Advertisement

A Proteolytic Cascade of Kallikreins in the Stratum Corneum

      Serine proteases belonging to the kallikrein group may play a central role in desquamation. We have identified human kallikreins 5, 7, and 14 (hK5, hK7, hK14) in catalytically active form in stratum corneum. All three enzymes are produced as inactive precursors. In this work, we prepared recombinant enzymes and enzyme precursors and characterized the catalytic properties of hK5 and hK14. With peptide substrates hK5 and hK14 both showed trypsin-like specificity and alkaline pH-optima. For the substrates tested, hK14 was superior to hK5 as regards maximum catalytic rate as well as catalytic efficiency. hK5, but not hK14, could activate pro-hK7 in a reaction which was optimal at pH 5–7. hK5 could activate its own precursor as well as pro-hK14. This was in contrast to hK14, which could activate pro-hK5 but not its own precursor. The activation of pro-hK5 either by auto-activation or by hK14 occurred at maximum rate at neutral or weakly alkaline pH, whereas activation of pro-hK14 by hK5 was optimal at pH 6–7. We conclude that the enzymes studied may be part of a protease cascade in the stratum corneum, and that the observed pH effects may have physiological relevance.

      Keywords

      Abbreviations

      hK
      human tissue kallikrein protein
      KLK
      tissue kallikrein gene
      PAGE
      polyacrylamide gel electrophoresis
      In simplistic terms, the stratum corneum can be described as a stockade made up by corneocytes held together by corneodesmosomes and surrounded by lipids. Its continuous production as a result of epidermal differentiation is balanced by desquamation. Regulated transition from a state of strong intercellular cohesion in deeper layers to a state where individual cells are being shed at the skin surface implies a well-tuned proteolytic degradation of adhesive structures. The responsible proteolytic system may be predicted to consist of pro-enzymes, enzymes, enzyme activators, and inhibitors. Since corneocytes are non-viable, production and deposition of these components must take place before or concomitant with the cornification process. This “proteolytic programming” of the stratum corneum, most likely of central importance for epidermal homeostasis, is still far from being fully understood (
      • Egelrud T.
      Desquamation in the stratum corneum.
      ).
      The stratum corneum has been shown to contain several proteases with suggested roles in desquamation. To these belongs a thiol protease (
      • Watkinson A.
      Stratum corneum thiol protease (SCTP): A novel cysteine protease of late epidermal differentiation.
      ) as well as the aspartic protease Cathepsin D (
      • Horikoshi T.
      • Arany I.
      • Rajaraman S.
      • et al.
      Isoforms of cathepsin D and human epidermal differentiation.
      ). In addition stratum corneum contains a number of serine proteases, recently shown to belong to the kallikrein family (
      • Komatsu N.
      • Takata M.
      • Otsuki N.
      • Toyama T.
      • Ohka R.
      • Takehara K.
      • Saijoh K.
      Expression and localization of tissue kallikrein mRNAs in human epidermis and appendages.
      ). We have presented evidence that stratum corneum chymotryptic enzyme (SCCE; human kallikrein 7; hK7) as well as stratum corneum tryptic enzyme (SCTE; hK5) may both be involved in the desquamation process (
      • Egelrud T.
      • Lundström A.
      A chymotrypsin-like proteinase that may be involved in desquamation in plantar stratum corneum.
      ;
      • Lundström A.
      • Egelrud T.
      Stratum corneum chymotryptic enzyme: A proteinase which may be generally present in the stratum corneum and with a possible involvement in desquamation.
      ;
      • Egelrud T.
      Purification and preliminary characterization of stratum corneum chymotryptic enzyme: A proteinase that may be involved in desquamation.
      ;
      • Ekholm E.
      • Brattsand M.
      • Egelrud T.
      Stratum corneum tryptic enzyme in normal epidermis: A missing link in the desquamation process?.
      ;
      • Caubet C.
      • Jonca N.
      • Brattsand M.
      • et al.
      Degradation of corneodesmosome proteins by two serine proteases of the kallikrein family, SCTE/KLK5/hK5 and SCCE/KLK7/hK7.
      ).
      Quantification of proteases based on enzyme activity yields different results depending on which assay system is used. When extracts of stratum corneum are analyzed with casein zymography hK5 and hK7 appear to be the two dominating enzymes, responsible for essentially all detectable activity in crude extracts (
      • Lundström A.
      • Egelrud T.
      Stratum corneum chymotryptic enzyme: A proteinase which may be generally present in the stratum corneum and with a possible involvement in desquamation.
      ). With chromogenic peptide substrates essentially all chymotrypsin-like activity can be ascribed to hK7 (
      • Egelrud T.
      Purification and preliminary characterization of stratum corneum chymotryptic enzyme: A proteinase that may be involved in desquamation.
      ). With chromogenic peptide substrates for trypsin-like enzymes, on the other hand, we recently found that at most 50% of the total activity in stratum corneum extracts was due to hK5, whereas a major part of the remaining activity was associated with an enzyme that, after purification and amino acid sequence analysis, could be identified as hK14 (manuscript in preparation).
      We have previously shown that hK5 and hK7 are both present in stratum corneum extracts in active as well as precursor forms (
      • Ekholm E.
      • Brattsand M.
      • Egelrud T.
      Stratum corneum tryptic enzyme in normal epidermis: A missing link in the desquamation process?.
      ). This raises the possibility that zymogen activation may be an important part of the regulation of mechanisms in the stratum corneum dependent on proteolysis, including desquamation. Enzymes present in the tissue may form a proteolytic cascade in which already activated enzymes serve as activators of precursors of other enzymes, resulting in amplification effects as well as a multitude of targets for regulatory influences. In this work we investigated hK5 and hK14 with regard to their catalytic properties and their ability to act as activators of their own and each other's precursors as well as of pro-hK7.

      Results

      Purified recombinant proteins from insect cells showed some heterogeneity, probably due to variations in degree of glycosylation (Figure 1). Production of catalytically active enzymes was verified by activity measurements with chromogenic peptide substrates and casein zymography, and by the appearance of new components with expected molecular masses (Figure 1). hK14 protein produced by insect cells was catalytically inactive if not treated with an activating enzyme whereas hK14 produced in the yeast system showed full catalytic activity after purification. This was most likely due to post-translational removal of the propeptide, as suggested by amino acid sequence analyses (not shown). The yield of the two variants of pro-hK5, and of pro-hK14 from transfected insect cells was in the range of 1–5 mg per liter of conditioned growth medium. Around 15 mg of hK14 protein per liter of medium was obtained from yeast cells. No difference in activity between hK14 protein with or without tag, produced in insect or yeast cells, could be detected (data not shown).
      Figure thumbnail gr1
      Figure 1Preparations of recombinant human tissue kallikrein protein (hK)5 and hK14; precursors and active enzymes. SDS-PAGE, coomassie staining. From left to right: Mutated pro-hK5 (proEK-hK5) and hK5EK activated by enterokinase treatment; wild-type pro-hK5 (pro-hK5) and hK5 activated by auto-activation; pro-hK14 and hK14 produced in insect cells and activated by enterokinase treatment, without and with tag (pro-hK14V5His, hK14V5His); hK14 produced in Pichia pastoris (hK14P).

      Activity toward chromogenic peptide substrates

      The catalytic properties of hK5 and hK14 toward a set of chromogenic peptide substrates are shown in Table II. Trypsin was included for comparison. Although S-2586 is a substrate for chymotrypsin-like proteases, the other substrates tested (see Table I for structures) are all substrates for enzymes with trypsin-like primary substrate specificity. As predicted by their amino acid sequence, hK5 and hK14 both preferentially cleave trypsin-like substrates. In contrast to hK5, hK14 also had a low but significant chymotryptic activity. Comparison of hK5 and hK14 showed that hK14 was superior for all substrates tested. The difference in maximum catalytic rate (kcat) was smallest, around 10-fold, with S-2302, and largest, around 80-fold, with S-2288. hK14 was even more superior to hK5 when catalytic efficiency (kcat/Km) were calculated. When we compared hK14 and trypsin we found that kcat for the two enzymes were of the same order of magnitude with substrates S-2288 and S-2302, but two orders of magnitude higher for trypsin with S-2222. When Km values were taken into account in addition to kcat, trypsin was much more efficient than hK14 toward all three substrates (Table II).
      Table IIKinetic parameters of recombinant human tissue kallikrein protein (hK)5, hK14 and trypsin
      Enzyme
      hK5 and tagged hK14 were produced in insect cells and activated by recombinant enterokinase.
      Substrate
      See Table I for structures.
      Enzyme concentration
      Enzyme concentration was calculated from three or more active site titration for each preparation.
      (nM)
      Km (mM)Vmax (nM per s)kcat (per s)kcat/Km (per s per M)
      rhK5S-22881500.76.04.0 × 10−25.7 × 101
      S-22221500.51.61.1 × 10−22.2 × 101
      S-23021502.4281.9 × 10−17.9 × 101
      S-2586150
      rhK14S-228810.30.3323.11.0 × 104
      S-222210.30.23.23.1 × 10−11.5 × 103
      S-230210.30.2242.31.2 × 104
      S-25866500.7375.7 × 10−28.1 × 101
      TrypsinS-22880.420.0085.7141.8 × 106
      S-22220.420.0411266.5 × 105
      S-230210.043.53.58.8 × 104
      S-2586ndndndndnd
      –, no enzyme activity could be detected.
      nd, not determined.
      Values for Km and Vmax represent means of results from at least three separate experiments.
      a hK5 and tagged hK14 were produced in insect cells and activated by recombinant enterokinase.
      b See Table I for structures.
      c Enzyme concentration was calculated from three or more active site titration for each preparation.
      Table IStructures of chromogenic substrates
      SubstrateAmino acid sequence
      S-2288H-D-Ile-Pro-Arg-pNA·2HCl
      S-2222Bz-Ile-Glu(γ-OR)-Gly-Arg-pNA·HCl
      S-2302H-D-Pro-Phe-Arg-pNA·2HCl
      S-2586MeO-Suc-Arg-Pro-Tyr-pNA·HCl
      With chromogenic peptide substrates hK5 and hK14 both had maximum activity at pH 8–9 (Figure 2). The pH profile of hK14 differed slightly from that of hK5; hK14 having an activity versus pH curve that was skewed toward weakly acidic pH.
      Figure thumbnail gr2
      Figure 2Effect of pH on the activity of human tissue kallikrein protein (hK)5 and hK14 towards S-2302. In a total volume of 82.5 μL, 350 ng of hK5 (auto-activated wild-type; A) or 6.25 ng hK14 (produced in Pichia pastoris; B) was incubated at 37°C in universal buffer (
      • Britton H.T.S.
      • Welford G.J.
      ) with pH as indicated. The initial concentration of S-2302 was 1 mM. 100% activity corresponds to maximal activity obtained in each experiment.
      The effects of various inhibitors on hK5 and hK14 are shown in Table III. Aprotinin, zinc sulfate, soybean trypsin inhibitor, and leupeptin all inhibited both enzymes with high and approximately similar efficiencies. Also chymostatin could inhibit hK14, although with an IC50 that was around one order of magnitude higher than for leupeptin. Chymostatin had no effect on the activity of hK5 at the concentrations tested.
      Table IIIEffect of inhibitors on recombinant human tissue kallikrein protein (hK)5 and hK14
      InhibitorhK5 IC50 (μM)hK14 IC50 (μM)
      Aprotinin33
      ZnSO442
      Soybean trypsin inhibitor0.10.1
      Leupeptin103
      Chymostatin>10030
      Inactivation experiments were performed with enzyme concentration 0.3 μM for hK5 and 1.4 nM for hK14 as described (
      • Egelrud T.
      Purification and preliminary characterization of stratum corneum chymotryptic enzyme: A proteinase that may be involved in desquamation.
      ) but with no KCl in the reaction mixture. The substrate S-2288 (initial concentration 1.0 mM) was used to measure residual hK5 and hK14 activity. The same enzyme preparations as in Table II were used.

      Activation of pro-hK7

      Since the propeptide of pro-hK7 ends with a tryptic cleavage site (Table IV), and activation can be obtained by treatment of the pro-enzyme with pancreatic trypsin (
      • Hansson L.
      • Strömqvist M.
      • Bäckman A.
      • Wallbrandt P.
      • Carlstein A.
      • Egelrud T.
      Cloning, expression, and characterization of stratum corneum chymotryptic enzyme. A skin-specific human serine proteinase.
      ), we wanted to elucidate whether hK5 and hK14 could act as pro-hK7 activators. Activation was followed by measuring activity toward the substrate S-2586, by immunoblotting and by casein gel zymography. hK14 could not activate pro-hK7 at any condition tested. This was in contrast to hK5, which could convert inactive precursor to catalytically active hK7 at a slow but significant rate (Figure 3a). The expected shift in molecular mass, suggesting removal of the pro-peptide from pro-hK7 by incubation with hK5, could be confirmed by immunoblotting (Figure 3b). Activation was confirmed also by zymography (data not shown). At molar ratio 1:1 30%–40% of pro-hK7 was converted to active enzyme after 24 h incubation (based on densitometric scanning of immunoblots, cf lane 3, Figure 3). The activation reaction had an acidic pH optimum (pH 5–7; Figure 4a). Control experiments showed that this could not be explained by decreased stability of hK5 at higher pH values (data not shown).
      Table IVStructures of pro-peptides and cleavage sites of pro-human tissue kallikrein protein (hK)5, pro-hK7 and pro-hK14
      EnzymeAmino acid sequence
      pro-hK5
      Only the seven amino acids closest to the activation site are shown for pro-hK5 and the enterokinase mutated proEK-hK5.
      SerAspAspSerSerSerArg-IleIleAsnGly
      proEK-hK5aSerAspAspAspAspAspLys-IleIleAsnGly
      pro-hK7GluGluAlaGlnGlyAspLys-IleIleAspGlyAla
      pro-hK14GlnGluAspGluAsnLys-IleIleGlyGly
      The N-terminal amino acid sequences of active enzymes are shown by italics.
      a Only the seven amino acids closest to the activation site are shown for pro-hK5 and the enterokinase mutated proEK-hK5.
      Figure thumbnail gr3
      Figure 3Activation of pro-human tissue kallikrein protein (hK)7 by hK5. pro-hK7, 5 μg, was incubated at room temperature with different amounts of auto-activated wild-type hK5 in 0.15 M NaCl, 0.1 M NaAc pH 5.6 in a total volume of 20 μL. Samples, 2.5 μL, were taken for analyses at time points as indicated. (A) Activity toward S-2586 at 25°C (1.2 mM in 0.075 mM Tris-HCl pH 8.0; incubation volume 77.5 μL). The ratio pro-hK7:hK5 was 1:0 (filled squares), 10:1 (open triangles), 3:1 (filled triangles), 1:1 (open circles), and 1:2 (filled circles). (B) Immunoblot of reduced samples stained with anti-hK7 antibody diluted 1:3000. Samples were taken after 24 h incubation with ratios pro-hK7:hK5 1:0 (lane 1), 3:1 (lane 2), 1:1 (lane 3), and 1:2 (lane 4). The bands in lanes 2–4 from top to bottom corresponds to (1) glycosylated pro-form, (2) non-glycosylated pro-form as well as glycosylated active form, and (3) non-glycosylated active form (
      • Hansson L.
      • Strömqvist M.
      • Bäckman A.
      • Wallbrandt P.
      • Carlstein A.
      • Egelrud T.
      Cloning, expression, and characterization of stratum corneum chymotryptic enzyme. A skin-specific human serine proteinase.
      ;
      • Ekholm E.
      • Brattsand M.
      • Egelrud T.
      Stratum corneum tryptic enzyme in normal epidermis: A missing link in the desquamation process?.
      ).
      Figure thumbnail gr4
      Figure 4The effect of pH on the activation rate of pro-kallikreins by kallikreins. In a total volume of 12 μL of universal buffer pH 4.5–10 auto-activated wild-type human tissue kallikrein protein (hK)5 or hK14 produced in Pichia pastoris was incubated at 37°C with pro-enzymes as follows: (A) hK5, 1 μg and pro-hK7, 3 μg. (B) hK5, 0.5 μg and wild-type pro-hK5, 5 μg. (C) hK14, 25 ng and wild-type pro-hK5, 4.2 μg. (D) hK5, 1 μg and pro-hK14 (without tag, produced in insect cells), 2.25 μg. After 3 h of incubation, protease activity was measured by mixing 2.5 μL (A, B, D) or 5 μL (C) of the respective incubation mixture with 75 μL of chromogenic substrate at 37°C (1.2 mM initial concentration S-2586 (A) or 1 mM S-2302 (B–D) in buffer 0.075 mM Tris-HCl pH 8.0). 100% activity corresponds to maximal activity obtained in each experiment.

      Activation of pro-hK5

      When purified native recombinant pro-hK5 was incubated without any additions at 37°C, pH 8.0, and its activity toward the substrate S-2302 measured at different time points, there was a time-dependent increase in catalytic activity. When activity was plotted against time a sigmoid curve was obtained, suggesting auto-activation. This could be confirmed by incubating pro-hK5 with small amounts of active hK5 (Figure 5a). The rate of auto-activation of hK5 was maximal at pH values around 8 (Figure 4b). Activation of pro-hK5 was efficiently catalyzed by hK14 (Figure 5a) with maximum rate at pH 7–9 (Figure 4c). On a molar basis hK14 appeared to be approximately 50-fold more efficient than hK5 in activating pro-hK5 (Figure 5a).
      Figure thumbnail gr5
      Figure 5Auto-activation of pro-human tissue kallikrein protein (hK)5 and pro-hK14; activation of pro-hK5 by hK14. Incubations were carried out in a total volume of 30 μL at 37°C. At time points as indicated, samples, 2.5 μL, were analyzed for protease activity with 75 μL chromogenic substrate (initial concentration 1 mM S-2302, in buffer 0.075 mM Tris-HCl pH 8.0). (A) Wild-type pro-hK5, 15 μg, was incubated in 0.1 M Tris-HCl pH 8.0, 0.15 M NaCl without additions (circles) or with either 1.5 μg of auto-activated hK5 (diamonds) or 30 ng of Pichia pastoris-produced hK14 (squares). (B) Non-tagged insect cell produced pro-hK14, 10 μg was incubated in universal buffer pH 6.5, without additions (circles) or with either 1 μg (squares) or 2.5 μg (diamonds) of auto-activated hK5. (C) Non-tagged insect cell produced pro-hK14 was incubated in universal buffer pH 6.5 without additions (circles) or with 1 μg of P. pastoris-produced hK14 (diamonds). In A“100% activity” was calculated from the observed activity of hK5 for which auto-activation was judged to be complete, based on mobility shift on SDS-PAGE (see ). In B and C“100% activity” was calculated from the activity of corresponding amounts of yeast-produced hK14.

      Activation of pro-hK14

      Recombinant pro-hK14 could be readily activated by hK5 (Figure 5b). As for pro-hK7, activation of pro-hK14 by hK5 occurred at maximum rate at slightly acidic pH (Figure 4d). We found no evidence of efficient auto-activation of hK14, neither when pro-hK14 was incubated alone for prolonged periods of time, nor when the precursor was incubated with previously activated hK14 (Figure 5b and c).

      Discussion

      The number of human serine proteases for which a specific physiological function has not yet been elucidated is very large. This holds true also for the 15 kallikreins encoded by genes clustered in a small part of chromosome 19 (19q13.3–q13.4). With sensitive RNA-techniques expression of the majority of the kallikreins can be detected in most tissues, including skin (
      • Clements J.
      • Hooper J.
      • Dong Y.
      • Harvey T.
      The expanded human kallikrein (KLK) gene family: Genomic organisation, tissue-specific expression and potential functions.
      ;
      • Yousef G.M.
      • Diamandis E.P.
      The new human tissue kallikrein gene family: Structure, function, and association to disease.
      ;
      • Komatsu N.
      • Takata M.
      • Otsuki N.
      • Toyama T.
      • Ohka R.
      • Takehara K.
      • Saijoh K.
      Expression and localization of tissue kallikrein mRNAs in human epidermis and appendages.
      ). So far it has been assumed that the high expression of, e.g., hK5 and hK7 in keratinizing epithelia, as compared with other tissues, is reflecting a specific function of these enzymes in processes involved in the formation and turnover of cornified keratinocytes (
      • Sondell B.
      • Dyberg P.
      • Anneroth G.K.
      • Ostman P.O.
      • Egelrud T.
      Association between expression of stratum corneum chymotryptic enzyme and pathological keratinization in human oral mucosa.
      ;
      • Ekholm E.
      • Brattsand M.
      • Egelrud T.
      Stratum corneum tryptic enzyme in normal epidermis: A missing link in the desquamation process?.
      ;
      • Hansson L.
      • Bäckman A.
      • Ny A.
      • et al.
      Epidermal overexpression of stratum corneum chymotryptic enzyme in mice: A model for chronic itchy dermatitis.
      ). This does not exclude, however, the possibility that hK5 and/or hK7 may have specific functions in other tissues (
      • Tanimoto H.
      • Underwood L.J.
      • Shigemasa K.
      • Yan Y.
      • Clarke J.
      • Parmley T.H.
      • O'Brien T.J.
      The stratum corneum chymotryptic enzyme that mediates shedding and desquamation of skin cells is highly overexpressed in ovarian tumor cells.
      ;
      • Dong Y.
      • Kaushal A.
      • Brattsand M.
      • Nicklin J.
      • Clements J.
      Differential splicing of KLK5 and KLK7 in epithelial ovarian cancer produces novel variants with potential as cancer biomarkers.
      ). Similarly, other serine proteases, including kallikreins, which are expressed at low levels in the epidermis, may exert important functions in the skin. In this work we have concentrated on the three human kallikreins that have so far been identified in catalytically active form in the stratum corneum.
      Our aim, besides elucidating basic catalytic properties of hK5 and hK14, was to investigate possible functions of hK5 and hK14 as activators of each other and of pro-hK7. The precursors of hK5 and hK7 can both be converted to active enzymes by treatment with trypsin (
      • Hansson L.
      • Strömqvist M.
      • Bäckman A.
      • Wallbrandt P.
      • Carlstein A.
      • Egelrud T.
      Cloning, expression, and characterization of stratum corneum chymotryptic enzyme. A skin-specific human serine proteinase.
      ;
      • Brattsand M.
      • Egelrud T.
      Purification, molecular cloning, and expression of a human stratum corneum trypsin-like serine protease with possible function in desquamation.
      ). Since trypsin is a very efficient enzyme and cleaves a wide variety of substrates, and since it is difficult to efficiently remove all trypsin once it has been added to a preparation of protease precursor, we explored alternative means to obtain catalytically active recombinant hK5 and hK14. For hK5 we produced a mutated precursor that contained an activation site specifically recognized by enterokinase, an enzyme with very high substrate specificity (
      • Huang C.
      • Wong G.W.
      • Ghildyal N.
      • et al.
      The tryptase, mouse mast cell protease 7, exhibits anticoagulant activity in vivo and in vitro due to its ability to degrade fibrinogen in the presence of the diverse array of protease inhibitors in plasma.
      ). In addition to the mutated pro-hK5 we found that also native pro-hK14 could be efficiently activated by enterokinase, most likely due to the high functional amino acid sequence similarity of its activation site to the natural enterokinase site (see Table IV). Active hK14 was also obtained by expressing the cDNA for pro-hK14 in yeast cells, most likely due to the presence of yeast enzymes in conditioned media that could cleave off the propeptide.
      The comparison of the catalytic properties of hK5 and hK14 toward a set of chromogenic peptide substrates showed that hK14 was superior to hK5, as regards maximum catalytic efficiency as well as substrate affinity. When substrate affinity (Km values) was taken into account, both skin enzymes were inferior to pancreatic trypsin. From the figures shown in Table II it can be calculated that, using, e.g., S-2288 at “optimal concentration” as substrate, and assuming that hK5 and hK14 were the only enzymes present in an extract to be analyzed, the same contribution to the total activity measured would be given by hK5 and hK14 at a concentration of active hK14 two orders of magnitude lower than that of active hK5. This should be taken into account when analyzing skin preparations for “trypsin-like” and “chymotrypsin-like” activities (
      • Egelrud T.
      • Lundström A.
      A chymotrypsin-like proteinase that may be involved in desquamation in plantar stratum corneum.
      ;
      • Egelrud T.
      Purification and preliminary characterization of stratum corneum chymotryptic enzyme: A proteinase that may be involved in desquamation.
      ;
      • Suzuki Y.
      • Koyama J.
      • Moro O.
      • Horii I.
      • Kikuchi K.
      • Tanida M.
      • Tagami H.
      The role of two endogenous proteases of the stratum corneum in degradation of desmoglein-1 and their reduced activity in the skin of ichthyotic patients.
      ;
      • Hachem J.P.
      • Crumrine D.
      • Fluhr J.
      • Brown B.E.
      • Feingold K.R.
      • Elias P.M.
      pH directly regulates epidermal permeability barrier homeostasis, and stratum corneum integrity/cohesion.
      ;
      • Komatsu N.
      • Takata M.
      • Otsuki N.
      • Toyama T.
      • Ohka R.
      • Takehara K.
      • Saijoh K.
      Expression and localization of tissue kallikrein mRNAs in human epidermis and appendages.
      ). It should also be noted that if there is an enzyme in the epidermis with catalytic properties similar to pancreatic trypsin, which has been suggested,
      Nakanishi J, Sato J, Koyama J, Nakayama Y: Expression of enteropeptidase is restricted in uppermost granular cells in human epidermis. J Invest Dermatol 110:611, 1998 (abstr).
      this enzyme could be present in very low concentrations and still play a significant role due to its high catalytic efficiency.
      We found that hK5, but not hK14, could act as an activator of hK7. This was in spite of the fact that hK14 had similar primary substrate specificity as hK5, and was a much more efficient catalyst with other substrates tested. In addition, hK5 could activate pro-hK14. hK14, on the other hand, could activate pro-hK5 but not pro-hK14. These results show that there may be considerable differences among “trypsin-like” serine proteases produced by the epidermis, which are likely to have implications not only for activation of protease precursors, but also for the concerted actions of this group of enzymes on, e.g., structural proteins in the stratum corneum.
      In this respect, the observed effects of pH may also be relevant. We reported previously that hK5 and hK7 both have significant activity toward protein substrates at the low pH of the stratum corneum (
      • Ekholm E.
      • Brattsand M.
      • Egelrud T.
      Stratum corneum tryptic enzyme in normal epidermis: A missing link in the desquamation process?.
      ). In this work we found that activation of hK7 and hK14 by hK5 proceeded faster at acidic pH-values close to those reported for the stratum corneum (
      • Öhman H.
      • Vahlquist A.
      In vivo studies concerning a pH gradient in human stratum corneum and upper epidermis.
      ) than at the alkaline pH, which was optimal for cleavage of small peptide substrates by hK5. Auto-activation of hK5, on the other hand, as well as activation of pro-hK5 by hK14 was optimal at pH 7–9. These differences in effects of pH may eminate from the specific protein-protein interactions occurring between each pair of precursor (substrate) and activator (enzyme).
      hK7 is responsible for a major part of the total proteolytic activity in the stratum corneum. hK5 is the only enzyme found in the epidermis so far that can activate hK7. Since hK5 can activate also its own precursor as well as pro-hK14, it may play a key role in a postulated proteolytic cascade in the stratum corneum, including hK5, hK7, hK14, and possibly other proteases. Our results allow us to speculate further on how hK5 and hK7 may be involved in desquamation. Precursors of both enzymes are being secreted to the extracellular space at the transition between the granular and cornified layers. At the close-to-neutral pH of the deepest layers of the stratum corneum hK5 undergoes auto-activation (or activation by other proteases such as hK14). Active hK5 starts to activate pro-hK7 at a rate, which, due to the pH tolerance of this reaction, is maintained throughout the stratum corneum. Due to the slow rate of hK7 activation significant amounts of pro-hK7 will remain also in layers close to the skin surface (
      • Ekholm E.
      • Brattsand M.
      • Egelrud T.
      Stratum corneum tryptic enzyme in normal epidermis: A missing link in the desquamation process?.
      ). As a consequence there will be a concentration gradient of active hK7 between deep and superficial layers of the stratum corneum, and hence also an increase in the rate of degradation of intercellular cohesive structures as the corneocytes move toward the skin surface.
      For a further understanding of the role played by proteases in various aspects of epidermal biology and pathophysiology, expression studies will be insufficient. Individual enzymes should be studied with regard to their catalytic properties, and we should learn more about precursor–activator relationships among the many proteases now known to be expressed by epidermal cells.

      Materials and Methods

      All studies were performed according to Declaration of Helsinki Principles and approved by the medical ethical committee of the University of Umeå.

      Production of recombinant hK5 and hK14

      Recombinant pro-hK5 was produced in native form as well as a mutated form with the activation cleavage site replaced by an enterokinase cleavage site (
      • Huang C.
      • Wong G.W.
      • Ghildyal N.
      • et al.
      The tryptase, mouse mast cell protease 7, exhibits anticoagulant activity in vivo and in vitro due to its ability to degrade fibrinogen in the presence of the diverse array of protease inhibitors in plasma.
      ). For the native form, a 954 bp KLK5 cDNA fragment (GenBank AF168768) spanning the entire open reading frame, including the stop codon and 72 bp upstream of the putative translation start codon (
      • Brattsand M.
      • Egelrud T.
      Purification, molecular cloning, and expression of a human stratum corneum trypsin-like serine protease with possible function in desquamation.
      ), was amplified using the primer pair P12 (5′ GTC TCA GCG CAG TGC CGA TGG T 3′)-rLJ4 (5′ GAG GAG AAG CCC GGT TCA GGA GTT GGC CTG GAT GGT TT 3′) and inserted into the expression vector pIZ/V5-His (Invitrogen, Groningen, the Netherlands), resulting in the plasmid 12N22/pIZ. pro-hK5 with an enterokinase site (proEK-hK5) was produced by site directed mutagenesis, using the Quickchange Site-Directed Mutagenesis kit (Stratagene, La Jolla, California) with 50 ng of plasmid (12N22/pIZ) as template, and the primers rEKs (5′ GCC CGG TCG GAT GAC GAC GAC GAC AAG ATC ATC AAT GGA TCC GAC 3′) and rEKas (5′ GTC GGA TCC ATT GAT GAT CTT GTC GTC GTC GTC ATC CGA CCG GGC 3′). Both constructs were transfected into High Five cells according to the user manual InsectSelect System (Invitrogen) and propagated in High Five Serum-Free medium (Invitrogen).
      For production of recombinant pro-hK14 in insect cells a cDNA containing the coding sequence (
      • Hooper J.D.
      • Bui L.T.
      • Rae F.K.
      • Harvey T.J.
      • Myers S.A.
      • Ashworth L.K.
      • Clements J.A.
      Identification and characterization of KLK14, a novel kallikrein serine protease gene located on human chromosome 19q13.4 and expressed in prostate and skeletal muscle.
      , GenBank AF283670), including the signal sequence, was amplified from an RNA-preparation of human epidermis (
      • Brattsand M.
      • Egelrud T.
      Purification, molecular cloning, and expression of a human stratum corneum trypsin-like serine protease with possible function in desquamation.
      ) using primer pairs KLK14-F1g (5′ AGC CCC TAA AAT GGT CCT CCT GCT 3′) and KLK14-as (5′ GTC CCG CAT CGT TTC CTC AAT CC 3′) or KLK14-R1 (5′ TCA TTT GTC CCG CAT CGT TTC CTC 3′) and cloned into the pMT/V5-His-TOPO vector (Invitrogen). Since the primer KLK14-as did not contain stop codon, the resulting pro-hK14 protein contains V5- and His-tags. The resulting constructs were used to transfect Drosophila Schneider 2 cells using the Cellfectin Reagent Kit (Invitrogen). Cells were co-transfected with the pCoBlast vector (Invitrogen) and stable cell lines were selected according to instructions in the DES Blasticidin Support kit (Invitrogen). Stable cell lines were adapted to serum-free media (Drosophila SFM, Invitrogen) supplemented with 5 μg per mL blasticidin. Protein expression was induced for 2 wk by adding 500 μM CuSO4 to the media.
      To produce hK14 in yeast cells, the coding sequence of pro-hK14 without the signal sequence and including the stop codon, amplified using the primer pair KLK14-F1 (5′ CAA GAG GAT GAG AAC AAG ATA ATT GG 3′) and KLK14-R1 (see above), was cloned into the pPICZαA vector in frame with the α-factor secretion signal and transformed into Pichia pastoris KM71H using the Easy Comp Transformation kit (Invitrogen). One liter of BMGY (1% yeast extract, 2% peptone, 100 mM K2HPO4 pH 6.0, 1.34% yeast nitrogen base, 4 × 10−5% biotin and 1% glycerol) was inoculated and incubated for 3 days. Cells were harvested by centrifugation at 1500 ×g for 5 min and resuspended in BMMY (as BMGY, see above, but with 1% methanol instead of glycerol) at a cell density corresponding to OD600=160. The cells were induced for 72 h, and every 24 h methanol was added to a final concentration of 1%.
      The nucleotide sequences of inserts of all constructs were verified by sequencing using the Big Dye Terminator Cycle Sequencing kit and analyzed using an ABI377 automated DNA sequencer (Applied Biosystems, Foster City, California).
      Production and purification of recombinant pro-hK7 has been published (
      • Hansson L.
      • Strömqvist M.
      • Bäckman A.
      • Wallbrandt P.
      • Carlstein A.
      • Egelrud T.
      Cloning, expression, and characterization of stratum corneum chymotryptic enzyme. A skin-specific human serine proteinase.
      ).

      Purification of recombinant hK5 and hK14

      Recombinant protein was purified from conditioned cell medium by batch adsorption to CM Sepharose FF (Amersham Pharmacia Biotech, Freiburg, Germany) followed by reversed phase chromatography (RPC) (
      • Brattsand M.
      • Egelrud T.
      Purification, molecular cloning, and expression of a human stratum corneum trypsin-like serine protease with possible function in desquamation.
      ). When needed, further purification was obtained by gel exclusion chromatography followed by a second run of RPC (
      • Caubet C.
      • Jonca N.
      • Brattsand M.
      • et al.
      Degradation of corneodesmosome proteins by two serine proteases of the kallikrein family, SCTE/KLK5/hK5 and SCCE/KLK7/hK7.
      ). Purification was considered satisfactory if all components detected by coomassie-stained SDS-PAGE had the expected molecular mass and were labeled by the respective specific antibody. Purified proteins were stored at 4°C until use. Before use in activation or kinetic experiments the enzymes were transferred to the appropriate buffers with Micro Biospinn Columns (BioRad Laboratories, Hercules, California).
      Mutated pro-hK5 and pro-hK14 produced in insect cells were activated by treatment with recombinant enterokinase (Novagen, MERCK Eurolab, Darmstadt, Germany), 0.04 U per μg recombinant protein, at 37°C overnight in 20 mM Tris-HCl pH 8.0, 50 mM NaCl, 2 mM CaCl2. Activated enzymes were stored at 4°C after acidification to pH 4 with 0.5 M HCl.
      Protein concentration was determined with the DC Protein Assay (BioRad). SDS-PAGE and immunoblotting of reduced samples were performed according to standard protocols. Polyclonal rabbit antibodies were raised against synthetic peptides; for hK5 a mixture of RIRPTKDVRPINVSSHC and CEDAYPRQIDDTMF, and for hK14 a mixture of QEDENKIIGGHTC, CRQVTHPNYNSRTHD, and CKYRSWIEETMRDK. For hK7 a polyclonal antiserum raised against rpro-hK7, labeling precursor as well as active enzyme, was used (
      • Sondell B.
      • Dyberg P.
      • Anneroth G.K.
      • Ostman P.O.
      • Egelrud T.
      Association between expression of stratum corneum chymotryptic enzyme and pathological keratinization in human oral mucosa.
      ). Extracts of plantar stratum corneum in acetic acid were prepared as described (
      • Ekholm E.
      • Brattsand M.
      • Egelrud T.
      Stratum corneum tryptic enzyme in normal epidermis: A missing link in the desquamation process?.
      ).

      Enzyme activity and kinetics

      Casein zymography was performed as described (
      • Horie H.
      • Fukuyama K.
      • Ito Y.
      • Epstein W.L.
      Detection and characterization of epidermal proteinases by polyacrylamide gel electrophoresis.
      ). Active site titration with α1-antitrypsin (SIGMA, St Louis, Missouri) was performed essentially according to
      • Salvesen G.
      • Nagase H.
      Inhibition of proteolytic enzymes.
      . Known amounts of bovine pancreatic trypsin (SIGMA, cat. no. T-8253) were used for standardization. Chromogenic peptide substrates (see Table I for structures) were purchased from Haemochrom Diagnostica AB (Mölndal, Sweden). Mixtures containing substrate, enzyme, and buffer, as specified in legends to tables and figures, were incubated in duplicate or triplicate at 25°C or 37°C. When the enzyme under study had been activated with enterokinase, blank values were obtained from control incubations with the appropriate concentrations of enterokinase run in parallel. Km and Vmax were calculated using a Hanes plot. Aprotinin, soybean trypsin inhibitor, leupeptin, and chymostatin were purchased from Roche (Mannheim, Germany).

      ACKNOWLEDGMENTS

      This work was supported by the Swedish Medical Research Council (GrantK2001-71X-11206-07C), Lions Cancer Research Foundation, Umeå University, the Welander-Finsen Foundations and Arexis AB. The technical assistance of Astrid Lundgren and Bo Glas is gratefully acknowledged.

      References

        • Brattsand M.
        • Egelrud T.
        Purification, molecular cloning, and expression of a human stratum corneum trypsin-like serine protease with possible function in desquamation.
        J Biol Chem. 1999; 274: 30033-30040
        • Britton H.T.S.
        • Welford G.J.
        J Chem Soc. 1937: 1848-1852
        • Caubet C.
        • Jonca N.
        • Brattsand M.
        • et al.
        Degradation of corneodesmosome proteins by two serine proteases of the kallikrein family, SCTE/KLK5/hK5 and SCCE/KLK7/hK7.
        J Invest Dermatol. 2004; 122: 1235-1244
        • Clements J.
        • Hooper J.
        • Dong Y.
        • Harvey T.
        The expanded human kallikrein (KLK) gene family: Genomic organisation, tissue-specific expression and potential functions.
        J Biol Chem. 2001; 382: 5-14
        • Dong Y.
        • Kaushal A.
        • Brattsand M.
        • Nicklin J.
        • Clements J.
        Differential splicing of KLK5 and KLK7 in epithelial ovarian cancer produces novel variants with potential as cancer biomarkers.
        Clin Cancer Res. 2003; 9: 1710-1720
        • Egelrud T.
        Purification and preliminary characterization of stratum corneum chymotryptic enzyme: A proteinase that may be involved in desquamation.
        J Invest Dermatol. 1993; 101: 200-204https://doi.org/10.1111/1523-1747.ep12363804
        • Egelrud T.
        Desquamation in the stratum corneum.
        Acta Derm Venereol Suppl. 2000; 208: 44-45https://doi.org/10.1080/000155500750042853
        • Egelrud T.
        • Lundström A.
        A chymotrypsin-like proteinase that may be involved in desquamation in plantar stratum corneum.
        Arch Dermatol Res. 1991; 283: 108-112https://doi.org/10.1007/BF00371618
        • Ekholm E.
        • Brattsand M.
        • Egelrud T.
        Stratum corneum tryptic enzyme in normal epidermis: A missing link in the desquamation process?.
        J Invest Dermatol. 2000; 114: 56-63https://doi.org/10.1046/j.1523-1747.2000.00820.x
        • Hachem J.P.
        • Crumrine D.
        • Fluhr J.
        • Brown B.E.
        • Feingold K.R.
        • Elias P.M.
        pH directly regulates epidermal permeability barrier homeostasis, and stratum corneum integrity/cohesion.
        J Invest Dermatol. 2003; 121: 345-353https://doi.org/10.1046/j.1523-1747.2003.12365.x
        • Hansson L.
        • Bäckman A.
        • Ny A.
        • et al.
        Epidermal overexpression of stratum corneum chymotryptic enzyme in mice: A model for chronic itchy dermatitis.
        J Invest Dermatol. 2002; 118: 444-449https://doi.org/10.1046/j.0022-202x.2001.01684.x
        • Hansson L.
        • Strömqvist M.
        • Bäckman A.
        • Wallbrandt P.
        • Carlstein A.
        • Egelrud T.
        Cloning, expression, and characterization of stratum corneum chymotryptic enzyme. A skin-specific human serine proteinase.
        J Biol Chem. 1994; 269: 19420-19426
        • Hooper J.D.
        • Bui L.T.
        • Rae F.K.
        • Harvey T.J.
        • Myers S.A.
        • Ashworth L.K.
        • Clements J.A.
        Identification and characterization of KLK14, a novel kallikrein serine protease gene located on human chromosome 19q13.4 and expressed in prostate and skeletal muscle.
        Genomics. 2001; 73: 117-122https://doi.org/10.1006/geno.2000.6490
        • Horie H.
        • Fukuyama K.
        • Ito Y.
        • Epstein W.L.
        Detection and characterization of epidermal proteinases by polyacrylamide gel electrophoresis.
        Comp Biochem Physiol B. 1984; 77: 349-353https://doi.org/10.1016/0305-0491(84)90342-0
        • Horikoshi T.
        • Arany I.
        • Rajaraman S.
        • et al.
        Isoforms of cathepsin D and human epidermal differentiation.
        Biochimie. 1998; 80: 605-612
        • Huang C.
        • Wong G.W.
        • Ghildyal N.
        • et al.
        The tryptase, mouse mast cell protease 7, exhibits anticoagulant activity in vivo and in vitro due to its ability to degrade fibrinogen in the presence of the diverse array of protease inhibitors in plasma.
        J Biol Chem. 1997; 272: 31885-31893https://doi.org/10.1074/jbc.272.50.31885
        • Komatsu N.
        • Takata M.
        • Otsuki N.
        • Toyama T.
        • Ohka R.
        • Takehara K.
        • Saijoh K.
        Expression and localization of tissue kallikrein mRNAs in human epidermis and appendages.
        J Invest Dermatol. 2003; 121: 542-549https://doi.org/10.1046/j.1523-1747.2003.12363.x
        • Lundström A.
        • Egelrud T.
        Stratum corneum chymotryptic enzyme: A proteinase which may be generally present in the stratum corneum and with a possible involvement in desquamation.
        Acta Derm Venereol. 1991; 71: 471-474
        • Salvesen G.
        • Nagase H.
        Inhibition of proteolytic enzymes.
        in: Reynon R.J. Bond J.S. Proteolytic Enzymes, a Practical Approach. 3rd edn. Oxford University Press, Oxford. England1994: 83-104
        • Sondell B.
        • Dyberg P.
        • Anneroth G.K.
        • Ostman P.O.
        • Egelrud T.
        Association between expression of stratum corneum chymotryptic enzyme and pathological keratinization in human oral mucosa.
        Acta Derm Venereol. 1996; 76: 177-181
        • Suzuki Y.
        • Koyama J.
        • Moro O.
        • Horii I.
        • Kikuchi K.
        • Tanida M.
        • Tagami H.
        The role of two endogenous proteases of the stratum corneum in degradation of desmoglein-1 and their reduced activity in the skin of ichthyotic patients.
        Br J Dermatol. 1996; 134: 460-464https://doi.org/10.1046/j.1365-2133.1996.31794.x
        • Tanimoto H.
        • Underwood L.J.
        • Shigemasa K.
        • Yan Y.
        • Clarke J.
        • Parmley T.H.
        • O'Brien T.J.
        The stratum corneum chymotryptic enzyme that mediates shedding and desquamation of skin cells is highly overexpressed in ovarian tumor cells.
        Cancer. 1999; 86: 2074-2082
        • Watkinson A.
        Stratum corneum thiol protease (SCTP): A novel cysteine protease of late epidermal differentiation.
        Arch Dermatol Res. 1999; 291: 260-268https://doi.org/10.1007/s004030050406
        • Yousef G.M.
        • Diamandis E.P.
        The new human tissue kallikrein gene family: Structure, function, and association to disease.
        Endocrine Rev. 2001; 22: 184-204https://doi.org/10.1210/er.22.2.184
        • Öhman H.
        • Vahlquist A.
        In vivo studies concerning a pH gradient in human stratum corneum and upper epidermis.
        Acta Derm Venereol. 1994; 74: 375-379