Skip Navigation


Human Molecular Genetics Advance Access originally published online on December 8, 2004
Human Molecular Genetics 2005 14(2):335-346; doi:10.1093/hmg/ddi030
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
14/2/335    most recent
ddi030v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (19)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Hewett, D. R.
Right arrow Articles by McKenzie, A. N.J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hewett, D. R.
Right arrow Articles by McKenzie, A. N.J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Human Molecular Genetics, Vol. 14, No. 2 © Oxford University Press 2005; all rights reserved

Lethal, neonatal ichthyosis with increased proteolytic processing of filaggrin in a mouse model of Netherton syndrome

Duncan R. Hewett1,*, Alison L. Simons1, Niamh E. Mangan2, Helen E. Jolin1, Shelia M. Green1, Padraic G. Fallon2 and Andrew N.J. McKenzie1

1MRC Laboratory of Molecular Biology, Hills Road, Cambridge CB2 2QH, UK and 2Department of Biochemistry, Trinity College, Dublin 2, Ireland

* To whom correspondence should be addressed. Tel: + 44 1223402350; Fax: +44 1223412178; Email: drhewett{at}mrc-lmb.cam.ac.uk

Received September 16, 2004; Accepted November 25, 2004


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Netherton syndrome is an autosomal recessive multisystemic disorder characterized by congenital ichthyosiform erythroderma, hair shaft defects and atopy, caused by mutations within the human SPINK5 gene. To investigate the development of this disease, we have cloned mouse spink5 and created mice with a mutated premature stop codon at amino acid R820X, to produce an allele that closely mimics a point mutation (E827X) in human SPINK5. Newborn spink5R820X/R820X mice develop a lethal, severe ichthyosis with a loss of skin barrier function and dehydration, resulting in death within a few hours of birth, similar to that observed in patients with severe Netherton syndrome. Epidermal barrier function is compromised because of the stratum corneum becoming spontaneously detached in the newborn mice, and this is probably compounded by the reduced mechanical strength detected in the cornified envelopes. Biochemical analysis of skin from newborn wild-type and spink5R820X/R820X mice revealed a substantial increase in the proteolytic processing of profilaggrin into its constituent filaggrin monomers. Filaggrin functions to organize keratin filaments into highly ordered macrofibrils that crisscross the cornified cells of the stratum corneum imparting structural integrity, and defects in filaggrin processing occur in a number of forms of congenital ichthyosis. These data suggest that in the absence of the serine protease inhibitor spink5, there is an abnormal increase in the processing of profilaggrin, resulting in an overabundance of filaggrin monomers, and that this may play a direct role in the observed deficit in the adhesion of the stratum corneum and the severely compromised epidermal barrier function.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Netherton syndrome (MIM 256500) is an autosomal recessive multisystemic disorder characterized by congenital ichthyosiform erythroderma, hair shaft defects and atopy (1Go). The syndrome classically presents at birth with a generalized erythema, together with scaling and peeling of the skin. Many newborns develop hypernatraemic dehydration, failure to thrive, recurrent infections and difficulties in thermoregulation (2Go,3Go). The generalized exfoliative erythroderma may persist throughout a patient's life, or it can evolve into a distinctive ichthyosis linearis circumflexa. The incidence of syndrome has been cited as ranging from one in 200 000 to one in 50 000 live births (4Go,5Go), although it could account for ~20% of congenital erythrodermas (6Go). Most patients later display characteristic hair shaft abnormalities, particularly trichorrhexis invaginata (bamboo hair) and torsion twists (pili torti). Netherton syndrome patients also suffer from a high incidence of atopic manifestations, including elevated IgE levels, eosinophilia, atopic dermatitis, hay fever and food allergies. No specific treatment is available for the disorder.

Mutations in the gene for the serine protease inhibitor SPINK5 were found to cause Netherton syndrome (7Go). The majority of mutations cause premature stop codons and reduced or null expression from the mutated allele. A marked reduction in SPINK5 RNA levels has, indeed, been observed in patients (7Go), suggestive of nonsense-mediated decay of the transcripts containing premature stop codons. Furthermore, SPINK5 protein cannot be detected in cultured keratinocytes from patients (4Go,8Go).

SPINK5 may also play a role in protection against more common cutaneous and atopic diseases. Genetic associations have been found between SPINK5 polymorphisms and atopy (9Go), asthma (9Go,10Go) and atopic dermatitis (9Go,11Go,12Go). The majority of these studies concern a Glu420-Lys variant that introduces an extra basic residue into a probable linker region between inhibitory domains 6 and 7 of SPINK5 and raises the possibility of differential cleavage of this linker region. Microarray analysis has also revealed a significant (~7-fold) reduction in SPINK5 expression levels in squamous cell carcinomas (SCC) compared with non-malignant tissue (13Go). Moreover, SPINK5 transcripts were undetectable in all established SCC cell lines tested (13Go).

SPINK5 was originally identified as a lympho-epithelial Kazal-type-related inhibitor and was recognized as the common precursor protein for two inhibitory peptides isolated from human blood filtrate, one of which had demonstrated an inhibitory effect on trypsin in vitro (14Go). The SPINK5 gene encodes a protein with 15 potential serine protease inhibitory domains. Two of these domains have the conserved pattern of six cysteine residues that are characteristic of Kazal-type serine protease inhibitors (15Go). The remaining 13 domains have a related pattern of four conserved cysteines and were termed non-Kazal-type inhibitory domains. High SPINK5 expression was originally noted in the thymus, oral mucosa, Bartholin's glands and vaginal epithelium. This led to the speculation of its involvement in anti-inflammatory or antimicrobial protection of mucoid epithelia and T-cell development (14Go). The expression in the thymus is limited to the Hassall's corpuscles, the terminally differentiated part of the thymic epithelium (8Go). SPINK5 is also expressed in the upper-spinous and granular layers of the epidermis as well as the hair follicle and sebaceous gland and cultured epidermal keratinocytes (2Go,7Go,8Go).

SPINK5 appears to undergo extensive proteolytic processing of the multidomain precursor. Individual domains one, five and six have been isolated from hemofiltrate (14Go,16Go). A peptide with an N-terminal sequence identical to that of domain 8 has been isolated from human epidermal keratinocytes (17Go). The exact extent of the processing of SPINK5 remains to be defined, as does the biological activity of the released domains. Furin, a subtilisin-like proprotein convertase expressed in the epidermis, may be involved in this processing. In vitro cleavage of recombinant SPINK5 into 13 peptides by furin has been reported (18Go), although less intensive SPINK5 processing by the same enzyme has also been described (8Go).

Trypsin-inhibiting activity by SPINK5 has been described for a number of native and recombinant domains (14Go,16Go,19Go). Full-length recombinant SPINK5 was also capable of inhibiting the serine proteases plasmin, subtilisin A, cathepsin G, human neutrophil elastase and trypsin (18Go). Whether any of these enzymes are the physiological targets for inhibition by SPINK5 is unknown. Indeed, given the wide tissue distribution pattern of SPINK5 and the pleiotropic nature of Netherton syndrome, from ichthyosis to atopic disorders, a number of proteases have been proposed as targets for inhibition by SPINK5. Such speculation has included: stratum corneum trypsin-like enzyme, stratum corneum chymotrypsin-like enzyme, mast cell tryptase, granzymes, kallikreins, major house dust mite allergens and activators of proteinase-activated receptor-2 (2Go,8Go,16Go,20Go). However, to date, in patients with Netherton syndrome, the only changes in enzyme activity that have been reported were observed in the skin and represent an increase in trypsin-like activity (2Go) and a more widespread transglutaminase 1 activity (21Go).

We set out to characterize the murine orthologue of SPINK5 and produce a mouse model of Netherton syndrome. The spectrum of mutations seen in human patients suggested that a complete knockout or null allele of SPINK5 may be incompatible with mouse survival. Three patients found to be homozygous for the 153delT mutation, which introduces a premature termination codon (PTC) into exon 3 of SPINK5, died within the first 3 months of birth, despite extensive treatment in intensive care (22Go). We have therefore used gene targeting to generate mice homozygous for a PTC in exon 26 of the gene in a region encoding the linker region preceding the truncated 13th inhibitory domain. The position of the PTC in these mice is very similar to that of a recurrent homozygous PTC mutation seen in some patients with Netherton syndrome: an E827X mutation in exon26 [referred to as 2468insA in Chavanas et al. (7Go)]. Significantly, spink5R820X/R820X mice developed symptoms comparable with those observed in newborn Netherton syndrome patients, with neonatal ichthyosis, spontaneous shedding of the stratum corneum, impaired epidermal barrier function and death. We have also shown that affected mice displayed a dramatic increase in the proteolytic processing of filaggrin, a major component of the transitional cell layer. This is the first implication of a role for the serine protease inhibitor spink5 in the proteolytic cascade controlling filaggrin processing.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Cloning and gene targeting of the mouse spink5 gene
The mouse orthologue of human SPINK5 was cloned by a combination of database searching and RT–PCR. The mouse gene has 32 exons and encodes a protein of 1017 amino acids with 49% homology to the human SPINK5. Like human SPINK5, the mouse spink5 protein is predicted to be a serine protease inhibitor proprotein containing multiple putative inhibitory domains. It contains two classical Kazal-type domains and 12 non-Kazal-type domains. The main difference from the human protein is the presence of a truncated 13th non-Kazal-type domain in the mouse spink5 polypeptide (Fig. 1A and B). The mouse orthologue has since been independently cloned by other researchers (GenBank accession no. XP_283487, RIKEN cDNA clone 2310065D10).



View larger version (42K):
[in this window]
[in a new window]
  		Figure 1. Characterization of mouse spink5. (A) Schematic comparison of mouse spink5 with human SPINK5 proteins. The putative protease inhibitor domains are represented by the 15 barrels, which are separated from each other by short linker sequences. The shaded barrels are the two Kazal-type domains. The position of the PTC introduced into the mouse gene using gene targeting is shown in relation to an analagous human mutation. (B) The predicted amino acid sequence of mouse spink5. The sequence has been arranged to more easily display the predicted protease inhibitor domains and the short N-terminal signal peptide. The conserved cysteines of the inhibitor domains are emboldened. An asterisk indicates the position of the R820X mutation. (C) Tissue distribution of spink5. Real time PCR was used to determine the relative levels of spink5 RNA in total RNA extracted from the indicated tissues. The levels in lung and thymus are indicated by the written values. Levels in heart and kidney were below detection limits.

 
Using Taqman PCR, we determined that the highest levels of spink5 transcription were detected in the oral mucosa and oesophagus (Fig. 1C). Significant expression was also detectable in the lung and skin, with lower levels of transcription present in the thymus and liver (Fig. 1C). We were unable to detect spink5 transcripts in heart or kidney (Fig. 1C). It is noteworthy that this tissue expression pattern of spink5 mRNA in mice closely resembles that reported for human SPINK5 (14Go,16Go).

To investigate the functional importance of mouse spink5 and assess whether gene targeting could be used to generate a mouse model of Netherton syndrome, we produced a replacement gene targeting construct designed to disrupt exon 26 of the mouse spink5 gene and introduce a premature stop codon, R820X mutation, into the part of the gene encoding the linker region separating the 12th inhibitory domain from the truncated 13th inhibitory domain (Figs 1A and 2A). The location of this premature stop codon was selected to closely emulate that of a recurrent homozygous premature stop codon mutation reported in some patients with Netherton syndrome: E827X [described as 2468insA mutation in Chavanas et al. (7Go)] (Fig. 1A). The R820X mutation was introduced into themouse spink5 gene following homologous recombination in embryonic stem (ES) cells. The correctly targeted allelecontained a novel BamHI site for use in screening for recombination events, and a loxP-flanked neomycin resistance gene cassette immediately downstream of this premature stop codon (Fig. 2A). Chimeric mice were generated from correctly targeted ES cells, and germline transmission from these chimeras was detected by screening for the 4.8 kb BamHI fragment at the targeted allele (Fig. 2B).



View larger version (28K):
[in this window]
[in a new window]
  		Figure 2. Disruption of the murine spink5 gene. (A) The structure of the spink5 gene and the targeting vector and the predicted homologous recombination event. B, BamHI; N, NotI; neo, neomycin resistance cassette; DT, diptheria toxin cassette; filled arrow heads, loxP sites; STOP, introduced PTC. (B) Southern blot of BamHI-digested DNA from F2 genomic DNA from newborn mice. Probe A detects a 9 kb fragment at the wild-type allele and a 4.8 kb fragment at the targeted allele. (C) Reduced spink5 transcript levels in spink5R820X/R820X mice. Levels of spink5 RNA detectable by real time PCR assays to two different regions of the mouse spink5 RNA. Assayed using real time PCR analysis and normalized to hprt transcript levels.

 
Heterozygous animals developed normally with no apparent phenotypic abnormalities. Cohorts of heterozygous mice were kept for up to 12 months and showed no predisposition to ill health compared with age and sex matched wild-type controls. Specifically, the mice showed no outward cutaneous phenotype or increased tendency to scratch themselves. No differences were seen in the levels of circulating immunoglobulins, including IgE. The percentages (and total counts) of CD4/CD8 single positive and double positive thymocytes as determined by FACS analysis were comparable (data not shown).

spink5R820X/R820X Mice develop a neonatal, lethal ichthyosis
Interbreeding of multiple spink5+/R820X heterozygous animals failed to produce any spink5R820X/R820X pups of weaning age. Therefore, timed matings were established and spink5R820X/R820X embryos of normal appearance were detected at days 16 and 19 of gestation. Further analysis revealed that spink5R820X/R820X mice were born alive but died soon after birth. Real time PCR analysis with primer sets to different regions of the spink5 coding sequence revealed a substantial reduction in the levels of spink5 transcript in total skin from spink5R820X/R820X mice (Fig. 2C).

The presence of milk spots in the newborn spink5R820X/R820X pups indicated that they were capable of feeding, but despite this ability, the pups failed to thrive and expired within 1–2 h after birth. Although the spink5R820X/R820X pups appeared phenotypically normal immediately after birth [histological examination of the neonates failed to show any obvious abnormality of any major organs (data not shown)], within 30–60 min of birth they developed a dry and wrinkled appearance that was accompanied with a generalized shedding of the outer layer of the skin and the development of a red, shiny appearance (Fig. 3A). Significantly, infants with severe congenital Netherton syndrome develop a severe, generalized exfoliative erythroderma and hypernatremic dehydration that may prove fatal (1Go,22Go). Indeed, the visual appearance of spink5R820X/R820X pups (Fig. 3A) is similar to the cutaneous phenotype of congenital Netherton syndrome (2Go,3Go). The whiskers of the neonatal spink5R820X/R820X miceshow no apparent hair shaft abnormalities compared with those of wild-type mice (Fig. 3D). This is not altogether unexpected as the characteristic hair defects of Netherton syndrome are often not observed in newborn children (1Go). Thespink5R820X/R820X phenotype was unaltered following the removal of the neomycin gene cassette by intercrossing of spink5+/R820X mice with mice transgenic for the Cre recombinase.



View larger version (45K):
[in this window]
[in a new window]
  		Figure 3. Gross morphology of spink5R820X/R820X neonates. (A) Newborn spink5R820X/R820X pups rapidly develop a red, shiny appearance with apparent shedding of the outer layers of the skin. Note the progressive development of the skin condition, from shedding at limbs and ventral side (left-most spink5R820X/R820X mouse), through generalized shedding (middle spink5R820X/R820X mouse), to severe erythroderma and complete loss of outer layer of skin (right-most spink5R820X/R820X mouse). (B) Abnormality in skin barrier function in newborn spink5R820X/R820X mice. Toluidine blue dye penetration assay performed on wild-type and spink5R820X/R820X embryos mice at different stages of development. (C) Haematoxylin and eosin-stained sections of skin showing complete detachment of the stratum corneum (SC) in spink5R820X/R820X newborn mice. (D) Shafts of whiskers taken from spink5R820X/R820X and wild-type littermates.

 
Epidermal barrier function is impaired in spink5R820X/R820X mice
Given that the spink5R820X/R820X mice developed a generalized exfoliative erythroderma immediately after birth, we wished to determine whether this resulted in impaired epidermal barrier function or whether barrier function was compromised before birth. A functional skin epidermal barrier is normally fully developed by day E17 of mouse gestation (23Go), and to test whether the absence of spink5 altered the onset of epidermal barrier function, we determined the ability of spink5R820X/R820X and wild-type mice to prevent the penetration of external solutions before and after birth. The penetration of the dye toluidine blue was used to assess barrier function in both spink5R820X/R820X and wild-type littermates. As reported previously (23Go), the epidermis of wild-type mice was permeable to dye at day E16 of gestation, becoming impermeable by day E19 of gestation, with dye exclusion continuing after birth (Fig. 3B). The spink5R820X/R820X mice showed comparable dye permeability at day E16 of gestation and impermeability to dye by day E19 of gestation as wild-type mice (Fig. 3B), indicating normal embryonic development of skin epidermal barrier function. However, newborn spink5R820X/R820X mice showed extensive penetration of toluidine blue and the extent of dye penetration (which varied from limbs and ventral side only to total skin) correlated with skin shedding (Fig. 3B). We also quantified transepidermal water loss (TEWL) through the skin as a measure of skin barrier function. We found no difference in TEWL between wild-type and spink5R820X/R820X mice at day E19 of gestation (mean TEWL for spink5R820X/R820X mice, 10.2 g/hm2; mean TEWL for wild-type littermates, 9.3 g/hm2). These results demonstrate that although epidermal barrier function develops normally in the spink5R820X/R820X mice through gestation, there is a loss of epidermal barrier function immediately after birth, and this impairment is coincident with exfoliative erythroderma. This breakdown in epidermal barrier function leads to rapid dehydration of the newborn spink5R820X/R820X mice resulting in neonatal lethality.

Histopathology demonstrated the dramatic loss of adhesion at the transitional layer between the stratum corneum and the underlying granular layer of the epidermis of the spink5R820X/R820X mice when compared with the wild-type mice (Figs 3C and 4A). With the exception of some mild hyperkeratosis in spink5–/– epidermis, there were no further differences between the appearances of the epidermal layers from spink5R820X/R820X and wild-type mice. Furthermore, immunohistochemistry revealed no major differences in either the amounts or the distribution of the basal cell-specific keratin 14 (Fig. 4B), the suprabasal cell-specific keratin 1 (Fig. 4C) or the stratum corneum protein, loricrin (Fig. 4D). No epidermal expression of keratin 6 was detectable in either wild-type or spink5R820X/R820X mice (data not shown). Keratin 6 is not normally expressed in epithelial cells in the absence of hyperproliferation/wound healing (with the exception of the outer root sheath of hair follicles). We noted that staining of filaggrin, a normally abundant protein of the granular layer, was more diffuse in the spink5R820X/R820X and lacked the strict zonal distribution apparent in the wild-type sections (Fig. 4E). Thus, in the absence of spink5, there is a structural impairment that fails to maintain the integrity of the stratum corneum.



View larger version (74K):
[in this window]
[in a new window]
  		Figure 4. Histopathology of skin from neonatal spink5R820X/R820X mice. (A) Haematoxylin and eosin-stained sections of skin from whole mounted neonatal mice. The arrowhead indicates the detachment of the stratum corneum (B–E). Immunohistochemistry of skin sections from neonatal mice with antibodies to keratin 14 (B), keratin 1 (C), loricrin (D) and filaggrin (E).

 
Cornified envelopes of spink5R820X/R820X mice have reduced mechanical strength
In the stratum corneum of stratified squamous epithelia, a tough, insoluble layer termed the cornified envelope (CE) is assembled at the inner cell surface of the corneocyte. The CE is a complex mixture of proteins and lipids, cross-linked by characteristic N{varepsilon}-({gamma}-glutamyl) lysine isopeptide bonds formed by transglutaminases. A properly assembled CE surrounded by an ordered lipid lamellae is essential for effective barrier function (reviewed in 24Go). CEs can be purified as the insoluble material that remains after exhaustive boiling of excised skin in the presence of detergent and reducing agent (25Go–27Go). The CEs isolated from spink5R820X/R820X mice and wild-type littermates were typically smooth and polygonal in appearance and showed no significant difference in size or shape (Fig. 5). Despite this morphological similarity, the CEs demonstrated remarkably different mechanical strengths, as measured by sensitivity to fracture by sonication. After 30 s of sonication, ~90% of the CEs from spink5R820X/R820X neonates displayed a broken or ruptured appearance compared with no detectable fragility in the CEs isolated from newborn wild-type mice at this time point (Fig. 5). These results indicate that the absence of spink5 also impairs thenormal structural composition of the CEs making them more susceptible to mechanical damage, which in turn would compromise the functional integrity of the epidermal barrier.



View larger version (36K):
[in this window]
[in a new window]
  		Figure 5. Reduced CE strength in spink5R820X/R820X mice. Phase contrast images of CEs purified from day 19 spink5R820X/R820X and wild-type mice. Differential sensitivity of CEs to sonication is shown.

 
spink5R820X/R820X mice display increased filaggrin processing
As spink5 is a serine protease inhibitor, we predicted that the targeting of spink5 would disturb the balance of protease activities in the epidermis and possibly lead to an increase inthe proteolytic processing/degradation/activation of proteins in this tissue. To identify the potential differences in the protein profiles expressed in the skin of newborn spink5R820X/R820X and wild-type mice, proteins were extracted from total dorsal skin of these animals and separated by SDS–PAGE. Strikingly, a band of just under 30 kDa was consistently observed to be more intense in extracts from spink5R820X/R820X mice compared with either wild-type or heterozygous littermates (Fig. 6A). To identify the ~30 kDa protein, we excised the band and subjected it to in-gel digestion with trypsin. The resultant peptides were resolved by MALDI-TOF mass spectrometry to accurately determine their molecular weights. A comparison was then made between the experimental peptide masses and those predicted from in silico digestion of the sequences deposited on a public protein database. This analysis identified that the excised protein was mouse filaggrin [with a probability based Mowse score of 177; this is the –10 log(P), where P is the probability that the observed match is a random event; protein scores greater than 75 are considered significant (P<0.05)]. Eleven out of the 44 peptides could be assigned to mouse filaggrin, which has a molecular weight of 27 kDa (Table 1). To confirm the identity of the ~27 kDa protein, western blot analysis was performed using a polyclonal antibody specific to the mouse filaggrin monomer repeat (DSQVHSGVQVEGRRGH) (Fig. 6B). Increased levels of filaggrin monomer were evident in extracts from spink5R820X/R820X mice compared with those from wild-type littermates. Control western blots, using antibodies to the abundant epidermal proteins keratin 14, keratin 1 and loricrin, revealed no differences in the level of immunoreactive material (Fig. 6B). The 27 kDa filaggrin monomer is produced by proteolytic processing of a larger precusor protein called profilaggrin. We found no evidence of any increase in the levels of profilaggrin gene expression, as determined by real time PCR analysis (data not shown).



View larger version (37K):
[in this window]
[in a new window]
  		Figure 6. Increased proteolytic processing of filaggrin in skin of spink5R820X/R820X mice. (A) SDS–PAGE separation of total proteins extracted from neonatal dorsal skin of four littermates. Genotypes are indicated above lanes. Molecular weight standards (ladder) are in kDa. (B) Western blot analysis of total skin proteins from paired spink5R820X/R820X and spink5+/+ neonates derived from two different litters. Sizes of immunoreactive bands are as indicated.

 

View this table:
[in this window]
[in a new window]
  		Table 1. Identification of filaggrin by mass spectrometry
 
These results suggest that in the absence of the serine protease inhibitor spink5, there is increased processing of profilaggrin leading to an increase in the abundance of filaggrin monomers in the skin from spink5R820X/R820X mice. Significantly, filaggrin is a major structural protein important for the maintenance of the structural integrity of the stratum corneum. Despite this evidence for an increase in profilaggrin proteolysis, no difference in total protease activity between the stratum corneums of spink5R820X/R820X and wild-type mice was observed using in situ zymography (Fig. 7). This assay gives an estimate of the total caseinolytic activity in the stratum corneum, which is attributable to many types of proteases. This result does not preclude a difference in the activity of specific proteases involved in profilaggrin processing.



View larger version (57K):
[in this window]
[in a new window]
  		Figure 7. Protease activity in the stratum corneum of newborn spink5R820X/R820X and spink5+/+ mice as measured by in situ zymography. Green fluorescence indicates proteolytically released BODIPY moiety, nuclei were counterstained in red with propidium iodide. A control reaction which included the serine protease inibitor, aprotonin, is shown.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
We have cloned the coding sequence of the mouse orthologue of human SPINK5 and created mice homozygous for a disruption of exon 26 of the gene, which introduces a PTC at R820X. This produces an allele that closely mimics a point mutation(also a PTC) in the human SPINK5 gene that has been identified as a causative mutation in Netherton syndrome (7Go). Newborn spink5R820X/R820X mice develop a lethal, severe ichthyosis with a loss of skin barrier function and dehydration, resulting in death within 1–2 h of birth. The observed exfoliative erythroderma and dehydration is very similar to that observed in patients with Netherton syndrome, which is caused by a range of mutations (mainly PTCs) within the human SPINK5 gene (7Go). We demonstrate that the barrier function is compromised because of the stratum corneum becoming spontaneously detached in the newborn mice, and this is probably compounded by the reduced mechanical strength we detected in the CEs. Although the aetiology of Netherton syndrome is currently unknown, it is likely that the absence of the protease inhibitory function of SPINK5 plays a major role in disease progression. The use of these new spink5R820X/R820X mice permitted biochemical identification of specific protein species undergoing increased proteolysis in the absence of spink5. Strikingly, we found that the skin of newborn spink5R820X/R820X mice displayed a substantial increase in the proteolytic processing of profilaggrin into its constituent filaggrin monomers. Significantly, previous reports have linked defects in filaggrin processing with the development of a number of forms of congenital ichthyosis (28Go–31Go), identifying filaggrin as a critical component in skin development and barrier function. Our data now demonstrate that in the absence of the serine protease inhibitor spink5, normally restrained proteases are no longer inhibited efficiently, leading to an abnormal increase in the processing of profilaggrin, resulting in an overabundance of filaggrin monomers. These alterations in protease activity may play a direct role in the observed deficit in the adhesion of the stratum corneum and the severely compromised epidermal barrier function.

Filaggrin and loricrin are two of the most abundant proteins expressed in the granular layer of the epidermis. Their precursor proteins are the major constituents of the cytoplasmic aggregates (called keratohyalin granules), which are characteristic of this layer of the skin. Profilaggrin is initially synthesized as a large (>300 kDa), insoluble, highly phosphorylated precursor that contains many tandemly arranged repeats of filaggrin monomers. Upon terminal differentiation of the granular layer, and coincident with keratohyalin granule dissolution, profilaggrin is dephosphorylated and rapidly processed proteolytically into functional filaggrin (27 kDa in mouse and 35 kDa in human). Filaggrin was originally recognized for its ability to aggregate keratin filaments in vitro (32Go). Mature filaggrin (but not its precursor) interacts with these filaments to form highly ordered structures known as macrofibrils that crisscross the cornified cells of thestratum corneum thereby imparting structural integrity. Filaggrin has also been reported to become incorporated into CEs (33Go). Filaggrin is subsequently degraded to its constituent amino acids that are modified to provide a mixture of hygroscopic compounds important for maintaining epidermal hydration (reviewed in 34Go).

The potential importance of the altered filaggrin processing we have observed in the spink5R820X/R820X mice is also supported by the role of filaggrin in both human disease and mouse models. As mentioned earlier, the disruption of profilaggrin/filaggrin production and processing has been described in a number of human skin disorders including Harlequin ichthyosis (28Go,29Go), epidermolysis hyperkeratosis (30Go) and ichthyosis vulgaris (31Go). Furthermore, it has also been reported that the overexpression of filaggrin in epidermal keratinocytes results in decreased cell-to-cell adhesion associated with altered distribution of desmosomal components (35Go).

Abnormal profilaggrin processing has also been seen in a number of experimental mouse models. Mice overexpressing the tight junction protein claudin 6 in the suprabasal layers of the epidermis die within 48 h from a defective skin barrier (36Go). The animals are erythematous at birth becoming dehydrated and displaying cracked skin within a few hours of birth. The epidermis is disorganized, often lacking keratohyalin granular cells, and the stratum corneum is thicker, often fragmented. Disturbance of a number of markers of epidermal differentiation is seen in the newborn skin, but one of the most striking is a 13-fold increase in the amount of processed filaggrin (36Go). Defective processing of filaggrin is also seen in the epidermis of tissue specific Pig-a null mice (37Go). These mice have a lethal impairment of epidermal function and exhibit dry, wrinkled skin and prominent hyperkeratosis associated with the virtual absence of the filaggrin monomer.

More recently, loss of processed filaggrin has been described in mice deficient for the transmembrane serine protease matriptase/MT-SP1 that is expressed by epithelial cells including keratinocytes (27Go,38Go). These mice die within 48 h of birth, developing abnormally dry, wrinkled, red andshiny skin with the defective epithelium leading to rapiddehydration of the neonates. A striking aspect of the impaired desquamation observed in the MT-SP1–/– mice was the increased adherence of a more compact stratum corneum (27Go).

Thus, in the absence of the serine protease, matriptase, there is severely impaired filaggrin processing associated with a more compact and adherent stratum corneum, whereas absence of the spink5, a serine protease inhibitor, results in increased filaggrin processing and detachment of the stratum corneum. Given the reciprocal nature of the phenotypes of the MT-SP1–/– mice and spink5R820X/R820X mice, and the opposite effects of the two proteins on profilaggrin processing, it is tempting to speculate that both proteins may be involved in the same proteolytic pathway. A number of other proteases have been implicated in this pathway, including profilaggrin endoproteinase 1 (39Go), calpain (40Go,41Go) and furin (42Go). Changes in the activity of the protein phosphatase PP2A have also been associated with altered profilaggrin processing (43Go). Further studies will be required to determine the potentialfunctional relationship between matriptase (and/or these other enzymes) and spink5 in regulating filaggrin processing.

Yang et al. (44Go) have recently reported the creation of a null allele at the murine spink5 locus. Using a random mutagenesis approach, a 66 kb of mouse chromosome 18 was deleted and replaced with two marker transgenes. This deletion removed the whole coding region of the spink5 gene. Mice homozygous for this deletion exhibit a lethal, perinatal blistering of the skin associated with fragile desmosomal junctions and premature proteolysis of corneodesmosin. This phenotype is similar to the one observed in our spink5R820X/R820X mice. Spink5 is therefore implicated in controlling the correct programmed proteolysis of at least two epithelial components, profilaggrin and corneodesmosin. The critical difference between our study and that of Yang et al. (44Go) is that Netherton syndrome patients have never been demonstrated to harbour large-scaledeletions of the SPINK5 gene. The vast majority of the SPINK5 mutations described in patients result in the production of PTCs. Our study shows that the introduction of PTCs into the mouse spink5 gene is sufficient to produce a phenotype that models the cutaneous manifestations of severe neonatal cases of Netherton syndrome. Our phenotype is also retained in mice that have had the neomycin resistance cassette excised. These mice have minimal genetic alteration at the spink5 locus, other than the introduction of a premature stop codon and a single residual loxP site. Yang et al. (44Go) have succeeded in removing the spink5 gene, but the large deletion (or the introduction of the two exogenous transgenes) may have an effect on the transcription of other known and unidentified genes in this region of mouse chromosome 18. No differences were reported in the closest genes upstreamor downstream of spink5, but the latest version of Ensembl genomic database shows the downstream gene (Ttid) to be 300 kb downstream from spink5 and not the 101 kb stated by Yang et al. (44Go). Another gene, Npy6r, is located between spink5 and Ttid. One cannot be sure that spink5 is the only gene with altered gene expression in these mice.

The neonatal lethality of the spink5R820X/R820X mice has prevented our analysis of other characteristics of Netherton syndrome, such as the associated symptoms of atopic disease, which manifest later in a patient's lifetime. Investigation of the role of spink5 in these aspects of the disease will require the generation of mice harbouring conditional and/or tissue specific gene modifications. However, the generation of conventional spink5R820X/R820X mice has produced a model of severe Netherton syndrome, and analysis of skin extracts from these animals suggests that filaggrin lies in a potential pathway in the aetiology of Netherton syndrome.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Cloning of the mouse spink5 cDNA
Total RNA was prepared from thymus using RNA-Bee solution (Biogenesis, Poole, UK), according to the manufacturer's instructions. First strand cDNA was prepared using Super RT reverse transcriptase (HT Biotechnology Ltd) and an oligo-dT primer. Putative spink5 exons were recognized by searching the mouse genome (www.ensembl.org) for sequences homologous to the human spink5 cDNA. The full-length spink5 cDNA was generated in overlapping fragments using RT–PCR with primers designed to these exons.

Generation of spink5 targeted mice
The insert from IMAGE clone 1210333 (GenBank accession no. AA727418) was used to isolate PAC 519m17 from RP21 library (HGMP Resource Centre, Hinxton). A 9 kb BamHI plasmid containing exons 22–30 of the spink5 gene was isolated using the same probe. The 3.5 kb 3' arm of the targeting vector was generated by PCR of this plasmid with primers 5'-GGT ACC (KpnI) CAG CCC ACA AAT GAA GCA AAG-3' and 5'-ATG AGA AGC ATG AGT CTT CAC-3'. The 3.2 kb 5' arm of the vector was generated by PCR of the plasmid with primers 5'-TTT TGC GGC CGC (NotI) GCT ACA TCA AGC CAG GCG TTA-3' and 5'-TTG GTA CC (KpnI) T ATA GGA TCC TTT CAT TAG CTT CAC GCT CA-3'. A negative selection, diphtheria toxin, cassette was placed into the HindIII site of the plasmid pUCMB21. The two vector arms were then sequentially ligated into this plasmid. The replacement vector was completed by introduction of a loxP-flanked neomycin resistance gene into the KpnI site between the two arms of homology. The targeting vector was linearized and electroporated into E14.1 ES cells. A total of 480 G418 resistant clones were isolated and screened by Southern analysis using a probe external to the vector. The probe for Southern screening was a combination of two PCR products of ~500 bp generated using the primers 5'-CTA GGA ACT ACC TGC CTT CGT-3' and 5'-ATG GCA CAC TTG TTG CCA TGC-3' and 5'-GCA TGG CAA CAA GTG TGC CAT-3' and 5'-TAA CGC CTG GCT TGA TGT AGC-3'. Three clones were found to be targeted correctly. Probing with neomycin gene sequences confirmed this targeting and showed that only a single integration of the vector had occurred. Chimeric mice were generated by the injection of targeted ES clones into 3.5 day C57BL/6 blastocysts. Male chimeras were mated with C57BL/6 females and transmission of the ES cell genotype through the germline was determined by Southern blot analysis. Mice homozygous for the spink5 gene disruption were generated by interbreeding heterozygotes. Routine genotyping of subsequent mice was carried out by multiplex PCR using primers 5'-CTA TCA GGA CAT AGC GTT GGC TACC-3' and 5'-GAG TCT TCA GAC AAT AGT-3' and 5'-GTA GGA GAG TTC TGT AAG-3' (wild-type allele: 419 bp; targeted allele: 529 bp).

Quantitative PCR analysis
Total RNA was prepared from excised skin following tissue homogenization in RNA-Bee (Biogenesis) solution according to the supplier's instructions. cDNA was prepared using an oligo-dT primer and Super RT reverse transcriptase (HT Biotechnology). TaqmanTM real time PCR analysis was used to quantitate gene expression of spink5 and filaggrin using the standard curve method. Expression levels were normalized to HPRT levels. Standard curves were generated by serial dilution of a cDNA sample prepared from a normal, wild-type thymus. The following primer and probe sets were used: for filaggrin, 5'-GGA CAA CTA CAG GCA GTC TTG AAG A-3' and 5'-CAT TTG CAT GAA GAC TTC AGC G-3' and 5'-FAM-TCC AGA TGA CCA AGA CA-TAMRA-3'; between exons 7–8 of spink5, 5'-CCA CTC GTA ACA GAG AGA GCA GAA-3' and 5'-ATT CCT CAC TTG GTT TTC AAA TTC C-3' and 5'-FAM-CAA AGT TCC TTC TCG GCA TCT CTC CGA-TAMRA-3'; between exons 30–31 of spink5, 5'-TGG CAC ACC TGC ATC TGA AA-3' and 5'-GGC AGT TCA GAA TCG AGG CT-3' and 5'-FAM-AGC TTT TTC TCT GTA ATT TCA CCA ATC ATT GCA TCA-TAMRA-3'. The reactions were performed on an ABI PRISMTM 7700 sequence detector (Applied Biosystems) and results analysed using Sequence Detection Software 1.9.

Pathological examination of spink5R820X/R820X mice
Whole mice were sacrificed and fixed in Accustain 10% formalin (Sigma-Aldrich). Mice were embedded in paraffin, and sections were stained and examined by a veterinary pathologist (FINN Laboratories, Diss).

Histological and immunohistochemical analysis
Newborn mice were fixed in 10% formalin fixative (Sigma-Aldrich) before embedding in paraffin. Sections were stained using haematoxylin and eosin or analysed by immunohistochemistry. The following rabbit polyclonal antibodies were used: keratin-1 (1 : 500), keratin-14 (1 : 500), loricrin (1 : 500) and filaggrin monomer repeat (1 : 100) (Covance). Antibody binding was detected using StreptABComplex/HRP Duet peroxidase kit (DakoCytomation) using DAB as the chromogenic substrate, with the exception that antibodies to loricrin and filaggrin were detected using goat anti-rabbit immunoglobulins (1 : 500) (DakoCytomation). All sections were treated at 96°C for 20 min in Target Retrieval Solution (DakoCytomation) prior to the addition of antibodies.

Barrier permeability assays
TEWL was measured on dorsal skin from newborn mice for 1 min using a Tewameter TM210 (Courage and Khazaka, UK Ltd), according to the manufacturer's instructions. Skin permeability was also measured using a toluidine blue penetration assay (26Go). Newborn mice were dehydrated for 1 min each in 25, 50, 75 and 100% methanol/PBS, prior to rehydration in PBS. The mice were stained for 5 min in 0.0125% toluidine blue O (Sigma-Aldrich)/PBS and then destained in a large volume of PBS prior to being photographed.

Cornified envelope extraction and sonication
CEs were prepared by boiling dorsal skin of newborn mice for 15 min in CE extraction buffer: 100 mM Tris–HCl pH 8.5, 5 mM EDTA, 20 mM DTT and 2% SDS (26Go). The insoluble CEs were pelleted by centrifugation at 9000g for 15 min, and then subjected to a second round of boiling and centrifugation. Purified CEs were resuspended in extraction buffer and were photographed using a Zeiss microscope. Sonication experiments were performed in a Misonix bath sonicator at 4°C on power setting 7. These results are representative of n=3 replicates.

Skin protein and western analysis
Dorsal skin from neonatal mice was homogenized in 500 µl of lysis buffer (2% SDS, 80 mM Tris–HCl pH 7.6, 10% glycerol, 100 nM DTT) containing complete EDTA-free protease inhibitor cocktail (Roche Diagnostics). After boiling for 5 min, the proteins were examined by SDS–PAGE on a Novex 4–20% Tris–Glycine gel (Invitrogen), followed by staining with Coomassie blue R-250 (BDH). Bands for mass spectrometry were subjected to in-gel trypsin digestion prior to analysis on a Voyager DE-STR MALDI-TOF instrument. Proteins were identified by comparison of peptide masses with predicted tryptic digests of proteins held on the National Center for Biotechnology Information protein database. Western analysis was performed using rabbit polyclonal antibodies to keratin-1 (1 : 2000), keratin-14 (1 : 2000), loricrin (1 : 1000) and filaggrin monomer (1 : 2000) (Covance). Bound antibodies were detected using HRP-conjugated goat anti-rabbit immunoglobulin (1 : 1000) (Santa Cruz Biotechnology), followed by chemiluminescence with ECL western blotting detection reagents (Amersham Biosciences).

In situ zymography
Cryostat sections of newborn mice were rinsed in 1% Tween-20, prior to incubation with 2 µg/ml of BODIPY FL casein (Enzchek Protease Assay Kit, Molecular Probes) and 0.5 µg/ml of propidium iodide for 10–60 min at 37°C. The sections were then rinsed in 1% Tween-20 and dH2O, prior to visualization of the released BODIPY using a fluorescence microscope. Co-incubation with the serine protease inhibitor aprotonin (2 mg/ml) was used as a negative control.


   ACKNOWLEDGEMENTS
 
We thank members of SABU for help with animal breeding. D.H. and A.S. are supported by the Leukemia Research Fund, UK. P.G.F. is supported by Science Foundation Ireland.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Judge, M.R., Morgan, G. and Harper, J.I. (1994) A clinical and immunological study of Netherton's syndrome. Br. J. Dermatol., 131, 615–621.[CrossRef][ISI][Medline]

  2. Komatsu, N., Takata, M., Otsuki, N., Ohka, R., Amano, O., Takehara, K. and Saijoh, K. (2002) Elevated stratum corneum hydrolytic activity in Netherton syndrome suggests an inhibitory regulation of desquamation by SPINK5-derived peptides. J. Invest. Dermatol., 118, 436–443.[CrossRef][ISI][Medline]

  3. Muller, F.B., Hausser, I., Berg, D., Casper, C., Maiwald, R., Jung, A., Jung, H. and Korge, B.P. (2002) Genetic analysis of a severe case of Netherton syndrome and application for prenatal testing. Br. J. Dermatol., 146, 495–499.[CrossRef][ISI][Medline]

  4. Bitoun, E., Chavanas, S., Irvine, A.D., Lonie, L., Bodemer, C., Paradisi, M., Hamel-Teillac, D., Ansai, S., Mitsuhashi, Y., Taieb, A. et al. (2002) Netherton syndrome: disease expression and spectrum of SPINK5 mutations in 21 families. J. Invest. Dermatol., 118, 352–361.[CrossRef][ISI][Medline]

  5. Traupe, H., Macher, E., Hamm, H. and Happle, R. (1989) Mutation rate estimates are not compatible with autosomal dominant inheritance of the dysplastic nevus ‘syndrome’. Am. J. Med. Genet., 32, 155–157.[CrossRef][ISI][Medline]

  6. Pruszkowski, A., Bodemer, C., Fraitag, S., Teillac-Hamel, D., Amoric, J.C. and de Prost, Y. (2000) Neonatal and infantile erythrodermas: a retrospective study of 51 patients. Arch. Dermatol., 136, 875–880.[Abstract/Free Full Text]

  7. Chavanas, S., Bodemer, C., Rochat, A., Hamel-Teillac, D., Ali, M., Irvine, A.D., Bonafe, J.L., Wilkinson, J., Taieb, A., Barrandon, Y. et al. (2000) Mutations in SPINK5, encoding a serine protease inhibitor, cause Netherton syndrome. Nat. Genet., 25, 141–142.[CrossRef][ISI][Medline]

  8. Bitoun, E., Micheloni, A., Lamant, L., Bonnart, C., Tartaglia-Polcini, A., Cobbold, C., Al Saati, T., Mariotti, F., Mazereeuw-Hautier, J., Boralevi, F. et al. (2003) LEKTI proteolytic processing in human primary keratinocytes, tissue distribution and defective expression in Netherton syndrome. Hum. Mol. Genet., 12, 2417–2430.[Abstract/Free Full Text]

  9. Walley, A.J., Chavanas, S., Moffatt, M.F., Esnouf, R.M., Ubhi, B., Lawrence, R., Wong, K., Abecasis, G.R., Jones, E.Y., Harper, J.I. et al. (2001) Gene polymorphism in Netherton and common atopic disease. Nat. Genet., 29, 175–178.[CrossRef][ISI][Medline]

  10. Kabesch, M., Carr, D., Weiland, S.K. and von Mutius, E. (2004) Association between polymorphisms in serine protease inhibitor, kazal type 5 and asthma phenotypes in a large German population sample. Clin. Exp. Allergy, 34, 340–345.[CrossRef][ISI][Medline]

  11. Kato, A., Fukai, K., Oiso, N., Hosomi, N., Murakami, T. and Ishii, M. (2003) Association of SPINK5 gene polymorphisms with atopic dermatitis in the Japanese population. Br. J. Dermatol., 148, 665–669.[CrossRef][ISI][Medline]

  12. Nishio, Y., Noguchi, E., Shibasaki, M., Kamioka, M., Ichikawa, E., Ichikawa, K., Umebayashi, Y., Otsuka, F. and Arinami, T. (2003) Association between polymorphisms in the SPINK5 gene and atopic dermatitis in the Japanese. Genes Immun., 4, 515–517.[CrossRef][ISI][Medline]

  13. Gonzalez, H.E., Gujrati, M., Frederick, M., Henderson, Y., Arumugam, J., Spring, P.W., Mitsudo, K., Kim, H.W. and Clayman, G.L. (2003) Identification of 9 genes differentially expressed in head and neck squamous cell carcinoma. Arch. Otolaryngol. Head Neck Surg., 129, 754–759.[Abstract/Free Full Text]

  14. Magert, H.J., Standker, L., Kreutzmann, P., Zucht, H.D., Reinecke, M., Sommerhoff, C.P., Fritz, H. and Forssmann, W.G. (1999) LEKTI, a novel 15-domain type of human serine proteinase inhibitor. J. Biol. Chem., 274, 21499–21502.[Abstract/Free Full Text]

  15. Green, L.J. and Bartelt, D.C. (1969) The structure of the bovine pancreatic secretory trypsin inhibitor—Kazal's inhibitor. II. The order of the tryptic peptides. J. Biol. Chem., 244, 2646–2657.[Abstract/Free Full Text]

  16. Magert, H.J., Kreutzmann, P., Drogemuller, K., Standker, L., Adermann, K., Walden, M., John, H., Korting, H.C. and Forssmann, W.G. (2002) The 15-domain serine proteinase inhibitor LEKTI: biochemical properties, genomic organization, and pathophysiological role. Eur. J. Med. Res., 7, 49–56.[ISI][Medline]

  17. Magert, H.J., Kreutzmann, P., Standker, L., Walden, M., Drogemuller, K. and Forssmann, W.G. (2002) LEKTI: a multidomain serine proteinase inhibitor with pathophysiological relevance. Int. J. Biochem. Cell Biol., 34, 573–576.[CrossRef][ISI][Medline]

  18. Mitsudo, K., Jayakumar, A., Henderson, Y., Frederick, M.J., Kang, Y., Wang, M., El-Naggar, A.K. and Clayman, G.L. (2003) Inhibition of serine proteinases plasmin, trypsin, subtilisin A, cathepsin G, and elastase by LEKTI: a kinetic analysis. Biochemistry, 42, 3874–3881.[CrossRef][Medline]

  19. Walden, M., Kreutzmann, P., Drogemuller, K., John, H., Forssmann, W.G. and Hans-Jurgen, M. (2002) Biochemical features, molecular biology and clinical relevance of the human 15-domain serine proteinase inhibitor LEKTI. Biol. Chem., 383, 1139–1141.[CrossRef][ISI][Medline]

  20. Moffatt, M.F. (2004) SPINK5: a gene for atopic dermatitis and asthma. Clin. Exp. Allergy, 34, 325–327.[CrossRef][ISI][Medline]

  21. Raghunath, M., Tontsidou, L., Oji, V., Aufenvenne, K., Schurmeyer-Horst, F., Jayakumar, A., Stander, H., Smolle, J., Clayman, G.L. and Traupe, H. (2004) SPINK5 and Netherton syndrome: novel mutations, demonstration of missing LEKTI, and differential expression of transglutaminases. J. Invest. Dermatol., 123, 474–483.[CrossRef][ISI][Medline]

  22. Bitoun, E., Bodemer, C., Amiel, J., de Prost, Y., Stoll, C., Calvas, P. and Hovnanian, A. (2002) Prenatal diagnosis of a lethal form of Netherton syndrome by SPINK5 mutation analysis. Prenat. Diagn., 22, 121–126.[CrossRef][ISI][Medline]

  23. Hardman, M.J., Sisi, P., Banbury, D.N. and Byrne, C. (1998) Patterned acquisition of skin barrier function during development. Development, 125, 1541–1552.[Abstract]

  24. Kalinin, A.E., Kajava, A.V. and Steinert, P.M. (2002) Epithelial barrier function: assembly and structural features of the cornified cell envelope. Bioessays, 24, 789–800.[CrossRef][ISI][Medline]

  25. Segre, J.A., Bauer, C. and Fuchs, E. (1999) Klf4 is a transcription factor required for establishing the barrier function of the skin. Nat. Genet., 22, 356–360.[CrossRef][ISI][Medline]

  26. Koch, P.J., de Viragh, P.A., Scharer, E., Bundman, D., Longley, M.A., Bickenbach, J., Kawachi, Y., Suga, Y., Zhou, Z., Huber, M. et al. (2000) Lessons from loricrin-deficient mice: compensatory mechanisms maintaining skin barrier function in the absence of a major cornified envelope protein. J. Cell Biol., 151, 389–400.[Abstract/Free Full Text]

  27. List, K., Szabo, R., Wertz, P.W., Segre, J., Haudenschild, C.C., Kim, S.Y. and Bugge, T.H. (2003) Loss of proteolytically processed filaggrin caused by epidermal deletion of Matriptase/MT-SP1. J. Cell Biol., 163, 901–910.[Abstract/Free Full Text]

  28. Dale, B.A., Holbrook, K.A., Fleckman, P., Kimball, J.R., Brumbaugh, S. and Sybert, V.P. (1990) Heterogeneity in harlequin ichthyosis, an inborn error of epidermal keratinization: variable morphology and structural protein expression and a defect in lamellar granules. J. Invest. Dermatol., 94, 6–18.[CrossRef][ISI][Medline]

  29. Dale, B.A. and Kam, E. (1993) Harlequin ichthyosis. Variability in expression and hypothesis for disease mechanism. Arch. Dermatol., 129, 1471–1477.[Abstract]

  30. Ishida-Yamamoto, A., Eady, R.A., Underwood, R.A., Dale, B.A. and Holbrook, K.A. (1994) Filaggrin expression in epidermolytic ichthyosis (epidermolytic hyperkeratosis). Br. J. Dermatol., 131, 767–779.[CrossRef][ISI][Medline]

  31. Sybert, V.P., Dale, B.A. and Holbrook, K.A. (1985) Ichthyosis vulgaris: identification of a defect in synthesis of filaggrin correlated with an absence of keratohyaline granules. J. Invest. Dermatol., 84, 191–194.[CrossRef][ISI][Medline]

  32. Dale, B.A., Resing, K.A. and Lonsdale-Eccles, J.D. (1985) Filaggrin: a keratin filament associated protein. Ann. N.Y. Acad. Sci., 455, 330–342.[ISI][Medline]

  33. Steven, A.C. and Steinert, P.M. (1994) Protein composition of cornified cell envelopes of epidermal keratinocytes. J. Cell Sci., 107, 693–700.[Abstract]

  34. Rawlings, A.V., Scott, I.R., Harding, C.R. and Bowser, P.A. (1994) Stratum corneum moisturization at the molecular level. J. Invest. Dermatol., 103, 731–741.[CrossRef][ISI][Medline]

  35. Presland, R.B., Kuechle, M.K., Lewis, S.P., Fleckman, P. and Dale, B.A. (2001) Regulated expression of human filaggrin in keratinocytes results in cytoskeletal disruption, loss of cell–cell adhesion, and cell cycle arrest. Exp. Cell. Res., 270, 199–213.[CrossRef][ISI][Medline]

  36. Turksen, K. and Troy, T.C. (2002) Permeability barrier dysfunction in transgenic mice overexpressing claudin 6. Development, 129, 1775–1784.[Abstract/Free Full Text]

  37. Hara-Chikuma, M., Takeda, J., Tarutani, M., Uchida, Y., Holleran, W.M., Endo, Y., Elias, P.M. and Inoue, S. (2004) Epidermal-specific defect of GPI anchor in Pig-a null mice results in Harlequin ichthyosis-like features. J. Invest. Dermatol., 123, 464–469.[CrossRef][ISI][Medline]

  38. List, K., Haudenschild, C.C., Szabo, R., Chen, W., Wahl, S.M., Swaim, W., Engelholm, L.H., Behrendt, N. and Bugge, T.H. (2002) Matriptase/MT-SP1 is required for postnatal survival, epidermal barrier function, hair follicle development, and thymic homeostasis. Oncogene, 21, 3765–3779.[CrossRef][ISI][Medline]

  39. Resing, K.A., Thulin, C., Whiting, K., al-Alawi, N. and Mostad, S. (1995) Characterization of profilaggrin endoproteinase 1. A regulated cytoplasmic endoproteinase of epidermis. J. Biol. Chem., 270, 28193–28198.[Abstract/Free Full Text]

  40. Resing, K.A., al-Alawi, N., Blomquist, C., Fleckman, P. and Dale, B.A. (1993) Independent regulation of two cytoplasmic processing stages of the intermediate filament-associated protein filaggrin and role of Ca2+ in the second stage. J. Biol. Chem., 268, 25139–25145.[Abstract/Free Full Text]

  41. Yamazaki, M., Ishidoh, K., Suga, Y., Saido, T.C., Kawashima, S., Suzuki, K., Kominami, E. and Ogawa, H. (1997) Cytoplasmic processing of human profilaggrin by active mu-calpain. Biochem. Biophys. Res. Commun., 235, 652–656.[CrossRef][ISI][Medline]

  42. Dale, B.A., Nirunsuksiri, W., Rehemtulla, A., Lewis, S.P. and Presland, R.B. (1997) Profilaggrin is proteolytically cleaved by furin during epidermal differentiation (abstract). J. Invest. Dermatol., 108, 554.[ISI]

  43. Kam, E., Nirunsuksiri, W., Hager, B., Fleckman, P. ans Dale, B.A. (1997) Protein phosphatase activity in human keratinocytes cultured from normal epidermis and epidermis from patients with harlequin ichthyosis. Br. J. Dermatol. 137, 874–882.[CrossRef][ISI][Medline]

  44. Yang, T., Liang, D., Koch, P.J., Hohl, D., Kheradmand, F. and Overbeek, P.A. (2004) Epidermal detachment, desmosomal dissociation, and destabilization of corneodesmosin in Spink5–/– mice. Genes Dev., 18, 2354–2358.[Abstract/Free Full Text]


Add to CiteULike CiteULike