New mechanisms of skin innate immunity: ASK1-mediated keratinocyte differentiation regulates the expression of β-defensins, LL37, and TLR2

Authors


Abstract

Epidermal keratinocytes differentiate and form a multilayered epidermis, which is the primary barrier between the body and the outer environment. As the epidermis is constantly exposed to a variety of microbial pathogens, its function of resisting microbial pathogens is vital. This characteristic feature is formed during differentiation. Immunohistochemical analysis revealed that the upper epidermis of normal human skin expresses β-defensins 1–3 and LL37. We hypothesized that epidermal keratinocytes develop an innate immune barrier based on human β-defensins (hBD) and LL37 during differentiation. To prove this, we introduced an active form of the apoptosis signal-regulating kinase-1 (ASK1), an intracellular regulator of keratinocyte differentiation, into cultured normal human keratinocytes. Transfection of this active form, ASK1-ΔN, significantly enhanced the expression of hBD1–3 and LL37. In addition, a p38 inhibitor abolished this induction, indicating that the ASK1-p38 cascade regulates the expression of hBD1–3 and LL37. Furthermore, the ASK1-p38 pathway also regulated the expression of Toll-like receptor (TLR)2 in keratinocytes. Contact between S. aureus and keratinocytes resulted in the phosphorylation of p38 and induced the expression of hBD2 and hBD3. Moreover, the p38 inhibitor reduced this induction. In conclusion, the ASK1-p38 cascade regulates the innate immunity of the skin by forming an immune barrier consisting of hBD, LL37, and TLR2 during epidermal differentiation.

Abbreviations:
AD:

Atopic dermatitis

ASK1:

Apoptosis signal-regulating kinase-1

BHE:

Bovine hypothalamic extract

GAPDH:

Glyceraldehyde 3-phosphate dehydrogenase

hBD:

Human β-defensins

JNK:

Jun N-terminal kinase

LTA:

Lipoteichoic acid

Introduction

The epidermis is the primary barrier between the body and the outer environment. As the epidermis is constantly exposed to a variety of microbial pathogens, its function of resisting microbial pathogens is vital. Epidermal keratinocytes differentiate and form a physical barrier consisting of a multilayered epidermis. This multilayered structure is characteristic of keratinocytes and is totally different from the simple epithelia that cover most of the gastrointestinal tract, the respiratory tract, and the urinary system. In addition to this physical barrier, the epidermis functions as an innate immune barrier to resist microbial pathogens. This characteristic feature also arises during differentiation. Dysregulation of innate immunity in the epidermis may increase susceptibility to skin infection, such as in atopic dermatitis (AD) 1. Few recent studies have considered the regulation of the immune function of epidermal keratinocytes by differentiation and the alterations with abnormal differentiation in the pathogenesis of disease.

Epidermal keratinocytes produce antimicrobial peptides, such as β-defensins and LL37. Antimicrobial peptides are the effector molecules of the innate immune system. The major human antimicrobial peptides are defensin and cathelicidin 25. Defensins are cysteine-rich cationic peptides and are further classified into α- and β-defensins by their structure. α-Defensins (HNP1–4) are produced by neutrophils 2, 6, and α-defensins 5 and 6 are found in Paneth cells of the gastrointestinal tract 7, 8. Human β-defensins (hBD)1–3 are also produced in the epithelium 911. LL37 is the only human antimicrobial peptide identified in the cathelicidin family; it is produced in the epithelium and in neutrophils 5. LL37 is secreted in human eccrine sweat to deliver innate effector molecules in the absence of inflammation 12. Although these peptides were originally identified as effector molecules against microorganisms, it was recently found that they also mediate inflammatory responses, similar to the action of cytokines or chemokines. hBD are chemotactic for dendritic cells; LL37 is chemotactic for neutrophils, monocytes, and T cells, but not for dendritic cells 4, 13. In addition to its effects on leukocytes, LL37 activates endothelial cells directly, resulting in increased proliferation and formation of vessel-like structures in cultivated endothelial cells and neovascularization 14. Furthermore, LL37 binds and neutralizes LPS and lipoteichoic acids (LTA) 5, which are mediators of inflammation.

Previously, we showed that apoptosis signal-regulating kinase-1 (ASK1), a member of the MAP kinase superfamily, regulates the late-phase differentiation of epidermal keratinocytes 15. ASK1 was identified as a MAP kinase kinase kinase involved in the stress-induced apoptosis signaling cascade that activates the SEK1-Jun N-terminal kinase (JNK) and MKK-p38 MAP kinase cascade 16. ASK1 is activated by death receptor ligands, such as TNF-α and Fas ligand, and by various cytotoxic stresses, such as hydrogen peroxide, anticancer drugs, and growth factor deprivation 17. Furthermore, endoplasmic reticulum stress and G protein-coupled receptor signaling were recently shown to activate ASK1 17. In normal human keratinocytes, ceramide increases the expression and activity of ASK1, and in turn, ASK1 induces keratinocyte differentiation. As ASK1 is expressed in the upper epidermis, it has been suggested that ASK1 regulates late-phase differentiation 15. Recently, a tie between ASK1 and innate immunity has been reported. C. elegans has an ASK1 homologue, NSY-1, and a p38 homologue, PMK-1. Mutations of these genes resulted in enhanced susceptibility to microbial pathogens in C. elegans 18. Therefore, we hypothesized that the ASK1-p38 cascade regulates the innate immune response in epidermal keratinocytes via differentiation. To prove this, we transfected ASK1 into cultured normal human keratinocytes and asked whether ASK1 regulates the skin innate immunity.

Results

Expression of hBD1–3, LL37, and ASK1 in normal human epidermis

Polyclonal anti-hBD1–3 and anti-LL37 antibodies were raised against synthetic peptides. The specificity of the antibodies was evaluated by Western blot (Fig. 1). Neither of the antibodies reacted to the other peptide.

Figure 1.

Specificity of anti-hBD1–3 and LL37 antibodies. hBD1–3 and LL37 (100 ng/well) were applied to 15% SDS-PAGE with Tris/tricine buffer and transferred to PVDF membranes. After blocking, the membranes were reacted with anti-hBD1–3 or anti-LL37 antibodies overnight at 4°C. The signals were detected with an ECL detection kit.

We first analyzed the localization of antimicrobial peptides in normal human epidermis. hBD1, hBD2, hBD3, and LL37 were expressed in the upper epidermis (Fig. 2), suggesting that the expression is induced by differentiation. ASK1 is also expressed in the upper epidermis, as previously described 15.

Figure 2.

Expression of hBD1–3, LL37, and ASK1 in normal human epidermis. Frozen sections of normal human skin were reacted overnight with anti-hBD1–3, anti-LL37, and anti-ASK1 antibodies. The first antibodies were detected with the corresponding second antibody using streptavidin-biotin-peroxidase and visualized with AEC. The nucleus was counterstained with hematoxylin. (A) hBD1, (B) hBD2, (C) hBD3, (D) LL37, (E) ASK1, (F) mouse pre-immune serum, and (G) rabbit pre-immune serum (600×).

Induction of hBD1–3 and LL37 by the ASK1-p38 pathway

Therefore, we hypothesized that ASK1 regulates the expression of hBD and LL37. To prove this, we introduced an active form of ASK1 (ASK1-ΔN) into cultured normal human keratinocytes using an adenovirus vector (Ax-ASK1-ΔN). Ax-β-gal carrying β-galactosidase was used as a control vector. After transfecting ASK1-ΔN into normal human keratinocytes, the mRNA expression of antimicrobial peptides was analyzed using an RNase protection assay. Transfection of ASK1-ΔN significantly enhanced the expression of hBD1, hBD2, hBD3, and LL37 mRNA (Fig. 3). The control vector had no effect on the mRNA expression. To further confirm the induction of enhanced production of antimicrobial peptides by ASK1, culture supernatants were concentrated and applied to membranes for dot blot analysis (Fig. 4). The hBD1-3 signals in supernatants of keratinocytes transfected with Ax-ASK1-ΔN were significantly higher than the signals of the control (Ax-β-gal). However, the LL37 signal of cells transfected with Ax-ASK1-ΔN was almost the same as that of the control. The discrepancy between the data for mRNA expression (Fig. 3) and the dot blot analysis for LL37 may be attributable to the mechanism of production of LL37. As LL37 is produced by extracellular cleavage of full-length cathelicidin hCAP18 by proteolytic activity after secretion 5, enhanced mRNA expression and translation of the premature form alone may not be sufficient to increase the level of LL37 in the culture supernatant in vitro. The supernatant of the control (Ax-β-gal) was positive for both anti-hBD1–3 and anti-LL37 antibodies when tested at higher concentrations, suggesting that keratinocytes produce small amounts of these peptides even at non-differentiated conditions.

Figure 3.

Induction of hBD1-3 and LL37 by ASK1. Cultured normal human keratinocytes were infected with Ax-ASK1-ΔN or Ax-β-gal at an MOI of 5. Ax-β-gal is a control vector. After infection with the adenovirus vectors for 24 h, mRNA expression of hBD1–3 and LL37 was analyzed using an RNase protection assay. GAPDH is an internal standard.

Figure 4.

Dot blot analysis of hBD1–3 and LL37. After transfection of the adenovirus vectors into normal human keratinocytes in 10-cm dishes at an MOI of 5, the medium was replaced with 5 ml medium without BHE, and the cells were cultured for an additional 2 days. After adjusting the pH of the supernatants to 7.2, 1.0 ml of pre-treated MacroPrep beads was added to 100 ml of the culture supernatants, gently shaken overnight at 4°C, washed with 25 mM ammonium acetate buffer at pH 7.4 and then eluted with 1.0 ml 5% acetic acid. The eluates were freeze-dried and redissolved in 1.0 ml distilled water. Samples (1 µl) of serially diluted, reconstituted supernatants or control solutions containing synthetic peptides (hBD1–3 and LL37) were applied to nitrocellulose membranes and air-dried. After blocking, the membranes were reacted overnight at 4°C with anti-hBD1–3 or anti-LL37 antibodies diluted 1:1,000. After extensive washing, the signals were detected using an ECL kit.

As p38 is located downstream of ASK1 and is activated by ASK1 in keratinocytes 15, we next analyzed whether p38 is involved in the induction of antimicrobial peptides by ASK1. Normal human keratinocytes were pretreated with a p38 inhibitor (SB203580), a JNK inhibitor (SP600125), and a MAPK/ERK kinase (MEK) inhibitor (PD098059), and then ASK1-ΔN was transfected. mRNA expression was analyzed using quantitative RT-PCR. The p38 inhibitor almost completely abolished the induction of hBD1, hBD2, hBD3, and LL37 (Fig. 5). Conversely, the JNK and MEK inhibitors did not reduce the induction of hBD or LL37. Therefore, the ASK1-p38 pathway regulates the expression of hBD and LL37.

Figure 5.

Inhibition of ASK1-induced hBD and LL37 by MAP kinase inhibitors. The involvement of p38, JNK, and MEK was analyzed using specific inhibitors: SB203580 (200 nM), SP600125 (20 µM), and PD098059 (30 µM), respectively. DMSO was the vehicle. Normal human keratinocytes were pretreated with the inhibitors for 1 h. Then, the cells were transfected with the adenovirus vectors. After 24 h, hBD and LL37 mRNA expression was analyzed using quantitative RT-PCR. The data are adjusted to the internal standard GAPDH and represent the fold increase compared with the control. There were three samples in each group. Three independent experiments were carried out, and similar results were obtained in each.

Regulation of TLR2 expression by the ASK1-p38 pathway

Next, we tested whether the ASK1-p38 pathway regulates the expression of TLR2, a receptor for the cell wall components of gram-positive bacteria 19. Although peptidoglycans were previously shown to be a ligand for TLR2, it has recently been suggested that the sensing of peptidoglycans by TLR2 was attributable to the binding of contaminating lipoproteins or LTA from bacterial cell walls that were present in the peptidoglycan preparations 20. ASK1 transfection enhanced TLR2 mRNA expression (Fig. 6A). The p38 inhibitor, SB203580, abolished this enhancement (Fig. 6B). Therefore, in addition to hBD and LL37, the ASK1-p38 cascade regulates TLR2 expression in human keratinocytes. However, ASK1 transfection did not enhance the expression of TLR3, 4, 5, 7, and 9 (data not shown).

Figure 6.

Regulation of TLR2 expression by the ASK1-p38 pathway. (A) Induction of TLR2 expression by ASK1. Induction of TLR2 expression by ASK1 was analyzed by transfecting cultured normal human keratinocytes with Ax-ASK1-ΔN at an MOI of 5. Ax-β-gal was the control vector. After 24 h, TLR2 mRNA expression was analyzed using an RNase protection assay. GAPDH was the internal standard. (B) Inhibition of ASK1-induced TLR2 expression by MAP kinase inhibitors. The involvement of p38, JNK, and MEK was analyzed using specific inhibitors: SB203580 (200 nM), SP600125 (20 µM), and PD098059 (30 µM), respectively. DMSO was the vehicle. Normal human keratinocytes were pretreated with the inhibitors for 1 h. Then, the cells were transfected with the adenovirus vectors. After 24 h, TLR2 mRNA expression was analyzed using quantitative RT-PCR. The data are adjusted to the internal standard GAPDH and represent the fold increase compared with the control. There were three samples in each group. Three independent experiments were carried out, and similar results were obtained in each.

Inhibition of S. aureus-induced hBD2 and hBD3 by SB203580

To study the role of p38 further, we used the model of induction of antimicrobial peptides by S. aureus. Direct contact of S. aureus with keratinocytes induces antimicrobial peptides 21. We studied whether p38 is involved in this induction. After contact with S. aureus, the phosphorylation of p38 started to increase at 0.5 h and lasted until 1 h (Fig. 7A). Furthermore, SB203580 inhibited S. aureus-induced hBD2 and hBD3 expression (Fig. 7B). These results indicate that p38 is involved in S. aureus-induced hBD2 and hBD3 production.

Figure 7.

Induction of hBD2 and hBD3 by S. aureus. Overnight cultures of S. aureus (209P) grown in TSB at 37°C were washed with PBS and then incubated at 68°C for 30 min to inactivate the bacteria. The heat-inactivated bacteria (final concentration, 108 cells/ml) were added to subconfluent keratinocytes in MCDB153 medium without supplements, and the keratinocytes were cultured and harvested for analysis of mRNA expression and p38 phosphorylation. (A) Phosphorylation of p38 was analyzed using Western blotting with anti-phospho-p38 and anti-p38 antibodies at the indicated times for up to 6 h. (B) Normal human keratinocytes were pretreated with the inhibitor SB203580 (200 nM) for 1 h; heat-killed S. aureus was then added to the cultures. The induction of hBD2 and hBD3 was analyzed at 12 h using quantitative RT-PCR. Three independent experiments were carried out, and similar results were obtained in each.

Discussion

The epithelium of the skin (epidermis), the respiratory tract, the gastrointestinal tract, and the urinary system serves as the frontline for immune system resistance to microbial pathogens. The epithelium produces antimicrobial peptides, such as hBD and LL37. The mechanisms most commonly regulating the production of antimicrobial peptides are contact with the pathogenic bacteria and cytokines. Colon epithelial cells produce hBD2 upon stimulation by bacteria 22. In the small intestine, Paneth cells secrete α-defensins in response to bacteria or bacterial antigens 23. Tracheobronchial epithelial cells produce hBD2 after contact with bacterial lipopeptide 24. In epidermis, bacterial contact 9, 21, 25, cytokines including TNF-α and IL-1 9, 25, 26, and wounding 27 are reported to stimulate keratinocytes to produce antimicrobial peptides. Our study clarified a new mechanism: ASK1-p38 cascade-mediated differentiation regulates the production of antimicrobial peptides in epidermal keratinocytes.

Although several reports have shown the expression of hBD2 28, 29 and LL37 30 in normal human epidermis, the expression patterns of these peptides remain controversial, because it has also been reported that the expression levels of hBD2 and LL37 are low in the epidermis in organotypic cultures 26 or absent in normal epidermis 27, 31, 32. This discrepancy may be at least partially attributable to the low levels of expression of hBD2 and LL37 in normal epidermis and to the immunostaining methods used by the different groups. Some authors used frozen sections 30, 31 and others used samples processed without freezing, such as paraffin-embedded tissue sections 2629, 32. Furthermore, different antibodies were used in these studies. Apparently, hBD2, hBD3, and LL37 are inducible by inflammation or infection 26, 28, 3032, as we have also shown in Fig. 7 and in a previous report 21. Therefore, the relatively low expression of these peptides in normal tissues, as compared with the expression levels in the epidermis with inflammation or infection, may not be detectable using some techniques. Taken together, the findings indicate that the expression levels of hBD2, hBD3, and LL37 in the normal epidermis, shown in Fig. 2, should be considered the basal level, which may be further increased by stimulation by inflammation or infection.

A low level of hBD1 mRNA expression was seen in keratinocytes before ASK1 transfection (Fig. 3). This suggests that regulatory mechanisms other than ASK1 affect hBD1 expression. As hBD1 expression in keratinocytes is reported to be regulated by calcium 33 and cell confluency 34, it is most probable that ASK1 is not the sole factor regulating hBD1 expression. In the present study, ASK1 was shown to participate only in the induction phase of hBD1 expression and may not be involved in the regulation of the basal expression of hBD1.

Keratinocytes produce antimicrobial peptides in response to two different stimuli: differentiation and bacterial contact. Although ASK1-mediated differentiation induces hBD1–3, bacterial contact induces only hBD2 and 3 21. This indicates that the mechanisms to induce antimicrobial peptides by differentiation differ from that by bacterial contact. Therefore, keratinocytes have at least two different mechanisms to induce antimicrobial peptides. However, both pathways involve p38, suggesting that signals downstream of p38 are the same in both pathways.

In contrast to the idea that differentiation regulates the expression of the antimicrobial peptides, there is an interesting report suggesting that hBD1 drives the differentiation of keratinocytes 34. As hBD1 is localized to the upper epidermis, it is possible that hBD1 is involved in ASK1-mediated differentiation mechanisms in keratinocytes. Recently it became evident that the roles of the hBD and LL37 are not limited only to their antimicrobial activities but include certain roles in adaptive and innate immunity 4, 5, 13, 14. The concept that hBD1 drives keratinocyte differentiation suggests a new aspect of the functions of antimicrobial peptides.

Bacterial contact activates p38, as shown in Fig. 7. Although p38 regulates the expression of involucrin and differentiation in keratinocytes 15, 35, bacterial contact does not induce involucrin expression or differentiation (data not shown). Similar inconsistencies are also observed with cellular stresses, such as UV irradiation. UV irradiation activates p38, but does not induce involucrin or differentiation (data not shown). However, ASK1-induced p38 activation differentiates keratinocytes 15. Although p38 phosphorylation occurs with UV irradiation, bacterial contact, and differentiation, the strength or duration of the signal may be important in the differential reaction to each stimulus.

Although ASK1 enhanced TLR2 mRNA expression in keratinocytes, we failed to detect TLR2 on keratinocytes by cell sorter analysis (data not shown) using the monoclonal antibody TL2.1, the most commonly used antibody against TLR2, as previously described 36. However, several groups have shown TLR2 expression on primary cultured keratinocytes 37 and normal human epidermis 38 using polyclonal goat anti-TLR2 antibody (Santa Cruz Biotechnology, Santa Cruz, CA). In light of the observations that keratinocytes react with gram-positive bacteria 9, 21, 25 and peptidoglycans 37 (or contaminating lipoproteins or LTA in peptidoglycan preparations 20), the presence of TLR2 on keratinocytes can be assumed. This issue should be studied further, assisted by the development of new antibodies against or ligands for TLR2.

Since skin innate immunity is regulated by keratinocyte differentiation as shown in the present study, abnormal keratinocyte differentiation may disturb skin innate immunity. Among the diseases with abnormal differentiation and innate immunity, AD is the most common skin disease. AD is a highly pruritic chronic inflammatory skin disease. With its increased prevalence, AD has become a major public health problem. AD is complicated by recurrent skin infections, including bacterial, fungal, and viral infections 1. As a skin infection triggers the exacerbation of AD, there is considerable interest in the mechanisms underlying the increased colonization of bacteria on the AD skin lesion. The mechanism that eliminates S. aureus from the epidermis is impaired in AD. This immune abnormality is the result, at least in part, of the reduced expression of antimicrobial peptides in the epidermis 31, 39. The mechanism of reduced expression of antimicrobial peptides is thought to consist in increased levels of IL-4 and IL-13 31. These Th2-type cytokines inhibit the expression of antimicrobial peptides in human keratinocytes 31. In addition, epidermal differentiation is impaired in AD. A hallmark of epidermal differentiation is the cornified envelope in the stratum corneum, an insoluble structure that replaces the plasma membrane of differentiated keratinocytes 40. The cornified envelope in the involved area of AD is immature 41. This physical barrier is formed during keratinocyte differentiation; thus, the barrier disruption in AD 42 is due to impaired differentiation, making AD a disease involving abnormal keratinocyte differentiation. As shown here, an innate immune barrier is formed during differentiation; consequently, the dysregulation of epidermal differentiation in AD would increase the susceptibility to skin infection. This hypothesis should be further studied as a new pathogenesis of AD.

In conclusion, the ASK1-p38 cascade forms an innate epidermal immune barrier consisting of hBD, LL37, and TLR2 during epidermal differentiation.

Materials and methods

Synthesis of antimicrobial peptides

hBD1, 2, and 3 and LL37 were synthesized using a peptide synthesizer (Shimazu, Tokyo, Japan), as described 21. The peptides were purified using reversed-phase high-performance liquid chromatography with an octadecyl-4PW column (Tosoh, Tokyo, Japan) and a linear gradient from aqueous 0.05% trifluoroacetic acid (TFA) to 100% acetonitrile containing 0.05% TFA. The peptides were then lyophilized to completely remove the organic solvent. To confirm the purity and quality of the peptides, mass spectrometry using the MALDI/TOF-MS method was performed with Voyager (PerSeptive Biosystems, MA).

Antibodies

Antiserum to ASK1, called PEL, was raised in rabbits against the peptide sequence PELRPHFSLASESDTAD (amino acids 1172–1188 of human ASK1), as described 16. Rabbit anti-hBD2 and anti-LL37 antibodies were raised against synthetic hBD2 and LL37, respectively. Mouse polyclonal anti-hBD1 and anti-hBD3 antibodies were also raised against synthetic hBD1 and hBD3, respectively.

Immunohistochemical staining

Frozen skin sections were fixed with methanol containing 3% H2O2 at 4°C for 5 min, washed with PBS and blocked with 10% goat serum. The sections were reacted overnight at 4°C with rabbit or mouse first antibody diluted 1:1,000 in PBS. After washing with PBS, the first antibodies were detected using a streptavidin-biotin-peroxidase staining kit (Nichirei Co. Inc., Tokyo, Japan) and visualized with 3-amino-9-ethyl-carbazole (AEC) according to the manufacturer's instructions. Control staining with pre-immune serum showed no positive signal. The nucleus was counterstained with hematoxylin.

Adenovirus vectors

An adenovirus vector (Ax) encoding a constitutively active form of ASK1 (Ax-ASK1-ΔN) was prepared as described 15. The adenovirus vector expressing a bacterial β-galactosidase gene (Ax-β-gal) was used as a control to exclude the effect of the vector itself. Virus stocks were prepared using a standard procedure. Concentrated, purified virus stocks were prepared using the CsCl gradient method, and the virus titer was checked with the plaque formation assay. Normal human keratinocytes were infected with adenovirus vector at a concentration of multiplicity of infection (MOI) equaling 5. Protein expression was confirmed by Western blot analysis.

Keratinocyte culture

Human skin samples were obtained after plastic surgery under a protocol approved by the Institutional Review Board of Ehime University School of Medicine. Primary normal human keratinocytes were isolated from neonatal surgical-discard skin. Normal human keratinocytes were cultured with MCDB153 medium supplemented with insulin (1 µg/ml), hydrocortisone (0.5 µg/ml), ethanolamine (0.1 mM), phosphoethanolamine (0.1 mM), bovine hypothalamic extract (BHE) (50 µg/ml), and Ca2+ (0.1 mM), as described 43.

RNase protection assay

Total RNA was prepared using Isogen (Nippon Gene Co., Tokyo, Japan). The analysis was performed using the Multi-Probe RNase Protection Assay System (PharMingen Co.) according to the manufacturer's guidelines. Oligonucleotide probes were prepared by inserting PCR-amplified human cDNA corresponding to oligonucleotides 151–265 of hBD1 (GenBank accession no. U73945), 44–200 of hBD2 (NM004942), 65–204 of hBD3 (AF295370), 321–517 of CAP18/LL37 (NM004345), and 2204–2513 of TLR2 (AF051152) into the Eco RI and Hind III sites of the pPMG vector. Total RNA (5 µg) was hybridized with a 32P-labeled riboprobe and digested with RNase. The hybridization products were separated on 5% polyacrylamide-8 M urea gels and exposed to film. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was the internal standard.

Quantitative RT-PCR analysis

To quantify the mRNA expression in vivo, we performed quantitative RT-PCR using the ABI PRISM 7700 sequencer detection system (Perkin-Elmer Applied Biosystems, Foster City, CA). RT-PCR mixtures were prepared according to the manufacturer's instructions for the TaqMan One-Step RT-PCR Master Mix Reagent kit (Perkin-Elmer Applied Biosystems). Briefly, 50 ng total RNA were added to each 50-µl reaction mixture containing Master Mix, MultiScribe, and RNase Inhibitor Mix, 200 nM of each primer, and 100 nM hybridization probe specific for the target cDNA. The probe was labeled with a reporter fluorescent dye [6-carboxyfluorescein (FAM)] at the 5′ end. For GAPDH detection, Pre-Developed TaqMan Assay Reagent (Perkin-Elmer Applied Biosystems) was added. The thermal conditions were 48°C for 30 min for reverse transcription and 95°C for 10 min, followed by 45 amplification cycles of 95°C for 15 s for denaturing and 55°C for 1.5 min for annealing and extension. The PCR products were sequenced to confirm proper amplification. To compare mRNA expression, the results were determined as relative values using GAPDH as an internal reference. There were n=3 samples in each group. The sense and antisense primers and probes for hBD1–3 and CAP18/LL37 were as follows: HBD1–67F: TCG CCA TGA GAA CTT CCT ACC T, HBD1–195R: CTC CAC TGC TGA CGC AAT TGT A, HBD1–136T: CCA CCT GAG GCC ATC TCA GAC AAA AGT AAG, HBD2–16F: TGA AGC TCC CAG CCA TCA G, HBD2–144R: GGC TCC ACT CTT CCA AAG GA, HBD2–107T: CAC CAA AAA CAC CTG GAA GAG GCA TCA, HBD3–346F: TCA GCT GCC TTC CAA AGG A, HBD3–414R: TTC TTC GGC AGC ATT TTC G, HBD3–367T: AAC AGA TCG GCA AGT GCT CGA CGC, CAP18–284F: CAC AGC AGT CAC CAG AGG ATT G, CAP18–366R: GGC CTG GTT GAG GGT CAC T, and CAP18–341T: ATA CAC CGC TTC ACC AGC CCG TCC. The primers and probe for TLR2 were purchased from Applied Biosystems (Assays-on-Demand).

Western blotting

The analysis of p38 was performed using a Vistra ECF kit (Amersham Life Science, Inc., Arlington Heights, IL) according to the manufacturer's instructions. Proteins (20 µg) were separated by 10% SDS-PAGE and transferred to a nitrocellulose membrane. The membrane was blocked with 5% non-fat dry milk in Tris-HCl pH 7.4, 0.15 M NaCl, and 0.05% Tween-20, and incubated with rabbit anti-p38 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) and anti-phospho-p38 (Cell Signaling Technology, Inc., Beverly, MA) antibodies. After washing, the membrane was incubated with fluorescein-labeled goat anti-mouse IgG (diluted 1:2,500) for 1 h. The signal was amplified using an anti-fluorescein antibody conjugated with alkaline phosphatase, followed by the fluorescent substrate AttoPhos (Amersham Life Science). Then, the membrane was scanned using a FluoroImager (Molecular Dynamics, Inc., Sunnyvale, CA).

For Western blotting of hBD1–3 and LL37, the synthetic peptides (100 ng) were applied to 15% SDS-PAGE with Tris/tricine buffer and transferred to PVDF membranes. After blocking with 5% non-fat dry milk, the membranes were reacted overnight at 4°C with anti-hBD or anti-LL37 antibodies diluted 1:1,000. After extensive washing, the signals were detected using an ECL Advance Western blotting detection kit (Amersham Biosciences, Piscataway, NJ).

Dot blot analysis

Antimicrobial peptides were prepared for dot blot analysis by partial purification from the culture supernatants. MacroPrep CM Support beads (Bio-Rad Laboratories, Richmond, CA) were washed with distilled water and then pretreated with 25 mM ammonium acetate buffer, pH 7.4. After transfection of the adenovirus vectors into normal human keratinocytes in 10-cm dishes at an MOI of 5, the medium was replaced with 5 ml medium without BHE, and the cells were cultured for an additional 2 days. The culture supernatants were centrifuged and filtered through 0.22-µm filters to remove the cellular debris. After adjusting the pH of the supernatants to 7.2, 1.0 ml of pretreated MacroPrep beads was added to 100 ml of the culture supernatants, gently shaken overnight at 4°C, washed with 25 mM ammonium acetate buffer at pH 7.4 and then eluted with 1.0 ml 5% acetic acid. The eluates were freeze-dried and redissolved in 1.0 ml distilled water.

Samples (1 µl) of serially diluted, partially purified supernatants or control solutions containing synthetic peptides (hBD1–3 and LL37) were applied to nitrocellulose membranes and air-dried. After blocking with 5% non-fat dry milk, the membranes were reacted overnight at 4°C with anti-hBD or anti-LL37 antibodies diluted 1:1,000. After extensive washing, the signals were detected using an ECL Advance Western blotting detection kit (Amersham Biosciences).

Induction of antimicrobial peptides by S. aureus

Induction of antimicrobial peptides by S. aureus was analyzed as described 21. We used FDA 209P (ATCC6538), a laboratory strain, for S. aureus. Overnight cultures of S. aureus (209P) grown in tryptic soy broth (TSB) at 37°C were harvested and washed twice with PBS. Then, the S. aureus cells were suspended in PBS and incubated at 68°C for 30 min to kill the bacteria. The S. aureus suspension was subjected to mild sonication, and the dissociation of clumps was confirmed by microscopic observation. Before bacterial contact (12 h), the confluent keratinocyte medium was replaced with MCDB153 medium without supplements, and then the heat-killed bacteria were added (final concentration, 108 cells/ml) to the medium. The culture was incubated for an appropriate time period, and mRNA expression and p38 phosphorylation were analyzed. This experiment was carried out three times independently.

Inhibitors

SB203580, PD098059, and SP600125 were purchased from Calbiochem-Novabiochem International Co. (San Diego, CA), dissolved in dimethyl sulfoxide (DMSO) at 2, 30, and 20 mM, respectively, as stock solutions, and used at final concentrations of 200 nM, 30 µM, and 20 µM, respectively.

Acknowledgements

This work was supported by grants from the Ministries of Health, Labor, and Welfare and Education, Culture, Sports, Science, and Technology of Japan. We gratefully acknowledge Ms. Teruko Tsuda and Ms. Eriko Tan for their technical assistance.

Footnotes

  1. 1

    WILEY-VCH

  2. 2

    WILEY-VCH

  3. 3

    WILEY-VCH

  4. 4

    WILEY-VCH

  5. 5

    WILEY-VCH

  6. 6

    WILEY-VCH

  7. 7

    WILEY-VCH

Ancillary