ARTICLE
ejd.2011.1264
Auteur(s) : Nathalie JONCA, Emilie A. LECLERC, Cécile
CAUBET, Michel SIMON, Marina GUERRIN, Guy SERRE guy.serre@udear.cnrs.fr
UMR 5165 CNRS - Université de Toulouse, Hôpital Purpan, Place du
Dr Baylac, TSA 40031, 31059 Toulouse Cedex 9, France
Reprints: G. SERRE
The epidermis constitutes a constantly renewed multifunctional
barrier separating the body from the environment. Several of its
barrier functions are fulfilled by the stratum corneum (SC),
the most superficial layer of the epidermis. This layer is composed
of corneocytes, dead and flattened “mummified” cells resulting from
the cornification of granular keratinocytes, which are endowed with
a cornified envelope that makes them extremely resistant. Moreover,
the strong cohesion of the corneocytes gives to the layer a high
resistance to mechanical stresses. Owing to its relative
impermeability to water and water soluble substances, the SC also
provides an efficient “inside-out” and “outside-in” barrier for
these components. Lastly, it constitutes a barrier against
physical, chemical and microbial agents from the environment.
In the eighties, the predominant model to describe the
properties of the SC was the “brick wall” model proposed by P.
Elias. In the model, the corneocytes were the bricks, embedded
within an intercellular lipid substance which was the mortar,
considered as responsible for cell cohesion. In the early nineties,
emerging data suggested that the lipids, crucial for impermeability
to water, played only a limited part in cohesion which depended
mostly on the corneocyte-associated intercellular junctions that we
proposed to call corneodesmosomes. In particular, the finding that
protease activity was necessary for desquamation to take place
[1, 2], as well as the identification of corneodesmosin (CDSN)
as the only protein specifically localized in the extracellular
part of the corneodesmosomes [3], contributed to reconsider the
role of protein junctions in the SC cohesion.
This review presents the data from the discovery of CDSN to the
most recent findings concerning its structure and its function. In
particular, the important benefits of mouse models and human
diseases for unravelling the role of CDSN in the epidermis and hair
follicle integrity are reported in details.
A new protein specific to the late stages of terminal epidermis
differentiation: corneodesmosin
In the early nineties, our group raised mouse monoclonal
antibodies against human plantar SC in order to identify new late
markers of terminal epidermis differentiation. This led to the
identification of CDSN, specifically recognized by two monoclonal
antibodies, G36-19 and F28-27. Immunohistological studies showed
that CDSN is expressed in the cornified epithelia, and absent from
the uncornified epithelia of the vagina, uterine cervix, or
oesophagus [3]. In addition, CDSN is present in the thymus Hassal's
bodies, the epithelial cells of which are very close to epidermis
keratinocytes, and using RT-PCR the CDSN gene was shown to
be also transcribed in the placenta [4]. CDSN is also
expressed in the hair follicle where the protein appears
asynchronously in the 3 compartments of the inner root sheath,
first in the Henle's layer, then in the cuticle, and lastly in the
Huxley's layer [3, 5]. CDSN is also detected in the cells of
the hair follicle medulla [4].
In the epidermis, as shown by immunohistology, CDSN is
cytoplasmic in the lower stratum granulosum (SG) then
becomes pericellular and progressively disappears in the lower SC.
In contrast, the protein persists up to the desquamating
keratinocytes in the palmoplantar epidermis and hard palate
epithelium, with a discontinuous pericorneocyte microgranular
distribution. Similarly, CDSN is detected as surface spots on
corneocytes isolated by scraping the hard palate or the epidermis.
Immunoelectron microscopy reveals that CDSN is present in the
lamellar bodies from their emergence in the upper spinous cells, in
the extracellular part of the desmosomes of the granular
keratinocytes, and lastly in the core of the modified desmosomes of
the SC. Finally, CDSN was shown to be largely preserved at the
surface of purified cross-linked envelopes, where it is located on
tips gathered over the external side. Moreover, as the envelopes
were purified under reducing and denaturing conditions, these data
demonstrate that CDSN is covalently linked to the cornified
envelope not only by disulfide bonds but also by another type of
covalent links. The nature of these covalent links, the molecular
partner – protein or lipid – involved in the links, as
well as the enzymes responsible for the crosslink, remain unknown.
Transglutaminases 1, 3 and 5 are responsible for the crosslink of
the envelope protein precursors in the granular keratinocytes.
However, these enzymes are located in the cytosol or anchored to
the internal side of the cytoplasmic membrane. No transglutaminase
isoforms were described to be secreted in the extracellular space,
where it could use CDSN as a substrate. CDSN is the first and until
now the only protein specifically localized in the extracellular
core of corneodesmosomes. This discovery contributed to the idea
that corneodesmosomes are not only desmosome remnants, as
considered until then, but fulfill a precise function in the upper
layers of the epidermis.
CDSN is a 52-56 kDa basic phosphoprotein. Deglycosylation
experiments, reactivity to lectins, and chromatography on
concanavalin A-sepharose indicated that it is N-glycosylated, with
the oligosaccharide moiety comprising ∼10% of the protein mass [6].
Cloning of its encoding cDNA revealed that CDSN is located
on chromosome 6, in PSORS1, the major locus for psoriasis
susceptibility. The mRNA encodes a 529-aminoacid protein with a
N-terminal signal peptide and one putative N-glycosylation site,
consistent with the demonstration that it is secreted and
glycosylated [7].
As a step in elucidating the mechanisms of tissue-specific
expression, the human CDSN promoter was characterized using
transgenic mice. The transgene consisted on the 4.2 kb upstream
region of the human CDSN transcription initiation site
linked to the LacZ gene. The reporter gene expression was
detected by histoenzymology in special areas of the inner root
sheath of the hair follicles and the hair medulla, with an
expression pattern that perfectly matched that of the endogenous
protein. The 4.2 kb DNA fragment is thus sufficient to correctly
drive the expression of the reporter gene in the hair follicle.
Surprisingly, no β-galactosidase activity was detected in the
granular keratinocytes of the epidermis. However, induction of
epidermal hyperproliferation either by topic pharmacological agents
or by wounding, led to strong expression of the reporter gene in
the keratinocytes of the SG and in the parakeratotic corneocytes of
the SC. This suggests that the transgene did not comprise all the
elements necessary for the expression of CDSN in normal epidermis
[4].
Corneodesmosin presents homophilic adhesion properties mediated
by its N-terminal glycine-loop domain
A striking feature of CDSN is its very high serine and glycine
content (27.5% and 16%, respectively), particularly at both termini
of the protein, i.e. amino acids 60-171 and 375-450. Similar
serine- and glycine-rich domains described at both ends of keratins
have been proposed to form structural motifs called “glycine loops”
as a consequence of the association of interspersed aromatic or
aliphatic residues. These glycine loop domains have been suggested
to mediate intermolecular adhesion by acting like a Velcro [8].
Such molecular interactions could be essential for skin homeostasis
since mutations in the glycine loops of keratins and loricrin lead
to various human cutaneous diseases, e.g., ichtyosis hystrix and
palmoplantar keratodermas [9-12].
The potential role of CDSN in homophilic cell-cell interactions
was assessed with mouse L-fibroblasts expressing a chimeric protein
composed of the human CDSN and the transmembrane and cytoplasmic
domains of mouse E-cadherin. CDSN thus anchored at the cell surface
mediates cellular aggregation. Overlay binding assays and
quantitative analysis by surface plasmon resonance using bacterial
recombinant forms of full-length CDSN confirmed the homophilic
adhesion properties of the protein [13, 14]. Moreover,
recombinant CDSN associates into large homooligomers of at least
three subunits that are only partially dissociated in 8 M urea.
These very stable oligomers obtained in vitro possibly
correspond to cis- and/or trans-interactions in
vivo. Similar experiments of size-exclusion chromatography and
surface plasmon resonance performed with truncated recombinant
CDSNs showed that the N-terminal glycine loop domain, that
perfectly matches features of the Steinert's glycine loop model,
was necessary and sufficient, and thus responsible for both the
oligomerization of CDSN and its homophilic adhesive properties.
Altogether, these data clearly establish that CDSN displays
adhesive properties mediated by its N-terminal glycine loop
domain.
The progressive proteolysis of corneodesmosin in the stratum
corneum is a prerequisite for desquamation
A tight balance between keratinocyte proliferation in the basal
layer and cell shedding at the surface of the epidermis is
essential for skin homeostasis and renewing of the SC.
Ultrastructural studies demonstrated that the degradation of
corneodesmosomes in the outermost corneocytes is concomitant to
desquamation, and showed that proteolytic cleavage of the
extracellular part of these cell-cell junctional structures is a
key event in the process. In non-palmoplantar SC, normal
exfoliation of corneocytes results from 2 steps of corneodesmosomal
degradation. First, the non-peripheral corneodesmosomes are
degraded at the interface between the stratum compactum and
the stratum disjunctum. Second, the persistent peripheral
corneodesmosomes are broken at the skin surface allowing corneocyte
exfoliation. Desquamation in palmoplantar skin is quite different,
as the non-peripheral corneodesmosomes appear to be preserved
against proteolysis in the lower SC accounting for the formation of
a thick cornified layer, characteristic of these anatomical
regions. This persistence is also a common feature of
hyperkeratosis and ichthyoses [15, 16].
The presence of cleavage products of CDSN in extracts of
superficial SC attests to the degradation of corneodesmosomes. When
extracted from viable layers of human epidermis, CDSN showed an
apparent molecular mass of around 52-56 kDa, whereas a molecular
form of 33 kDa is the major form extracted from the most
superficial and less firmly attached corneocytes [3]. The presence
of CDSN in the extracellular core of corneodesmosomes, and the
in vitro demonstration of the homophilic adhesive properties
of the protein, strongly suggest that, in vivo, CDSN
contributes to the close and strong cohesion between corneocytes.
Consequently, its proteolysis was proposed as one of the major
biochemical changes that lead to desquamation [17].
Deglycosylation experiments and reactivity with lectins
demonstrated that the CDSN carbohydrate moiety does not prevent the
protein proteolysis. Using a set of affinity-purified anti-peptide
antibodies and monoclonal antibodies recognizing the five
structural domains of CDSN, a refined characterization of its
proteolysis during terminal differentiation of epidermis was
realized. Immunoblotting, immunohistochemistry and immunoelectron
microscopy experiments showed that CDSN processing begins in the
SG. In an initial proteolysis step, the N- and C-terminus of the
protein are eliminated, giving rise to a 48-46 kDa CDSN fragment
that incorporates into desmosomes and might contribute to their
transformation into corneodesmosomes. This fragment is still
endowed with the glycine loop-related domains. Later, at the
transition between the stratum compactum and the stratum
disjunctum, cleavage of the N-terminal adhesive glycine loop
domain of the protein leads to a 36-30 kDa fragment. This step
seems to be correlated with the abrupt decrease in cohesion of the
SC. Finally, a CDSN fragment of only 15 kDa, mainly corresponding
to the central part of the protein and devoid of both the N- and
C-terminal glycine-rich domains, is present at the surface of the
non-cohesive corneocytes [6, 18].
A number of different proteases of the serine, cysteine, or
aspartic protease families, as well as protease inhibitors of these
classes of enzymes, have been identified in the SC and play a major
role in desquamation (for a review see [19-21]). Among these
proteases, KLK7 and KLK5 – also known as stratum
corneum chymotryptic enzyme (SCCE) and stratum corneum
tryptic enzyme (SCTE), respectively – are serine proteases of
the kallikrein family [22]. They are both highly expressed in the
granular keratinocytes and present in the intercellular spaces of
the SC. Synthesized as inactive proforms, they are both activated
by cleavage of a short N-terminal domain performed by an enzyme
with the characteristics of a trypsin-like protease [23, 24].
They were proposed to be actors of a cascade of activated proteases
and inactive pro-enzymes that could control desquamation [25] (for
a review see [21]). The importance of pH for SC cohesion and
desquamation was also evidenced using superbases topically applied
on hairless mouse skin [26]. CDSN was demonstrated to be a
preferred substrate of both serine proteases in vitro [18].
A detailed analysis of the proteolysis by KLK5 and KLK7 of CDSN,
and also of two other major components of the extracellular core of
corneodesmosomes, namely desmocollin 1 (DSC1) and desmoglein 1
(DSG1), was performed. KLK7 directly cleaved CDSN and DSC1 but was
unable to degrade DSG1. Incubation with KLK5 induced degradation of
the three corneodesmosomal components. Moreover, it was shown that
KLK7 and KLK5 conserved their activity at pH 5.6, close to that of
physiological conditions in the SC. This study also suggested that
KLK5 was able to activate the proform of KLK7 [27].
It is highly probable that, besides serine proteases, the
involvement of other classes of proteases, also present in the SC,
contribute to degradation of the corneodesmosomal components. Among
these are the 2 cysteine proteases, Cathepsin L2 (CTSL2, also named
Cathepsin V) and Cathepsin L-like (CTSL-like), and one aspartic
acid protease Cathepsin D (CTSD). Synthesized as proenzymes,
cathepsins undergo proteolytic maturation, sometimes in an
autocatalytic way, and are active in an acidic environment. They
have long been classified as hydrolases with exclusive functions in
terminal degradation of proteins in the lysosomal compartment.
Active forms of CTSL-like and CTSD have been shown to cleave CDSN
in vitro ([28, 29] and Jonca et al.,accepted for
publication). A detailed description of proteases and protease
inhibitors, as well as recent data on the degradation of the
corneodesmosomal components and on desquamation, leading to a model
of regulation of corneodesmosomal components degradation, has been
recently reviewed elsewhere [20].
Corneodesmosin plays an essential role in maintaining epidermis
integrity: lessons from animal models
The demonstration that CDSN is an adhesive protein covalently
linked to the cornified envelope as well as the data showing that
its proteolysis is a prerequisite to desquamation, strongly suggest
that CDSN is primordial for the cohesive function of
corneodesmosomes. But how exactly the protein fulfils its function
remains unclear. CDSN was proposed to be necessary for
corneodesmosome morphogenesis. It could also protect the other
corneodesmosomal components from premature degradation by the
proteases involved in desquamation. Finally, CDSN could reinforce
corneocyte cohesion by its own adhesive properties. Analysis of
mouse models with inactivated Cdsn gene was of particular
interest to answer these questions. The first publication of
Cdsn inactivation in mice, obtained by insertional mutation,
concluded that Cdsn is necessary for corneodesmosome formation
[30]. We also realized conditional ablation of Cdsn using a
K14-driven Cre-mediated recombination. The detailed analysis
of our Cdsn-deficient mice confirmed that Cdsn is vital for
epidermis and hair follicle integrity, and our conclusion supported
the idea that Cdsn is mainly an adhesive protein [31]. Indeed,
immediately after birth, KO neonates showed a severe detachment of
the SC starting from abdominal area and extremities (paws and
snoot), then died within 1 h. In contrast, heterozygous newborns
did not develop any skin phenotypes, demonstrating that ablation of
only one allele had no consequences. Surprisingly, KO pups obtained
after caesarian delivery at day E18.5 were indistinguishable from
their littermates. However, during grooming by surrogate mothers,
they rapidly developed a skin phenotype similar to that observed
after natural birth. The effect of the mutation on epidermis
permeability was analyzed by a dye penetration assay and
transepidermal water loss measurements. The degree of SC cohesion
was appreciated by tape-stripping, the material obtained being
evaluated by protein quantification. Therefore, although the
greatly reduced mechanical resistance of the SC is an intrinsic
feature caused by Cdsn deficiency, the epidermal tear
leading to the lethal barrier defect, occurred only under
mechanical stresses encountered after birth.
Intact dorsal skin of Cdsn KO newborns (i.e. before
appearance of the macroscopic phenotype) observed by transmission
electron microscopy revealed a structural organization and a
thickness of the granular layer and the SC similar to that of wild
type (WT) skin. In particular, we did not detect any significant
differences in the number of transitional desmosomes between WT and
KO neonates, unlike Matsumoto and coworkers’ published data [30].
Moreover, the electron density of the corneodesmosomes appeared
unchanged, suggesting no premature degradation of these structures.
Thus, Cdsn does not seem indispensable for the morphogenesis of
corneodesmosomes nor their protection from SC proteases. On the
other hand, the histological analysis of the dorsal KO skin
revealed the presence of blisters as soon as cornification occurs,
that is, at the SG/SC transition. At the ultrastructural level, the
main part of numerous split junctions typically remain attached to
the granular keratinocytes, suggesting that the cohesive defect
lies in the upper side of the junction at the limit between
desmoglea and cornified envelope of the upper transitional cell.
Thus, although Cdsn was already present in the extracellular core
of desmosomes from the SG, it seemed to play its fundamental role
only when cornification is complete. No thinning out of single or
bundled corneocytes throughout the SC was observed but detachment
of the whole SC from the subjacent SG. This suggested that the
SG/SC interface may actually be the most fragile zone as it links
two epidermal layers with different junctional organizations and
mechanical characteristics: desmosomes and keratin intermediate
filaments organized in taut cables in the SG, rigid cornified
envelopes linked by corneodesmosomes in the SC.
In order to assess long-term consequences of Cdsn
inactivation, we performed grafting of skin from
Cdsn-deficient newborns onto nude mice. The grafted
epidermis first developed acanthosis and hyperkeratosis. Increased
expression of various differentiation markers (involucrin, K10),
and induced expression of the hyperproliferative keratin K6,
confirmed the altered differentiation and the hyperproliferative
state of the grafted epidermis. However, these compensatory
mechanisms appeared to be ineffective, and were followed by a
progressive but finally complete disappearance of the epidermis.
Cdsn was thus necessary for maintaining the integrity and water
impermeability of the postnatal epidermis. The consequences of Cdsn
loss in adult mice was investigated using an additional model
consisting on K14-driven Cre-mediated loxP
recombination system with a chimeric CreERT2 recombinase that can
be induced by tamoxifen. Inducible Cdsn KO resulted in
histological abnormalities similar to those of the Cdsn KO
skin graft model. Barrier maintaining became unsuccessful,
compromising the vital prognosis of the mice when the whole skin
was affected. Thus, Cdsn proved to play a vital role in maintaining
epidermis integrity and functions in adult too.
These somatic and inducible Cdsn KO mice also allowed to
investigate the consequences of Cdsn loss on hair follicles.
Analysis of skin sections from newborn Cdsn KO mice showed a
similar morphology and number of hair follicles to those of skin
from WT littermate, suggesting that Cdsn seemed to be dispensable
for hair follicle morphogenesis. Long term consequences of
Cdsn excision in hair follicle was analyzed after grafing of
dorsal Cdsn KO skin onto nude mice and in adult mouse skin
induced for Cdsn excision. In both cases, hair follicles
first showed an altered morphology, developed cysts, and finally
disappeared. This suggested that Cdsn is necessary for maintaining
normal hair follicle architecture.
Altogether, these data obtained in vivo demonstrated that
Cdsn plays a vital role in the structural and functional integrity
of the epidermis and the hair follicle integrity by preventing the
rupture of corneodesmosome.
The corneodesmosin gene is highly polymorphic and some of its
haplotypes could be involved in psoriasis susceptibility
A common feature of numerous skin disorders is scaling, usually
associated with thickening of the SC. Hyperkeratosis may be due to
increased cell proliferation, impaired desquamation, or a
combination of both. It has been invariably associated with the
persistence of both peripheral and non-peripheral corneodesmosomes
in the outer SC, a feature reminiscent of normal palmoplantar
epidermis. This has been accurately described in the case of
several congenital ichthyoses [32, 33] but also winter xerosis
[16] or soap-induced xerosis [19]. In parallel, overexpression of
CDSN was detected in all the analyzed hyperkeratotic lesions,
including inflammatory diseases [15, 34]. As a defect in the
barrier functions is commonly encountered in hyperkeratosis, CDSN
overexpression may rather result from a compensatory mechanism to
the barrier defect.
CDSN is of particular interest in relation to psoriasis. Indeed,
as previously mentioned, CDSN is localized to chromosome
6p21 at PSORS1, the major susceptibility locus of this chronic
inflammatory skin disorder. CDSN is highly polymorphic, with
more than one SNP every 100 bp in the coding region [35]. Some
of these SNPs have been associated with psoriasis in many genetic
studies (reviewed in [36]). However, PSORS1 spans a ∼300 kb region
containing at least 11 other genes including HLA-C, and the
exact identity of the PSORS1 gene remains controversial owing to
extensive linkage disequilibrium across this region. A recent study
revealed that only 2 genes from the locus, HLA-C and
CDSN, yield alleles encoding protein unique to risk
haplotypes [37]. However, this study and others, using ancestral
haplotype analysis, resulted in contradictory findings for the
PSORS1 gene: HLA-C, CDSN, or both genes
[37-41]. Ultrastructural analyses of psoriatic skin show that
desmosomes are not dramatically transformed into corneodesmosomes
at the SG/SC transition, but rather stay at an intermediate state
between the two structures. As already stated, persistence of
corneodesmosomes accompanied by an increased amount of CDSN in the
SC, is a feature common to hyperkeratotic lesions. However, the
earlier synthesis and secretion of CDSN in the upper stratum
spinosum, is specific to psoriasis among several
hyperproliferative disorders [34]. Finally, persistence of
full-length CDSN and of other corneodesmosomal proteins in the
upper SC was observed in psoriatic lesions [42]. The
psoriasis-associated SNPs of CDSN could affect the function
of the corresponding proteins. To date, only one study has shown
that an intragenic SNP (SNP*971T) present in 3 common haplotypes
associated with psoriasis in various ethnic groups, is responsible
for an increased stability of the encoded mRNAs [43]. Although
these results are in agreement with the known overexpression of
CDSN in psoriatic lesions [15, 34], they pointed a discrepancy
between CDSN haplotypes at the mRNA level only. The CDSN*TTC allele
(SNP*619T, SNP*1236T, SNP*1243C), associated with the disease in
many different ethnic groups, was identified as the smallest
combination of protein-altering variations unique to CDSN
carried by risk haplotypes [37]. Assessment of the functional
consequences of these amino-acid substitutions, particularly
regarding CDSN sensitivity to proteolysis, should provide insight
into the putative involvement of the protein in psoriasis (Jonca
et al., accepted for publication).
Nonsense mutations of the corneodesmosin gene are responsible
for two different genodermatoses
Intriguingly, the first monogenic disease associated with
mutations in CDSN that has been identified revealed a hair
phenotype and did not affect the epidermis. Indeed, nonsense
mutations in CDSN were found to be responsible for a rare
autosomal dominant disease, hypotrichosis simplex of the scalp
(HSS; OMIM 146520) [44]. Affected individuals experience
progressive loss of scalp hairs beginning in the middle of the
first decade and almost complete baldness by the third decade. The
body hairs, beard, eyebrows and axillary hairs are normal. To date,
3 different nonsense mutations have been reported in 4 families
from Israel, Denmark and Mexico [45]. All of them produce a
truncated form of CDSN, mainly corresponding to the N-terminal
glycine loop domain. An antibody directed against this domain
labeled irregularly-sized “deposits” located in ridges of the
superficial dermis and at the periphery of the hair follicles
deeper in the dermis. These deposits were not stained by antibodies
directed against epithelial proteins (keratins K1, K2, K10, K11,
(pro)filaggrin, involucrin) strongly implying that the truncated
CDSN is not associated with any of these proteins in the deposits.
By western blot analysis of dermis extracts, the truncated CDSN was
detected as SDS- and boiling-resistant dimers, trimers and larger
oligomers [44].
We noticed the histological similarities between HSS and the two
main forms of primary localized cutaneous amyloidosis, namely
lichen amyloidosis and macular amyloidosis, two rare chronic
pruritic skin disorders. In both diseases, amyloid deposits that
could be derived from epidermal keratins accumulates in the dermal
papillae [46-52]. These similarities led us to wonder whether HSS
was an amyloidosis? Amyloidoses form a large heterogeneous group of
misfolding diseases associated with the conversion of soluble
peptides/proteins into highly organized fibrillar aggregates [46].
Shape, tinctorial properties and secondary structure of amyloid
fibrils are characteristic: the fibrils are unbranched, 7.5-20-nm
thick, bind Congo Red and Thioflavine T dyes, exhibit yellow-green
birefringence in polarized light upon binding of Congo Red, and
have a significant content of β-sheets that are systematically
hydrogen-bonded. We recently demonstrated that HSS is a new
amyloidosis [53]. Indeed, consistent with the possibility that the
mutant CDSN may adopt an amyloid conformation, the dermal deposits
from skin biopsies of affected individuals were stained by
Thioflavine T and Congo Red. The presence of the SAP component, a
protein that colocalizes with amyloid plaques irrespective of their
chemical nature and the clinical type of amyloidosis, was also
evidenced within the CDSN deposits. In agreement with these in
vivo observations, analysis by electron microscopy demonstrated
that recombinant forms of the truncated mutated CDSN forms
expressed in the skin of HSS patients, as well as the recombinant
N-terminal glycine/serine rich domain of CDSN alone, did assemble
into ring-shaped oligomeric structures and into fibrils. The
amyloid-like structure of the fibrils was evidenced by tinctorial
affinity for Congo Red and Thioflavine T dyes. The recorded Fourier
transform infrared spectra suggested that the recombinant
N-terminal glycine/serine rich domain of CDSN assembly into fibrils
was accompanied by a conformational change with a very significant
increase in β-sheet content typical of the assembly of native
unfolded polypeptides into amyloid fibrils. Finally, the truncated
CDSN forms expressed in the skin of HSS patients, and assembled
into prefibrillar oligomers, were toxic to primary keratinocytes
grown in vitro. The latter result is consistent with two
recent reports suggesting that soluble oligomeric species rather
than mature amyloid fibrils were the toxic entity in
amyloid-related diseases [54, 55].
Mouse models with inactivation of Cdsn showed that
heterozygous Cdsn + /- mice were
indistinguishable from their wild type littermate, in particular,
they did not develop any hair phenotype up to 8 months. This
reinforced the hypothesis that HSS is not caused by Cdsn
haploinsufficiency [30, 31]. Given our previous finding that a
recombinant truncated form of CDSN is able to bind the entire CDSN
[13, 14], a dominant negative interaction between the mutant
and wild-type proteins may account for the loss of cohesion in the
inner root sheath of the hair follicles. This could affect the
functionality of the inner root sheath, resulting in perturbation
of hair cycle re-initiation. However, in view of the delayed onset
of alopecia and the fact that lost hairs are not regenerated,
another possibility is that the observed CDSN aggregates are toxic
for the hair follicle cells. The aggregates released from defective
hair follicles probably migrate through the dermis until they reach
the dermo-epidermal junction under which they accumulate. As
similar aggregates are not detected in the epidermis, these are
either unformed in this tissue context, or degraded by SC
proteases.
Very recently, another nonsense mutation in the CDSN gene
was identified as responsible for the peeling skin syndrome type B,
associating peeling, pruritus and atopy [56]. The new autosomal
recessive monogenic disease was therefore named “peeling skin
disease” (PSD, MIM 270300), this rare ichthyosiform erythroderma
being mainly characterized by lifelong peeling of the skin. Four
individuals from a large consanguineous Roma family from Germany
have been studied. As well as superficial skin peeling since birth,
the patients presented severe pruritus, atopic manifestations, food
allergies and susceptibility to bacterial infections. Histological
and ultrastructural analyses of their skin revealed a detachment of
the SC at the interface between the granular and cornified layers.
This syndromic ichthyosis shares numerous similarities with the
Netherton “syndrome” but no mutations have been found in the
SPINK5 gene. Based on a genome-wide linkage analysis, the
authors defined a candidate region on the chromosome 6 and tested
CDSN as a candidate gene, due to the phenotype of the
Cdsn KO mice [30, 31]. Sequencing of CDSN in the
4 patients revealed the presence of a nonsense mutation at the
position 175 (A > T), leading to a stop codon just before
encoding of the first glycine loop domain of the protein. Western
blot and immunofluorescence experiments showed an absence of
expression of CDSN in the skin whereas other components of the
desmosomes and corneodesmosomes (i.e., desmocollin 1 and desmoglein
1) showed an almost regular expression. The protease inhibitor
LEKTI and other epidermal differentiation markers (filaggrin,
involucrin, loricrin, etc.) were immunohistologically found to be
hyperexpressed. Contrary to HSS where a normal CDSN allele
is still expressed, the loss of CDSN expression is complete in PSD.
The absence of protein in the hairs only leads to their easy
plugging, a mild hair phenotype compared to HSS.
Finally, using 3-dimensional skin models produced with
fibroblasts and keratinocytes of PSD patients, elevated levels of
KLK5 and filaggrin were found, and also a permeability barrier
defect supposed to account for predisposition to atopy. Overall,
these results confirmed the role of CDSN as an essential epidermis
adhesion molecule. Therefore, PSD represents a novel
pathophysiological model for atopic disorders, besides the
Netherton “syndrome” or ichtyosis vulgaris.
Conclusion – Future prospects
The data accumulated from the discovery of CDSN to the more
recent works using mouse models or studying human diseases,
demonstrated the primordial role of CDSN in the epidermis and hair
follicle integrity. However, some functional aspects concerning
CDSN are not fully understood, especially at the molecular level.
For instance, immunoelectron microscopy experiments clearly
demonstrated that CDSN exclusively localizes to the extracellular
core of corneodesmosomes in the SC. But how is this specific
localization achieved? It is likely that a preexisting component of
the extracellular part of the junction interacts with CDSN. We were
unable to evidence any interactions between CDSN and desmosomal
cadherins (unpublished results). The molecular partners for CDSN
thus remain to be identified. Similarly, the molecular event
resulting in the covalent link of CDSN to the cornified envelope,
and probably involving transglutaminase-like enzymatic
activity, has not yet been characterized. Finally, the underlying
molecular mechanism by which CDSN assumes its adhesive function
within the corneodesmosome has not been fully elucidated. The
homophilic adhesive properties of CDSN and the strong resistance to
highly denaturing conditions of the aggregates formed by
recombinant CDSN [13, 14] suggest that, in vivo, CDSN
reinforces cohesion by its own adhesive properties. However,
adhesion provided by glycine-loop domains has been suggested to
mediate reversible and constantly adjustable intermolecular links
[8]. Consistent with this, CDSN could confer elasticity to the
desmosomal junctions, an essential property to prevent breaking as
soon as the cell envelope rigidifies. Clarification of those CDSN
properties closely dealing with the molecular physiology of the SC
constitutes the task for the future.
Disclosure
Financial support: none. Conflict of interest: none.
References
1 DL Bissett, JF McBride, L.F. Patrick Role of protein and
calcium in stratum corneum cell cohesion Arch Dermatol Res
1987; 279: 184-189.
2 A Lundström, T. Egelrud Stratum corneum chymotryptic enzyme: a
proteinase which may be generally present in the stratum corneum
and with a possible involvement in desquamation Acta Derm
Venereol 1991; 71: 471-474.
3 G Serre, V Mils, M Haftek et al. Identification of late
differentiation antigens of human cornified epithelia, expressed in
re-organized desmosomes and bound to cross-linked envelope J
Invest Dermatol 1991; 97: 1061-1072.
4 H Gallinaro, N Jonca, L Langbein et al. A 4.2 kb
upstream region of the human corneodesmosin gene directs
site-specific expression in hair follicles and hyperkeratotic
epidermis of transgenic mice J Invest Dermatol 2004; 122:
730-738.
5 V Mils, C Vincent, F Croute, G. Serre The expression of
desmosomal and corneodesmosomal antigens shows specific variations
during the terminal differentiation of epidermis and hair follicle
epithelia J Histochem Cytochem 1992; 40: 1329-1337.
6 M Simon, M Montézin, M Guerrin, JJ Durieux, G. Serre
Characterization and purification of human corneodesmosin, an
epidermal basic glycoprotein associated with corneocyte-specific
modified desmosomes J Biol Chem 1997; 272: 31770-31776.
7 M Guerrin, M Simon, M Montézin, M Haftek, C Vincent, G. Serre
Expression cloning of human corneodesmosin proves its identity with
the product of the S gene and allows improved characterization of
its processing during keratinocyte differentiation J Biol
Chem 1998; 273: 22640-22647.
8 PM Steinert, JW Mack, BP Korge, SQ Gan, SR Haynes, A.C. Steven
Glycine loops in proteins: their occurrence in certain intermediate
filament chains, loricrins and single-stranded RNA binding proteins
Int J Biol Macromol 1991; 13: 130-139.
9 E Maestrini, AP Monaco, JA McGrath et al. A molecular
defect in loricrin, the major component of the cornified cell
envelope, underlies Vohwinkel's syndrome Nat Genet 1996; 13:
70-77.
10 E Sprecher, A Ishida-Yamamoto, OM Becker et al.
Evidence for novel functions of the keratin tail emerging from a
mutation causing ichthyosis hystrix J Invest Dermatol 2001;
116: 511-519.
11 A Terrinoni, P Puddu, B Didona et al. A mutation in
the V1 domain of K16 is responsible for unilateral palmoplantar
verrucous nevus J Invest Dermatol 2000; 114: 1136-1140.
12 NV Whittock, FJ Smith, H Wan et al. Frameshift
mutation in the V2 domain of human keratin 1 results in striate
palmoplantar keratoderma J Invest Dermatol 2002; 118:
838-844.
13 C Caubet, N Jonca, F Lopez, JP Estève, M Simon, G. Serre
Homo-oligomerization of human corneodesmosin is mediated by its
N-terminal glycine loop domain J Invest Dermatol 2004; 122:
747-754.
14 N Jonca, M Guerrin, K Hadjiolova et al.
Corneodesmosin, a component of epidermal corneocyte desmosomes,
displays homophilic adhesive properties J Biol Chem 2002;
277: 5024-5029.
15 M Haftek, M Simon, J Kanitakis et al. Expression of
corneodesmosin in the granular layer and stratum corneum of normal
and diseased epidermis Br J Dermatol 1997; 137: 864-873.
16 M Simon, D Bernard, AM Minondo et al. Persistence of
both peripheral and non-peripheral corneodesmosomes in the upper
stratum corneum of winter xerosis skin versus only peripheral in
normal skin J Invest Dermatol 2001; 116: 23-30.
17 A Lundström, G Serre, M Haftek, T. Egelrud Evidence for a
role of corneodesmosin, a protein which may serve to modify
desmosomes during cornification, in stratum corneum cell cohesion
and desquamation Arch Dermatol Res 1994; 286: 369-375.
18 M Simon, N Jonca, M Guerrin et al. Refined
characterization of corneodesmosin proteolysis during terminal
differentiation of human epidermis and its relationship to
desquamation J Biol Chem 2001; 276: 20292-20299.
19 A. Rawlings Trends in stratum corneum research and the
management of dry skin conditions International Journal of
Cosmetic Science 2003; 25: 63-95.
20 N Jonca, C Caubet, M Guerrin, M Simon, G. Serre
Corneodesmosomal proteins AV Rawlings JJ Leyden Skin
moisturization, Second Edition 2009; Informa Healthcare USA, Inc
New York 99-122.
21 P Ovaere, S Lippens, P Vandenabeele, W. Declercq The emerging
roles of serine protease cascades in the epidermis Trends
Biochem Sci 2009; 34: 453-463.
22 EP Diamandis, GM Yousef, LY Luo, A Magklara, C.V. Obiezu The
new human kallikrein gene family: implications in carcinogenesis
Trends Endocrinol Metab 2000; 11: 54-60.
23 M Brattsand, T. Egelrud Purification, molecular cloning, and
expression of a human stratum corneum trypsin-like serine protease
with possible function in desquamation J Biol Chem 1999;
274: 30033-30040.
24 L Hansson, M Stromqvist, A Backman, P Wallbrandt, A
Carlstein, T. Egelrud Cloning expression, and characterization of
stratum corneum chymotryptic enzyme. A skin-specific human serine
proteinase J Biol Chem 1994; 269: 19420-19426.
25 IE Ekholm, M Brattsand, T. Egelrud Stratum corneum tryptic
enzyme in normal epidermis: a missing link in the desquamation
process? J Invest Dermatol 2000; 114: 56-63.
26 JP Hachem, D Crumrine, J Fluhr, BE Brown, KR Feingold, P.M.
Elias pH directly regulates epidermal permeability barrier
homeostasis, and stratum corneum integrity/cohesion J Invest
Dermatol 2003; 121: 345-353.
27 C Caubet, N Jonca, M Brattsand et al. Degradation of
corneodesmosome proteins by two serine proteases of the kallikrein
family, SCTE/KLK5/hK5 and SCCE/KLK7/hK7 J Invest Dermatol
2004; 122: 1235-1244.
28 D Bernard, B Mehul, A Thomas-Collignon et al. Analysis
of proteins with caseinolytic activity in a human stratum corneum
extract revealed a yet unidentified cysteine protease and
identified the so-called “stratum corneum thiol protease” as
cathepsin l2 J Invest Dermatol 2003; 120: 592-600.
29 S Igarashi, T Takizawa, T Takizawa et al. Cathepsin D,
but not cathepsin E, degrades desmosomes during epidermal
desquamation Br J Dermatol 2004; 151: 355-361.
30 M Matsumoto, Y Zhou, S Matsuo et al. Targeted deletion
of the murine corneodesmosin gene delineates its essential role in
skin and hair physiology Proc Natl Acad Sci USA 2008; 105:
6720-6724.
31 EA Leclerc, A Huchenq, NR Mattiuzzo et al.
Corneodesmosin gene ablation induces lethal skin-barrier disruption
and hair-follicle degeneration related to desmosome dysfunction
J Cell Sci 2009; 122: 2699-2709.
32 H Elsayed-Ali, S Barton, R. Marks Stereological studies of
desmosomes in ichthyosis vulgaris Br J Dermatol 1992; 126:
24-28.
33 R Ghadially, ML Williams, SY Hou, P.M. Elias Membrane
structural abnormalities in the stratum corneum of the autosomal
recessive ichthyoses J Invest Dermatol 1992; 99:
755-763.
34 M Allen, A Ishida-Yamamoto, J McGrath et al.
Corneodesmosin expression in psoriasis vulgaris differs from normal
skin and other inflammatory skin disorders Lab Invest 2001;
81: 969-976.
35 M Guerrin, C Vincent, M Simon, R Tazi Ahnini, M Fort, G.
Serre Identification of six novel polymorphisms in the human
corneodesmosin gene Tissue Antigens 2001; 57: 32-38.
36 F Capon, G Novelli, S Semprini et al. Searching for
psoriasis susceptibility genes in Italy: genome scan and evidence
for a new locus on chromosome 1 J Invest Dermatol 1999;
112: 32-35.
37 RP Nair, PE Stuart, I Nistor et al. Sequence and
haplotype analysis supports HLA-C as the psoriasis susceptibility 1
gene Am J Hum Genet 2006; 78: 827-851.
38 F Capon, IK Toal, JC Evans et al. Haplotype analysis
of distantly related populations implicates corneodesmosin in
psoriasis susceptibility J Med Genet 2003; 40: 447-452.
39 C Helms, NL Saccone, L Cao et al. Localization of
PSORS1 to a haplotype block harboring HLA-C and distinct from
corneodesmosin and HCR Hum Genet 2005; 118: 466-476.
40 S Orrù, E Giuressi, C Carcassi, M Casula, L. Contu Mapping of
the major psoriasis-susceptibility locus (PSORS1) in a 70-Kb
interval around the corneodesmosin gene (CDSN) Am J Hum
Genet 2005; 76: 164-171.
41 CD Veal, F Capon, MH Allen et al. Family-based
analysis using a dense single-nucleotide polymorphism-based map
defines genetic variation at PSORS1, the major
psoriasis-susceptibility locus Am J Hum Genet 2002; 71:
554-564.
42 M Simon, R Tazi-Ahnini, N Jonca, C Caubet, MJ Cork, G. Serre
Alterations in the desquamation-related proteolytic cleavage of
corneodesmosin and other corneodesmosomal proteins in psoriatic
lesional epidermis Br J Dermatol 2008; 159: 77-85.
43 F Capon, MH Allen, M Ameen et al. A synonymous SNP of
the corneodesmosin gene leads to increased mRNA stability and
demonstrates association with psoriasis across diverse ethnic
groups Hum Mol Genet 2004; 13: 2361-2368.
44 E Levy-Nissenbaum, RC Betz, M Frydman et al.
Hypotrichosis simplex of the scalp is associated with nonsense
mutations in CDSN encoding corneodesmosin Nat Genet 2003;
34: 151-153.
45 JE Lai-Cheong, K Arita, J.A. McGrath Genetic diseases of
junctions J Invest Dermatol 2007; 127: 2713-2725.
46 M. Stefani Protein misfolding and aggregation: new examples
in medicine and biology of the dark side of the protein world
Biochim Biophys Acta 2004; 1739: 5-25.
47 R Apaydin, Y Gurbuz, D Bayramgurler, B Muezzinoglu, N. Bilen
Cytokeratin expression in lichen amyloidosus and macular
amyloidosis J Eur Acad Dermatol Venereol 2004; 18:
305-309.
48 YT Chang, HN Liu, WJ Wang, DD Lee, S.F. Tsai A study of
cytokeratin profiles in localized cutaneous amyloids Arch
Dermatol Res 2004; 296: 83-88.
49 H Kobayashi, K. Hashimoto Amyloidogenesis in organ-limited
cutaneous amyloidosis: an antigenic identity between epidermal
keratin and skin amyloid J Invest Dermatol 1983; 80:
66-72.
50 M Kumakiri, K. Hashimoto Histogenesis of primary localized
cutaneous amyloidosis: sequential change of epidermal keratinocytes
to amyloid via filamentous degeneration J Invest Dermatol
1979; 73: 150-162.
51 H Maeda, S Ohta, Y Saito, H Nameki, H. Ishikawa Epidermal
origin of the amyloid in localized cutaneous amyloidosis Br J
Dermatol 1982; 106: 345-351.
52 S Masu, M Hosokawa, M. Seiji Amyloid in localized cutaneous
amyloidosis: immunofluorescence studies with anti-keratin antiserum
especially concerning the difference between systemic and localized
cutaneous amyloidosis Acta Derm Venereol 1981; 61:
381-384.
53 C Caubet, L Bousset, O Clemmensen et al. A new
amyloidosis caused by fibrillar aggregates of mutated
corneodesmosin Faseb J 2010; 24: 3416-3426.
54 S Baglioni, F Casamenti, M Bucciantini et al.
Prefibrillar amyloid aggregates could be generic toxins in higher
organisms J Neurosci 2006; 26: 8160-8167.
55 M Bucciantini, G Calloni, F Chiti et al. Prefibrillar
amyloid protein aggregates share common features of cytotoxicity
J Biol Chem 2004; 279: 31374-31382.
56 V Oji, KM Eckl, K Aufenvenne et al. Loss of
corneodesmosin leads to severe skin barrier defect, pruritus, and
atopy: unraveling the peeling skin disease Am J Hum Genet
2010; 87: 274-281.
|
Version imprimable
Version PDF