Identification of Lympho-Epithelial Kazal-Type Inhibitor 2 in Human Skin as a Kallikrein-Related Peptidase 5-Specific Protease Inhibitor

Kallikreins-related peptidases (KLKs) are serine proteases and have been implicated in the desquamation process of the skin. Their activity is tightly controlled by epidermal protease inhibitors like the lympho-epithelial Kazal-type inhibitor (LEKTI). Defects of the LEKTI-encoding gene serine protease inhibitor Kazal type (Spink)5 lead to the absence of LEKTI and result in the genodermatose Netherton syndrome, which mimics the common skin disease atopic dermatitis. Since many KLKs are expressed in human skin with KLK5 being considered as one of the most important KLKs in skin desquamation, we proposed that more inhibitors are present in human skin. Herein, we purified from human stratum corneum by HPLC techniques a new KLK5-inhibiting peptide encoded by a member of the Spink family, designated as Spink9 located on chromosome 5p33.1. This peptide is highly homologous to LEKTI and was termed LEKTI-2. Recombinant LEKTI-2 inhibited KLK5 but not KLK7, 14 or other serine proteases tested including trypsin, plasmin and thrombin. Spink9 mRNA expression was detected in human skin samples and in cultured keratinocytes. LEKTI-2 immune-expression was focally localized at the stratum granulosum and stratum corneum at palmar and plantar sites in close localization to KLK5. At sites of plantar hyperkeratosis, LEKTI-2 expression was increased. We suggest that LEKTI-2 contributes to the regulation of the desquamation process in human skin by specifically inhibiting KLK5.


Introduction
The skin protects us from water loss and mechanical damage. The surface-exposed epidermis, a self-renewing stratified squamous epithelium composed of several layers of keratinocytes, is most important for the barrier defense against these challenges. Keratinocytes in the outmost stratum corneum (SC) of the epidermis are shed off and replaced by newly differentiated cells originating from epidermal stem cells located in the basal layer. They undergo a specific differentiation process and form the cornified envelope, which is a rigid and insoluble protein and lipid structure with essential properties of the barrier function [1,2]. Recent discoveries have highlighted the importance of proteaseinhibitors and proteases as key players in the desquamation process and in epidermal barrier function.
Human tissue kallikreins, or kallikrein-related peptidases (KLK), are the largest family of trypsin or chymotrypsin-like secreted serine proteases encoded by 15 genes on chromosome region 19q13.4 [3]. At least eight KLKs are expressed in normal skin, among which KLK5, KLK7, KLK8 and KLK14 have been reported to be most important [4][5][6]. KLKs are capable of cleaving corneodesmosomes [7][8][9][10] and are thought to be key regulators of the desquamation process. Epidermal overexpression of KLK7 resulted in pathologic skin changes with increased epidermal thickness, hyperkeratosis, dermal inflammation, and severe pruritus [11]. The activity of the KLKs is regulated by the pH and specific protease inhibitors in human skin. The importance of epithelial protease inhibitors has been revealed impressively in Netherton Syndrome (NS; OMIM 256500), an autosomal recessive disorder caused by mutations in the serine protease inhibitor Kazal-type 5 (Spink5) gene [12]. NS presents as an ichthyosiform dermatosis with variable erythroderma, hair-shaft defects (bamboo hair), atopic features, and growth retardation [13]. Lymphoepithelial Kazal-type-related inhibitor (LEKTI) [14], the product of Spink5, includes in its primary structure 15 different serine protease inhibitory domains [14]. The inhibitory functions of LEKTI are highly diverse. Inhibitory activities are directed toward trypsin, plasmin, subtilisin A, cathepsin G, and human neutrophil elastase [15]. Though LEKTI is absent, NS patients can still develop hyperkeratosis -a clinical sign of inhibited desquamation.
Therefore, we speculated that more KLK inhibitors are present in human skin generating a complex network of KLKs and their inhibitors to control the desquamation process. Since KLK5 is thought to be one of the most important enzymes involved in this process, we started a preparative attempt to identify KLK5 inhibitors in human stratum corneum. Herein we report the identification of a new protease inhibitor LEKTI-2 and its gene Spink9, which specifically inhibits KLK5.

Identification of a new KLK5-inhibiting peptide in human Stratum corneum
To follow the hypothesis that specific inhibitors for KLK5 exist in human skin, extracts from healthy persons' SC were analyzed for KLK5-inhibiting activity. Preparative reverse-phase HPLC (RP-HPLC) was used to separate heparin-bound cationic peptides. Results from KLK5-inhibiting activity revealed a fraction (Fig. 1A), which was further purified by analytical RP-HPLC using a C2C18-column (data not shown). SDS-PAGE analysis (Fig. 1B) of these HPLC fractions, eluting at low acetonitrile concentration and containing KLK5-inhibitory activity showed the presence of a 7-kDa peptide. Electrospray-ionisation mass spectrometry (ESI-MS) analysis (Fig. 1C) resulted in a principal ion corresponding to a mass of 7058.19 Da. Da. N-terminal sequencing of the dominant fraction yielded a sequence of 25 residues (TKQMVDXSHYKKLPPGQQRFX-HHMY; Fig. 1D). A blast search using the 25-residue sequence retrieved no matches in any protein/gene/EST databases, suggesting a novel human gene may encode this sequence.

The KLK5-inhibiting peptide is encoded by Spink9
To identify the gene corresponding to the amino acid sequence, a BLAT search with the N-terminal 25-residue sequence of the novel peptide (where6was replaced by the cysteine residue) against the April 2003 human genome assembly localized this sequence to a chromosome 5 clone RP11-373N22 on 5p33.1 ( Fig. 2A). Subsequent analysis of the retrieved RP11-373N22 DNA sequence identified two putative exons exactly encoding the isolated peptide (Fig. 1). Based on the generated theoretical partial DNA sequence, gene-specific primers were designed to perform 39-and 59-RACE. The full-length cDNA sequence (453 bp) was completed by combining the overlapping sequences from each PCR product and then confirmed by a long distance PCR ( Fig. 2C; GenBank accession No. AY396740). Alignment of the mRNA sequence against human genome sequences clearly indicated that each unique mRNA segment represents an individual exon and that all introns are flanked by the consensus donor and acceptor splice sites conforming to the GT/AG rule ( Fig. 2B; data not shown). This gene contains an open reading frame of 261 nucleotides encoding a protein of 86 amino acids; a polyadenylation signal (AAUAAA) is situated 13 nucleotides 59 of the polyadenine tail (Fig. 2C). By SMART analysis, the 16 residues from the first Met are a leader sequence containing a signal peptide while the last 55 residues correspond to a typical Kazal domain. A BLAST search revealed that this Kazal domain is about 33% identical (50% similar) and 32% identical (40% similar) to domains 2 and 15 of human Lekti, respectively, (Fig. 2D) that is encoded by Spink5, the defective gene in Netherton syndrome [12]. These three Kazal domains possess similar domain patterns, including a conserved tyrosine residue, disulfide bonds and the residue numbers spacing the cysteine residues. Only the P1 residue of the putative active site is different, suggesting they might have different substrate binding modes. Therefore, we designated this novel gene as serine protease inhibitor Kazal-type 9 with the gene symbol Spink9, which was approved later by the HUGO gene nomenclature committee, while its protein product was named lympho-epithelial Kazal-type inhibitor 2 (LEKTI-2).

Spink9 is expressed in human skin and in cultured keratinocytes
To investigate the cellular source of LEKTI-2, both RT-PCR and real-time RT-PCR were used to determine its mRNA expression. Expression of Spink9 mRNA was detected in skin samples from foreskin and cultured primary keratinocytes (Fig. 3A).
In addition, its expression was also detected in thymus, tonsils, testis, placenta and brain but not in other tissue samples tested (Fig. 3A). In cultured primary keratinocytes, the expression level of Spink9 mRNA was increased up to 10-fold over the time course during calcium-induced differentiation, suggesting that Spink9 is produced by epithelial terminally differentiating keratinocytes.

LEKTI-2 is expressed at palmar and plantar sites
To analyze LEKTI-2 protein expression, we generated affinitypurified polyclonal LEKTI-2 antibodies. Westernblot analyses performed with rLEKTI-2, purified natural LEKTI-2 and stratum corneum extracts revealed antigen specificity of the antibodies (Fig. 4), which was further confirmed by blocking experiments using recombinant LEKTI-2. Subsequently, LEKTI-2 immunohistochemistry was performed to localize LEKTI-2 expression in human skin samples. LEKTI-2 immunoreactivity was detected in the stratum granulosum and SC of human skin at the palms (inner sides) of hands and feet (n = 8) but no visible immunoreactivity was detected at other sites of healthy human skin (n = 16) (Fig. 5).

Recombinant LEKTI-2 is a specific KLK5-inhibitor
To verify the protease inhibition of the purified LEKTI-2, recombinant LEKTI-2 was tested for its protease inhibitory activity. KLK5 was inhibited in a dose dependent manner. Assuming full competitivity of binding (alpha = infinitive) Baici Model was used to calculate K i (approximately 250 nM, Fig. 6). Interestingly, no other tested serine proteases, which include KLK7, KLK14, trypsin and chymotrypsin were inhibited by LEKTI-2 (Table 1, Fig. 6).

LEKTI-2 and KLK5 are in close localization in vivo
To visualize whether LEKTI-2 and KLK5 localize to the same site in human skin, fluorescent microscopic analyses were performed. LEKTI-2 fluorescent staining revealed granular-like structures inside the keratinocytes at the stratum granulosum and an intercellular staining in this area and in the SC (Fig. 7). KLK5staining revealed intercellular expression pattern at the same area. However, a clear co-localization was not observed.

LEKTI-2 is highly expressed at sites of hyperkeratosis
Hyperkeratosis at the palmar sites can occur due to increased local mechanical pressure and lead to hyperkeratosis in the form of clavus. Interestingly, LEKTI-2 immunoreactivity was shown to be markedly induced at lesions of clavi (Fig. 8). Since KLK5 is one of the major proteases for desquamation, increased LEKTI-2 expression at the sites of clavi might contribute to the hyperkeratosis of these lesions.

Discussion
In this study we aimed to identify major substances that might contribute to the epithelial barrier shield by inhibiting the epidermal serine protease KLK5. We identified a new peptide termed LEKTI-2 as a specific inhibitor for KLK5, which is encoded by Spink9, a novel member of the Spink gene family. Our findings give evidence for the importance of LEKTI-2 in epidermal desquamation and provide new insight to the complex protease-protease inhibitor interaction in human skin.
LEKTI-2 expression shows some similarities to the expression of LEKTI, which was demonstrated to be expressed in lamellar bodies, likely the granular-like structures in our fluorescent staining, and secreted into the intercellular space, in the uppermost stratum granulosum [16][17][18]. Electron microscopy studies revealed that LEKTI and KLK7 are transported separately in the lamellar  granule system and are co-localized in the extracellular spaces [18]. Our findings of LEKTI-2 and KLK5 expression are accordable to those results but need further evaluation by electron microscopy. However, LEKTI-2 expression was only detected in our studies at palmar and plantar sites where a rigid SC is needed to protect the hands and feet from mechanical damage. The fact that we did not find LEKTI-2 immunoreactivity at other sites, though low mRNA expression was detectable in skin samples, points to a minor role of LEKTI-2 in non-plantar skin compared to LEKTI, which is expressed throughout the entire skin. The circumstance that we used plantar human callus as the natural source of KLK5 inhibitors was therefore beneficial for the identification LEKTI-2. The enhanced expression of LEKTI-2 in plantar clavus corroborates the hypothesis that LEKTI-2-mediated KLK5 inhibition results in suppressed desquamation. Clavi are often induced by abnormal local mechanical pressure due to malformation of feet bones or tight footgear. It will be interesting to study how LEKTI-2 expression is induced by these mechanical forces. Mechanical stress represents an important part of signaling in skin. Indeed, in vitro it induces phosphorylation of keratin K6 and EGFR [19] and clustering of beta1-integrins [20], and activates ERK1/2 [19] as well as Akt, one of the kinases known to suppress apoptosis [21].
Most notably, LEKTI-2 exhibited only inhibiting activity against tryptic KLK5 but not against the chymotryptic KLK7, tryptic KLK14 or all other serine proteases tested including trypsin and chymotrypsin. LEKTI-2 activity differs in this respect from that of LEKTI, which contains multiple Kazal domains exhibiting highly diverse inhibitory functions beyond others against trypsin, plasmin, subtilisin A, cathepsin G, and human neutrophil elastase [15]. Therefore, the functions of both Kazal-type inhibitors are suspected to be different. Overall trypsin-like and/or chymotrypsin-like activities resulting mainly from KLKs are considerably elevated in the skin of Spink5-deficient mice [22] and in NS patients [4,23]. The elevated activities cause increased degradation of corneodesmosomal cadherins in NS patients [24]. It was shown that KLKs are capable of cleaving corneodesmosomes, [7][8][9][10]. Furthermore, Kallikrein-mediated proteolysis regulates the antimicrobial effects of cathelicidins in skin [25] and contributes to the pathogenesis of rosacea [26]. Moreover, KLK5 and KLK14 haven been reported to activate the protease activated receptor (PAR)-2 [27], a signaling receptor in epidermal inflammation [28] and regulator of epidermal barrier function [29]. Altogether, these accumulating data strongly suggest that in skin, KLKs are desquamation-related serine proteases and that the balance between serine proteases and inhibitors may be essential, not only for steady desquamation but also for skin barrier function and inflammation. Regulation of KLKs by endogenous proteinase inhibitors like LEKTI and LEKTI-2 might therefore have therapeutic potential in inflammatory skin diseases.
The K i of LEKTI-2 against KLK5 observed herein can be considered as preliminary since more in depth analysis of inhibition was not possible in this study due to limited access to KLK5. Several LEKTI domains have been reported to inhibit KLK5 in various reports. In comparison, the determined K i of LEKTI domains to inhibit KLK5 was in the range of 3 nM (domain 8-11) to 120 nM (domain 9-15) [9,30,31]. It will be very interesting to study the inhibition of other KLKs by LEKTI-2 in future investigations. KLKs are tryptic and chymotryptic enzymes though a detailed comparison of their activities, substrates and specific inhibitors has not been done systematically. As the most intensively studied KLK, KLK3 (also known as prostate-specific antigen), is a chymotryptic enzyme, similar in this respect to KLK7, which was not inhibited by LEKTI-2. Since Spink9 mRNA was detected in other organs like tonsils, testis, placenta and brain, but not in the prostate, it remains speculative what function LEKTI-2 might have in other organs. The only LEKTI-2 target enzyme identified so far is KLK5, which was reported to be expressed at high levels (100-1000 ng/g tissue) in breast, testis, salivary glands and thyroids but in the skin at highest reported levels (above 10,000 ng/g tissue) [32]. Detailed investigations of LEKTI-2 function in other organs are required to clarify its role outside the skin.
In summary, we identified a new Kazal-type inhibitor LEKTI-2 and its gene Spink9 and showed its expression in human skin. LEKTI-2 exhibited specific inhibition against KLK5, which might be important in desquamation especially at palmar and plantar sites.

Material
All experiments were performed according to the Declaration of Helsinki protocols and under protocols approved by the ethics committees of the Medical Faculty of the Christians Albrechts University Kiel, (Schwanenweg 20, D-24105 Kiel). Normal skin specimens were taken from routine clinical work at the Department of Dermatology, UKSH Kiel, and represent tumorfree margins of benign melanocytic tumors surgically removed from patients. Specimens of clavi were removed therapeutically before written informed consents were received from the patients. Restriction endonucleases were from New England Biolabs (Frankfurt, Germany). KLK5, 7, and 14 were purchased from R&D Systems, Minneapolis, MN. All other proteases, primers, substrates and chemicals were purchased from Sigma-Aldrich (Taufkirchen, Germany), if not otherwise indicated.

Protease Inhibition Assays
All protease assays were performed measuring chromogenic substrate release by proteases. Following buffers were used: KLK5:   (Table 1) were measured in the buffer recommended by the manufacturer.
Specific concentrations of protease, substrate and inhibitor are indicated in Table 1. The changes of absorbance at 405 nm were followed up to 16 h in comparison with enzyme-free controls. Inhibition of proteases was measured after preincubating the enzyme with inhibitor or HPLC fraction for 15 min at 21uC. K m was determined for KLK5 using various concentrations of substrate. K i was calculated using Baici Model [33] assuming full competitivity of binding.

Isolation of KLK5 inhibitor from human stratum corneum
Total proteins were extracted from heel stratum corneum, a clinical circumstance when desquamation might be inhibited by KLK-inhibitors, with an acidic buffer. Briefly, pooled heel stratum corneum (80-120 grams) was extracted with acidic ethanolic citrate buffer as described [34]. After diafiltration (Amicon filters, cut off: 3 kDa) against 10 mM Tris/citrate buffer, pH 8.0, extracts were applied to a heparin-sepharose cartridge (1065 mm, Pharmacia, Freiburg, FRG), previously equilibrated with the diafiltration buffer. After washing, bound proteins were eluted with 2 ml 2 M NaCl in 0.1 M Tris/citrate buffer and the heparinbound material was further diafiltrated against 0.1% (v/v) trifluoroacetic acid (TFA) in HPLC grade water. Heparin-bound material was purified by preparative wide-pore reversed phase high-performance liquid chromatography (RP-HPLC) using a column (30067 mm, C8 Nucleosil, 250612.6 mm, Macherey and Nagel, Düren, Germany) that was previously equilibrated with 0.1% (v/v) TFA in HPLC grade water containing 20% (v/v) acetonitrile. Proteins were eluted with a gradient of increasing concentrations of acetonitrile containing 0.1% (v/v) TFA (flow rate: 2 ml/min). Aliquots (10-30 ml) of each fraction were lyophilized, dissolved in 5 ml 0.1% (v/v) aqueous acetic acid and tested for protease-inhibiting activity.
Fractions containing KLK5-inhibiting activity, eluting at low (25%) acetonitrile were further purified by micro-C2/C18-RP-HPLC and tested for KLK5-inhibiting activity. Protease-inhibiting activity-containing HPLC fractions were further analyzed by Electrospray-ionization mass spectrometry (ESI-MS) in the positive ionization mode with a quadrupole orthogonal accelerating time-of-flight mass spectrometer (QTOF-II hybrid mass spectrometer; Micromass, Manchester, United Kingdom). Concentrations of proteins present in HPLC fractions were estimated via UV-absorbance integration at 215 nm using ubiquitine for calibration. The N-terminal amino acid sequence of the principal protein was determined using a pulsed-liquid-phase 776 automated protein sequencer (Perkin Elmer Applied Biosystems, Massachusetts, USA).

Rapid amplification of cDNA ends (RACE)
Total RNA was obtained from cultured human foreskin-derived keratinocytes using TRIzol reagent (Invitrogen, Hamburg, Germany). After treatment with RNase-free DNase I (Roche Diagnostics, Mannheim, Germany) to exclude contamination with genomic DNA, 3 mg of DNA-free total RNA was used for the first-strand cDNA synthesis for RACE using SMART RACE cDNA Amplification Kit (BD Bioscience Clontech, Heidelberg, Germany) according to the manufacturer's protocol. To obtain the 59-end of Spink9 cDNA, a 59-RACE was performed with a gene-specific antisense primer (59-TGC CAT CAG ATC CAC AAA TTG GAT CAT AC-39) and a universal primer mix (106 UPM) essentially according to the manufacturer's protocol. 59-RACE PCR reaction cycles were performed with an annealing temperature of 68uC and 35 cycles. To obtain the 39-end of Spink9 cDNA, the first round PCR was performed with a gene-specific sense primer (59-GCC AAA CAG ACG AAA CAG ATG GTT GAC T-39) and 106 UPM. Subsequently, 0.5 ml of PCR products was used as a template for a nested PCR with a nested gene-specific sense primer (59-ACC ACC AGG ACA ACA GAG ATT TTG TCA TC-39) and a nested universal primer (NUP) under the following conditions: 1 min at 95uC, 30 cycles of 20 s at 95uC and 3 min at 70uC, and a final extension of 10 min at 70uC. The amplified fragment was gel purified and subcloned into the pGEM-T vector (Promega, Mannhein, Germany) followed by fully sequencing in both directions.

mRNA expression analyses
A total of 2 mg of total RNA from human skin samples or cultured foreskin-derived keratinocytes was reverse transcribed  As an internal control of cDNA templates, the housekeeping gene GAPDH (glyceraldehyde phosphodehydrogenase) was assessed with each cDNA in a separate PCR reaction. For quantitative real-time RT-PCR, assay was carried out with the first primer pair as above and the SYBRH Premix Ex Taq TM Kit (Takara Bio, Heidelberg, Germany) in a fluorescence thermocycler following the instructions of the manufacturer (LightCycler, Roche Molecular Biochemicals, Hamburg, Germany). During the evaluation phase of the assay, amplicons were analyzed by 2.0% agarose gel electrophoresis and, where necessary purified and sequenced to confirm their identity. For calculation of the relative transcripts amplification, the housekeeping gene GAPDH was performed with each cDNA in a separate PCR reaction. The data from triplicate samples were analyzed with software (GraphPad Prism 4) and expressed as mean6SD of mRNA in question relative to that of GAPDH. The statistical analyses were performed with One-way ANOVA method and p,0.05 was considered significant while p,0.01 was considered very significant.

Recombinant protein production
The recombinant expression of Spink9 cDNA in E. coli was performed by molecular subcloning of Spink9 cDNA into the prokaryotic expression vectors pET-32a (Novagen, North Ryde, Australia). The purified form of LEKTI2 (amino acid 26 to 86) was generated as PCR fragments (Primer sequences are available upon request) by using Pfu DNA polymerase (Promega, Mannhein, Germany). PCR products were double-digested with BglII and NotI prior to be cloned into the similarly double-digested pET-32a vectors. Clones were sequenced to check for any mutation that might have been misincorporated during the amplification. The expression construct was transformed into Escherichia coli (E. coli) BL21trxB(DE3)pLysS cells (Novagen) and selected on Luria-Bertani (LB) agar plates containing carbenicillin (100 mg/ml), chloramphenicol (34 mg/ml) and kanamycin (15 mg/ml). Protein expression was induced with 1 mM IPTG (isopropyl thio-b-D-galactoside) for a 3 h. After incubation, cells were harvested by centrifugation and resuspended in 16LEW buffer (50 mM NaH 2 PO 4 , 300 mM NaCl, pH 8.0). Resuspended cells were subjected to one cycle of freeze-thawing and sonicated on ice until complete lyses. After centrifugation at 15,5006g for 30 min, the clarified supernatant was applied to ProtinoHNi prepared columns (Macherey-Nagel, Dueren, Germany) and the polyhistidine-tagged protein was eluted with 16elution buffer (50 mM NaH 2 PO 4 , 300 mM NaCl, 250 mM imidazole, pH 8.0). The further  . LETKI-2 and KLK5 are expressed at the stratum granulosum of palmar skin. Immunofluorescence localization of LEKTI-2 (red) and KLK5 (green) in human skin. Nuclei staining was done using Hoechst 33258 reagent. (B-E) shows the magnification of the white-square area of (A). LEKTI-2 staining showed granular structures inside the keratinocytes at the stratum granulosum (C) with a faint intercellular staining pattern and remaining immunoreactivity inside the stratum corneum. KLK5 staining (D) exhibited only intercellular staining at the stratum granulosum. Comparative localization of LEKTI-2 and KLK5 is shown in the merged image (D). (E) shows the control omitting the first antibody. doi:10.1371/journal.pone.0004372.g007 purification of the fusion protein was achieved by reversed phase high-performance liquid chromatography (RP-HPLC) using preparative wide-pore C8 RP-HPLC with a column (SP250/10 Nucleosil 300-7 C8; Macherey-Nagel) that was previously equilibrated with 0.1% (v/v) TFA in HPLC-grade water containing 10% acetonitrile. Proteins were eluted with a gradient of increasing concentrations of acetonitrile containing 0.1% (v/v) TFA (flow rate, 3 ml/min). Fractions containing UV (215 nm)absorbing material were collected, lyophilized and analyzed by ESI-QTOF-mass spectrometry (Micromass, Manchester, U.K.). The His-tagged fusion protein purified from HPLC-RP8 was cleaved from its His-tag with SUMO protease according to the manufacturer's suggestion (Lifesensors Inc., Pennsylvania, USA). The digestion mixture contained 1 unit of SUMO protease per 100 mg of the fusion protein in a volume of 500 ml of 16PBS buffer and was incubated for 2 h at 30uC. The dialyzed sample was adjusted to a pH value of 3.0 to 4.0 and then purified by centrifugation. The supernatant was collected and injected onto a Jupiter-5m-C4-300A HPLC column (Phenomenex, Aschaffenburg, Germany) equilibrated with 0.1% TFA in water. Peptides were eluted with a gradient of increasing concentrations of acetonitrile containing 0.1% (v/v) TFA (flow rate, 0.5 ml/min). Fractions of each peak were collected, lyophilized and analyzed by ESI-QTOF-mass spectrometry.

Production of antibodies
Polyclonal antiserum was generated in a goat against a full length peptide with the amino-acid sequence of human LEKTI-2 (amino acid 26 to 86). A total of 1.0 mg of protein mixture including 500 mg of fusion protein (pET-32a-LEKTI-2-3) and 500 mg of HPLC-purified recombinant peptide was conjugated by the glutaraldehyde method to maleimide-activated keyhole limpet hemocyanin (KLH) (protein-KLH 1 : 1, w/w) and subsequently mixed with 500 mg of pET-32a-lekri-2-3 for use as immunogens. Immunization of a goat was carried out four times on days 0, 14, 28 and 35. Goats were bled 2 weeks after the last booster. The serum was separated from the clot and stored at 270uC until required. Antisera were affinity-purified by absorption against rLEKTI-2-3 that was covalently bound to HiTrap NHS-activated HP 1 ml columns (Amersham Biosciences, Freiburg, Germany). Specificity was tested by immuno-dot analyses and Western blot analyses using purified rLEKTI-2, purified natural LEKTI-2 and stratum corneum extracts.

Cell culture
Foreskin-derived primary keratinocytes were prepared from neonatal foreskin after surgery following established methods [40] and were cultured in Epilife medium in 75-cm 2 flasks (BD Biosciences, Heidelberg, Germany) in a humidified atmosphere with 5% CO 2 . For stimulation and RNA isolation, cells were grown in 12-well tissue culture plates (BD Biosciences) and were used after the second passage at a confluence of 70-80%. Stimulation was performed for the indicated time with 1.0 mM of freshly prepared CaCl 2 for the indicated time.

Immunohistochemistry
Fixation of the tissue samples was performed in 4% paraformaldehyde. Paraffin sections (5 mm) of the tissue samples were deparaffinised and rehydrated before heat-induced antigen retrieval was performed in 0.01 M citrate buffer (pH 6.0). The slides were blocked with normal rabbit serum (1:75, Dako Cytomation, Glostrup, Denmark) before staining. Immunohistochemical staining was performed at room temperature for one hour using with affinity-purified polyclonal goat anti-lekti-2 antibody (1:200 dilutions. A biotinylated secondary rabbit antigoat IgG (1:100, Dako Cytomation) antibody was used, followed by incubation with Vector Universal ABC Alkaline Phophatase Substrate Kit (Vector, Burlingame, CA, USA) developed with Vector NovaRED Substrate (Vector) and counterstained with hematoxylin. Specificity test of the anti-lekti-2 antibody was performed by using recombinant lekti-2 peptides to block the primary antibody. Negative controls were established by using preimmune goat sera to stain sections.

Immunofluorescence
To show the potential spatial relationships between LEKTI-2 and KLK5, the paraffin-embedded skin sections (5 mm) were blocked with 10% normal goat and rabbit sera (Vector) in TBS containing 0.1% BSA and 0.2% glycine after standard rehydration. Sections were incubated with a mixture of anti-LEKTI2 antibody (1:100 dilution) and a rabbit anti-KLK5 antibody (1:200 dilution; Jackson R&D systems, Minneapolis, MN). After washing, they were incubated with a mixture of secondary antibodies (Cy3coupled pig-antigoat IgG and Cy2-coupled chicken-antirabbit IgG, diluted 1:400 each; Dianova, Hamburg, Germany) for 1 h at room temperature. Sections were counterstained with the DNAselective bisbenzimide dye (blue; Hoechst 33258). To exclude artificial autofluorescence secondary to the preparation of the sections, control sections were stained without primary antibodies and no unspecific labeling was observed following incubation with secondary antibodies (data not shown). Slides were analyzed using a confocal laser scanning microscopy (Zeiss, LSM 510 UV, Jena, Germany).