Advertisement

Par2 Inactivation Inhibits Early Production of TSLP, but Not Cutaneous Inflammation, in Netherton Syndrome Adult Mouse Model

      Netherton syndrome (NS) is a severe genodermatosis characterized by abnormal scaling and constant atopic manifestations. NS is caused by mutations in SPINK5 (Serine Protease INhibitor Kazal-type 5), which encodes LEKTI (LymphoEpithelial Kazal Type-related Inhibitor). Lack of LEKTI causes stratum corneum detachment secondary to epidermal proteases hyperactivity. Whereas a skin barrier defect is generally regarded as a major cause for atopy, we previously identified a cell-autonomous signaling cascade that triggers pro-Th2 cytokine thymic stromal lymphopoietin (TSLP) production in LEKTI-deficient epidermis. This signaling is initiated by unrestricted kallikrein 5 (KLK5) activity, which directly activates proteinase-activated receptor 2 (PAR2)-mediated expression of TSLP and favors a cutaneous proallergic microenvironment independently of the environment and of the adaptive immune system. To further confirm these results in vivo, we generated Spink5/Par2 double knockout (DKO) mice. At embryonic day 19.5, these mice display a dramatic decrease in TSLP expression, although stratum corneum detachment persists, confirming the role of the KLK5–PAR2 cascade in TSLP-mediated early proallergic signaling. However, deletion of Par2 in adult DKO-grafted skin does not rescue the inflammatory phenotype probably resulting from stratum corneum detachment. We conclude that several mechanisms trigger and maintain the inflammatory phenotype in NS. These include skin barrier impairment, mechanical stress secondary to stratum corneum detachment, as well as protease-induced proinflammatory and proallergic pathways, including PAR2-mediated overexpression of TSLP.

      Abbreviations

      AD
      atopic dermatitis
      DKO
      double knockout
      KLK5
      kallikrein 5
      LEKTI
      LymphoEpithelial Kazal Type-related Inhibitor
      NS
      Netherton syndrome
      PAR2
      proteinase-activated receptor 2
      SPINK5
      Serine Protease INhibitor Kazal-type 5
      TNFα
      tumor necrosis factor alpha
      TSLP
      thymic stromal lymphopoietin
      WT
      wild type

      Introduction

      The epidermis is a pluristratified epithelium, mainly composed of keratinocytes, which provides the first barrier protecting the body against penetration of harsh environmental agents. Skin injuries, danger signals, and epidermal barrier dysfunction lead to complex mechanisms triggering activation of keratinocytes, which produce a wide range of proinflammatory molecules (
      • Nickoloff B.J.
      • Naidu Y.
      Perturbation of epidermal barrier function correlates with initiation of cytokine cascade in human skin.
      ). Loss of this skin barrier function is believed to favor the development of atopic dermatitis (AD) by facilitating allergen penetration, as illustrated by patients with eczema due to null mutations in the filaggrin gene (
      • O’Regan G.M.
      • Sandilands A.
      • McLean W.H.
      • et al.
      Filaggrin in atopic dermatitis.
      ), and by the spontaneous mice mutant bearing a stop mutation in the murin Flg gene, which develop severe dermatitis after exposition to experimental allergen (
      • Fallon P.G.
      • Sasaki T.
      • Sandilands A.
      • et al.
      A homozygous frameshift mutation in the mouse Flg gene facilitates enhanced percutaneous allergen priming.
      ).
      Netherton syndrome (NS, OMIM 256500) is a rare and severe inherited skin disorder characterized by congenital scaly erythroderma, a specific hair shaft defect (Trichorrhexis invaginata), and constant atopic manifestations (
      • Comel M.
      Ichtyosis Linearis circumflexa.
      ;
      • Netherton E.W.
      A unique case of trichorrhexis nodosa; bamboo hairs.
      ). Allergic manifestations in NS include “atopic dermatitis”-like lesions associated with hyperimmunoglobulinemia E, allergic rhinitis, and food allergies (
      • Traupe H.
      ;
      • Judge M.R.
      • Morgan G.
      • Harper J.I.
      A clinical and immunological study of Netherton's syndrome.
      ). This is in sharp contrast with other ichthyoses, which exhibit skin barrier impairment but do not develop constant atopic manifestations. This observation suggests that specific mechanisms in NS underlie an allergic response. We previously showed that NS is caused by mutations in SPINK5 (Serine Protease INhibitor Kazal-type 5), encoding the serine protease inhibitor LEKTI (LymphoEpithelial Kazal Type-related Inhibitor) (
      • Chavanas S.
      • Bodemer C.
      • Rochat A.
      • et al.
      Mutations in SPINK5, encoding a serine protease inhibitor, cause Netherton syndrome.
      ). LEKTI tissue distribution pattern is restricted to the most differentiated viable layers of stratified epithelial tissues and the thymic Hassall's corpuscles (
      • Bitoun E.
      • Micheloni A.
      • Lamant L.
      • et al.
      LEKTI proteolytic processing in human primary keratinocytes, tissue distribution and defective expression in Netherton syndrome.
      ). In the epidermis, LEKTI is mainly restricted to the granular layer where it is expressed as high-molecular-weight precursors, which are rapidly processed into several proteolytic fragments secreted in the intercellular space (
      • Bitoun E.
      • Micheloni A.
      • Lamant L.
      • et al.
      LEKTI proteolytic processing in human primary keratinocytes, tissue distribution and defective expression in Netherton syndrome.
      ;
      • Deraison C.
      • Bonnart C.
      • Lopez F.
      • et al.
      LEKTI fragments specifically inhibit KLK5, KLK7, and KLK14 and control desquamation through a pH-dependent interaction.
      ). We along with others have shown that LEKTI fragments can efficiently and specifically inhibit epidermal kallikrein (KLK)5, KLK7, and KLK14 (
      • Egelrud T.
      • Brattsand M.
      • Kreutzmann P.
      • et al.
      hK5 and hK7, two serine proteinases abundant in human skin, are inhibited by LEKTI domain 6.
      ;
      • Schechter N.M.
      • Choi E.J.
      • Wang Z.M.
      • et al.
      Inhibition of human kallikreins 5 and 7 by the serine protease inhibitor lympho-epithelial Kazal-type inhibitor (LEKTI).
      ;
      • Deraison C.
      • Bonnart C.
      • Lopez F.
      • et al.
      LEKTI fragments specifically inhibit KLK5, KLK7, and KLK14 and control desquamation through a pH-dependent interaction.
      ).
      Spink5 knockout mice (Spink5−/-) faithfully reproduced key features of NS (
      • Yang T.
      • Liang D.
      • Koch P.J.
      • et al.
      Epidermal detachment, desmosomal dissociation, and destabilization of corneodesmosin in Spink5-/- mice.
      ;
      • Descargues P.
      • Deraison C.
      • Bonnart C.
      • et al.
      Spink5-deficient mice mimic Netherton syndrome through degradation of desmoglein 1 by epidermal protease hyperactivity.
      ;
      • Hewett D.R.
      • Simons A.L.
      • Mangan N.E.
      • et al.
      Lethal, neonatal ichthyosis with increased proteolytic processing of filaggrin in a mouse model of Netherton syndrome.
      ), and provided good models to gain insights into the cutaneous pathophysiological pathways of the disease. We showed that epidermal LEKTI deficiency results in KLK5, KLK7, and elastase 2 (ELA2) unrestricted protease activity, leading to premature desmosome cleavage and abnormal filaggrin maturation in the upper granular layer (
      • Descargues P.
      • Deraison C.
      • Bonnart C.
      • et al.
      Spink5-deficient mice mimic Netherton syndrome through degradation of desmoglein 1 by epidermal protease hyperactivity.
      ;
      • Bonnart C.
      • Deraison C.
      • Lacroix M.
      • et al.
      Elastase 2 is expressed in human and mouse epidermis and impairs skin barrier function in Netherton syndrome through filaggrin and lipid misprocessing.
      ). This induces accelerated stratum corneum shedding and impaired epidermal barrier function (
      • Descargues P.
      • Deraison C.
      • Bonnart C.
      • et al.
      Spink5-deficient mice mimic Netherton syndrome through degradation of desmoglein 1 by epidermal protease hyperactivity.
      ;
      • Bonnart C.
      • Deraison C.
      • Lacroix M.
      • et al.
      Elastase 2 is expressed in human and mouse epidermis and impairs skin barrier function in Netherton syndrome through filaggrin and lipid misprocessing.
      ). More recently, we demonstrated that unrestricted KLK5 activity also directly triggers the expression of proinflammatory (tumor necrosis factor alpha (TNFα), ICAM1, and IL8) and proallergic (thymic stromal lymphopoietin, TSLP) molecules through proteinase-activated receptor 2 (PAR2) activation in NS keratinocytes (
      • Briot A.
      • Deraison C.
      • Lacroix M.
      • et al.
      Kallikrein 5 induces atopic dermatitis-like lesions through PAR2-mediated thymic stromal lymphopoietin expression in Netherton syndrome.
      ). The pro-Th2 cytokine TSLP is a major actor of allergic inflammation, overexpressed in the AD lesional skin (
      • Soumelis V.
      • Reche P.A.
      • Kanzler H.
      • et al.
      Human epithelial cells trigger dendritic cell mediated allergic inflammation by producing TSLP.
      ). TSLP is necessary and sufficient to induce eczematous lesions and a systemic Th2 response, in the absence of exogenous allergen (
      • Yoo J.
      • Omori M.
      • Gyarmati D.
      • et al.
      Spontaneous atopic dermatitis in mice expressing an inducible thymic stromal lymphopoietin transgene specifically in the skin.
      ). We have shown that TSLP induction, already present in Spink5−/- embryos, is a cell-autonomous process, which is likely to have a key role in the development of AD-like lesions in NS.
      To extend these results in vivo, we generated mice inactivated for both Spink5 and Par2 genes. This double knockout (DKO) mouse model allowed us to identify PAR2 as an essential actor of TSLP expression early in life, although Par2 deletion is not sufficient to prevent inflammation in the adult skin where other mechanisms such as stratum corneum detachment could contribute to the inflammatory process.

      Results

      Spink5/Par2 DKO newborn mice display cutaneous features similar to Spink5−/- animals

      To further explore the implication of the KLK5–PAR2 pathway in the Spink5−/- phenotype, we generated DKO mice for both Spink5 and Par2 genes. We crossed heterozygous Spink5−/+ mice with Par2−/- mice (
      • Lindner J.R.
      • Kahn M.L.
      • Coughlin S.R.
      • et al.
      Delayed onset of inflammation in protease-activated receptor-2-deficient mice.
      ) and then intercrossed viable Spink5+/-/Par2−/- animals to obtain DKO mice. DKO newborn mice exhibited skin erosions and vibrissae abnormalities as seen in Spink5−/- mice and underwent rapid postnatal lethality due to severe dehydration consequently to stratum corneum detachment (Figure 1a). These results suggested that Par2 deletion, in a Spink5−/- context, does not abrogate the severe cutaneous phenotype of Spink5−/- newborn mice.
      Figure thumbnail gr1
      Figure 1Par2 inactivation does not rescue stratum corneum detachment but inhibits TSLP expression in Spink5−/- mice embryos. (a) At birth, Spink5−/- and double knockout (DKO) mice exhibit vibrissae abnormalities (left enlargement) and skin erosions (arrows and right enlargement). (b) At embryonic day 19.5 (E19.5), hematoxylin–eosin coloration reveals that the Spink5−/- and DKO epidermis show slight acanthosis and stratum corneum detachment. (c) Immunostaining shows that TSLP is not detected in the WT and Par2−/- epidermis and is poorly detected in the DKO epidermis in contrast to the Spink5−/- epidermis. (d) ICAM1 staining, restricted to basal keratinocytes in the WT and Par2−/- epidermis, is extended to suprabasal keratinocytes in the Spink5−/- and DKO epidermis. (e) Quantitative RT-PCR was performed on RNA extracted from the E19.5 skin. mRNA amount is expressed as a percentage of Hprt. Graphs represent the mean of three independent embryos (n=3). *P<0.05; ***P<0.01. (Panels bd) Bar=9.2μm. Par2, proteinase-activated receptor 2; RT-PCR, real-time PCR; Spink5, Serine Protease INhibitor Kazal-type 5; TSLP, thymic stromal lymphopoietin; WT, wild type.

      Par2 deficiency represses TSLP-induced expression in Spink5−/- embryos

      We next analyzed the skin from embryos delivered by Caesarean section at embryonic day 19.5 (E19.5) to disregard the effect of the environment on disease expression. Histological examination of skin cross-sections after hematoxylin–eosin coloration showed stratum corneum detachment and moderate acanthosis in both Spink5−/- and DKO embryos. In contrast, the Par2−/- skin appeared similar to the wild-type (WT) skin (Figure 1b). These observations suggested that the lack of PAR2 does not abolish epidermal protease hyperactivity leading to stratum corneum detachment and confirmed the severe cutaneous features observed in Spink5−/- newborns (Figure 1a). We nevertheless explored the effect of PAR2 inactivation on TSLP and ICAM1 expression, which we previously showed to be induced by KLK5-PAR2 signaling as early as E19.5 in the context of Spink5 deficiency (
      • Briot A.
      • Deraison C.
      • Lacroix M.
      • et al.
      Kallikrein 5 induces atopic dermatitis-like lesions through PAR2-mediated thymic stromal lymphopoietin expression in Netherton syndrome.
      ). We confirmed that TSLP was not detected in the WT or in the Par2−/- epidermis, whereas it was strongly produced in the epidermis of skin sections from Spink5−/- embryos (Figure 1c). In striking contrast, TSLP staining was poorly detected in the DKO epidermis (Figure 1c). Quantitative real-time-PCR analyses of the skin from E19.5 embryos showed that Tslp transcripts were highly overexpressed in the Spink5−/- skin as observed previously (
      • Briot A.
      • Deraison C.
      • Lacroix M.
      • et al.
      Kallikrein 5 induces atopic dermatitis-like lesions through PAR2-mediated thymic stromal lymphopoietin expression in Netherton syndrome.
      ) and could be compared with an on/off induction relative to the WT mice skin. In the DKO skin, induction of Tslp mRNA reached only 10% of that observed in the Spink5−/- skin (Figure 1e), confirming the immunohistological observation. ICAM1 staining, restricted to the basal layer in the WT and Par2−/- skin, was extended to suprabasal layers in the Spink5−/- and DKO epidermis (Figure 1d). Quantitative real-time-PCR confirmed that Icam1 mRNA amount was similar between the Spink5−/- and DKO embryo skin (data not shown). Finally, we analyzed transcriptional expression levels of Tnfα, which we previously identified as another target gene of KLK5-PAR2 signaling (
      • Briot A.
      • Deraison C.
      • Lacroix M.
      • et al.
      Kallikrein 5 induces atopic dermatitis-like lesions through PAR2-mediated thymic stromal lymphopoietin expression in Netherton syndrome.
      ). Tnfα mRNA increase in DKO was 60% of that seen in Spink5−/- (15.8- and 26.6-fold induction, respectively, relative to WT) (Figure 1e). Taken together, these results demonstrate that, in the NS mouse model, TSLP expression is highly dependent on the KLK5–PAR2 pathway in vivo as early as E19.5.

      Par2 inactivation does not improve cutaneous inflammation in the Spink5−/- adult skin

      To determine whether inhibition of the KLK5–PAR2 cascade is sufficient to reverse the inflammatory phenotype previously observed in the Spink5−/- adult skin (
      • Briot A.
      • Deraison C.
      • Lacroix M.
      • et al.
      Kallikrein 5 induces atopic dermatitis-like lesions through PAR2-mediated thymic stromal lymphopoietin expression in Netherton syndrome.
      ), we grafted the E19.5 skin onto athymic immunodeficient nude mice. Six weeks after grafting, WT and Par2−/- grafts were covered with normal hair, whereas Spink5−/-- and DKO-grafted skin appeared erythematous, scaly, and showed no hair (Figure 2a). Histological examination showed that both the Spink5−/-- and the DKO-grafted skin exhibit acanthosis with papillomatosis, hyperkeratosis (thickening of the stratum corneum), and parakeratosis (persistence of the nuclei in the stratum corneum) (Figure 2b). Luna staining of skin graft cross-sections showed rare polynuclear eosinophilic cells, colored in red, in the dermis of WT and Par2−/- grafts. In contrast, these cells were numerous in the dermis and infiltrated the epidermis of DKO- and Spink5−/--grafted skin (Figure 2c). Toluidin blue staining showed that mast cells were more numerous (4.5- and 3.8-fold increased, respectively) in Spink5−/- and DKO grafts than in WT and Par2−/- grafts dermis (Figure 2d and g).
      Figure thumbnail gr2
      Figure 2Par2 deficiency does not improve the Spink5−/- adult skin phenotype or inflammatory features. (a) The skin from embryonic day 19.5 (E19.5) mice was grafted onto nude mice for 6 weeks. (b) After hematoxylin–eosin coloration, WT- and Par2−/--grafted skin appears similar to the normal adult skin. In contrast, the Spink5−/- and DKO-grafted epidermis exhibit acanthosis, papillomatosis, and detachment of the stratum corneum, which is hyperkeratosic and parakeratosic. Inflammatory infiltrates were present in the dermis. (c) Luna staining reveals rare polynuclear eosinophilic cells (arrowhead) in the dermis of WT and Par2−/- grafts. In contrast, numerous polynuclear eosinophilic cells infiltrated the dermis and the epidermis of both Spink5−/- and DKO grafts. (d) Mast cells, stained with toluidin blue (arrowhead), are rare in the dermis of WT and Par2−/- grafts as compared with a higher number in the dermis of Spink5−/- and DKO grafts. (e) Immunohistochemistry reveals no TSLP detection in WT and Par2−/- adult-grafted epidermis, whereas it is induced in Spink5−/- mice and also in DKO animals. (f) ICAM1 expression, restricted to basal keratinocytes in the WT and Par2−/- epidermis, is extended to suprabasal keratinocytes in both the Spink5−/- and the DKO epidermis. (g) Mast cells were counted per dermal surface unit. Graphs represent the mean of 3 independent mice (n=3) for which 10 microscopic fields were analyzed. **P<0.02. (Panel b) bar=36.8μm; (panel c) bar=5.75μm; (panels de) bar=9.2μm. DKO, double knockout; Par2, proteinase-activated receptor 2; Spink5, Serine Protease INhibitor Kazal-type 5; TSLP, thymic stromal lymphopoietin; WT, wild type.
      We next analyzed TSLP and ICAM1 expression in the adult grafted skin. TSLP was not detected in the WT and Par2−/- skin in contrast to the Spink5−/- epidermis where it was strongly enhanced. TSLP was also detected in the DKO adult skin (Figure 2e). Immunostaining of ICAM1, restricted to the basal layer in WT or Par2−/- adult epidermis, extended to suprabasal layers in Spink5−/- and DKO (Figure 2f).
      Taken altogether, these data indicate that Par2 deletion is not sufficient to abolish inflammation in the LEKTI-deficient adult skin.

      Discussion

      Patients with NS exhibit abnormal desquamation and constant atopic manifestations. We previously demonstrated that in the absence of LEKTI, unrestricted KLK5 activity causes both stratum corneum detachment (
      • Descargues P.
      • Deraison C.
      • Bonnart C.
      • et al.
      Spink5-deficient mice mimic Netherton syndrome through degradation of desmoglein 1 by epidermal protease hyperactivity.
      ) and expression of proallergic and proinflammatory molecules by keratinocytes (
      • Briot A.
      • Deraison C.
      • Lacroix M.
      • et al.
      Kallikrein 5 induces atopic dermatitis-like lesions through PAR2-mediated thymic stromal lymphopoietin expression in Netherton syndrome.
      ). Indeed, KLK5 is able to cleave and activate PAR2 (
      • Oikonomopoulou K.
      • Hansen K.K.
      • Saifeddine M.
      • et al.
      Kallikrein-mediated cell signalling: targeting proteinase-activated receptors (PARs).
      ), leading to the induction of TSLP, ICAM1, and TNFα expression in human primary keratinocytes (
      • Briot A.
      • Deraison C.
      • Lacroix M.
      • et al.
      Kallikrein 5 induces atopic dermatitis-like lesions through PAR2-mediated thymic stromal lymphopoietin expression in Netherton syndrome.
      ). Thus, the KLK5–PAR2–TSLP signaling pathway seems to have a key role in atopic manifestations in NS. To further confirm these in vitro data, we developed Spink5/Par2 DKO mice. At birth, DKO mice displayed skin erosions similar to Spink5−/- mice and underwent rapid postnatal lethality. Histological examination of the skin at E19.5 confirmed abnormal detachment of the stratum corneum in DKO embryos as seen in Spink5−/- embryos. This observation implied that PAR2 inactivation does not abolish epidermal protease hyperactivity and consequently does not prevent stratum corneum detachment. This is consistent with previous data showing that KLK5 is a direct target of LEKTI (
      • Egelrud T.
      • Brattsand M.
      • Kreutzmann P.
      • et al.
      hK5 and hK7, two serine proteinases abundant in human skin, are inhibited by LEKTI domain 6.
      ;
      • Schechter N.M.
      • Choi E.J.
      • Wang Z.M.
      • et al.
      Inhibition of human kallikreins 5 and 7 by the serine protease inhibitor lympho-epithelial Kazal-type inhibitor (LEKTI).
      ;
      • Deraison C.
      • Bonnart C.
      • Lopez F.
      • et al.
      LEKTI fragments specifically inhibit KLK5, KLK7, and KLK14 and control desquamation through a pH-dependent interaction.
      ). Thus, enhanced protease activities in Spink5−/- newborn mice is due to the absence of LEKTI, with no increased transcriptional expression of these proteases (
      • Descargues P.
      • Deraison C.
      • Bonnart C.
      • et al.
      Spink5-deficient mice mimic Netherton syndrome through degradation of desmoglein 1 by epidermal protease hyperactivity.
      ).
      To disregard an indirect effect of environmental stimuli secondary to skin barrier impairment, we studied the expression of proinflammatory and proallergic molecules in the skin of DKO E19.5 embryos. Par2 deletion, in a context of LEKTI deficiency, dramatically inhibited Tslp induction at the mRNA and the protein levels, and decreased Tnfα expression by 40% in the skin of DKO embryos compared with Spink5−/- embryos. However, Icam1 expression seemed similar between the DKO and the Spink5−/- embryo skin. Although ICAM1 was demonstrated to be induced by PAR2 activation in human primary keratinocytes (
      • Buddenkotte J.
      • Stroh C.
      • Engels I.H.
      • et al.
      Agonists of proteinase-activated receptor-2 stimulate upregulation of intercellular cell adhesion molecule-1 in primary human keratinocytes via activation of NF-kappa B.
      ;
      • Briot A.
      • Deraison C.
      • Lacroix M.
      • et al.
      Kallikrein 5 induces atopic dermatitis-like lesions through PAR2-mediated thymic stromal lymphopoietin expression in Netherton syndrome.
      ), this key molecule is probably activated through other various and numerous stimuli, leading to immune cell recruitment in our mouse model. Taken together, these results demonstrate that Tslp expression is highly dependent on the KLK5–PAR2 pathway in vivo and that Tnfα expression is less dependent on this pathway, at least in the initial steps of the inflammatory response. This supports our previous data on TSLP overexpression in the epidermis of E19.5 Spink5−/- mice and in NS primary keratinocytes (
      • Briot A.
      • Deraison C.
      • Lacroix M.
      • et al.
      Kallikrein 5 induces atopic dermatitis-like lesions through PAR2-mediated thymic stromal lymphopoietin expression in Netherton syndrome.
      ). These data obtained in vivo, confirm the role of the KLK5/PAR2 axis in triggering a proallergic microenvironment through strong Tslp induction as early as E19.5, and consequently independently of the adaptive immune system and environmental stimuli.
      To further explore whether inhibition of the KLK5–PAR2 cascade could reduce inflammation previously observed in the Spink5−/- adult skin (
      • Briot A.
      • Deraison C.
      • Lacroix M.
      • et al.
      Kallikrein 5 induces atopic dermatitis-like lesions through PAR2-mediated thymic stromal lymphopoietin expression in Netherton syndrome.
      ), we grafted skin from E19.5 DKO onto athymic immunodeficient nude mice. Six weeks after grafting, the DKO adult skin exhibited a severe ichthyosiform phenotype similar to Spink5−/- grafts. At the histological level, the skin from DKO grafts displayed epidermal acanthosis and stratum corneum detachment as seen in the Spink5−/- skin, highlighting that these NS features are independent of PAR2. Moreover, although Par2−/- mice exhibit accelerated barrier recovery after tape stripping (
      • Hachem J.P.
      • Houben E.
      • Crumrine D.
      • et al.
      Serine protease signaling of epidermal permeability barrier homeostasis.
      ), Par2 deletion was not sufficient to rescue the effects of premature desquamation in the DKO adult epidermis.
      The adult DKO-grafted skin also displayed an inflammatory infiltrate composed of a high number of mast cells and polynuclear eosinophilic cells. This could be secondary to persistent SC detachment favoring sensitization by environmental allergens. However, DKO mice already suffer from stratum corneum detachment as early as E19.5. Considering the fact that our mice are maintained under specific pathogen free conditions and that other mouse models such as flaky tail mice do not exhibit skin inflammation without allergen challenge under this condition (
      • Fallon P.G.
      • Sasaki T.
      • Sandilands A.
      • et al.
      A homozygous frameshift mutation in the mouse Flg gene facilitates enhanced percutaneous allergen priming.
      ), it is likely that cutaneous inflammation in the DKO-grafted skin is also due to intrinsic signals secondary to LEKTI deficiency. Unregulated proteases activities, in the absence of LEKTI, could induce signals independently of PAR2, promoting expression of inflammatory molecules such as thymus and activation-regulated chemokine and macrophage-derived chemokine that are overproduced by NS primary keratinocytes cultured under basal conditions (
      • Briot A.
      • Deraison C.
      • Lacroix M.
      • et al.
      Kallikrein 5 induces atopic dermatitis-like lesions through PAR2-mediated thymic stromal lymphopoietin expression in Netherton syndrome.
      ). The role of other unrestricted protease activities in a LEKTI-deficient context is also supported by Ela2 (
      • Bonnart C.
      • Deraison C.
      • Lacroix M.
      • et al.
      Elastase 2 is expressed in human and mouse epidermis and impairs skin barrier function in Netherton syndrome through filaggrin and lipid misprocessing.
      ) and Klk7 (
      • Hansson L.
      • Backman A.
      • Ny A.
      • et al.
      Epidermal overexpression of stratum corneum chymotryptic enzyme in mice: a model for chronic itchy dermatitis.
      ) transgenic mice, which both exhibit spontaneous cutaneous inflammation. In addition, as demonstrated in the literature, other mechanisms such as mechanical stress caused by stratum corneum detachment could also induce skin inflammation (
      • Wood L.C.
      • Jackson S.M.
      • Elias P.M.
      • et al.
      Cutaneous barrier perturbation stimulates cytokine production in the epidermis of mice.
      ;
      • Nickoloff B.J.
      • Naidu Y.
      Perturbation of epidermal barrier function correlates with initiation of cytokine cascade in human skin.
      ). Indeed, tape striping induces IL1β expression and secretion by keratinocytes (
      • Nickoloff B.J.
      • Naidu Y.
      Perturbation of epidermal barrier function correlates with initiation of cytokine cascade in human skin.
      ). IL1β, which is strongly expressed in the skin of Spink5−/- mice (
      • Briot A.
      • Deraison C.
      • Lacroix M.
      • et al.
      Kallikrein 5 induces atopic dermatitis-like lesions through PAR2-mediated thymic stromal lymphopoietin expression in Netherton syndrome.
      ), could then be processed and activated by unrestricted KLK7 (
      • Nylander-Lundqvist E.
      • Egelrud T.
      Formation of active IL-1 beta from pro-IL-1 beta catalyzed by stratum corneum chymotryptic enzyme in vitro.
      ) or caspase 1 (
      • Hosomi N.
      • Fukai K.
      • Nakanishi T.
      • et al.
      Caspase-1 activity of stratum corneum and serum interleukin-18 level are increased in patients with Netherton syndrome.
      ), and participated in the recruitment of innate immune cells.
      Although IL1β is not considered as a Th2 cytokine, it is not excluded that residual production of TSLP and TNFα in the DKO epidermis could be sufficient to switch inflammation toward an AD-like phenotype. Independently of the PAR2 pathway, residual TSLP expression could be due to other unidentified mechanisms or secondary to IL1β and TNFα production, which have been reported to induce its expression (
      • Lee H.C.
      • Ziegler S.F.
      Inducible expression of the proallergic cytokine thymic stromal lymphopoietin in airway epithelial cells is controlled by NfkappaB.
      ). In our grafted skin model, the recruitment of polynuclear eosinophilic cells and mast cells could be first a consequence of IL1β production and residual TNFα and TSLP. Together, these molecules are known to induce the secretion of Th2 cytokines by activated mast cells (
      • Allakhverdi Z.
      • Comeau M.R.
      • Jessup H.K.
      • et al.
      Thymic stromal lymphopoietin is released by human epithelial cells in response to microbes, trauma, or inflammation and potently activates mast cells.
      ).
      Interestingly, these innate immune cells could arise not only from the donor skin but also from host nude mice bearing the functional PAR2 receptor. Indeed, our skin graft model is somehow different from Par2−/- mice, in which Par2 is inactivated in the epidermal, dermal, and immune compartments. Par2−/- mice exhibit significant reduction of immune cell infiltration after experimental induced-allergic dermatitis compared with WT mice (
      • Kawagoe J.
      • Takizawa T.
      • Matsumoto J.
      • et al.
      Effect of protease-activated receptor-2 deficiency on allergic dermatitis in the mouse ear.
      ;
      • Seeliger S.
      • Derian C.K.
      • Vergnolle N.
      • et al.
      Proinflammatory role of proteinase-activated receptor-2 in humans and mice during cutaneous inflammation in vivo.
      ). Thus, with time, phenotype aggravation of the DKO-grafted skin could be due to PAR2 activation at the surface of cells from the host, such as mast cells and eosinophilic cells (
      • Shpacovitch V.
      • Feld M.
      • Hollenberg M.D.
      • et al.
      Role of protease-activated receptors in inflammatory responses, innate and adaptive immunity.
      ).
      Six weeks after grafting, we observed an aggravated cutaneous inflammation probably resulting from the “vicious circle” promoted and maintained by hyperactivated proteases, mechanical stress, skin barrier impairment, and activation of innate immune cells, which also secrete cytokines and proteases. In an immune competent model of DKO, one can argue that the secretion of the proallergic chemokines, macrophage-derived chemokine and thymus and activation-regulated chemokine, overexpressed in NS patient keratinocytes and the Spink5−/- epidermis independently of the KLK5–PAR2 pathway (
      • Briot A.
      • Deraison C.
      • Lacroix M.
      • et al.
      Kallikrein 5 induces atopic dermatitis-like lesions through PAR2-mediated thymic stromal lymphopoietin expression in Netherton syndrome.
      ), together with residual expression of TSLP could still promote a Th2 response in the adult skin.
      To date, there is no evident point demonstrating that either intrinsic or extrinsic signals are promoting cutaneous allergic-type inflammation in AD. The fact that only 37–50% of patients with Ichthyosis vulgaris develop atopic manifestations (
      • Kuokkanen K.
      Ichthyosis vulgaris. A clinical and histopathological study of patients and their close relatives in the autosomal dominant and sex-linked forms of the disease.
      ;
      • Smith F.J.
      • Irvine A.D.
      • Terron-Kwiatkowski A.
      • et al.
      Loss-of-function mutations in the gene encoding filaggrin cause ichthyosis vulgaris.
      ), supports the notion that in NS in which all patients develop allergic manifestations, additional mechanisms to primary skin barrier defect have a key role in immune response polarization toward a Th2 response. This study highlights the fact that the KLK5–PAR2–TSLP pathway, already activated in utero in Spink5−/- mice, is an early inducer of proinflammatory and proallergic signals seen in our models (
      • Briot A.
      • Deraison C.
      • Lacroix M.
      • et al.
      Kallikrein 5 induces atopic dermatitis-like lesions through PAR2-mediated thymic stromal lymphopoietin expression in Netherton syndrome.
      ). This observation could be relevant to the development of “intrinsic atopy” before sensitization to environmental allergens. However, skin integrity is essential to protect our body against allergens that could trigger “extrinsic atopy” when the epidermal barrier is defective.
      This study also provides strong evidence that several mechanisms trigger skin inflammation in NS models in vivo. These include skin barrier defect and keratinocyte mechanical stress induced by stratum corneum detachment as well as protease-induced proinflammatory and proallergic pathways, including KLK5-PAR2-mediated overexpression of TSLP (
      • Briot A.
      • Deraison C.
      • Lacroix M.
      • et al.
      Kallikrein 5 induces atopic dermatitis-like lesions through PAR2-mediated thymic stromal lymphopoietin expression in Netherton syndrome.
      ) and the consequences of other hyperactivated proteases (
      • Hansson L.
      • Backman A.
      • Ny A.
      • et al.
      Epidermal overexpression of stratum corneum chymotryptic enzyme in mice: a model for chronic itchy dermatitis.
      ;
      • Descargues P.
      • Deraison C.
      • Bonnart C.
      • et al.
      Spink5-deficient mice mimic Netherton syndrome through degradation of desmoglein 1 by epidermal protease hyperactivity.
      ;
      • Bonnart C.
      • Deraison C.
      • Lacroix M.
      • et al.
      Elastase 2 is expressed in human and mouse epidermis and impairs skin barrier function in Netherton syndrome through filaggrin and lipid misprocessing.
      ). Taken together, unregulated biological cascades lead to a severe clinical phenotype in NS. These specific pathophysiological mechanisms triggering atopic manifestation in NS may serve as a model for more complex unknown biological cascades involved in the more common disease, AD.
      Therapeutic developments for NS should consider multiple sides of the disease. Reducing inflammation through TSLP repression together with restoring skin barrier function through inhibition of unrestricted protease activities would probably be a promising approach to significantly improve the clinical phenotype.

      Materials and Methods

      Mouse housing conditions and skin grafting

      This study was approved by the Commission de Génie Génétique (23 November 2003, agreement number 3987) and by the Local Ethical Committee (DEC 04005). All experiments were conducted in accordance with the relevant guidelines and regulations. Mice embryos were delivered by Caesarean section at E19.5.
      Total dorsal skin from E19.5 embryos was transplanted onto athymic immunodeficient nude mice using the skin-flap technique (
      • Barrandon Y.
      • Li V.
      • Green H.
      New techniques for the grafting of cultured human epidermal cells onto athymic animals.
      ) and analyzed after 6 weeks.
      Gestate, newborn, and host grafted mice were maintained under specific pathogen free housing conditions during every experiment.

      Histological and immunohistochemical analysis

      Mouse skin samples from E19.5 embryos or skin grafts were fixed in 10% formalin and embedded in paraffin. Sagittal sections of 4μm were stained with hematoxylin–eosin for histological examination or toluidin blue for mast cell counting. Luna staining was performed to reveal polynuclear eosinophilic cells. For immunohistochemistry experiments, antigen retrieval of paraffin-embedded cross-sections were performed in citrate buffer (1mM, pH6) boiled for 20minutes. TSLP and ICAM (R&D Systems, Minneapolis, MN) primary antibodies were used and immunodetection was performed using appropriate Horseradish peroxidase (diaminobenzidine)-conjugated polymers (EnVision System, Dakocytomation, Glostrup, Denmark).

      RNA extraction, reverse transcription, and quantitative real-time PCR

      Total RNA were extracted from the E19.5 skin using RNeasy mini kit (Qiagen, Courtaboeuf, France) and cDNA synthesis from 2 to 3μg of total RNA was performed using the SuperScript III First-Strand Synthesis System (Invitrogen, Carlsbad, CA). Quantitative real-time PCR was realized on ABI prism 7000 Sequence Detection System (Applied Biosystems, Carlsbad, CA) using the qPCR MasterMix Plus SYBR Green I kit (Eurogentec, Angers, France). Results were normalized with the Hprt gene and analyzed using Sequence Detection System version 1.2 (Applied Biosystems). The following primers were used: Tslp forward 5′-CGACAGCATGGTTCTTCTCA-3′ and reverse 5′-CGATTTGCTCGAACTTAGCC-3′; Icam forward 5′-AGGGCTGGCATTGTTCTCTA-3′ and reverse 5′-CTTCAGAGGCAGGAAACAGG-3′; Hprt forward 5′-CTGGTTAAGCAGTACAGCCCCAA-3′ and reverse 5′-CGAGAGGTCCTTTTCACCAGC-3′. P-values were calculated using the Student–Fischer test.

      ACKNOWLEDGMENTS

      We are grateful to Talal Al Saati, Florence Capilla, and Delphine Lestrade from the experimental histopathology platform of IFR30 for technical assistance and the staff of mouse facilities for animal care (INSERM, Institut Claude de Préval, IFR30, Histopathologie Expérimentale, Toulouse, F-31300 France). We thank Fabienne Tocque for experimental assistance. This work was supported by grants from the National Agency for Research (Programme GENOPAT, NS2AD: ANR-08-GENO-033), the Patient Association “Athina Ichtyoses Monaco,” Weston-Haven Foundation (MS), the French Ministry of Research and Technology, and the Foundation for Medical Research (FRM: DAL20051205066).

      REFERENCES

        • Allakhverdi Z.
        • Comeau M.R.
        • Jessup H.K.
        • et al.
        Thymic stromal lymphopoietin is released by human epithelial cells in response to microbes, trauma, or inflammation and potently activates mast cells.
        J Exp Med. 2007; 204: 253-258
        • Barrandon Y.
        • Li V.
        • Green H.
        New techniques for the grafting of cultured human epidermal cells onto athymic animals.
        J Invest Dermatol. 1988; 91: 315-318
        • Bitoun E.
        • Micheloni A.
        • Lamant L.
        • et al.
        LEKTI proteolytic processing in human primary keratinocytes, tissue distribution and defective expression in Netherton syndrome.
        Hum Mol Genet. 2003; 12: 2417-2430
        • Bonnart C.
        • Deraison C.
        • Lacroix M.
        • et al.
        Elastase 2 is expressed in human and mouse epidermis and impairs skin barrier function in Netherton syndrome through filaggrin and lipid misprocessing.
        J Clin Invest. 2010; 120: 871-882
        • Briot A.
        • Deraison C.
        • Lacroix M.
        • et al.
        Kallikrein 5 induces atopic dermatitis-like lesions through PAR2-mediated thymic stromal lymphopoietin expression in Netherton syndrome.
        J Exp Med. 2009; 206: 1135-1147
        • Buddenkotte J.
        • Stroh C.
        • Engels I.H.
        • et al.
        Agonists of proteinase-activated receptor-2 stimulate upregulation of intercellular cell adhesion molecule-1 in primary human keratinocytes via activation of NF-kappa B.
        J Invest Dermatol. 2005; 124: 38-45
        • Chavanas S.
        • Bodemer C.
        • Rochat A.
        • et al.
        Mutations in SPINK5, encoding a serine protease inhibitor, cause Netherton syndrome.
        Nat Genet. 2000; 25: 141-142
        • Comel M.
        Ichtyosis Linearis circumflexa.
        Dermatologica. 1949; 98: 133-136
        • Deraison C.
        • Bonnart C.
        • Lopez F.
        • et al.
        LEKTI fragments specifically inhibit KLK5, KLK7, and KLK14 and control desquamation through a pH-dependent interaction.
        Mol Biol Cell. 2007; 18: 3607-3619
        • Descargues P.
        • Deraison C.
        • Bonnart C.
        • et al.
        Spink5-deficient mice mimic Netherton syndrome through degradation of desmoglein 1 by epidermal protease hyperactivity.
        Nat Genet. 2005; 37: 56-65
        • Egelrud T.
        • Brattsand M.
        • Kreutzmann P.
        • et al.
        hK5 and hK7, two serine proteinases abundant in human skin, are inhibited by LEKTI domain 6.
        Br J Dermatol. 2005; 153: 1200-1203
        • Fallon P.G.
        • Sasaki T.
        • Sandilands A.
        • et al.
        A homozygous frameshift mutation in the mouse Flg gene facilitates enhanced percutaneous allergen priming.
        Nat Genet. 2009; 41: 602-608
        • Hachem J.P.
        • Houben E.
        • Crumrine D.
        • et al.
        Serine protease signaling of epidermal permeability barrier homeostasis.
        J Invest Dermatol. 2006; 126: 2074-2086
        • Hansson L.
        • Backman 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: 444-449
        • Hewett D.R.
        • Simons A.L.
        • Mangan N.E.
        • et al.
        Lethal, neonatal ichthyosis with increased proteolytic processing of filaggrin in a mouse model of Netherton syndrome.
        Hum Mol Genet. 2005; 14: 335-346
        • Hosomi N.
        • Fukai K.
        • Nakanishi T.
        • et al.
        Caspase-1 activity of stratum corneum and serum interleukin-18 level are increased in patients with Netherton syndrome.
        Br J Dermatol. 2008; 159: 744-746
        • Judge M.R.
        • Morgan G.
        • Harper J.I.
        A clinical and immunological study of Netherton's syndrome.
        Br J Dermatol. 1994; 131: 615-621
        • Kawagoe J.
        • Takizawa T.
        • Matsumoto J.
        • et al.
        Effect of protease-activated receptor-2 deficiency on allergic dermatitis in the mouse ear.
        Jpn J Pharmacol. 2002; 88: 77-84
        • Kuokkanen K.
        Ichthyosis vulgaris. A clinical and histopathological study of patients and their close relatives in the autosomal dominant and sex-linked forms of the disease.
        Acta Derm Venereol Suppl (Stockh). 1969; 62: 1-72
        • Lee H.C.
        • Ziegler S.F.
        Inducible expression of the proallergic cytokine thymic stromal lymphopoietin in airway epithelial cells is controlled by NfkappaB.
        Proc Natl Acad Sci USA. 2007; 104: 914-919
        • Lindner J.R.
        • Kahn M.L.
        • Coughlin S.R.
        • et al.
        Delayed onset of inflammation in protease-activated receptor-2-deficient mice.
        J Immunol. 2000; 165: 6504-6510
        • Netherton E.W.
        A unique case of trichorrhexis nodosa; bamboo hairs.
        AMA Arch Derm. 1958; 78: 483-487
        • Nickoloff B.J.
        • Naidu Y.
        Perturbation of epidermal barrier function correlates with initiation of cytokine cascade in human skin.
        J Am Acad Dermatol. 1994; 30: 535-546
        • Nylander-Lundqvist E.
        • Egelrud T.
        Formation of active IL-1 beta from pro-IL-1 beta catalyzed by stratum corneum chymotryptic enzyme in vitro.
        Acta Derm Venereol. 1997; 77: 203-206
        • O’Regan G.M.
        • Sandilands A.
        • McLean W.H.
        • et al.
        Filaggrin in atopic dermatitis.
        J Allergy Clin Immunol. 2008; 122: 689-693
        • Oikonomopoulou K.
        • Hansen K.K.
        • Saifeddine M.
        • et al.
        Kallikrein-mediated cell signalling: targeting proteinase-activated receptors (PARs).
        Biol Chem. 2006; 387: 817-824
        • Schechter N.M.
        • Choi E.J.
        • Wang Z.M.
        • et al.
        Inhibition of human kallikreins 5 and 7 by the serine protease inhibitor lympho-epithelial Kazal-type inhibitor (LEKTI).
        Biol Chem. 2005; 386: 1173-1184
        • Shpacovitch V.
        • Feld M.
        • Hollenberg M.D.
        • et al.
        Role of protease-activated receptors in inflammatory responses, innate and adaptive immunity.
        J Leukoc Biol. 2008; 83: 1309-1322
        • Seeliger S.
        • Derian C.K.
        • Vergnolle N.
        • et al.
        Proinflammatory role of proteinase-activated receptor-2 in humans and mice during cutaneous inflammation in vivo.
        FASEB J. 2003; 17: 1871-1885
        • Smith F.J.
        • Irvine A.D.
        • Terron-Kwiatkowski A.
        • et al.
        Loss-of-function mutations in the gene encoding filaggrin cause ichthyosis vulgaris.
        Nat Genet. 2006; 38: 337-342
        • Soumelis V.
        • Reche P.A.
        • Kanzler H.
        • et al.
        Human epithelial cells trigger dendritic cell mediated allergic inflammation by producing TSLP.
        Nat Immunol. 2002; 3: 673-680
        • Traupe H.
        The Ichthyosis. A Guide to Clinical Diagnosis, Genetic Counselling, and Therapy. Springer-Verlag Ed, Berlin Heidelberg1989
        • Wood L.C.
        • Jackson S.M.
        • Elias P.M.
        • et al.
        Cutaneous barrier perturbation stimulates cytokine production in the epidermis of mice.
        J Clin Invest. 1992; 90: 482-487
        • Yang T.
        • Liang D.
        • Koch P.J.
        • et al.
        Epidermal detachment, desmosomal dissociation, and destabilization of corneodesmosin in Spink5-/- mice.
        Genes Dev. 2004; 18: 2354-2358
        • Yoo J.
        • Omori M.
        • Gyarmati D.
        • et al.
        Spontaneous atopic dermatitis in mice expressing an inducible thymic stromal lymphopoietin transgene specifically in the skin.
        J Exp Med. 2005; 202: 541-549