× SSC (1× SSC = 150 mM NaCl, 15 mM sodium citrate, pH 7.0), 0.1% SDS followed twice for 15 min in 2× SSC, 0.1% SDS, once for 15 min in 0.2× SSC, 0.1% SDS, and twice for 15 min in 0.1× SSC, 0.1% SDS at 42°C. The blot was developed and then quantified by a phosphoimager (Bio-Rad). The sizes of the mRNAs were determined by reference to 18 S and 28 S ribosomal RNA, which were visualized by staining of membranes with methylene blue. The membranes were stripped by boiling in 0.1% SDS before rehybridization.

cDNA was synthesized from 200 ng of purified RNA using iScript cDNA synthesis kit (Bio-Rad) according to the instructions by the manufacturer. hBD-1, hBD-2, and hBD-3 together with G3DPH expression were analyzed using iQ SYBR Green Supermix (Bio-Rad). The primers were as follows: hBD-1, 5′-GTCGCCATGAGAACTTCCTACC-3′ and 5′-CATTGCCCTCCACTGCTGAC-3′; hBD-2, 5-CCTGTTACCTGCCTTAAGAGTG-3′ and 5′-GAATCCGCATCAGCCACAG-3′; hBD-3, 5′-CTTCTGTTTGCTTTGCTCTTCC-3′ and 5′-CACTTGCCGATCTGTTCCTC-3′; human G3PD, 5′-TGGTATCGTGGAAGGACTC-3′ and 5′-AGTAGAGGCAGGGATGATG-3′; EGFR, 5′-CCGTCGCTATCAAGGAATTAAG-3′ and 5′-GTGGAGGTGAGGCAGATG-3′; TGF-α, 5′-CTGGCTGTCCTTATCATCAC-3′ and 5′-AGCGGTTCTTCCCTTCAG-3′; HB-EGF, 5′-TGCCAAGTCTCAGAAGAGG-3′ and 5′-GGAGTAGCACCAGAAGAATG-3′; amphiregulin, 5′-GTCTCCACTCGCTCTTCC-3′ and 5′-GGGCTCTCATTGGTCCTTC-3′; EGF, 5′-TGCGTGGTGGTGCTTGTC-3′ and 5′-GCCTGCGACTCCTCACATC-3′; betacellulin, 5′-CCAAGCAATACAAGCATTACTG-3′ and 5′-TGTCCTCTGTCTCCTCTTAG-3′; epiregulin, 5′-GGAGGAGGATGGAGATGC-3′ and 5′-CCTGGGATACATGATGGAATC-3′; IL-1α, 5′-GTGCTGCTGAAGGAGATG-3′ and 5′-ACAAGTGAGACAAGTGAGAC-3′; IL-1β, 5′-AGGCACAAGGCACAACAG-3′ and 5′-GTAGTGGTGGTCGGAGATTC-3′.

Amplification was performed at 40 cycles of 95°C for 30 s followed by 58°C for 30 s in iCycler Thermal Cycler (Bio-Rad), and data were analyzed with iCycler iQ Optical System software. The relative expression in each sample was calculated adjusting for the real-time RT-PCR efficiencies (19). All experiments were performed in triplicate, and the expression of the hBDs was normalized to the expression of G3PD.

The epidermal cultures or skin slices were fixed in 10% formalin, dehydrated, and embedded in paraffin. Sections of 2 μm were placed on poly-lysine-coated glass slides, deparaffinized in xylene, and rehydrated in graded alcohol. For Ag retrieval the slides were incubated for 20–40 min at 97°C in DAKOTarget Retrieval solution. The slides were subsequently washed, blocked for 20 min with 20% goat serum, washed again, and incubated for 24 h at room temperature in 1/1000 dilution of rabbit polyclonal Ab against hBD-1 or hBD-2 or a 1/666 dilution of rabbit polyclonal Abs against hBD-3. The Abs were diluted in TBS with 1% albumin, 0.05% Tween 20 (Sigma-Aldrich), and 0.01% thimerosal. After three 20-min washes in TBS with 0.05% Tween 20, the slides were incubated with alkaline phosphatase-conjugated goat anti-rabbit IgG (Pierce) diluted 1/1000 in the same buffer as the first Ab and incubated for another 24 h followed by three 20-min washes. Color was developed with Fast Red chromogen (Sigma-Aldrich), and the slides were counterstained with Harris hematoxylin (EM Science).

TGF-α was assayed by a TGF-α ELISA (Oncogene Science) according to the manufacturer’s instructions.

S. pyogenes (ATCC no. 700460) were grown in a shaker-incubator at 37°C until log phase in brain heart infusion (BD Biosciences). The bacteria were washed once and resuspended in 10 mM phosphate buffer (pH 7.2) with 0.01× brain heart infusion and OD620 was adjusted to 0.57. The bacteria were incubated in 10 mM sodium phosphate (pH 7.2) with peptide dissolved in 0.01% acetic acid for 3 h at 37°C. The bacterial solutions were plated on blood agar plates overnight at 37°C, and the colonies were counted. All experiments were performed in triplicate.

To study epidermal responses to conserved microbial macromolecules we challenged organotypic epidermal keratinocyte cultures with LPS, PGN, or SpeB (a virulence factor from S. pyogenes) in the presence of 5% serum. As seen previously for hBD-2 (12, 20), this had only minor effects on the expression of the β-defensins compared with controls cultured in medium only (data not shown). However, it was noted previously that monocyte/macrophage-like cells greatly enhance epithelial responses to LPS both in the skin (12) and in the lungs (20). Many epithelia contain resident monocyte-derived cells (macrophages or dendritic cells) such as the Langerhans cells in the skin, Kupffer cells in the liver, and alveolar macrophages in the lung together with intraepithelial lymphocytes. Furthermore, additional blood-derived cells, including monocytes and lymphocytes, are recruited to the epithelia during inflammation. To determine whether blood-derived cells enhance epidermal production of β-defensins we stimulated isolated mononuclear blood leukocytes (MBLs) with LPS, SpeB, or PGN. The supernatants from unstimulated and stimulated MBL (termed control-MBL supernatant, LPS-MBL supernatant, PGN-MBL supernatant, and SpeB-MBL supernatant, respectively) were subsequently used to treat organotypic cultures of keratinocytes.

Control-MBL supernatant did not change the expression of hBD-1, hBD-2, and hBD-3 in organotypic epidermal cultures (Fig. 1). However, supernatants from MBL stimulated with LPS, PGN, or SpeB induced the mRNA expression of the three hBDs (Fig. 1). These results were confirmed by independent experiments with real-time PCR (Table I). The induction of hBDs production at the peptide level was investigated by immunohistochemistry of epidermal cultures and by Western blot analysis of extracted material from epidermal cultures and culture supernatants (Fig. 2).

FIGURE 1.

Induced expression of hBD-1, hBD-2, and hBD-3 in epidermal cultures stimulated with MBL supernatants. Northern blot of total RNA from epidermal cultures stimulated either with MBL supernatant or directly with LPS, PGN, and SpeB. The blots were hybridized with probes for hBD-1, hBD-2, hBD-3, and G3PD. With each Northern blot is shown graphs of the hBD-1, hBD-2, and hBD-3 mRNA normalized to the expression of G3PD mRNA. hBD mRNA concentrations in epidermal tissue treated with MBL-control supernatant were assigned the value 1.

FIGURE 1.

Induced expression of hBD-1, hBD-2, and hBD-3 in epidermal cultures stimulated with MBL supernatants. Northern blot of total RNA from epidermal cultures stimulated either with MBL supernatant or directly with LPS, PGN, and SpeB. The blots were hybridized with probes for hBD-1, hBD-2, hBD-3, and G3PD. With each Northern blot is shown graphs of the hBD-1, hBD-2, and hBD-3 mRNA normalized to the expression of G3PD mRNA. hBD mRNA concentrations in epidermal tissue treated with MBL-control supernatant were assigned the value 1.

Close modal
Table I.

Induction of hBD mRNA in epidermal keratinocyte cultures by MBL supernatantsa

SupernatanthBD-1hBD-2hBD-3
LPS-MBL 3.2 (2.9–3.6) 127 (38–284) 41.3 (19–64) 
SpeB-MBL 3.9 (3.1–5.3) 201 (69–430) 40.1 (11–62) 
PGN-MBL 2.5 (1.9–3.0) 388 (274–1024) 33.9 (20–46) 
SupernatanthBD-1hBD-2hBD-3
LPS-MBL 3.2 (2.9–3.6) 127 (38–284) 41.3 (19–64) 
SpeB-MBL 3.9 (3.1–5.3) 201 (69–430) 40.1 (11–62) 
PGN-MBL 2.5 (1.9–3.0) 388 (274–1024) 33.9 (20–46) 
a

The induction of hBD mRNA in epidermal cultures incubated with supernatants from MBL was measured by real-time RT-PCR with RNA from three independent experiments. The stimulatory effect of supernatants from MBL pretreated with LPS, SpeB, or PGN was expressed as a ratio to the effect of supernatants from untreated MBL. The range of induction in the three experiments is shown in parentheses.

FIGURE 2.

Induction of hBD peptides in epidermal cultures. A, Epidermal cultures were stained for hBD-1, hBD-2, and hBD-3 after 2 days of stimulation with MBL supernatants. Color was developed with fast red chromogen, and Harris hematoxylin was used for counterstaining. B, hBDs were extracted with cation exchange beads from the cell culture medium and from the epidermal cultures 2 days after stimulation with PGN-MBL. The eluate from the beads were analyzed on acid urea-PAGE followed by immunoblotting with anti-hBD Abs. Synthetic hBDs (25 ng of each) were used as standards.

FIGURE 2.

Induction of hBD peptides in epidermal cultures. A, Epidermal cultures were stained for hBD-1, hBD-2, and hBD-3 after 2 days of stimulation with MBL supernatants. Color was developed with fast red chromogen, and Harris hematoxylin was used for counterstaining. B, hBDs were extracted with cation exchange beads from the cell culture medium and from the epidermal cultures 2 days after stimulation with PGN-MBL. The eluate from the beads were analyzed on acid urea-PAGE followed by immunoblotting with anti-hBD Abs. Synthetic hBDs (25 ng of each) were used as standards.

Close modal

After incubation with stimulated MBL supernatants, an increased amount of hBD-1 in the epidermal cultures was clearly detectable by immunohistochemistry. By Western blot analysis, we found hBD-1 in the medium from the stimulated cultures but not in the medium from nonstimulated cultures. The lack of detection of hBD-1 from the cultures may be due to the fact that this peptide was not extracted as efficiently from the cultures as the other peptides. However, the combined data from immunohistochemistry and Western blot analysis demonstrate that hBD-1 was clearly present in both medium and cells after stimulation with PGN-MBL supernatant. The induced hBD-1 comigrated with the 44 aa standard, demonstrating that the hBD-1 peptide was processed into one of the active forms.

After incubation with stimulated MBL supernatants hBD-2 was found in the epidermal cultures by immunohistochemistry. The staining for hBD-2 was predominantly localized in the upper epidermal layer as previously described (12). Also the membranes supporting the epidermal cultures were heavily stained after stimulation. This strongly suggested that hBD-2 was secreted into the medium. Western blot analysis found that the hBD-2 was mainly secreted from the epidermal cultures, and only a small fraction was found within the cells. The hBD-2 comigrated with the 41 aa standard, demonstrating that the hBD-2 was processed into the active form.

The induced production of hBD-3 was found both by immunohistochemistry and Western blot analysis. It was evident from the Western blot that the majority of the produced hBD-3 was retained within the epithelium and not secreted into the medium. EGF was withdrawn from the growth medium 2 days before stimulation with MBL supernatants. However, some EGF is necessary for the growth of these cultures in the initial phase, and EGF may account for the hBD-3 peptide we detected in the epidermal cell cultures not stimulated with MBL supernatants. The majority of the produced hBD-3 comigrated with the 45 aa mature form of the peptide.

Subsequently, we used various inhibitors and Abs to identify the specific mediators responsible for the induction of each β-defensin.

Although hBD-1 is constitutively expressed (14), we found that the expression of hBD-1 in the organotypic epidermal cultures increased after stimulation with MBL supernatants (Fig. 1). The induction was confirmed at the peptide level by immunostaining with a specific anti-hBD-1 Ab (Fig. 2). However, none of our Abs or inhibitors blocked the induction of hBD-1 by the PGN-MBL supernatant (Fig. 3). The increased expression of hBD-1 after treatment of the epidermis with LPS-MBL and SpeB-MBL supernatant was also not affected by any of our inhibitors or Abs (data not shown). In separate experiments we found that IFN-γ (but not IFN-α or IFN-β) increased the expression of hBD-1 in keratinocytes (data not shown). However, blocking Abs against the IFN-γ did not affect the induction of hBD-1 by stimulated MBL supernatants. As a control for the neutralization of IFN-inducible activity by the blocking Abs, we performed real-time PCR for IFN-inducible protein 10. Anti-IFN-β neutralizing Ab blocked the induction of IFN-inducible protein 10 induced by stimulated MBL supernatants, demonstrating that IFN-β was responsible for the IFN-inducible activity of the MBL supernatants. However, inhibition of the IFN-inducible activity failed to reduce the increased expression of hBD-1. In additional experiments to identify the mediator in MBL supernatants responsible for the increase of hBD-1 expression, we found that TGF-α, TGF-β1, IL-1, IL-2, IL-4, IL-6, IL-8, IL-17, or TNF-α did not increase the expression of hBD-1 in keratinocytes (data not shown).

FIGURE 3.

Inhibition of hBD-1 expression induced by PGN-MBL supernatant. Northern blot of total RNA from organotypic epidermal cultures stimulated for 48 h with PGN-MBL supernatant in the presence of various inhibitors and Abs. The blot was hybridized with probes for hBD-1 and G3PD. Top panel, A graph of the expression of hBD-1 mRNA normalized to the expression of G3PD mRNA. Expression of hBD-1 mRNA in cultures treated with MBL-control supernatant was assigned the value 1.

FIGURE 3.

Inhibition of hBD-1 expression induced by PGN-MBL supernatant. Northern blot of total RNA from organotypic epidermal cultures stimulated for 48 h with PGN-MBL supernatant in the presence of various inhibitors and Abs. The blot was hybridized with probes for hBD-1 and G3PD. Top panel, A graph of the expression of hBD-1 mRNA normalized to the expression of G3PD mRNA. Expression of hBD-1 mRNA in cultures treated with MBL-control supernatant was assigned the value 1.

Close modal

The induction of hBD-1 (3- to 4-fold) confirmed in additional independent experiments was less than for hBD-2 and hBD-3. This is partly due to the much higher basal expression of hBD-1 than for the two other β-defensins. By immunohistochemistry it was evident that the overall amount of hBD-1 was higher in stimulated than in nonstimulated cultures.

Both hBD-2 mRNA and protein were induced in the epidermal cultures after treatment with supernatants from stimulated MBLs (Figs. 1 and 2). The expression of hBD-2 in keratinocytes after treatment with supernatants from LPS-monocyte supernatants was previously found to be strictly IL-1 dependent (12). We found that only the IL-1R antagonist inhibited the hBD-2 expression in PGN-MBL supernatant stimulated epidermal culture (Fig. 4). The expression of hBD-2 induced in epidermal culture by LPS-MBL and SpeB-MBL supernatant was also inhibited solely by IL-1R antagonist (data not shown). Thus, stimulated MBL cells induce hBD-2 predominantly through IL-1 signaling. To examine whether keratinocyte-derived IL-1 contributed to the expression of hBD-2 we examined the induction of IL-1 in keratinocytes 48 h after stimulation with MBL supernatants by real-time PCR. When epidermal cultures were stimulated with LPS-MBL, PGN-MBL, and SpeB-MBL (three independent experiments with each supernatant), the IL-1α mRNA increased 3- to 6-fold and IL-1β mRNA 3- to 9-fold compared with epidermal cultures incubated with control MBL. Thus, keratinocyte-derived IL-1 may also contribute to the induction of hBD-2.

FIGURE 4.

Inhibition profile of hBD-2 mRNA induced by PGN-MBL supernatant. Northern blot of total RNA from organotypic epidermal cultures stimulated for 48 h with PGN-MBL supernatant in the presence of various inhibitors and Abs. The blot was hybridized with probes for hBD-2 and G3PD. Top panel, A graph of the expression of hBD-2 mRNA normalized to the expression of G3PDH mRNA. Expression of hBD-2 mRNA in cultures treated with MBL-control supernatant was assigned the value 1.

FIGURE 4.

Inhibition profile of hBD-2 mRNA induced by PGN-MBL supernatant. Northern blot of total RNA from organotypic epidermal cultures stimulated for 48 h with PGN-MBL supernatant in the presence of various inhibitors and Abs. The blot was hybridized with probes for hBD-2 and G3PD. Top panel, A graph of the expression of hBD-2 mRNA normalized to the expression of G3PDH mRNA. Expression of hBD-2 mRNA in cultures treated with MBL-control supernatant was assigned the value 1.

Close modal

The expression of hBD-3 in epidermis was induced by supernatant from stimulated MBLs (Fig. 1). The induction of hBD-3 was in all instances inhibited by a mAb against the EGFR (Fig. 5). This is in accordance with our previous finding that TGF-α, a ligand for EGFR, induces hBD-3 in keratinocytes in vitro (21). However, TGF-α neutralizing Abs were less effective inhibitors of the expression of hBD-3 than the EGFR blocking Ab (Fig. 5), indicating that other ligands of EGFR could contribute to the induction of hBD-3.

FIGURE 5.

Inhibition profile of hBD-3 mRNA induced by PGN-MBL supernatant. Northern blot of total RNA from organotypic epidermal cultures stimulated for 48 h with PGN-MBL supernatant in the presence of various inhibitor and Abs. The blot was hybridized with probes for hBD-3 and G3PD. Top panel, A graph of the expression of hBD-3 mRNA normalized to the expression of G3PDH mRNA. Expression of hBD-3 mRNA in cultures treated with nonstimulated MBL supernatant was assigned the value 1.

FIGURE 5.

Inhibition profile of hBD-3 mRNA induced by PGN-MBL supernatant. Northern blot of total RNA from organotypic epidermal cultures stimulated for 48 h with PGN-MBL supernatant in the presence of various inhibitor and Abs. The blot was hybridized with probes for hBD-3 and G3PD. Top panel, A graph of the expression of hBD-3 mRNA normalized to the expression of G3PDH mRNA. Expression of hBD-3 mRNA in cultures treated with nonstimulated MBL supernatant was assigned the value 1.

Close modal

TGF-α can be produced and secreted by monocytes (22). To determine the origin of TGF-α in our model, we assayed TGF-α released into the medium from stimulated MBL but it was undetectable (Fig. 6). Thus, TGF-α from MBLs did not participate in the induction of hBD-3. Epithelial cells release membrane-bound ligands of EGFR by a metalloprotease-dependent process (23). The released ligands then bind and activate the EGFR, a process known as transactivation of EGFR. If the MBL supernatants induced hBD-3 expression by transactivation of the EGFR, ligands of the receptor should be present in the medium after stimulation of the epidermal cultures with MBL supernatant. Indeed, we found that TGF-α was released in the medium of the epidermal cultures after treatment with supernatant from stimulated MBLs but not with supernatant from unstimulated MBL (Fig. 6). Due to the small volume of the epidermal tissue compared with the volume of the medium the epidermal tissue could have been exposed to a high concentration of TGF-α.

FIGURE 6.

TGF-α release by inflammatory mediators from epidermal cultures. TGF-α was assayed by ELISA in the PGN-MBL supernatant and in the medium from epidermal cultures stimulated with nonstimulated (control) MBL supernatant or PGN-MBL supernatant in the presence of IL-1R antagonist, IL-6 Ab, or galardin.

FIGURE 6.

TGF-α release by inflammatory mediators from epidermal cultures. TGF-α was assayed by ELISA in the PGN-MBL supernatant and in the medium from epidermal cultures stimulated with nonstimulated (control) MBL supernatant or PGN-MBL supernatant in the presence of IL-1R antagonist, IL-6 Ab, or galardin.

Close modal

To determine whether TGF-α was released into the medium by a metalloprotease-dependent process, the epidermal cultures were treated with MBL supernatants in the presence of a metalloprotease inhibitor, galardin. Galardin abolished the release of TGF-α into the medium, demonstrating that the release of TGF-α was indeed a metalloprotease-dependent process (Fig. 6). The expression of hBD-3 was markedly decreased by galardin in parallel with the inhibition of TGF-α shedding (Fig. 5). Galardin did not affect the induced expression of hBD-1 (Fig. 3) or hBD-2 (Fig. 4). Thus, the effect of galardin was specific for hBD-3 expression.

Because TGF-α is one of several known ligands of EGFR, we next tested the ability of other ligands of EGFR to induce the expression of hBD-3 in keratinocytes in vitro. Although TGF-α was the most potent, three other ligands of EGFR, EGF, amphiregulin, and HB-EGF, were also capable of inducing hBD-3 expression in keratinocytes (Fig. 7). Moreover, blocking Abs to HB-EGF and amphiregulin also inhibited hBD-3 expression (Fig. 5) but none of the neutralizing Abs to TGF-α, HB-EGF, or amphiregulin was capable of inhibiting the expression of hBD-3 to the same extent as blocking Abs to EGFR. These findings indicated that several EGFR ligands participated in the induction of hBD-3.

FIGURE 7.

Expression of hBD-3 in keratinocytes after stimulation with ligands of the EGFR. Total RNA Northern blot from keratinocytes stimulated with ligands for EGFR: TGF-α, HB-EGF, amphiregulin, and EGF. Blots were hybridized with probes for hBD-3 and G3PD. Top panel, A graph of the expression of hBD-3 mRNA normalized to the expression of G3PD mRNA. Expression of hBD-3 in keratinocytes stimulated with TGF-α is set to value 1.

FIGURE 7.

Expression of hBD-3 in keratinocytes after stimulation with ligands of the EGFR. Total RNA Northern blot from keratinocytes stimulated with ligands for EGFR: TGF-α, HB-EGF, amphiregulin, and EGF. Blots were hybridized with probes for hBD-3 and G3PD. Top panel, A graph of the expression of hBD-3 mRNA normalized to the expression of G3PD mRNA. Expression of hBD-3 in keratinocytes stimulated with TGF-α is set to value 1.

Close modal

MBL supernatant or concentrated medium from epidermal cultures stimulated with MBL supernatant were assayed for protease activity in sensitive casein and gelatin zymography assays. However, no protease activity was detected in the MBL supernatants or medium from stimulated epidermal cultures (data not shown). This is consistent with studies that implicate membrane-bound metalloproteases in the shedding of the EGFR ligands (23). In the aggregate, these data strongly support that hBD-3 is induced by transactivation of EGFR. Experiments with LPS-MBL and SpeB-MBL yielded similar results as with PGN-MBL.

What are the signals that triggered the transactivation of EGFR? IL-1R antagonist and IL-6 blocking Abs (IL-6 Ab) were consistently found to partially inhibit the expression of hBD-3 in epidermal cultures stimulated with MBL supernatants (Fig. 5). However, neither IL-1 nor IL-6 induced hBD-3 expression in keratinocytes (21). In separate experiments, we found that IL-1 and IL-6 did not amplify the expression of hBD-3 induced by TGF-α in epidermal cultures (data not shown). The early events in transactivation of EGFR are poorly understood, but many mediators are capable of initiating and influencing this process (23). Because IL-1R antagonist and IL-6 Ab inhibited hBD-3 expression, we asked whether IL-1R antagonist and IL-6 Ab could inhibit TGF-α release from PGN-MBL supernatant-treated epidermal cultures. Indeed, IL-1R antagonist and IL-6 Ab inhibited the shedding of TGF-α in the medium in parallel to the inhibition of hBD-3 expression (Figs. 5 and 6). Thus, the effects of IL-1R antagonist and IL-6 Ab on hBD-3 expression were due to inhibition of the transactivation of the EGFR by MBL supernatants.

We asked next whether PGN-MBL also induced the expression of EGFR or any of its ligands (TGF-α, EGF, HB-EGF, amphiregulin, betacellulin, or epiregulin) in epidermal cultures. By real-time PCR we found an increase of TGF-α (5-fold), HB-EGF (7-fold), and amphiregulin (4-fold) but not of EGFR, epiregulin, EGF, or betacellulin. This increase was not found in epidermal cultures stimulated with mononuclear supernatants in the presence of galardin. Stimulation of EGFR is known to cause increased expression of the ligands (24). The induced expression of TGF-α, HB-EGF, and amphiregulin could serve to replenish shed ligands in the plasma membrane of the keratinocytes.

To determine whether both monocytes and lymphocytes could participate in the induction of the β-defensins, monocytes, and lymphocytes were separated (see Fig. 8,A) and then separately stimulated with LPS, PGN, and SpeB. The supernatants were subsequently used for stimulation of epidermal cultures (Fig. 8 B). The functional separation of lymphocytes and monocytes was validated by examination of the hBD-2 expression in epidermal cultures after treatment with stimulated supernatants. Supernatant from LPS-stimulated lymphocytes produced only a small induction of hBD-2, but the supernatant from LPS-stimulated monocytes gave a much higher hBD-2 expression. Conversely, SpeB-stimulated lymphocytes were a strong inducer of hBD-2, but this was not the case with SpeB-stimulated monocytes. Thus, there are functional differences in the monocyte- and lymphocyte-mediated expression of antimicrobial peptides in response to specific microbe-derived molecules.

FIGURE 8.

Expression of β-defensins in the epidermis after stimulation with supernatants from isolated lymphocytes or monocytes. A, Cytospin of isolated lymphocytes and monocytes. B, Northern blot of total RNA from epidermal cultures stimulated with supernatants from purified lymphocytes and monocytes treated with LPS, PGN, or SpeB. The blots were hybridized with probes for hBD-1, hBD-2, hBD-3, and G3PD. Graphs shown at the top of the Northern blots are of the expression of hBD-1, hBD-2, and hBD-3 mRNA normalized to the expression of G3PD mRNA. Expression of hBDs in cultures treated with nonstimulated lymphocyte supernatant were assigned the value 1.

FIGURE 8.

Expression of β-defensins in the epidermis after stimulation with supernatants from isolated lymphocytes or monocytes. A, Cytospin of isolated lymphocytes and monocytes. B, Northern blot of total RNA from epidermal cultures stimulated with supernatants from purified lymphocytes and monocytes treated with LPS, PGN, or SpeB. The blots were hybridized with probes for hBD-1, hBD-2, hBD-3, and G3PD. Graphs shown at the top of the Northern blots are of the expression of hBD-1, hBD-2, and hBD-3 mRNA normalized to the expression of G3PD mRNA. Expression of hBDs in cultures treated with nonstimulated lymphocyte supernatant were assigned the value 1.

Close modal

The possible role for lymphocytes as modulators of the expression of epithelial antimicrobial peptides has not been previously described. Intraepithelial lymphocytes and those recruited during inflammation could contribute to the increased expression of β-defensins after microbial challenge.

Supernatant from unstimulated monocytes induced hBD-1 expression more than the supernatant from nonstimulated lymphocytes (Fig. 5). The more elaborate procedure for isolation of pure monocytes compared with mononuclear cells might cause some stimulation of the monocytes, which leads to secretion of mediators that promote hBD-1 expression.

Monocyte-derived IL-1 has been regarded as the principal inducer of hBD-2 expression (12, 20). To validate that isolated lymphocytes indeed induced hBD-2 by IL-1 signaling, we stimulated epidermal cultures with supernatant from SpeB-challenged lymphocytes in the presence of IL-1R antagonist. This antagonist strongly inhibited the expression of hBD-2, thus demonstrating that lymphocytes also induce hBD-2 expression through IL-1 (Fig. 8).

Whole human skin was obtained from breast or abdominal reduction surgery to examine the microbe-derived molecules could induce the expression of hBD-1, hBD-2, and hBD-3. After stimulation with LPS, PGN, or SpeB, mRNAs for all three β-defensins were increased (Table II). At the peptide level the induction of the β-defensins after stimulation was most evident in areas of the part of the skin not damaged by wounding as shown in Fig. 9 A.

Table II.

Induction of hBD mRNA in whole skin stimulated with LPS, PGN, and SpeBa

hBD-1hBD-2hBD-3
LPS 1.95 (1.4–2.5) 31.5 (8–55) 5.1 (4.5–5.7) 
SpeB 1.5 (1.1–1.9) 16.0 (9–23) 3.3 (1.9–4.6) 
PGN 1.1 (0.9–1.2) 39.5 (25–52) 3.8 (2.9–4.8) 
hBD-1hBD-2hBD-3
LPS 1.95 (1.4–2.5) 31.5 (8–55) 5.1 (4.5–5.7) 
SpeB 1.5 (1.1–1.9) 16.0 (9–23) 3.3 (1.9–4.6) 
PGN 1.1 (0.9–1.2) 39.5 (25–52) 3.8 (2.9–4.8) 
a

The induction of hBD mRNA in whole skin stimulated with LPS, PGN, and SpeB was verified by real-time RT-PCR with RNA from two independent experiments. Data are presented as fold induction of hBD mRNA in stimulated whole skin compared with unstimulated whole skin. The range of induction in the experiments is shown in parentheses.

FIGURE 9.

Expression of β-defensins in whole epidermis. A, Skin slices were stained for hBD-1, hBD-2, and hBD-3 after 2.5 days stimulation with LPS, PGN, or SpeB. Color was developed with fast red chromogen, and Harris hematoxylin was used for counterstaining. The induction of the β-defensins after stimulation was most evident in areas of the skin not damaged by wounding as shown. B, Northern blot of total RNA from whole skin stimulated with PGN for 48 h. The blots were hybridized with probes for hBD-2, hBD-3, and G3PD. At the top of each Northern blot, a graph shows the hBD-2 and hBD-3 mRNA normalized to the expression of G3PD mRNA. hBD mRNA concentrations in skin treated with PGN were assigned the value 1.

FIGURE 9.

Expression of β-defensins in whole epidermis. A, Skin slices were stained for hBD-1, hBD-2, and hBD-3 after 2.5 days stimulation with LPS, PGN, or SpeB. Color was developed with fast red chromogen, and Harris hematoxylin was used for counterstaining. The induction of the β-defensins after stimulation was most evident in areas of the skin not damaged by wounding as shown. B, Northern blot of total RNA from whole skin stimulated with PGN for 48 h. The blots were hybridized with probes for hBD-2, hBD-3, and G3PD. At the top of each Northern blot, a graph shows the hBD-2 and hBD-3 mRNA normalized to the expression of G3PD mRNA. hBD mRNA concentrations in skin treated with PGN were assigned the value 1.

Close modal

However, experiments in whole human skin require that skin be removed by surgery. This surgically removed skin is wounded and the wounding by itself influenced the expression of hBDs to varying degree (O. E. Sørensen, D. R. Thapa, J. Kim, A. A. Roberts, and T. Ganz, manuscript in preparation). By real-time PCR, we found a 30-fold induction of hBD-3 in wounded skin after 48 h of culture without additional stimulation and this induction was also very evident by Northern blotting (Fig. 9,B). The responses to wounding and ischemia in the surgically removed skin complicate experiments in whole skin. However, when surgically removed skin was stimulated with PGN, the induced expression of hBD-2 was clearly inhibited by IL-1R antagonist but not by neutralizing EGFR Abs (Fig. 9,B). Although substantial induction of hBD-3 expression was found in the wounded skin over time even without added stimuli, this expression was further enhanced by PGN. The additional PGN-induced hBD-3 expression was inhibited by IL-1R antagonist, EGFR Ab, and galardin but not by the isotype control Abs (Fig. 9 B). Hence, the additional PGN-induced hBD-3 expression had the same characteristics as the induced hBD-3 expression in epidermal keratinocyte cultures treated with stimulated MBL supernatants.

These data together with the results from organotypic epidermal keratinocyte cultures demonstrate that whole human skin responds to LPS, PGN, and SpeB by increasing β-defensin expression and that this response is likely mediated by monocyte-derived cells and by intraepithelial lymphocytes. Consequently, at least three mechanisms exist for induction of antimicrobial peptides in human skin following exposure to microbe-derived molecules: the induction of hBD-2 is IL-1-dependent, the induction of hBD-3 is a result of transactivation of the EGFR, and the mechanism that increases hBD-1 expression is distinct from the other two.

Because SpeB is produced by S. pyogenes, we examined the activity of hBD-1, hBD-2, and hBD-3 toward S. pyogenes by CFU assay. All three hBD types were active toward group A streptococcus (GAS), with hBD-3 being by far the most potent with a three log reduction of the CFU count at 0.5 μg/ml (data not shown).

In the last few years much has been learned about how microbes and microbe-derived molecules like LPS and PGN activate the innate immune response by interaction with specific receptors, including the TLRs (25). Despite the progress in understanding the functions of TLRs and their ligands and signaling pathways (26), it is less evident how activation of the innate immune responses leads to production of antimicrobial peptides in epithelial cells. It has previously been demonstrated that hBD-2 expression in response to LPS is mediated by IL-1 signaling from monocyte-derived cells (12). However, the skin produces many antimicrobial peptides not induced by IL-1 (21). Although hBD-3 has been found in inflamed epithelia (9, 13), the link between inflammation and hBD-3 expression was not defined on a molecular level.

We found that the three microbe-derived molecules LPS, PGN, and SpeB were all capable of inducing or increasing the expression of hBD-1, hBD-2, and hBD-3 in keratinocytes of whole human skin. Such induction of these antimicrobial peptides was not found in similar experiments with organotypic epidermal cultures consisting only of keratinocytes. Hence, additional cell types were necessary for the induction of these β-defensins. It has previously been found that monocyte/macrophage-like cells are necessary for microbe-induced epithelial production of hBD-2 (12, 20). In an extension of these studies, we found that the supernatants from stimulated MBLs induced the expression of hBD-1, hBD-2, and hBD-3 in organotypic epidermal keratinocyte cultures. In addition to monocyte-derived cells, lymphocytes also possess TLRs (27, 28), and intraepithelial lymphocytes are present in the skin and other epithelial surfaces (29). Additional lymphocytes are recruited to chronically inflamed epithelia where induced expression of hBD-2 and hBD-3 is found (13). We asked whether lymphocytes could contribute to the induction of antimicrobial peptides in epithelia during inflammation. Indeed, supernatants from purified lymphocytes induced expression of the three β-defensins. Thus, lymphocytes may play a role in the induction of epithelial antimicrobial peptides in vivo, and this effect might be particularly important in chronically inflamed epithelia. It is noteworthy that the lymphocytes and monocytes responded differently to SpeB and LPS. Thus, some microbe-derived molecules may induce epithelial antimicrobial peptides mainly by a lymphocyte-mediated response and others by a monocyte/macrophage/dendritic cell-mediated response.

Even though the expression of all three β-defensins was induced in epidermal cultures by the stimulated MBL supernatants, we found striking differences in the pathways involved. In accordance with previous findings (12, 18), we found that LPS induced hBD-2 through IL-1 signaling. PGN and SpeB also induced hBD-2 through IL-1 signaling. Regardless of stimuli, hBD-2 regulation was predominantly IL-1 dependent. Thus, keratinocytes produced hBD-2 as a direct response to a secreted cytokine from MBLs.

hBD-3 expression was induced by a very different mechanism in which hBD-3 production was dependent on transactivation of the EGFR. In this process EGFR is activated by membrane-bound ligands released in close vicinity to the receptor through a metalloprotease-dependent process. This leads to a rapid localized activation of the receptors by low concentrations of ligands (30). The initial events in transactivation of EGFR are presumed to involve G-protein receptors that activate the metalloproteases responsible for release of the membrane-bound EGFR ligands (23). Many signals thus converge and result in EGFR-dependent signaling. This process is important for various biological processes including normal growth and development (31), but this is the first time this process has been linked to inflammation or innate immune response. Interestingly we found the proinflammatory cytokines like IL-1 and IL-6 may modulate the transactivation of the EGFR leading to increased expression of hBD-3 during inflammation even though these cytokines did not induce hBD-3 expression directly.

Stimulated MBL supernatants increased the production of hBD-1 by a separate mechanism whose molecular details are yet to be identified. However, the fact that stimulated MBL supernatants increased the expression of hBD-1 establishes a clear molecular link between inflammation and increased production of hBD-1. It is interesting to note that we found a fairly consistent 3- to 4-fold increase in the hBD-1 expression. In bronchoalveolar lavage fluid from patients with inflammatory lung disease (diffuse panbronchiolitis), the hBD-1 levels were found to be ∼4-fold higher than in healthy volunteers (15). This increase in hBD-1 production is of the same order of magnitude as the increase of the hBD-1 expression in the skin.

LPS is a structural component of the cell wall of Gram-negative bacteria and varying amounts of PGN are found in cell walls of both Gram-negative and Gram-positive bacteria. The ligands for TLRs identified so far have mainly been structural molecules essential for the microbes. Some bacteria produce specific proteins to facilitate the infection or evade the immune response of the host, the so-called virulence factors. These proteins are not an indispensable structural component of the bacteria. Apart from LPS and PGN we therefore also examined whether nonstructural molecules from microbes such as a specific virulence factor, SpeB, from the important skin pathogen, GAS, could induce the expression of antimicrobial peptides in skin. SpeB plays an important role in the way GAS evades the immune response of the host for example by degrading Igs bound to the bacteria and thus evading the Ig-mediated killing by phagocytosis and complement activation by the host (32). Despite these properties there are conflicting reports on the specific role of SpeB during infection in vivo (32). We found that SpeB induced the expression of hBD in whole skin and that the hBDs, and in particular hBD-3, were very active against GAS (data not shown). Thus, the immunoevasive properties of SpeB are countered by previously unrecognized immunostimulatory activities leading to production of antimicrobial peptides that could act against GAS. In a transgenic mouse model, antimicrobial peptides have previously been found to play an important role in the clearance of SpeB-producing GAS in the skin (3). The finding that specific virulence factors from bacteria stimulated the production of antimicrobial peptides emphasizes their possible role in the intricate relationship between the host and specific microbes.

It is important to note that hBD-2 and also hBD-3 are not only antimicrobial peptides but display proinflammatory properties by stimulating the chemotaxis of immature dendritic cells, T cells, and neutrophils (33, 34, 35, 36). The same has not been found for hBD-1 (34). Thus, hBD seems to have separate “proinflammatory” properties in addition to the different antimicrobial activity of these peptides. The very distinct regulation of the expression of the hBD was furthermore accompanied by a distinct anatomical localization of the produced peptides. It is noteworthy that we found that the hBD-2 produced in the epidermal cultures is mainly secreted and thus capable of acting extracellularly as a chemoattractant. The differential regulation of the expression of each peptide may reflect the differing roles of the peptides both as antimicrobials and as proinflammatory agents.

In summary we found that microbe-derived molecules induced hBD-1, hBD-2, and hBD-3 in human epidermis through signaling involving either lymphocytes or monocyte-derived cells. Apart from structural pathogen-associated molecules like LPS and PGN the skin also induced the expression of hBD-1, hBD-2, and hBD-3 in response to a nonstructural virulence factor, SpeB. Thus, the skin and other epithelia might counter bacteria with increased virulence/production of virulence factors by increased production of antimicrobial peptides to eliminate these microbes. The induction of the β-defensins was mediated by three distinct molecular mechanisms and our study establishes for the first time a clear molecular link between inflammation and the induced expression of hBD-1 and hBD-3. The latter is induced by transactivation of the EGFR, a novel molecular mechanism for the induction of antimicrobial peptides in epithelial cells. It remains to be seen whether transactivation of EGFR could also locally induce antimicrobial peptides during other vulnerable states associated with growth or repair, such as pregnancy, the birth process, or wound repair.

The authors have no financial conflict of interest.

We acknowledge the valuable contributions of Erika Valore, Host Defense Laboratory (University of California, Los Angeles, CA) for generating polyclonal Ab to hBD-3 from hBD-3 synthesized chemically by Wuyuan Lu, University of Maryland (Baltimore, MD).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by National Institutes of Health Grants RO1 HL 46809 and P50 HL67665. O.E.S. is a recipient of an Alfred Benzon Fellowship from the Alfred Benzon Foundation.

3

Abbreviations used in this paper: hBD, human β-defensin; MBL, mononuclear blood leukocyte; PGN, peptidoglycan; EGF, epidermal growth factor; HB-EGF, heparin-binding EGF; PVDF, polyvinylidene difluoride; GAS, group A streptococcus.

1
Ganz, T..
2002
. Epithelia: not just physical barriers.
Proc. Natl. Acad. Sci. USA
99
:
3357
.
2
Zasloff, M..
2002
. Antimicrobial peptides of multicellular organisms.
Nature
415
:
389
.
3
Nizet, V., T. Ohtake, X. Lauth, J. Trowbridge, J. Rudisill, R. A. Dorschner, V. Pestonjamasp, J. Piraino, K. Huttner, R. L. Gallo.
2001
. Innate antimicrobial peptide protects the skin from invasive bacterial infection.
Nature
414
:
454
.
4
Moser, C., D. J. Weiner, E. Lysenko, R. Bals, J. N. Weiser, J. M. Wilson.
2002
. β-Defensin 1 contributes to pulmonary innate immunity in mice.
Infect. Immun.
70
:
3068
.
5
Morrison, G., F. Kilanowski, D. Davidson, J. Dorin.
2002
. Characterization of the mouse β defensin 1, Defb1, mutant mouse model.
Infect. Immun.
70
:
3053
.
6
Nomura, I., E. Goleva, M. D. Howell, Q. A. Hamid, P. Y. Ong, C. F. Hall, M. A. Darst, B. Gao, M. Boguniewicz, J. B. Travers, D. Y. Leung.
2003
. Cytokine milieu of atopic dermatitis, as compared to psoriasis, skin prevents induction of innate immune response genes.
J. Immunol.
171
:
3262
.
7
Ong, P. Y., T. Ohtake, C. Brandt, I. Strickland, M. Boguniewicz, T. Ganz, R. L. Gallo, D. Y. M. Leung.
2002
. Endogenous antimicrobial peptides and skin infections in atopic dermatitis.
N. Engl. J. Med.
347
:
1151
.
8
Harder, J., J. Bartels, E. Christophers, J.-M. Schröder.
1997
. A peptide antibiotic from human skin.
Nature
387
:
861
.
9
Harder, J., J. Bartels, E. Christophers, J. M. Schroder.
2001
. Isolation and characterization of human β-defensin-3, a novel human inducible peptide antibiotic.
J. Biol. Chem.
276
:
5707
.
10
Ganz, T..
2003
. Defensins: antimicrobial peptides of innate immunity.
Nat. Rev. Immunol.
3
:
710
.
11
Fulton, C., G. M. Anderson, M. Zasloff, R. Bull, A. G. Quinn.
1997
. Expression of natural peptide antibiotics in human skin.
Lancet
350
:
1750
.
12
Liu, L., A. A. Roberts, T. Ganz.
2003
. By IL-1 signaling, monocyte-derived cells dramatically enhance the epidermal antimicrobial response to lipopolysaccharide.
J. Immunol.
170
:
575
.
13
Wehkamp, J., J. Harder, M. Weichenthal, O. Mueller, K. R. Herrlinger, K. Fellermann, J. M. Schroeder, E. F. Stange.
2003
. Inducible and constitutive β-defensins are differentially expressed in Crohn’s disease and ulcerative colitis.
Inflamm. Bowel Dis.
9
:
215
.
14
Zhao, C., I. Wang, R. I. Lehrer.
1996
. Widespread expression of β-defensin hBD-1 in human secretory glands and epithelial cells.
FEBS Lett.
396
:
319
.
15
Hiratsuka, T., H. Mukae, H. Iiboshi, J. Ashitani, K. Nabeshima, T. Minematsu, N. Chino, T. Ihi, S. Kohno, M. Nakazato.
2003
. Increased concentrations of human β-defensins in plasma and bronchoalveolar lavage fluid of patients with diffuse panbronchiolitis.
Thorax
58
:
425
.
16
Zhu, B. D., Y. Feng, N. Huang, Q. Wu, B. Y. Wang.
2003
. Mycobacterium bovis bacille Calmette-Guerin (BCG) enhances human β-defensin-1 gene transcription in human pulmonary gland epithelial cells.
Acta Pharmacol. Sin.
24
:
907
.
17
Valore, E. V., C. H. Park, A. J. Quayle, K. R. Wiles, P. B. McCray, Jr, T. Ganz.
1998
. Human β-defensin-1: an antimicrobial peptide of urogenital tissues.
J. Clin. Invest.
101
:
1633
.
18
Liu, A. Y., D. Destoumieux, A. V. Wong, C. H. Park, E. V. Valore, L. Liu, T. Ganz.
2002
. Human β-defensin-2 production in keratinocytes is regulated by interleukin-1, bacteria, and the state of differentiation.
J. Invest. Dermatol.
118
:
275
.
19
Pfaffl, M. W..
2001
. A new mathematical model for relative quantification in real-time RT-PCR.
Nucleic Acids Res.
29
:
45
.
20
Tsutsumi-Ishii, Y., I. Nagaoka.
2003
. Modulation of human β-defensin-2 transcription in pulmonary epithelial cells by lipopolysaccharide-stimulated mononuclear phagocytes via proinflammatory cytokine production.
J. Immunol.
170
:
4226
.
21
Sørensen, O. E., J. B. Cowland, K. Theilgaard-Mönch, L. Liu, T. Ganz, N. Borregaard.
2003
. Wound healing and expression of antimicrobial peptides/polypeptides in keratinocytes, a consequence of common growth factors.
J. Immunol.
170
:
5583
.
22
Calafat, J., H. Janssen, B. M. Stahle, A. E. Zuurbier, E. F. Knol, A. Egesten.
1997
. Human monocytes and neutrophils store transforming growth factor-α in a subpopulation of cytoplasmic granules.
Blood
90
:
1255
.
23
Blobel, C. P..
2000
. Remarkable roles of proteolysis on and beyond the cell surface.
Curr. Opin. Cell Biol.
12
:
606
.
24
Coffey, R. J., R. Derynck, C. Wilcox, T. S. Bringman, A. S. Goustin, H. L. Moses, M. R. Pittelkow.
1987
. Production and auto-induction of transforming growth factor-α in human keratinocytes.
Nature
328
:
817
.
25
Hoffmann, J. A., F. C. Kafatos, C. A. Janeway, R. B. Ezekowitz.
1999
. Phylogenetic perspectives in innate immunity.
Science
284
:
1313
.
26
Beutler, B., K. Hoebe, X. Du, R. J. Ulevitch.
2003
. How we detect microbes and respond to them: the Toll-like receptors and their transducers.
J. Leukocyte Biol.
74
:
479
.
27
Caramalho, I., T. Lopes-Carvalho, D. Ostler, S. Zelenay, M. Haury, J. Demengeot.
2003
. Regulatory T cells selectively express Toll-like receptors and are activated by lipopolysaccharide.
J. Exp. Med.
197
:
403
.
28
Coutinho, A., A. Poltorack.
2003
. Innate immunity: from lymphocyte mitogens to Toll-like receptors and back.
Curr. Opin. Immunol.
15
:
599
.
29
Hayday, A., E. Theodoridis, E. Ramsburg, J. Shires.
2001
. Intraepithelial lymphocytes: exploring the third way in immunology.
Nat. Immunol.
2
:
997
.
30
Prenzel, N., E. Zwick, H. Daub, M. Leserer, R. Abraham, C. Wallasch, A. Ullrich.
1999
. EGF receptor transactivation by G-protein-coupled receptors requires metalloproteinase cleavage of proHB-EGF.
Nature
402
:
884
.
31
Chang, C., Z. Werb.
2001
. The many faces of metalloproteases: cell growth, invasion, angiogenesis and metastasis.
Trends Cell Biol.
11
:
S37
.
32
Pawel-Rammingen, U., L. Bjorck.
2003
. IdeS and SpeB: immunoglobulin-degrading cysteine proteinases of Streptococcus pyogenes.
Curr. Opin. Microbiol.
6
:
50
.
33
Yang, D., O. Chertov, N. Bykovskaia, Q. Chen, M. J. Buffo, J. Shogan, M. Anderson, J. M. Schroder, J. M. Wang, O. Z. Howard, J. J. Oppenheim. β-defensins: linking innate and adaptive immunity through dendritic and T cell CCR6.
Science
286
:
99 A.D.
525
.
34
Niyonsaba, F., H. Ogawa, I. Nagaoka.
2004
. Human β-defensin-2 functions as a chemotactic agent for tumour necrosis factor-α-treated human neutrophils.
Immunology
111
:
273
.
35
Niyonsaba, F., K. Iwabuchi, H. Matsuda, H. Ogawa, I. Nagaoka.
2002
. Epithelial cell-derived human β-defensin-2 acts as a chemotaxin for mast cells through a pertussis toxin-sensitive and phospholipase C-dependent pathway.
Int. Immunol.
14
:
421
.
36
Wu, Z. B., D. M. Hoover, D. Yang, C. Boulegue, F. Santamaria, J. J. Oppenheim, J. Lubkowski, W. Y. Lu.
2003
. Engineering disulfide bridges to dissect antimicrobial and chemotactic activities of human β-defensin 3.
Proc. Natl. Acad. Sci. USA
100
:
8880
.