Keratinocyte-Specific Mesotrypsin Contributes to the Desquamation

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector ORIGINAL ARTICLE

Keratinocyte-Speciﬁc Mesotrypsin Contributes to the Desquamation Process via Kallikrein Activation and LEKTI Degradation Masashi Miyai1,3, Yuuko Matsumoto1,3, Haruyo Yamanishi1, Mami Yamamoto-Tanaka1,2, Ryoji Tsuboi2 and Toshihiko Hibino1

Kallikrein-related peptidases (KLKs) have critical roles in corneocyte desquamation and are regulated by lymphoepithelial Kazal-type inhibitor (LEKTI). However, it is unclear how these proteases are activated and how activated KLKs are released from LEKTI in the upper cornified layer. Recently, we reported cloning of a PRSS3 gene product, keratinocyte-specific mesotrypsin, from a cDNA library. We hypothesized that mesotrypsin is involved in the desquamation process, and the aim of the present study was to test this idea by examining the effects of mesotrypsin on representative desquamation-related enzymes pro-KLK5 and pro-KLK7. Incubation of mesotrypsin and these zymogens resulted in generation of the active forms. KLK activities were effectively inhibited by recombinant LEKTI domains D2, D2–5, D2–6, D2–7, D5, D6, D6–9, D7, D7–9, and D10–15, whereas mesotrypsin activity was not susceptible to these domains, and in fact degraded them. Immunoelectron microscopy demonstrated that mesotrypsin was localized in the cytoplasm of granular cells and intercellular spaces of the cornified layer. Proximity ligation assay showed close association between mesotrypsin and KLKs in the granular to cornified layers. Age-dependency analysis revealed that mesotrypsin was markedly down- regulated in corneocyte extract from donors in their sixties, compared with younger donors. Collectively, our findings suggest that mesotrypsin contributes to the desquamation process by activating KLKs and degrading the intrinsic KLKs’ inhibitor LEKTI. Journal of Investigative Dermatology (2014) 134, 1665–1674; doi:10.1038/jid.2014.3; published online 30 January 2014

INTRODUCTION chymotrypsin-like enzyme (stratum corneum chymotryptic Epidermal differentiation is precisely controlled, and the thick- enzyme) (Lundstrom and Egelrud, 1991) have roles in the ness of the epidermis is maintained through constant shedding detachment of superﬁcial corneocytes. Later, these proteases of the outermost layers of corneocytes, a process known as were identiﬁed as KLK5 and KLK7, respectively (Caubet et al., desquamation (Watt, 1989). Desquamation is accomplished by 2004). Recent investigations have suggested that other KLK– the degradation of corneodesmosomes, which are involved in related enzymes, including KLKs 6, 8, 10, 11, 13, and 14, are corneocyte–corneocyte adherence (Caubet et al., 2004; Ishida- also expressed in human epidermis, and it was suggested that Yamamoto and Kishibe, 2011). Corneodesmosomes are com- they may form a proteolytic cascade that results in desquama- posed of three major proteins: desmoglein 1, desmocollin, and tion (Komatsu et al., 2007). All KLKs are synthesized as corneodesmosin. These proteins are proteolytically hydrolyzed proenzymes, and subsequently activated through mechanisms by a family of kallikrein-related peptidases (KLKs). Earlier that have not yet been established. KLK5 is thought to be the studies demonstrated that a trypsin-like enzyme (stratum initiator of this cascade, as it is capable of activating most of the corneum tryptic enzyme) (Suzuki et al., 1993) and a family members (Ishida-Yamamoto et al., 2011). Desquamation is regulated not only by proteases involved

1Shiseido Research Center, Kanazawa-ku, Yokohama, Japan and 2Department in the activation cascade but also by inhibitors such as of Dermatology, Tokyo Medical University, Shinjuku-ku, Tokyo, Japan lymphoepithelial Kazal-type related inhibitor (LEKTI), a pro- 3These authors contributed equally to this work. duct of the serine protease inhibitor Kazal-type 5 gene SPINK5 Correspondence: Toshihiko Hibino, Shiseido Research Center, 2-12-1 Fukuura, (Borgono et al., 2007; Roelandt et al., 2009). LEKTI consists of Kanazawa-ku 236-8643, Yokohama, Japan. 15 serine protease-inhibitory domains and is believed to be E-mail: [email protected] processed to single or multiple domains to facilitate specific Abbreviations: KLK, kallikrein-related peptidase; LEKTI, lymphoepithelial inhibition of particular proteases. Recently, LEKTI2 encoded Kazal-type inhibitor; PBS, phosphate-buffered saline; PLA, proximity ligation by SPINK9 was identified in the palmoplantar epidermis and assay was found to show KLK5–specific inhibition (Brannstrom Received 12 June 2013; revised 15 October 2013; accepted 15 October 2013; accepted article preview online 3 January 2014; published online et al., 2012). SPINK6 was isolated from human plantar 30 January 2014 stratum corneum extracts as a selective inhibitor of KLK5, 7,

& 2014 The Society for Investigative Dermatology www.jidonline.org 1665 MMiyaiet al. Involvement of Mesotrypsin in Desquamation

and 14 (Meyer-Hoffert et al., 2010). Thus, KLK activities seem mesotrypsin (Figure 1b). In addition, incubation of pro-KLK5 to be strongly suppressed under physiological conditions. with mesotrypsin resulted in a time-dependent linear increase It is suggested that active KLKs are dissociated from various of KLK5 activity (Figure 1c). Similar results were obtained for SPINK gene products under acidic conditions at the skin pro-KLK7 (Figure 1d–f). These results clearly indicate that surface (Pampalakis and Sotiropoulou, 2007). However, mesotrypsin generates active KLKs through limited proteolysis Ishida-Yamamoto et al. (2012) clearly demonstrated that of pro-KLKs. degradation of corneodesmosomes starts deeper in the cornified layers. Thus, it is plausible that dissociation of active KLKs Inhibition of KLK5 by LEKTI domains and the effect of pH on the from their inhibitors is not a cell-surface event, but is interaction controlled at least in part by mechanisms other than pH. Next, we generated various fragments of LEKTI (D2, D2–5, On the other hand, we have cloned an epidermis-specific D2–6, D2–7, D5, D6, D7, D7–9, D8–9, and D10–15), splicing variant of PRSS3 gene, trypsinogen 5, from a kerati- covering the whole length of the LEKTI protein (Figure 2a). nocyte cDNA library (Nakanishi et al., 2010). Trypsinogen 5 All of these LEKTI fragments were able to inhibit KLK5 activity possesses a very short N-terminal propeptide followed by the (Figure 2b). The inhibitory concentration 50 (IC50) values are enterokinase recognition sequence DDDDK-I. After cleavage shown in Figure 2c. D2 showed a high IC50 value for KLK5 by enterokinase, the same mesotrypsin would be produced as compared with other domains, indicating that D2 may not be from pancreatic trypsinogen 3 and brain-type trypsinogen 4. an effective inhibitor of KLK5. Similarly, all LEKTI fragments We confirmed that epidermal mesotrypsin is expressed in the were able to inhibit KLK7 activity (Supplementary Figure S1 upper epidermis, whereas its activating enzyme, enterokinase, online). Each domain construct showed somewhat weaker shows a more confined localization in the boundary area inhibition of KLK7, compared with KLK5, and none of these between the granular layer and cornified layer. Characteristic LEKTI domains showed a clear preference for KLK7 inhibition. features of mesotrypsin are as follows: (1) it cannot digest high- Next, we tested whether inhibition of KLK5 by LEKTI domains molecular-weight proteins, in contrast to anionic trypsin and was abrogated under acidic conditions (Figure 2d). KLK5 was cationic trypsin, the PRSS1 and PRSS2 gene products, incubated with D5 or D6 in buffers of different pH values. respectively, and (2) it is not susceptible to naturally KLK5 was active between pH 5.5 and pH 9, and D5 and D6 occurring trypsin inhibitors (Nyaruhucha et al., 1997). Thus, completely suppressed KLK5 activity in this pH range. These the physiological role of mesotrypsin is likely to be completely results indicate that loss of LEKTI inhibition of KLKs may not different from those of trypsins. Instead, it seems likely to be occur simply as a result of the acidic pH typically observed at involved in limited proteolysis, such as zymogen activation. the skin surface. Furthermore, it is capable of inactivating naturally occurring trypsin inhibitors by proteolytic cleavage. These unique pro- Mesotrypsin degrades various LEKTI domains perties of mesotrypsin led us to hypothesize that epidermal Next, we tested the effect of these LEKTI constructs on the mesotrypsin is involved in the desquamation process. To test activity of mesotrypsin (Figure 3a). None of these LEKTI this idea, we focused on two critical processes. One is the constructs inhibited mesotrypsin activity even in the presence activation of pro-KLK5 and the other is dissociation of active of 1,000-fold excess of recombinant LEKTIs. These results KLKs from LEKTI proteins. Furthermore, we examined the indicate that mesotrypsin is resistant to intrinsic serine pro- localization of mesotrypsin in human epidermis using tease inhibitors, including LEKTI. We then examined whether immunoelectronmicroscopy. Further, in situ proximity ligation the LEKTI constructs were degraded by mesotrypsin, using assay (PLA) was used to assess the interaction of mesotrypsin newly prepared antibodies against D2 and D12, and commer- with KLKs or LEKTI proteins in human epidermis. cially available anti-D4–8 antibody (F2). The N-terminal domains of LEKTI (D2 and D2–7), the inner domains (D5, RESULTS D6, D7, and D8–9), and the C-terminal domains (D10–15) all Mesotrypsin activates epidermal prokallikreins disappeared after incubation with mesotrypsin (Figure 3b). Epidermal kallikreins are all synthesized as inactive proen- Next, we investigated the effect of mesotrypsin on the zymes that require proteolytic activation by cleavage. Meso- recovery of KLK activities from the LEKTI-inhibited condition trypsin is expressed in the upper granular and transitional cell (Figure 3c). As shown in the Supplementary Information layers, where KLKs are also synthesized and secreted during online and Supplementary Figure S2a online, enzyme activ- terminal differentiation. Therefore, mesotrypsin and pro-KLKs ities of KLK5 and KLK7 were almost completely suppressed by may interact in this region. We focused on KLK5 and KLK7 as each of the LEKTI domains D2–7, D7–D9, and D10–15. We representative desquamation-related proteases, as KLK5 acti- selected these three recombinant proteins to cover the whole vates all other pro-KLKs so far tested (Bratts and et al., 2005) of the LEKTI inhibitory domains. Addition of active mesotryp- and KLK7 is one of the most abundant proteases expressed in sin to the KLK5–LEKTI mixtures resulted in gradual recovery of the epidermis (Ovaere et al., 2009). In the presence of KLK5 activity. Recovery of KLK activities was evident for each mesotrypsin, a 26-kDa band corresponding to active KLK5 of the LEKTI domains, although the rate varied depending on was generated from pro-KLK5 in a dose-dependent manner the particular inhibitory domain. Similar results were obtained (Figure 1a). In contrast, the incubation of pro-KLK5 alone for KLK7. Thus, not only is mesotrypsin not susceptible to resulted in no change. Time-course analysis confirmed con- LEKTI but also it has the ability to destroy LEKTI domains even version of pro-KLK5 to its active form by treatment with in the enzyme–inhibitor complex.

1666 Journal of Investigative Dermatology (2014), Volume 134 MMiyaiet al. Involvement of Mesotrypsin in Desquamation

Mesotrypsin Mesotrypsin 0 0.1 0.3 1 (nM) 0 0.1 0.3 31(nM) 52 kDa 38 kDa * 38 kDa 31 kDa 31 kDa 24 kDa Blot: KLK5 24 kDa Blot: KLK7 Blot: mesotrypsin Blot: mesotrypsin

Mesotrypsin Mesotrypsin –+ –+ 0 15 30 60 0 15 30 60 (min) 0 15 30 60 0 15 30 60 (min) 52 kDa 38 kDa * 31 kDa 38 kDa 31 kDa 24 kDa Blot: KLK5 24 kDa Blot: KLK7 Blot: mesotrypsin Blot: mesotrypsin

14,000 16,000 Pro-KLK5 pro-KLK7 14,000 12,000 Pro-KLK5+mesotrypsin pro-KLK7+mesotrypsin 12,000 10,000 10,000 8,000 8,000 6,000 6,000 4,000 4,000 Fluorescence emission 2,000 Fluorescence emission 2,000 0 0 0 153045607590105120 0 153045607590105120 Time (min) Time (min)

Figure 1. Activation of prokallikrein-related peptidase 5 (pro-KLK5) and pro-KLK7 by mesotrypsin. (a) Western blot analysis of conversion of pro-KLK5 to its active form. Pro-KLK5 (arrowhead) was incubated with different concentrations of mesotrypsin as indicated in the figures for 60 minutes. Pro-KLK5 was converted to the active form (arrow) by mesotrypsin in a dose-dependent manner (star indicates a nonspecific band). (b) Time-course analysis of pro-KLK5 conversion. Pro-KLK5 (arrowhead) was incubated with or without mesotrypsin for up to 60 minutes, and cleavage products were analyzed by western blot analysis. The active form of KLK5 (arrow) was detected after treatment with mesotrypsin in a time-dependent manner, whereas no active form was detected in the absence of mesotrypsin (star indicates a nonspecific band). (c) Measurement of enzymatic activity after the pro-KLK5 conversion. Pro-KLK5 was incubated with or without mesotrypsin for 30 minutes. Fluorescence emission was recorded after the addition of the fluorogenic substrate. Assay mixtures containing pro-KLK5 and mesotrypsin showed a marked increase of hydrolytic activity (closed circles), compared with pro-KLK5 alone (open diamonds). Experiments were performed in triplicate. Data are shown as mean fluorescence±SD. (d) Prepro-KLK7 (open arrowhead) decreased with time, and pro-KLK7 (filled arrowhead) was generated, being finally converted to the active form (arrow). (e) Time-course analysis of pro-KLK7 conversion. The incubation of prepro-KLK7 (open arrowhead) with mesotrypsin generated pro-KLK7 (filled arrowhead), which was converted to active KLK7 (arrow) in a time-dependent manner. (f) Measurement of enzymatic activity after pro-KLK7 conversion. Assay mixtures containing pro-KLK7 and mesotrypsin showed a linear increase of hydrolytic activity (closed circles), whereas pro-KLK7 alone showed no detectable activity during the 120-minute incubation time (open circles). Stability of mesotrypsin in the mixture was also confirmed by immunoblotting using the same samples, as shown in the lower panels of a, b, d,ande.

We also tested the presence of LEKTI fragments in human microscopy to examine its localization. Strong immunoreac- cornified cell extracts. Western blot analysis demonstrated tivity of mesotrypsin was associated with filamentous struc- three major bands of 65, 27, and 65 kDa that were visualized tures in the cytoplasm of granular cells, although not forming with anti-D2, anti-D4–8, and anti-D12 antibodies, respec- particular patterns, whereas immunoreactivity was sparsely tively (Supplementary Figure S2b online). Considering their distributed on spinous cells (Figure 4a). In addition, immunor- molecular weights and reactivity to antibodies, these bands eactivity was found broadly in interstitial spaces of cornified may correspond to D2–7, D7–9, and D10–15 fragments. layers, with some dense aggregates. We further examined There were also several additional bands that were detected whether mesotrypsin and KLKs are associated with each other with each antibody, although they seemed less abundant. in the human epidermis, using the in situ PLA method to visualize protein–protein interactions on tissue sections. PLA Spatiotemporal localization of mesotrypsin enables interaction indicated that mesotrypsin interacts with KLK5 and KLK7 with target molecules in the human epidermis (Figure 4b). Strong signals were detected at the boundary area To find out whether mesotrypsin can participate in the between the granular and cornified layers, and some dotted desquamation process in vivo,wecarriedoutimmunoelectron patterns were observed in the cornified layer. Interaction of

www.jidonline.org 1667 MMiyaiet al. Involvement of Mesotrypsin in Desquamation

Anti-D4-8 (F-2) Anti-D12 a Anti-D2 123456789101112131415 LEKTI

D2 D2-5 D2-6 D2-7 D5 D6 D7 D7-9 D8-9 D10-15

b 0.45 0.45 0.45 0.45 0.3 0.4 D2 0.4 D2-5 0.4 D2-6 0.4 D2-7 0.25 D5 0.35 0.35 0.35 0.35 mol) mol) mol) mol) mol) µ

µ 0.2 µ 0.3 0.3 µ 0.3 µ 0.3 0.25 0.25 0.25 0.25 0.15 0.2 0.2 0.2 0.2 0.15 0.15 0.15 0.15 0.1 0.1 0.1 0.1 0.1 Free MCA ( Free MCA ( Free MCA ( Free MCA ( 0.05 Free MCA ( 0.05 0.05 0.05 0.05 0 0 0 0 0 0 5 10 15 0 0.05 0.1 0 0.05 0.1 0.15 0 0.1 0.2 0.3 0.4 0 0.2 0.4 0.6 0.8 D5 (pM) D2 (nM) D2-5 (nM) D2-6 (nM) D2-7 (nM)

0.6 0.25 0.14 D7-9 D8-9 D10-15 0.5 D6 D7 0.12 0.4 0.2 0.2 0.35 mol) mol)

mol) 0.1 mol) mol) µ 0.4 µ µ µ

µ 0.3 0.15 0.08 0.15 0.3 0.25 0.06 0.1 0.1 0.2 0.2 0.04 0.15 Free MCA ( 0.1 0.05 Free MCA ( 0.05 0.1 Free MCA ( Free MCA (

0.02 Free MCA ( 0.05 0 0 0 0 0 0510 0 0.5 1 1.5 0 50 100 00.511.5 0 0.1 0.2 0.3 D6 (pM) D7-9 (nM) D7 (pM) D8-9 (nM) D10-15 (nM) c

IC50 (nM)

D2 4.25 d D5 D6 D2-5 0.027 12 120 12 120 ) ) D2-6 0.0004 10 100 10 100 ( ( ) 8 80 ) 8 80 D2-7 0.012 –1 –1 () () 6 60 6 60 D5 0.00018 4 40 4 40 D6 0.00025

2 20 20 Inhibition (%) Inhibition (%) 2 D7 0.03 Free MCA (nmol min Free MCA (nmol min 0 0 0 0 345678910 345678910 D7-9 0.0025 pH pH D8-9 0.0038

D10-15 0.012

Figure 2. Effect of recombinant lymphoepithelial Kazal-type inhibitor (LEKTI) domains on kallikrein-related peptidase 5 (KLK5). (a) Schematic illustration of recombinant LEKTI fragments. His-tag recombinants were expressed in Escherichia coli and were puriﬁed with Ni-Agarose, followed by anion-exchange Mono Q chromatography (GE Healthcare, Wauwatosa, WI). (b) Effect of recombinant LEKTI domains on KLK5. Inhibitor activity on KLK5 was tested for each LEKTI fragment. (c) Summary of inhibitory concentration 50 (IC50) values. Calculated IC50 values of recombinant LKETI fragments are listed. (d) Effect of pH on KLK5– LEKTI relationship. KLK5 was incubated with each LEKTI fragment in buffers of various pH values ranging from 4 to 9, and inhibitory activity was measured. Closed circles indicate KLK5 activities in various pH buffers. Open triangles and diamonds show inhibition on KLK5 in the presence of D5 (left) or D6 (right), respectively.

1668 Journal of Investigative Dermatology (2014), Volume 134 MMiyaiet al. Involvement of Mesotrypsin in Desquamation

a 4 4 4 4 4 D2 D2-5 D2-6 D2-7 D5 mol) mol) mol) mol) mol)

3 µ 3

µ µ 3 3 µ

µ 3 2 2 2 2 2 1 1 1 1 1 Free MCA ( Free Free MCA ( Free MCA ( Free Free MCA ( Free 0 0 0 0 MCA ( Free 0 0 5 10 15 0 0.5 1 0 0.5 1 1.5 0 0.2 0.4 0246 D2 (nM) D2-5 (nM) D2-6 (nM) D2-7 (nM) D5 (nM) 4 D6 4 4 4 2 D7 D7-9 D8-9 D10-15 mol) mol) mol) mol)

3 mol)

µ 3 3 3 µ µ µ µ 2 2 2 2 1 1 1 1 1 Free MCA ( Free Free MCA ( Free Free MCA ( Free Free MCA ( Free 0 0 0 0 MCA ( Free 0 0 0.5 1 0 0.5 1 0 0.05 0.1 0.15 0 0.05 0.1 0 0.01 0.02 0.03 D6 (nM) D7 (nM) D7-9 (nM) D8-9 (nM) D10-15 (nM)

b rhLEKTI-D2 rhLEKTI-D2-7 rhLEKTI-D5 0153060 (min) kDa 600 (min) kDa 600 (min) kDa 14.4 52 D5 D2 D2-7 7.7

rhLEKTI-D6 rhLEKTID-7 rhLEKTI-D8-9 rhLEKTI-D10-15 0153060 (min) kDa 600 (min) 600 (min) kDa 600 (min) kDa D6 kDa 17 7.7 D7 17 D8-9 52 D10-15

c D2-7 D7-9 D10-15 0.25 1.2

0.25 0.2 1 mol) mol) mol) µ 0.2 µ µ 0.8 0.15 0.15 0.6

KLK5 0.1 0.1 0.4 0.05

Free MCA ( Free 0.05 Free MCA ( Free Free MCA ( Free 0.2 0 0 0 0204060 0204060 0204060 Time (min) Time (min) Time (min)

1 0.4 0.4 0.8 mol) mol) mol) µ 0.3 µ 0.3 µ 0.6 0.2 0.2 KLK7 0.4

0.1 0.1 0.2 Free MCA ( Free Free MCA ( Free Free MCA ( Free 0 0 0 0 204060 0204060 0204060 Time (min) Time (min) Time (min)

Figure 3. Degradation of lymphoepithelial Kazal-type inhibitor (LEKTI) domains by mesotrypsin. (a) The same recombinant proteins were tested for inhibitory activities on mesotrypsin. None of these proteins showed any inhibitory effect on mesotrypsin. (b) Western blot analysis of mesotrypsin–LEKTI interaction. After incubation of the assay mixture for up to 60 minutes, LEKTI domains were analyzed by western blot analysis with anti-D2, anti-D4–8,and anti-D12 antibodies. (c) Recovery of kallikrein-related peptidase 5 (KLK5) (upper panel) and KLK7 activities (lower panel) from the LEKTI-inhibited state. The effect of mesotrypsin on KLK–LEKTI complexes was investigated under the condition of 100% KLK inhibition by LEKTI domains D2–7, D7–9, and D10–15. After enzyme–inhibitor (E–I) complex formation, mesotrypsin was added and recovery of KLK activities was measured at 15, 30, 45, and 60 minutes (closed circles). In separate experiments, KLK activities were monitored in the absence of mesotrypsin (open circles). mesotrypsin with KLK5 seems to start in upper spinous cells, mesotrypsin expression (about 0.4 ng ml), and this level was whereas that with KLK7 is more conﬁned to granular cells. maintained up to the age group of 50–59 years. However, tissue samples from the oldest age group (60–69 years) Age dependence of mesotrypsin expression showed a markedly lower concentration of mesotrypsin. It is suggested that desquamation ability decreases with advanced age. Therefore, we examined age-related changes DISCUSSION of mesotrypsin expression in extracts obtained from tape- In the present study, we focused on the physiological role of stripped samples of forearm skin (Figure 5). Amounts of mesotrypsin in the human epidermis. Intense staining of mesotrypsin were measured in each 10-year age group from mesotrypsin in the upper epidermis, especially in the granular 20 to 69 years. Younger age groups showed relatively high layer, and conﬁned localization of its activator enzyme,

www.jidonline.org 1669 MMiyaiet al. Involvement of Mesotrypsin in Desquamation

Granular cell Cornified layer

500 nm 100 µm

µ 200 m 200 nm

Mesotrypsin and normal IgG Mesotrypsin and KLK5 Mesotrpsin and KLK7

Figure 4. Ultrastructural localization of mesotrypsin and its interaction with kallikrein-related peptidases (KLKs). (a) Imunoelectronmicroscopy of mesotrypsin. (A) Upper epidermis, especially granular cells (indicated with dotted line), shows strong immunoreactivity for mesotrypsin. Bar ¼ 500 nm. (B) Immunoreactivity in the intercellular spaces of the cornified layer. Arrowheads indicate immunoreactivity for mesotrypsin in the cornified layer. Note the localization between detached corneocytes. Bar ¼ 100 mm. (C) Another field showing localization of mesotrypsin in the cornified layer. Bar ¼ 200 mm. (D) A higher-magnification view of the rectangular area marked in B. Dense aggregates of mesotrypsin immunoreactivity were frequently found in the intercellular spaces of the cornified layer. Bar ¼ 200 nm. (b) Interaction of mesotrypsin and KLKs evaluated with the PLA method. Close association between mesotrypsin and KLK5 is shown in the middle panel. Strong positive signals were detected at the granular layer. Less marked but still clear signals were found in the cornified layer (bottom). Similarly, mesotrypsin and KLK7 were closely associated in the granular layer and also to some degree in the cornified layer (right panel). Isotypic normal IgG did not show any positive reaction (left panel). An enlarged figure is shown at the bottom. Dotted line indicates the boundary of the skin surface. Broken line shows the epidermal–dermal junction. Bars ¼ 10 mm.

enterokinase, in granular cells strongly suggests that meso- pro-KLK7, and also through degradation of intrinsic trypsin has a role in keratinocyte terminal differentiation multidomain KLK inhibitors, LEKTIs. (Nakanishi et al., 2010). The present study is, to our know- Mesotrypsin is different from other trypsins or trypsin-like ledge previously unreported, to evaluate the possible role of enzymes, including KLKs, in having a unique amino-acid mesotrypsin in the desquamation process. Our results indicate substitution in the reaction pocket (Szepessy and Sahin-Toth, that mesotrypsin contributes to desquamation through 2006). It possesses the bulky amino-acid Arg instead of the activation of at least two key zymogens, pro-KLK5 and small amino-acid Gly, resulting in limited ability to hydrolyze

1670 Journal of Investigative Dermatology (2014), Volume 134 MMiyaiet al. Involvement of Mesotrypsin in Desquamation

1 As far as inhibitors are concerned, LEKTI, a SPINK5 gene ) –1 ** product, is implicated in regulation of the desquamation 0.8 protease cascade (Schechter et al., 2005; Borgono et al., N =11 2007). This is supported by the fact that SPINK5 deficiency ** 0.6 N =17 causes Netherton syndrome, an autosomal recessive ichthyosis- N =9 form disease characterized by dry skin, overdesquamation, N =14 0.4 hair abnormality, and atopy (Ishida-Yamamoto et al., 2005; Sun and Linden, 2006; Fortugno et al., 2011). LEKTI consists 0.2 N =13 of 15 Kazal-type serine protease domains and is processed to a series of single- or multidomain inhibitors. We prepared Conc. of mesotrypsin (ng ml Conc. 0 recombinant LEKTI domains of various sizes, covering the 20–29 30–39 40–49 50–59 60–69 whole length of LEKTI, and examined their inhibitory activities Age group toward KLK5, KLK7, and mesotrypsin. KLK5 was strongly inhibited by all the recombinant LEKTI domains except D2. Figure 5. Age-dependent change of mesotrypsin. Amounts of mesotrypsin The Kazal consensus sequence was best conserved in D2 and in extracts from tape-stripped corneocytes were measured using the mesotrypsin ELISA assay. Age groups of corneocyte donors covered each D15, and constructs containing these domains showed rather decade from the twenties to the sixties. weak inhibition of KLK5. The strongest inhibition of KLK5 was observed with D5 and D6, and fragments containing them were detected in corneocyte extracts in our study. Furthermore, high-molecular-weight proteins because of steric hindrance. we confirmed that at least D5, as well as D6, completely Substitution of Arg for Gly also makes mesotrypsin insuscep- suppressed KLK5 and KLK7 activities in the pH range of 5.5–9, tible to intrinsic serine protease inhibitors. These characteristic where KLKs are enzymatically active. Although it has been features of mesotrypsin distinguish it from other epidermal proposed that KLKs are dissociated under acidic conditions proteases that are tightly regulated by various inhibitors, and (around pH 5) at the skin surface (Deraison et al., 2007), it suggest that its role is to modulate physiological reactions by would be difficult to generate active KLKs at this relatively limited proteolysis of key protein(s). weak acidic pH, because (1) at pH 5.5 KLK activities were Increasing evidence supports a leading role of epidermal completely suppressed by LEKTI domains, and (2) at pH 5.0 KLKs in desquamation, specifically in the degradation of KLKs were inactivated. On the other hand, mesotrypsin was extracellular corneodesmosomal proteins. Interestingly, KLK5 completely resistant to all of these recombinant LEKTI is able to activate most, if not all, other KLK family members domains; in fact, it degraded these inhibitor proteins. This is (Ishida-Yamamoto and Kishibe, 2011). Furthermore, KLK5 in accordance with previous findings that human mesotrypsin degrades the three major corneodesmosomal proteins corneo- cleaves the reactive-site peptide bond of Kunitz-type soybean desmosin, desmoglein 1, and desmocollin 1 (Caubet et al., trypsin inhibitor and degrades Kazal-type pancreatic secretory 2004). Similar to KLK5, KLK7 also has strong proteolytic trypsin inhibitor (Szmola et al., 2003; Sahin-Toth, 2005). activity toward corneodesmosin and desmocollin 1 (Caubet Furthermore, we demonstrated that mesotrypsin is able to et al., 2004). KLK 6, 8, 10, 11, 13, and 14 protein levels in the degrade multiple LEKTI domains not only in the free state but stratum corneum have been quantified, and these molecules also in the enzyme–inhibitor complex. These lines of evidence were shown to degrade desmoglein 1 (Brattsand et al., 2005). suggest that mesotrypsin is likely to be an initiator of KLK14, which is highly expressed in plantar epidermis, could desquamation. activate pro-KLK5, but not pro-KLK7 or itself (Stefansson et al., The localization of mesotrypsin indicated that it would be 2006). involved in physiological reactions in the cytoplasm of On the basis of these findings, KLK5 appears to be a key granular cells and intercellular spaces of the cornified layer. enzyme in the desquamation protease cascade. Nevertheless, KLKs are incorporated into lamellar bodies and secreted from the activation mechanism of pro-KLK5 is still obscure. Although the apical surface of granular cells (Ishida-Yamamoto et al., pro-KLK5 can autoactivate under certain circumstances, this is 2004). Although it has been believed that KLKs are released a very slow process (Pampalakis and Sotiropoulou, 2007). from inhibitors and degrade corneodesmosomes in the stratum It was recently suggested that KLK5 is activated by matriptase, corneum, our recent study suggested that KLK5 is also present a type II membrane protease (Sales et al., 2010). Matriptase- in the cytoplasm of granular cells and may participate in the deficient mice showed a severe barrier defect, in addition processing of profilaggrin (Sakabe et al., 2013). This in turn to a defect of profilaggrin processing (List et al., 2003; may suggest that activation of KLKs occurs before secretion. Scharschmidt et al., 2009). In humans, loss-of-function Indeed, the results of PLA analysis indicated that KLK5 is mutation of matriptase causes autosomal recessive ichthyosis associated with mesotrypsin below the granular layer. with hypotrichosis, which is characterized by hyperkeratosis, The physiological roles of KLKs may not be limited to defective corneocyte maturation, and interlamellar lipid desquamation. KLKs are expressed in many tissues, including extrusion (Basel-Vanagaite et al., 2007; List et al., 2007). cancer cells, and aberrant expression of KLKs is often seen in Matriptase is a trypsin-like serine protease and is able to human malignancies. KLK3 or prostate-specific antigen is the activate pro-KLK5 in vitro (Netzel-Arnett et al., 2006). It may most widely known KLK owing to its application in diagnosis also be involved in desquamation. and monitoring of prostate cancer (LeBeau et al., 2010).

www.jidonline.org 1671 MMiyaiet al. Involvement of Mesotrypsin in Desquamation

Emerging evidence suggests a number of critical functional Preparation of anti-LEKTI-D2 and anti-LEKTI-D12 antibodies roles for KLKs in the pathogenesis of cancer, and they are Antibodies against human LEKTI D2 sequence (DDFKKGERDGDFIC) considered to be potential therapeutic targets (Andrade et al., and against human LEKTI D12 sequence (CTRESDPVRGPDGKTHGNK) 2010). It would be interesting to know whether mesotrypsin is were raised in rabbits (Supplementary Table S1 online). Monospecific expressed in cancer cells. anti-LEKTI D2 antibody and anti-LEKTI D12 antibody were purified Epidermal mesotrypsin showed a marked decrease in tissue from the respective sera using NHS-activated Sepharose 4B Fast Flow from subjects in the 60 to 69 age group. It is well known (GE Healthcare) conjugated with the antigen peptide. that with advancing age the skin shows reduced epidermal proliferation associated with slower desquamation of the Preparation of pro-KLK5 stratum corneum (Tagami, 2008). Although intercellular lipid Pro-KLK5 PCR products were generated in wheat germ lysate to production is also reduced in aged epidermis, the effect of this achieve efficient translation. The first PCR was performed using the is compensated by a thicker SC due to retardation of the adapter primers, a forward primer 50-CTTTAAGAAGGAGATATCA- desquamation process. These changes may be related to the TATGAACAACGACGTGAGCTGTG-30 and a reverse primer 50-TGA downregulation of mesotrypsin in aged skin. TGATGAGAACCCCCCCCTTAATGGTGGTGATGGTGGTGC-30,and On the basis of our findings, we propose the mechanism of the second amplification step was carried out by the RTS Wheat desquamation shown in Figure 6. Epidermal mesotrypsin is Germ Linear Template Generation Set, His-tag (Roche Diagnostics, activated by enteropeptidase in the granular layer and then Indianapolis, IN) according to the manufacturer’s instructions. The processes pro-KLK5 and pro-KLK7 to their active forms. It also PCR product was inserted into the pIVEX1.3WG vector. Recombinant degrades various forms of KLK-inhibitory LEKTI, allowing pro-KLK5 was expressed at 24 1C, 900 r.p.m. for 24 hours using the KLKs to initiate degradation of corneodesmosomes. Thus, RTS 500 Wheat- Germ CECF Kit (Roche Diagnostics). epidermis-specific mesotrypsin contributes to the desquamation process in two ways: via activation of pro-KLKs Preparation of various LEKTI constructs and via degradation of the intrinsic KLK inhibitor LEKTI. Constructs for LEKTI domains were generated by PCR amplification of Dysfunction of mesotrypsin regulation might therefore easily cDNA from a skin extract, using various primers (Supplementary affect terminal differentiation, resulting in various physiologi- Table S2 online). LEKTI fragments were cloned into pQE30, pQE100 cal disorders. Thus, the present findings throw light on (Qiagen, Valencia, CA), or pET30a (Novagen, Madison, WI). keratinocyte biology, especially in relation to diseases associated with aberrant keratinization, including deregulated Enzyme assays desquamation. Recombinant pro-KLK5 or pro-KLK7 was incubated alone or with mesotrypsin in the assay buffer (100 mM Tris-HCl, pH 8.0, 0.1% MATERIALS AND METHODS Tween-20, 1 mM CaCl2)at371C for 30 minutes. The fluorogenic Antibodies and proteins substrate Boc-Gln-Ala-Arg-AMC for KLK5 or Boc-Val-Pro-Arg-AMC Monoclonal anti-KLK5 antibody, recombinant active human (Peptide Institute, Osaka, Japan) for KLK7 was then added to a kallikrein 5, and recombinant active human trypsin 3/PRSS3 0.01 mM final concentration, and fluorescence was measured using a (mesotrypsin) were purchased from R&D Systems (Minneapolis, 2030 Multilabel Reader ARVO X3 (Perkin-Elmer, Waltham, MA) with MN). Monoclonal anti-KLK7 antibody and monoclonal anti-LEKTI excitation at 355 nm and emission at 460 nm. All experiments were (F-2) antibody were purchased from Santa Cruz Biotechnology (Santa performed in triplicate and repeated at least two times. For inhibitor Cruz, CA). assays, recombinant mature KLK5, KLK7, or active mesotrypsin was incubated with various LEKTI fragments at 37 1C for 30 minutes. The fluorogenic substrate was then added to the assay mixture (0.01 mM Desquamation final concentration), and fluorescence was measured as described above. Degradation of desmosome Recovery of KLK activities from LEKTI-inhibited condition Other KLKs Recombinant pro-KLK5 (10 ml, 5 ng ml 1)orpro-KLK7(10ml, Entropeptidase 1 KLK7 LEKTI 5ngml ) was mixed with LEKTI domains (D2–7, D7–9, or D10– Degradation 15) in the assay buffer and kept at room temperature for 10 minutes. KLK5 Activation Desmosome Amounts of recombinant LEKTI domains were determined from the Mesotrypsin inhibition curve, based on the calculated value of 100% inhibition for Pro-KLK5 Pro-KLK7 each recombinant. Then, 10 ml of active mesotrypsin (10 ng ml 1) and Boc-Gln-Ala-Arg-AMC (0.01 mM final concentration) were added Activation to the mixture, and recovery of KLK activities was monitored at 15, 30, 45, and 60 minutes. Figure 6. Proposed mechanism of desquamation. Epidermal mesotrypsin is activated by enteropeptidase at the granular layer, and processes prokallikrein- Western blot analysis related peptidase 5 (Pro-KLK5) and pro-KLK7 to their active forms. Furthermore, mesotrypsin degrades various forms of the KLK inhibitor LEKTI After electrophoresis, proteins were transferred from polyacrylamide (lymphoepithelial Kazal-type inhibitor), resulting in the appearance of KLK gel onto polyvinylidene difluoride membranes using the iBlot Dry activities. This initiates degradation of corneodesmosomes. Blotting System (Invitrogen, Carlsbad, CA). ImmunoBlock (DS Pharma

1672 Journal of Investigative Dermatology (2014), Volume 134 MMiyaiet al. Involvement of Mesotrypsin in Desquamation

Biomedical, Osaka, Japan) was used for blocking nonspecific reac- CONFLICT OF INTEREST tion. Membranes were subsequently incubated with anti-LEKTI D2 The authors state no conflict of interest. antibody, anti-LEKTI antibody (F-2), or anti-LEKTI D12 antibody for 1 hour at room temperature, followed by horseradish peroxidase– ACKNOWLEDGMENTS We are grateful to Mika Takeda for excellent technical assistance. conjugated anti-mouse IgG or horseradish peroxidase–conjugated anti-rabbit IgG (GE Healthcare). Immunoreactive proteins were visualized using ECL Plus (GE Healthcare). SUPPLEMENTARY MATERIAL Supplementary material is linked to the online version of the paper at http:// Tissue specimens www.nature.com/jid This study was approved by the Ethical Committee of Tokyo Medical University and the Shiseido Committee on Human Ethics. Human REFERENCES skin tissues were obtained at plastic surgery, in accordance with the Andrade D, Assis DM, Lima AR et al. (2010) Substrate specificity and inhibition of human kallikrein-related peptidase 3 (KLK3 or PSA) activated Declaration of Helsinki Principles. with sodium citrate and glycosaminoglycans. Arch Biochem Biophys 498: 74–82 Immunoelectron microscopy Basel-Vanagaite L, Attia R, Ishida-Yamamoto A et al. (2007) Autosomal Parts of the specimens were immediately cut into small blocks and recessive ichthyosis with hypotrichosis caused by a mutation in ST14, fixed in 4% paraformaldehyde in phosphate-buffered saline (PBS), encoding type II transmembrane serine protease matriptase. Am J Hum embedded in OTC compound (Sakura Finetek, Tokyo, Japan). Genet 80:467–77 Cryosections were immunostained with rabbit anti-trypsin antibody Borgono CA, Michael IP, Komatsu N et al. (2007) A potential role for multiple tissue kallikrein serine proteases in epidermal desquamation. JBiolChem (Athens Research & Technology, Athens, GA). After incubation with 282:3640–52 biotinylated anti-rabbit IgG antibody (Jackson ImmunoResearch Brannstrom K, Ohman A, von Pawel Rammingen U et al. (2012) Character- Laboratories, West Grove, PA), sections were reacted with AlexaFluor ization of SPINK9, a KLK5-specific inhibitor expressed in palmo-plantar 488-FluoroNanogold-Streptavidin (Nanoprobes, Yaphank, NY), Vec- epidermis. Biol Chem 393:369–77 tastain ABC Kit (Vector Laboratories, Burlingame, CA), and 3,30- Brattsand M, Stefansson K, Lundh C et al. (2005) A proteolytic cascade of diaminobenzidine, according to the manufacturer’s instructions. kallikreins in the stratum corneum. J Invest Dermatol 124:198–203 Following the silver enhancement step, specimens were postfixed Caubet C, Jonca N, Brattsand M et al. (2004) Degradation of corneodesmosome proteins by two serine proteases of the kallikrein family, SCTE/KLK5/ with 3% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4), followed hK5 and SCCE/KLK7/hK7. J Invest Dermatol 122:1235–44 by 1% osmium tetroxide, dehydrated in a graded series of alcohol, Deraison C, Bonnart C, Lopez F et al. (2007) LEKTI fragments specifically infiltrated with propylene oxide, and embedded in Epon–Araldite. inhibit KLK5, KLK7, and KLK14 and control desquamation through a pH- Sections were cut at 80 nm and examined with a transmission dependent interaction. Mol Biol Cell 18:3607–19 electron microscope H-7100 (Hitachi High-Tech, Tokyo, Japan). Fortugno P, Bresciani A, Paolini C et al. (2011) Proteolytic activation cascade of the Netherton syndrome-defective protein, LEKTI, in the Proximity ligation assay epidermis: implications for skin homeostasis. J Invest Dermatol 131:2223–32 PLA reaction was performed using combinations of monoclonal Ishida-Yamamoto A, Deraison C, Bonnart C et al. (2005) LEKTI is localized in antitrypsin antibody and anti-KLK5 antibody, and rabbit antitrypsin lamellar granules, separated from KLK5 and KLK7, and is secreted in the antibody and monoclonal anti-KLK7 antibody, respectively, accord- extracellular spaces of the superficial stratum granulosum. J Invest ing to the manufacturer’s instructions. Antibody dilution used was Dermatol 124:360–6 0.5 mgml 1 for each antibody. Ishida-Yamamoto A, Igawa S, Kishibe M (2011) Order and disorder in corneocyte adhesion. J Dermatol 38:645–54 Mesotrypsin ELISA assay Ishida-Yamamoto A, Kishibe M (2011) Involvement of corneodesmosome degradation and lamellar granule transportation in the desquamation Briefly, 4 mgml 1 of antitrypsin antibody in PBS was immobilized by process. Med Mol Morphol 44:1–6 overnight incubation on a 96-well plate. The plate was washed with Ishida-Yamamoto A, Kishibe M, Murakami M et al. (2012) Lamellar PBS containing 0.1% Tween-20 (PBS-T), nonspecific binding was granule secretion starts before the establishment of tight junction blocked with ImmunoBlock (Dainippon Sumitomo Pharma, Osaka, barrier for paracellular tracers in mammalian epidermis. PLoS One 7: Japan), and samples were incubated for 2 hours at 37 1C. The plate e31641 was washed with PBS-T, and then 0.9 mgml 1 of anti-mesotrypsin Ishida-Yamamoto A, Simon M, Kishibe M et al. (2004) Epidermal lamellar antibody (Lifespan Bioscience, Seattle, WA) was applied and the plate granules transport different cargoes as distinct aggregates. J Invest Dermatol 122:1137–44 was incubated for 1 hour at 37 1C. After incubation with horseradish Komatsu N, Saijoh K, Kuk C et al. (2007) Human tissue kallikrein expression in peroxidase–conjugated anti-mouse Fab (Sigma, St Louis, MO), color the stratum corneum and serum of atopic dermatitis patients. Exp was developed using the TMB Peroxidase EIA Substrate Kit (Bio-Rad Dermatol 16:513–9 Laboratories, Hercules, CA). Absorbance was read at 450 nm after LeBeau AM, Kostova M, Craik CS et al. (2010) Prostate-specific antigen: an stopping the reaction with 1 M sulfuric acid. Trypsin (Sigma) was used overlooked candidate for the targeted treatment and selective imaging of as a standard to quantify mesotrypsin. prostate cancer. Biol Chem 391:333–43 List K, Currie B, Scharschmidt TC et al. (2007) Autosomal ichthyosis with hypotrichosis syndrome displays low matriptase proteolytic activity and is Statistical analysis phenocopied in ST14 hypomorphic mice. J Biol Chem 282:36714–23 Data are expressed as the mean±SE. We used simple pairwise List K, Szabo R, Wertz PW et al. (2003) Loss of proteolytically processed comparison with Student’s t-test (two-tailed distribution with two- filaggrin caused by epidermal deletion of Matriptase/MT-SP1. J Cell Biol sample equal variance). Po0.05 was considered significant. 163:901–10

www.jidonline.org 1673 MMiyaiet al. Involvement of Mesotrypsin in Desquamation

Lundstrom A, Egelrud T (1991) Stratum corneum chymotryptic enzyme: a pro- Sales KU, Masedunskas A, Bey AL et al. (2010) Matriptase initiates activation of teinase which may be generally present in the stratum corneum and with a epidermal pro-kallikrein and disease onset in a mouse model of Netherton possible involvement in desquamation. Acta Dermatol Venereol 71:471–4 syndrome. Nat Genet 42:676–83 Meyer-Hoffert U, Wu Z, Kantyka T et al. (2010) Isolation of SPINK6 in human Scharschmidt TC, List K, Grice EA et al. (2009) Matriptase-deficient mice skin: selective inhibitor of kallikrein-related peptidases. JBiolChem exhibit ichthyotic skin with a selective shift in skin microbiota. J Invest 285:32174–81 Dermatol 129:2435–42 Nakanishi J, Yamamoto M, Koyama J et al. (2010) Keratinocytes synthesize Schechter NM, Choi EJ, Wang ZM et al. (2005) Inhibition of human kallikreins enteropeptidase and multiple forms of trypsinogen during terminal 5 and 7 by the serine protease inhibitor lympho-epithelial Kazal-type differentiation. J Invest Dermatol 130:944–52 inhibitor (LEKTI). Biol Chem 386:1173–84 Netzel-Arnett S, Currie BM, Szabo R et al. (2006) Evidence for a matriptase– Stefansson K, Brattsand M, Ny A et al. (2006) Kallikrein-related peptidase 14 prostasin proteolytic cascade regulating terminal epidermal differentia- may be a major contributor to trypsin-like proteolytic activity in human tion. J Biol Chem 281:32941–5 stratum corneum. Biol Chem 387:761–8 Nyaruhucha CN, Kito M, Fukuoka SI (1997) Identification and expression of Sun JD, Linden KG (2006) Netherton syndrome: a case report and review of the the cDNA-encoding human mesotrypsin(ogen), an isoform of trypsin with literature. Int J Dermatol 45:693–7 inhibitor resistance. JBiolChem272:10573–8 Suzuki Y, Nomura J, Hori J et al. (1993) Detection and characterization of Ovaere P, Lippens S, Vandenabeele P et al. (2009) The emerging roles of serine endogenous protease associated with desquamation of stratum corneum. protease cascades in the epidermis. Trends Biochem Sci 34:453–63 Arch Dermatol Res 285:372–7 Pampalakis G, Sotiropoulou G (2007) Tissue kallikrein proteolytic cascade Szepessy E, Sahin-Toth M (2006) Human mesotrypsin exhibits restricted S1’ pathways in normal physiology and cancer. Biochim Biophys Acta subsite specificity with a strong preference for small polar side chains. 1776:22–31 FEBS J 273:2942–54 Roelandt T, Thys B, Heughebaert C et al. (2009) LEKTI-1 in sickness and in Szmola R, Kukor Z, Sahin-Toth M (2003) Human mesotrypsin is a unique health. Int J Cosmetic Sci 31:247–54 digestive protease specialized for the degradation of trypsin inhibitors. Sahin-Toth M (2005) Human mesotrypsin defies natural trypsin inhibitors: JBiolChem278:48580–9 from passive resistance to active destruction. Protein Peptide Lett 12: Tagami H (2008) Functional characteristics of the stratum corneum in 457–64 photoaged skin in comparison with those found in intrinsic aging. Arch Sakabe JI, Yamamoto M, Hirakawa S et al. (2013) Kallikrein-related peptidase 5 Dermatol Res 300(Suppl 1):S1–6 functions in proteolytic processing of profilaggrin in cultured human Watt FM (1989) Terminal differentiation of epidermal keratinocytes. Curr Opin keratinocytes. JBiolChem288:17179–89 Cell Biol 1:1107–15

1674 Journal of Investigative Dermatology (2014), Volume 134