- Email: [email protected]
TGM5 Mutations Impact Epidermal Differentiation in Acral Peeling Skin Syndrome
ORIGINAL ARTICLE
TGM5 Mutations Impact Epidermal Differentiation in Acral Peeling Skin Syndrome Manuela Pigors1, Dimitra Kiritsi1, Cristina Cobzaru1, Agnes Schwieger-Briel1, Jose Sua´rez2, Flavio Faletra3, Heikki Aho4 , Leeni Ma¨kela¨5, Johannes S. Kern1, Leena Bruckner-Tuderman1,6 and Cristina Has1 Acral peeling skin syndrome (APSS) is an autosomal recessive skin disorder characterized by acral blistering and peeling of the outermost layers of the epidermis. It is caused by mutations in the gene for transglutaminase 5, TGM5. Here, we report on clinical and molecular findings in 11 patients and extend the TGM5 mutation database by four, to our knowledge, previously unreported mutations: p.M1T, p.L41P, p.L214CfsX15, and p.S604IfsX9. The recurrent mutation p.G113C was found in 9 patients, but also in 3 of 100 control individuals in a heterozygous state, indicating that APSS might be more widespread than hitherto expected. Using quantitative real-time PCR, immunoblotting, and immunofluorescence analysis, we demonstrate that expression and distribution of several epidermal differentiation markers and corneodesmosin (CDSN) is altered in APSS keratinocytes and skin. Although the expression of transglutaminases 1 and 3 was not changed, we found an upregulation of keratin 1, keratin 10, involucrin, loricrin, and CDSN, probably as compensatory mechanisms for stabilization of the epidermal barrier. Our results give insights into the consequences of TGM5 mutations on terminal epidermal differentiation. Journal of Investigative Dermatology (2012) 132, 2422–2429; doi:10.1038/jid.2012.166; published online 24 May 2012
INTRODUCTION The cornified layer constitutes the outermost compartment of the skin, thus forming a robust mechanical barrier against environmental insults. To accomplish this function, differentiating keratinocytes build the cornified envelope, which comprises a unique structure consisting of highly insoluble proteins, including loricrin (LOR), filaggrin, and involucrin (IVL), that are cross-linked by several transglutaminases (Candi et al., 2005). Genetic defects of the molecules associated with the cornified envelope lead to disorders of cornification such as ichthyoses and atopic dermatitis (Proksch et al., 2008; Oji et al., 2010b). Recently, the underlying genetic causes of particular types of ichthyoses, the peeling skin syndromes, have been elucidated: mutations in the corneodesmosin (CDSN) gene primarily impair the cohesion of corneocytes and lead to generalized 1
Department of Dermatology, University Freiburg Medical Center, Freiburg, Germany; 2Service of Dermatology, University Hospital Nuestra Sen˜ora de Candelaria Carretera del Rosario, Santa Cruz de Tenerife, Spain; 3Institute for Maternal and Child Health IRCCS ‘‘Burlo Garofolo’’, Trieste, Italy; 4 Department of Pathology, Turku University Hospital, Turku, Finland; 5 Department of Dermatology, Turku University Hospital, Turku, Finland and 6 Freiburg Institute for Advanced Studies, School of Life Sciences–LifeNet, University of Freiburg, Freiburg, Germany Correspondence: Cristina Has, Department of Dermatology, University Freiburg Medical Center, Hauptstrasse 7, 79104 Freiburg, Germany. E-mail: [email protected] Abbreviations: APSS, acral peeling skin syndrome; CDSN, corneodesmosin; IVL, involucrin; KRT, keratin; LOR, loricrin; TGase, transglutaminase protein; TGM, transglutaminase gene Received 21 December 2011; revised 21 March 2012; accepted 5 April 2012; published online 24 May 2012
2422 Journal of Investigative Dermatology (2012), Volume 132
peeling skin disease (Oji et al., 2010a; Bowden, 2011), whereas transglutaminase 5 (TGase 5), involved in crosslinking of the cornified envelope proteins, is affected in the acral peeling skin syndrome (APSS; Cassidy et al., 2005; Oji et al., 2010a; Pavlovic et al., 2011). Only little is known on the molecular disease mechanisms of APSS (Cassidy et al., 2005; Kiritsi et al., 2010). It is considered a rare autosomal recessive genodermatosis, characterized by painless, superficial blistering and peeling of hands and feet (Shwayder et al., 1997; Cassidy et al., 2005; Kiritsi et al., 2010). Thus far, three different TGM5 (transglutaminase 5 gene) missense mutations have been reported, including p.G113C, which is recurrent in the European population (Cassidy et al., 2005; Kharfi et al., 2009; Kiritsi et al., 2010; van der Velden et al., 2012). In this study, we describe 11 patients with APSS and extend the TGM5 mutation database by four, to our knowledge, previously unreported mutations: p.M1T, p.L41P, p.L214CfsX15, and p.S604IfsX9. The latter two are the first TGM5 mutations predicted to lead to premature termination codons. Furthermore, we show that TGM5 mutations significantly affect expression of epidermal differentiation markers and CDSN in vitro and in situ. RESULTS Clinical presentation and mutations
This study extends the number of 23 patients with TGM5 mutations, which have been reported previously in the literature, by 11 (Table 1; Kiritsi et al., 2010). Eight of eleven patients were children (patients 1–8 in Table 1, age between & 2012 The Society for Investigative Dermatology
M Pigors et al. TGM5 Mutations Impact Epidermal Differentiation
Table 1. Clinical features of the patients and mutations Age at diagnosis, origin of the patient
Age of onset
TGM5 mutations
Site of biopsy, level of skin cleavage
1
8 Months, Slovakian
Shortly after birth
c.[337G4T];[337G4T] p.[G113C];[G113C]
Lower leg, no cleavage
Peeling on volar aspects of hands and feet
2
1 Year, origin not known
7 Months
c.[2T4C];[2T4C] p.[M1T];[M1T]
Sole, intracorneal blister
Peeling, superficial blisters, and erosions on volar and dorsal aspects of hands and feet
3
1 Year, Italian
3 Months
c.[640delC]; [1811_1815delinsTCCTTCA] p.[L214CfsX15];[S604IfsX9]
No biopsy available
Peeling and erythematous patches on hands and feet; aggravation by heat, friction, perspiration, and exposure to water
4
2 Years, German
Not known
c.[337G4T];[337G4T] p.[G113C];[G113C]
Not known, no cleavage
Peeling and superficial blisters mainly on volar, but also on dorsal aspects of hands and feet
5
7 Years, German
6 Weeks
c.[122T4C];[337G4T] p.[L41P];[G113C]
Dorsum of the foot, no cleavage
Peeling and superficial blisters mainly on dorsal, but also on volar aspects of hands and feet; peeling on knees
6
7 Years, Spanish
11 Months
c.[337G4T];[337G4T] p.[G113C];[G113C]
Foot, intracorneal blister
Peeling and superficial blisters on volar and dorsal aspects of hands and feet; occasionally crusty erosions on knees and elbows; aggravation upon contact with water
7
11 Years, German Not known
c.[337G4T];[337G4T] p.[G113C];[G113C]
Foot, intracorneal blister
Not available
8
12 Years, German 3 Weeks
c.[337G4T];[640delC] p.[G113C];[L214CfsX15]
Not known, no cleavage
Large, superficial blisters on ankles; mild blistering and peeling on extremities
9
20 Years, Hungarian
Since birth
c.[337G4T];[337G4T] p.[G113C];[G113C]
Foot, intracorneal blister
Mild peeling on volar and dorsal aspects of hands and feet
10
27 Years, German Not known
c.[337G4T];[337G4T] p.[G113C];[G113C]
Palm, intracorneal Mild peeling on volar and dorsal aspects of hands and blister feet
11
33 Years, German Not known
c.[337G4T];[337G4T] p.[G113C];[G113C]
Lower arm, no cleavage
No.
8 months and 12 years) and presented with superficial blistering and peeling on volar and dorsal aspects of hands and feet (Figure 1a–c). It is noteworthy that patient 5 reported to have had erosions on knees after friction, and patient 8 had large, superficial blisters on the ankles (Figure 1c) and peeling on the legs, arms, and elbows. The three oldest patients (patients 9–11 in Table 1), 20, 27, and 33 years of age, only had peeling and were very mildly affected. In patients 1–4, 7, 8, and 11, epidermolysis bullosa simplex was initially suspected, and in patients 2, 3, and 7 keratin (KRT) 5 and 14 genes were analyzed and mutations excluded. We disclosed four previously unreported TGM5 mutations: c.2T4C, p.M1T in exon 1 (in patient 2), c.122T4C, p.L41P in exon 2 (in patient 5), c.640delC, p.L214CfsX15, in exon 5 (in patients 3 and 8), and c.1811_1815delinsTCCT TCA, p.S604IfsX9 in exon 11 (in patient 3; Figure 1b–d, Table 1). The first mutation affects the start codon and interferes with the initiation of translation, probably causing an N-terminal truncated TGase 5. This presumption is supported by the prediction of additional initiation sites in the complementary DNA sequence (http://atgpr.dbcls.jp/) and by the positive TGase 5 signal present in the skin of the patient (see below). The pathogenic role of the variation p.L41P is suggested by its absence in 100 control individuals
Clinical features
Mild peeling on volar and dorsal aspects of feet
and by the prediction to be ‘‘possibly damaging’’ (Polyphen, http://genetics.bwh.harvard.edu/pph/). The deletion c.640delC causes a frameshift and premature termination of translation, p.L214CfsX15, in the catalytic core domain. The last mutation, an insertion/deletion, leads to a frameshift and premature stop codon formation in the b-barrel 1 domain, p.S604IfsX9, probably leading to a C-terminal truncation of TGase 5. However, no material was available to verify whether frameshift mutations led to mRNA decay and/or truncated proteins. The recurrent mutation p.G113C was identified in seven patients in a homozygous state and in two patients in a heterozygous state (Table 1). Intriguingly, this mutation was also detected in a heterozygous state in 3 of 100 control individuals, together with the single-nucleotide polymorphism p.T109M in all patients and controls. The parents of patients 3, 5, 6, and 8 were available for analysis, and were heterozygous carriers of the mutations of their offspring. Impact of TGM5 mutations in keratinocytes
As data on TGase 5 functions mainly rely on experiments using recombinant protein (Candi et al., 2001, 2002; Pietroni et al., 2008), we used APSS as a genetic model to gain insight into the physiological functions of this enzyme in www.jidonline.org 2423
M Pigors et al. TGM5 Mutations Impact Epidermal Differentiation
a
b
2-Year-old
p.[L41P];[G113C]
T
390 G
T 53
260 G 53
G
G
T
C
395 G
C 53
265 G 53
G
260 A
A
C
c.337G>T, p.G113C G 53
G 53
12-Year-old p.[G113C];[L214CfsX15]
Control
Control G
c
7-Year-old
p.[G113C];[G113C]
T 53
C
265 C
T
Control A
C
C
G
C
G 44
115 C 44
T
c.122T>C, p.L41P A 56
A 44
C 44
C 35
C 35
250 T
G
C
G
G
120 C 29
G 29
G 44
c.640delC, p.L214CfsX15
135 G 53
C
A 46
C 56
C 53
T 26
T 20
G 29
d M1T L41P G113C
L214CfsX15 W255R
β-Sandwich
Catalytic core
1
S604IfsX9 K445N β-Barrel 1
β-Barrel 2 720
Figure 1. Clinical presentation of acral peeling skin syndrome and TGM5 mutations. (a–c) Patients presented with peeling and superficial blisters (arrows) mainly on volar and dorsal aspects of hands and feet. Note large erosions after blisters over the ankles in the 12-year-old patient who was wearing high-ankle leather shoes during summer. The lower panel shows chromatograms of controls and patients’ partial sequences of exon 3, 2, and 5 of TGM5 indicating the recurrent variation p.G113C, as well as the mutations p.L41P and p.L214CfsX15. (d) Schematic representation of transglutaminase 5 (TGase 5) and its domains (Grenard et al., 2001). The four, to our knowledge previously unreported, mutations are indicated in red. Both missense mutations, p.M1T and p.L41P, are located in the b-sandwich domain. The deletion/insertion mutations, p.L214CfsX15 and p.S604Ifsx9, are found in the catalytic core domain or b-barrel 1 domain of TGase 5, respectively. Previously reported mutations are shown in black. TGase 5 domain structures were designed with the DOG 2.0 software (Ren et al., 2009).
keratinocytes. We analyzed primary keratinocytes derived from two patients homozygous for the mutation p.G113C (designated as G113C-keratinocytes). Although cell culture is not an ideal model to study epidermal differentiation, treatment of keratinocytes with Ca2 þ for 7 days is known to induce expression of epidermal transglutaminases and main differentiation proteins (Pillai et al., 1990). These were analyzed at both mRNA and protein levels. Interestingly, the TGM5 mRNA levels were markedly increased (up to 6-fold) in G113C keratinocytes, as compared with normal keratinocytes (Figure 2a). In contrast, the expression of the genes for other transglutaminases present in the upper epidermal layers, transglutaminase 1 (TGM1) and 3 (TGM3), were comparable to the control. mRNA levels of TGase 5 substrates (Candi et al., 2005), LOR (4-fold), and IVL (3-fold), as well as KRT1 (up to 10-fold), were 2424 Journal of Investigative Dermatology (2012), Volume 132
upregulated, whereas that of KRT5, a marker of basal keratinocytes, remained unchanged (Figure 2a). Further, these results were validated at the protein level. TGase 5 is a highly insoluble protein present in several forms, including a monomeric form of 84 kDa and an autocatalytic, cross-linked form of higher molecular mass (Pietroni et al., 2008). As monomeric TGase 5 is only present to a low extent in differentiating keratinocytes in vitro (Candi et al., 2002), only a faint band was detected in all samples. In contrast, the autocatalytic, inactive form of TGase 5 was enhanced 2-fold in G113C keratinocytes (Figure 2b), presumably accounting for the mRNA upregulation. In agreement with the quantitative real-time PCR data, the level of transglutaminase 1 protein (TGase 1) was not changed to compensate for TGase 5 (Figure 2b). More importantly, the expression of IVL was increased 2-fold and that of CDSN, one of the most crucial
M Pigors et al.
a
Normalized fold expression
TGM5 Mutations Impact Epidermal Differentiation
Gene expression 10 8 6 4 2 0 TGM5
TGM1
TGM3
LOR
Control keratinocytes
b Control
TGase 5 G113C-KCs
Control
IVL
KRT1
KRT5
G113C-keratinocytes
TGase 1 G113C-KCs
Corneodesmosin Control G113C-KCs
250 95
72
55 95
Actin Involucrin G113C-KCs Control
72 130 55
36
Actin
Actin
Actin
Figure 2. Molecular characterization of acral peeling skin syndrome keratinocytes. (a) Keratinocytes from two patients homozygous for p.G113C (G113C-keratinocytes) and a control were treated for 7 days with Ca2 þ and used for RNA extraction for quantitative real-time PCR (a) and lysates for immunoblotting (b). Upregulation of the TGM5 transcript was observed in the G113C keratinocytes, whereas levels of TGM1 and TGM3 were comparable to that of control cells. The differentiation markers loricrin (LOR), involucrin (IVL), and keratin (KRT1) were strongly elevated in the patients’ cells. Keratin 5 (KRT5) remained unchanged. (b) Immunoblotting of control and G113C keratinocytes (G113C-KCs) lysates is shown. Transglutaminase 5 (TGase 5) monomeric form (lower arrow, 84 kDa) showed similar levels in control and patients’ lysates, whereas the autocatalytic form of TGase 5 (upper arrow) was markedly increased in the patients’ samples (2-fold). TGase 1 expression remained unchanged in patients’ cells. In contrast, involucrin and corneodesmosin were increased in the patients as compared with the control. Antibodies to b-actin were used to control loading. The numbers on the left side of each blot represent molecular weights in kilodaltons.
cell–cell adhesion molecules in the corneocytes, was increased up to 6-fold in G113C keratinocytes after 7 days of Ca2 þ stimulation (Figure 2b). Molecular characterization of APSS skin
Next, we validated the findings in situ by indirect immunofluorescence staining of the skin of patients 2, 5, and 10, all of them bearing different mutation constellations. All patients and control samples showed positive signals for TGase 5 confined to the cell membranes, suggesting that the enzymatic activity, rather than the expression or localization, is altered (Figure 3). To investigate this further, we performed an in situ TGase activity assay. Although this test does not discriminate among the activities of the epidermal TGases, it is known that TGase 1 is active at both pH 8.4 and 7.4, and the optimal pH for TGase 3 and 5 is considered to be B8.4 (Raghunath et al., 1998; Candi et al., 2001). We found a diminution of TGase activity at pH 8.4 in the skin of all patients, which was more pronounced in the patient
homozygous for p.M1T. At pH 7.4, TGase activity was preserved in the skin of all three patients (Figure 3). At both pH conditions, negative controls showed no fluorescent signal at the cell periphery. In agreement with the in vitro data, the signals for CDSN and LOR were increased in all patients, independent of the mutations. LOR, in particular, accumulated in the granular layer and shifted to the upper stratum corneum in the skin of all three patients, which showed a 2-fold increase of the intensity of the LOR signal, in contrast to the controls whose expression was lower and restricted to the granular layer. In addition, staining for KRT 10 appeared slightly increased in the patients’ skin. The abnormal signal within the stratum corneum is most probably due to false-positive antibody reaction. KRT 5/6 remained unchanged (Figure 4). In addition, transmission electron microscopy was performed to unravel more subtle morphological changes in the APSS skin with the homozygous mutation p.G113C. In particular, the granular cells contained stellate keratohyalin www.jidonline.org 2425
M Pigors et al. TGM5 Mutations Impact Epidermal Differentiation
p.[M1T];[M1T] -Sole-
p.[L41];[G113C] -Foot-
Control -Palm-
p.[G113C;[G113C] -Palm-
TGase activity pH 7.4
TGase activity pH 8.4
TGase 5
Control -Sole-
Figure 3. Transglutaminase 5 (TGase 5) expression and TGase activity assay in acral peeling skin syndrome skin. Indirect immunofluorescence showed TGase 5 expression confined to the cell membrane of keratinocytes in all three patients, predominantly in the granular layer of the epidermis. A TGase activity assay was performed at pH 8.4 and 7.4. The activity at pH 8.4 was reduced at the cell periphery in the upper spinous and granular layers of the skin of all patients, especially in the patient homozygous for p.M1T. At pH 7.4, TGase activity in patients’s skin was preserved. TGase 2 is active at the level of the basement membrane zone and dermis. Bars ¼ 50 mm.
Corneodesmosin
Control -Sole-
p.[M1T];[M1T] -Sole-
p.[L41P];[G113C] -Foot-
Control -Palm-
p.[G113C;[G113C] -Palm-
*
*
Keratin 5/6
Keratin 10
Loricrin
*
*
Figure 4. Indirect immunofluorescence staining of control and patients’ skin. In patients’ skin, expression of corneodesmosin, loricrin, and keratin 10 was increased. Particularly, loricrin showed strikingly elevated levels in the patients, and, in contrast to the control skin, it was also present in the cornified layer. The antibody to loricrin gave a weak signal at the basal layer of the epidermis, which is likely to represent unspecific staining. Keratin 5/6 was not changed. Skin cleavage occurred within the stratum corneum (asterisk). Bars ¼ 50 mm.
2426 Journal of Investigative Dermatology (2012), Volume 132
M Pigors et al. TGM5 Mutations Impact Epidermal Differentiation
a
b g
i
t
n
c
d *
t
t t
t d v
v
d d
d
Figure 5. Transmission electron microscope analysis of acral peeling skin syndrome skin. (a) A detailed electron micrograph of a granular keratinocyte is shown with stellate keratohyalin granules (g), which were inhomogenously electron-dense and less electron-dense tonofilaments (t), and nuclear material (n). Original magnification 5,000. Bar ¼ 1,000 nm. (b) The membranes of the cells in the lower cornified layer were normal, and the intercellular dense bodies (i) were homogenous. Original magnification 10,000. Bar ¼ 500 nm. (c) Cleavage (asterisk) was present between the cornified cells, and the cytoplasm next to the split showed loosened and disrupted tonofilament bundles (t), but the cell membrane and cornified envelope appeared normal (v). Original magnification 5,000. Bar ¼ 1,000 nm. (d) Desmosomes (d) were intact between degenerated cells with loosened tonofilaments (t). Original magnification 10,000. Bar ¼ 500 nm.
granules with abnormal varying electron-dense and lessdense areas, in addition to homogenously dense areas of tonofibrils (Figure 5a). The keratinization of the lower horny cells, as well as the intercellular dense bodies, appeared unchanged (Figure 5b). Cleavage was observed between the cornified cells, and the cytoplasm next to the split showed loosened and disrupted tonofilament bundles (Figure 5c). Desmosomes were normal and intercellular debris was not found. The cell membrane and cornified envelope were also intact (Figure 5d). DISCUSSION Here, we describe 11 patients with APSS and four previously unreported TGM5 mutations, and explore their impact on epidermal differentiation. Notably, we identified the first truncating mutations in TGM5, although the majority of the patients harbored the common mutation p.G113C in either a heterozygous or homozygous state. For the first time, we also disclosed p.G113C in patients from Eastern and Southern Europe, proving it to be a recurrent mutation of possibly ancestral origin disseminated in Europe (Cassidy et al., 2005; Kiritsi et al., 2010; van der Velden et al., 2012). But more importantly, we identified this variant in a heterozygous state in 3 of 100 control individuals, suggesting that the carrier frequency and, consequently, APSS may actually be more widespread than anticipated. As the phenotype is mild, it could remain undiagnosed or misdiagnosed in most cases.
Patients in this study had mild features, but were referred to us during episodes of aggravation induced by heat and/or mechanical trauma. On the basis of experiments using recombinant enzyme, TGase 5 has been predicted to be a major contributor to epidermal differentiation, especially to the formation and assembly of the cornified envelope (Candi et al., 2002). In vivo, however, APSS is associated with a limited and mild clinical picture, which is in contrast to the severe phenotype of lamellar ichthyosis caused by TGM1 mutations. This suggests that TGase 1 and 3 are sufficient for a relatively stable cornified envelope, but environmental insults determine the disease manifestations. Although TGase 1 and 3 expression levels are not increased, common TGase substrates, IVL and LOR, are upregulated. IVL is an early differentiation marker that associates, together with envoplakin–periplakin heterotetramers, with the cell membrane of the keratinocytes. The crosslinking of these proteins at this stage is primarily performed by TGase 1 in the upper spinous layer (Ishida-Yamamoto et al., 1996; Steinert and Marekov, 1997; Kalinin et al., 2001). High levels of IVL in APSS, and the fact that tissue separation occurs between the granular and cornified layers or higher, suggest that early processes of cornified envelope formation are established properly. LOR, on the other hand, constitutes the main component of the cornified envelope (Candi et al., 2005) and is expressed during late stages of cornification to reinforce the already existing scaffold of cornified envelope precursor proteins (Yoneda and Steinert, 1993; Ishida-Yamamoto et al., 1996; Kalinin et al., 2001). Before the cross-linking process, LOR is sequestered in granules, in which it is predicted to complex with profilaggrin, and possibly also with components of the desmosomes (Yoneda et al., 1992, 2012; Ishida-Yamamoto et al., 1996). When TGase 5 activity is missing, incorporation of LOR into the cornified envelope is impaired and it accumulates in the granular and cornified layer, which may also explain the presence of abnormal keratohyalin granules in the granular cells. Similar to LOR, cross-linking of other components of the corneocytes will be destabilized by shearing forces and exposure to water. The ultrastructural findings are very discreet and similar to those reported previously (Hashimoto et al., 2000; Cassidy et al., 2005). The level of cleavage in APSS with TGM5 mutations is unclear (Cassidy et al., 2005; Kharfi et al., 2009), but in our case it was observed between the corneocytes. Further, we hypothesize that owing to the subtle changes in stability of the cornified envelope in the horny cells KRTs are upregulated. In this context, the focal disruption of tonofilament bundles appears very interesting and remains to be elucidated. Similarly, high levels of CDSN should stabilize cell–cell contacts as a compensatory mechanism for the fragile cornified envelope. In conclusion, our results indicate that TGM5 mutations induce upregulation of genes for proteins involved in epidermal differentiation, including KRTs 1 and 10, IVL, and LOR. As assembly of the cornified cell envelope can be adequately established in the early stages, presumably by the www.jidonline.org 2427
M Pigors et al. TGM5 Mutations Impact Epidermal Differentiation
activity of other TGases, it seems that in patients with APSS only the terminal formation of the cornified envelope, mainly consisting of LOR, is disturbed. Consequently, APSS patients show relatively mild skin fragility at sites of high mechanical stress. In other body regions that are less burdened, we assume that other TGases are sufficient to establish a stable cornified envelope that can resist the mechanical forces of the environment. MATERIALS AND METHODS Patients and clinical samples The diagnosis was based on clinical assessment and morphological analysis of skin biopsies. After written informed consent, skin biopsies and EDTA blood were obtained from the patients, and whenever possible EDTA blood from their parents for mutation analysis was also obtained. The project was approved by the ethics committee of the University of Freiburg and was conducted according to the Declaration of Helsinki Principles.
Mutation detection Genomic DNA was extracted from EDTA blood and cultured keratinocytes using the QiAmp DNA Mini Kit, as well as from cryosections using the QiAmp DNA FFPE Tissue Kit (both from Qiagen, Hilden, Germany). Amplification of all 15 TGM5 exons and exon/intron boundaries was performed as described earlier (Cassidy et al., 2005). All PCR products were submitted to automated nucleotide sequencing in an ABI 3130XL genetic analyzer using Big Dye Terminator Chemistry (Applied Biosystems, Darmstadt, Germany). DNA sequences were compared with the reference sequence from NCBI Entrez Nucleotide database (transcript NM_201631.2 and NG_016124.1) using the Mutation SurveyorTM DNA variant analysis software (version 2.61 Softgenetics, State College, PA). Mutations were confirmed by resequencing. Mutation verification in 100 unrelated control individuals was carried out by sequencing as well.
Cell culture Primary keratinocytes from patients’ and control skin were cultured in serum-free keratinocyte medium supplemented with 25 mg ml1 bovine pituitary extract and 0.2 ng ml1 recombinant epidermal growth factor (Invitrogen, Darmstadt, Germany) using standard methods (Herz et al., 2006). Upon reaching 80–90% confluency, the cells were treated with 1.2 mM CaCl2 for 7 days.
RNA extraction and quantitative real-time PCR Total RNA was isolated from keratinocytes using the RNeasy Plus Mini Kit (Qiagen), transcribed into complementary DNA (Fermentas, St Leon-Rot, Germany), and subjected to quantitative real-time PCR using the iQTM SYBRw Green Supermix and Biorad CFX96 RealTime PCR Detection System (both from Bio-Rad, Munich, Germany). The data were analyzed using the BioRad CFX Manager Software (version 1.5). Expression levels were calculated relative to those of hypoxanthine phosphoribosyltransferase (HPRT1), 18s RNA, and plakoglobin. Efficiencies were determined for each marker and were shown to be close to the efficiency of the normalizing marker. Relative expression was determined as 2DDCT. The control sample was set to 1, and expression levels in the patient cells were indicated as fold change compared with the control. 2428 Journal of Investigative Dermatology (2012), Volume 132
Protein extraction and immunoblot analysis Keratinocytes were lysed in a buffer containing 25 mM Tris-HCl, pH 8.0, 100 mM NaCl, 1 mM EDTA, 1 mM dithiothreitol, 1% NP-40, 1% Triton X-100, 1 mM protease inhibitor cocktail set III (Merck, Darmstadt, Germany), and 10 mM Pefabloc SC (Merck) for 1 hour with agitation at 4 1C, and clarified by centrifugation at 20,000 g for 20 minutes (Pietroni et al., 2008). Immunoblot analysis was performed as described elsewhere (Has et al., 2009) using primary antibodies to TGase 5, TGase 1 (both Santa Cruz, Heidelberg, Germany), IVL (clone SY5; Sigma, Munich, Germany), CDSN (Santa Cruz), and b-actin (clone AC; Sigma) to control loading. The signals obtained for IVL and CDSN from western blot analysis were quantified with the ImageJ gel analysis program (version 1.38, http://rsbweb.nih.gov/ij/) and normalized to b-actin.
Indirect immunofluorescence staining Indirect immunofluorescence staining of the patients’ and sitematched control skin was performed on 5-mm cryosections, which were air-dried and incubated with primary antibodies at room temperature overnight. Primary antibodies to the following proteins were used: TGase 5, CDSN (both Santa Cruz), LOR (Abcam, Cambridge, UK), KRT 10 (clone LH2; Santa Cruz), and KRT 5/6 (clone D5/16 B4; Dako, Glostrup, Denmark). The secondary antibodies were Alexa-488 anti-mouse IgG, Alexa-488 anti-goat IgG, or Alexa-488 anti-rabbit IgG (Invitrogen). Nuclei were stained with 40 ,6-diamidino-2-phenylindole (Millipore, Temecula, CA). Stained sections were observed with an Axiophot fluorescence microscope (Carl Zeiss, Jena, Germany). Images were captured using the Zeiss internal software. To quantify LOR expression, the fluorescence signal of four skin sections of each patient and matching control was quantified using ImageJ after background subtraction.
Transglutaminase activity assay The TGase activity assay was performed as described elsewhere (Raghunath et al., 1998; Cheng et al., 2010). In brief, 5-mm cryosections from patients’ and site-matched control skin were blocked with 1% BSA in 0.1 M Tris/HCl, pH 8.4 or 7.4, for 30 minutes at room temperature, and then incubated with the substrate buffer (at the respective pH) containing either 5 mM CaCl2 or 20 mM EDTA (for the negative control) and biotinylated-X-cadaverine (AnaSpec, San Jose, CA) for 2 hours at room temperature. The reaction was stopped by incubating the sections with PBS/10 mM EDTA for 5 minutes. The sections were washed two times with PBS and then incubated with Streptavidin-conjugated Alexa-488 (Invitrogen) for 1 hour, and washed three times in PBS. Sections were visualized with an Axiophot fluorescence microscope (Carl Zeiss). Images were captured using the Zeiss internal software.
Transmission electron microscopy For electron microscopy, small pieces of tissue were fixed in 3% glutaraldehyde and postfixed in 1% osmium tetroxide according to routine procedures. Ultrathin sections were contrasted with uranyl acetate and lead citrate before being studied in a Jeol JEM 100SX electron microscope. CONFLICT OF INTEREST The authors state no conflict of interest.
M Pigors et al. TGM5 Mutations Impact Epidermal Differentiation
ACKNOWLEDGMENTS We thank the patients and their families who participated in this study, as well as Vera Morand, Margit Schubert, and Ka¨the Thoma for expert technical assistance. We acknowledge Dr Emilie Leclerc and Dr Nathalie Jonca (CHU Purpan, Toulouse) for excellent scientific support. This work was supported by the Network Epidermolysis Bullosa Grant from the Federal Ministry for Education and Research (BMBF) and by the Excellence Initiative of the German Federal Governments and the Freiburg Institute for Advanced Studies FRIAS, School of Life Sciences–LifeNet.
REFERENCES Bowden PE (2011) Peeling skin syndrome: genetic defects in late terminal differentiation of the epidermis. J Invest Dermatol 131:561–4 Candi E, Oddi S, Paradisi A et al. (2002) Expression of transglutaminase 5 in normal and pathologic human epidermis. J Invest Dermatol 119: 670–7 Candi E, Oddi S, Terrinoni A et al. (2001) Transglutaminase 5 cross-links loricrin, involucrin, and small proline-rich proteins in vitro. J Biol Chem 276:35014–23 Candi E, Schmidt R, Melino G (2005) The cornified envelope: a model of cell death in the skin. Nat Rev Mol Cell Biol 6:328–40 Cassidy AJ, van Steensel MA, Steijlen PM et al. (2005) A homozygous missense mutation in TGM5 abolishes epidermal transglutaminase 5 activity and causes acral peeling skin syndrome. Am J Hum Genet 77:909–17 Cheng X, Jin J, Hu L et al. (2010) TRP channel regulates EGFR signaling in hair morphogenesis and skin barrier formation. Cell 141:331–43 Grenard P, Bates MK, Aeschlimann D (2001) Evolution of transglutaminase genes: identification of a transglutaminase gene cluster on human chromosome 15q15. Structure of the gene encoding transglutaminase X and a novel gene family member, transglutaminase Z. J Biol Chem 276:33066–78 Has C, Herz C, Zimina E et al. (2009) Kindlin-1 is required for RhoGTPasemediated lamellipodia formation in keratinocytes. Am J Pathol 175:1442–52 Hashimoto K, Hamzavi I, Tanaka K et al. (2000) Acral peeling skin syndrome. J Am Acad Dermatol 43:1112–9 Herz C, Aumailley M, Schulte C et al. (2006) Kindlin-1 is a phosphoprotein involved in regulation of polarity, proliferation, and motility of epidermal keratinocytes. J Biol Chem 281:36082–90
Kharfi M, El Fekih N, Ammar D et al. (2009) A missense mutation in TGM5 causes acral peeling skin syndrome in a Tunisian family. J Invest Dermatol 129:2512–5 Kiritsi D, Cosgarea I, Franzke CW et al. (2010) Acral peeling skin syndrome with TGM5 gene mutations may resemble epidermolysis bullosa simplex in young individuals. J Invest Dermatol 130:1741–6 Oji V, Eckl KM, Aufenvenne K et al. (2010a) Loss of corneodesmosin leads to severe skin barrier defect, pruritus, and atopy: unraveling the peeling skin disease. Am J Hum Genet 87:274–81 Oji V, Tadini G, Akiyama M et al. (2010b) Revised nomenclature and classification of inherited ichthyoses: results of the First Ichthyosis Consensus Conference in Soreze 2009. J Am Acad Dermatol 63:607–41 Pavlovic S, Krunic AL, Bulj TK et al. (2011) Acral peeling skin syndrome: a clinically and genetically heterogeneous disorder. Pediatr Dermatol 29:258–63 Pietroni V, Di Giorgi S, Paradisi A et al. (2008) Inactive and highly active, proteolytically processed transglutaminase-5 in epithelial cells. J Invest Dermatol 128:2760–6 Pillai S, Bikle DD, Mancianti ML et al. (1990) Calcium regulation of growth and differentiation of normal human keratinocytes: modulation of differentiation competence by stages of growth and extracellular calcium. J Cell Physiol 143:294–302 Proksch E, Brandner JM, Jensen JM (2008) The skin: an indispensable barrier. Exp Dermatol 17:1063–72 Raghunath M, Hennies HC, Velten F et al. (1998) A novel in situ method for the detection of deficient transglutaminase activity in the skin. Arch Dermatol Res 290:621–7 Ren J, Wen L, Gao X et al. (2009) DOG 1.0: illustrator of protein domain structures. Cell Res 19:271–3 Shwayder T, Conn S, Lowe L (1997) Acral peeling skin syndrome. Arch Dermatol 133:535–6 Steinert PM, Marekov LN (1997) Direct evidence that involucrin is a major early isopeptide cross-linked component of the keratinocyte cornified cell envelope. J Biol Chem 272:2021–30 van der Velden JJ, Jonkman MF, McLean WH et al. (2012) A recurrent mutation in the TGM5 gene in European patients with acral peeling skin syndrome. J Dermatol Sci 65:74–6 Yoneda K, Hohl D, McBride OW et al. (1992) The human loricrin gene. J Biol Chem 267:18060–6
Ishida-Yamamoto A, Eady RA, Watt FM et al. (1996) Immunoelectron microscopic analysis of cornified cell envelope formation in normal and psoriatic epidermis. J Histochem Cytochem 44:167–75
Yoneda K, Nakagawa T, Lawrence OT et al. (2012) Interaction of the Profilaggrin N-terminal domain with loricrin in human cultured keratinocytes and epidermis. J Invest Dermatol 132:1206–14
Kalinin A, Marekov LN, Steinert PM (2001) Assembly of the epidermal cornified cell envelope. J Cell Sci 114:3069–70
Yoneda K, Steinert PM (1993) Overexpression of human loricrin in transgenic mice produces a normal phenotype. Proc Natl Acad Sci USA 90:10754–8
www.jidonline.org 2429
Copyright of Journal of Investigative Dermatology is the property of Nature Publishing Group and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
TGM5 Mutations Impact Epidermal Differentiation in Acral Peeling Skin Syndrome Manuela Pigors1, Dimitra Kiritsi1, Cristina Cobzaru1, Agnes Schwieger-Briel1, Jose Sua´rez2, Flavio Faletra3, Heikki Aho4 , Leeni Ma¨kela¨5, Johannes S. Kern1, Leena Bruckner-Tuderman1,6 and Cristina Has1 Acral peeling skin syndrome (APSS) is an autosomal recessive skin disorder characterized by acral blistering and peeling of the outermost layers of the epidermis. It is caused by mutations in the gene for transglutaminase 5, TGM5. Here, we report on clinical and molecular findings in 11 patients and extend the TGM5 mutation database by four, to our knowledge, previously unreported mutations: p.M1T, p.L41P, p.L214CfsX15, and p.S604IfsX9. The recurrent mutation p.G113C was found in 9 patients, but also in 3 of 100 control individuals in a heterozygous state, indicating that APSS might be more widespread than hitherto expected. Using quantitative real-time PCR, immunoblotting, and immunofluorescence analysis, we demonstrate that expression and distribution of several epidermal differentiation markers and corneodesmosin (CDSN) is altered in APSS keratinocytes and skin. Although the expression of transglutaminases 1 and 3 was not changed, we found an upregulation of keratin 1, keratin 10, involucrin, loricrin, and CDSN, probably as compensatory mechanisms for stabilization of the epidermal barrier. Our results give insights into the consequences of TGM5 mutations on terminal epidermal differentiation. Journal of Investigative Dermatology (2012) 132, 2422–2429; doi:10.1038/jid.2012.166; published online 24 May 2012
INTRODUCTION The cornified layer constitutes the outermost compartment of the skin, thus forming a robust mechanical barrier against environmental insults. To accomplish this function, differentiating keratinocytes build the cornified envelope, which comprises a unique structure consisting of highly insoluble proteins, including loricrin (LOR), filaggrin, and involucrin (IVL), that are cross-linked by several transglutaminases (Candi et al., 2005). Genetic defects of the molecules associated with the cornified envelope lead to disorders of cornification such as ichthyoses and atopic dermatitis (Proksch et al., 2008; Oji et al., 2010b). Recently, the underlying genetic causes of particular types of ichthyoses, the peeling skin syndromes, have been elucidated: mutations in the corneodesmosin (CDSN) gene primarily impair the cohesion of corneocytes and lead to generalized 1
Department of Dermatology, University Freiburg Medical Center, Freiburg, Germany; 2Service of Dermatology, University Hospital Nuestra Sen˜ora de Candelaria Carretera del Rosario, Santa Cruz de Tenerife, Spain; 3Institute for Maternal and Child Health IRCCS ‘‘Burlo Garofolo’’, Trieste, Italy; 4 Department of Pathology, Turku University Hospital, Turku, Finland; 5 Department of Dermatology, Turku University Hospital, Turku, Finland and 6 Freiburg Institute for Advanced Studies, School of Life Sciences–LifeNet, University of Freiburg, Freiburg, Germany Correspondence: Cristina Has, Department of Dermatology, University Freiburg Medical Center, Hauptstrasse 7, 79104 Freiburg, Germany. E-mail: [email protected] Abbreviations: APSS, acral peeling skin syndrome; CDSN, corneodesmosin; IVL, involucrin; KRT, keratin; LOR, loricrin; TGase, transglutaminase protein; TGM, transglutaminase gene Received 21 December 2011; revised 21 March 2012; accepted 5 April 2012; published online 24 May 2012
2422 Journal of Investigative Dermatology (2012), Volume 132
peeling skin disease (Oji et al., 2010a; Bowden, 2011), whereas transglutaminase 5 (TGase 5), involved in crosslinking of the cornified envelope proteins, is affected in the acral peeling skin syndrome (APSS; Cassidy et al., 2005; Oji et al., 2010a; Pavlovic et al., 2011). Only little is known on the molecular disease mechanisms of APSS (Cassidy et al., 2005; Kiritsi et al., 2010). It is considered a rare autosomal recessive genodermatosis, characterized by painless, superficial blistering and peeling of hands and feet (Shwayder et al., 1997; Cassidy et al., 2005; Kiritsi et al., 2010). Thus far, three different TGM5 (transglutaminase 5 gene) missense mutations have been reported, including p.G113C, which is recurrent in the European population (Cassidy et al., 2005; Kharfi et al., 2009; Kiritsi et al., 2010; van der Velden et al., 2012). In this study, we describe 11 patients with APSS and extend the TGM5 mutation database by four, to our knowledge, previously unreported mutations: p.M1T, p.L41P, p.L214CfsX15, and p.S604IfsX9. The latter two are the first TGM5 mutations predicted to lead to premature termination codons. Furthermore, we show that TGM5 mutations significantly affect expression of epidermal differentiation markers and CDSN in vitro and in situ. RESULTS Clinical presentation and mutations
This study extends the number of 23 patients with TGM5 mutations, which have been reported previously in the literature, by 11 (Table 1; Kiritsi et al., 2010). Eight of eleven patients were children (patients 1–8 in Table 1, age between & 2012 The Society for Investigative Dermatology
M Pigors et al. TGM5 Mutations Impact Epidermal Differentiation
Table 1. Clinical features of the patients and mutations Age at diagnosis, origin of the patient
Age of onset
TGM5 mutations
Site of biopsy, level of skin cleavage
1
8 Months, Slovakian
Shortly after birth
c.[337G4T];[337G4T] p.[G113C];[G113C]
Lower leg, no cleavage
Peeling on volar aspects of hands and feet
2
1 Year, origin not known
7 Months
c.[2T4C];[2T4C] p.[M1T];[M1T]
Sole, intracorneal blister
Peeling, superficial blisters, and erosions on volar and dorsal aspects of hands and feet
3
1 Year, Italian
3 Months
c.[640delC]; [1811_1815delinsTCCTTCA] p.[L214CfsX15];[S604IfsX9]
No biopsy available
Peeling and erythematous patches on hands and feet; aggravation by heat, friction, perspiration, and exposure to water
4
2 Years, German
Not known
c.[337G4T];[337G4T] p.[G113C];[G113C]
Not known, no cleavage
Peeling and superficial blisters mainly on volar, but also on dorsal aspects of hands and feet
5
7 Years, German
6 Weeks
c.[122T4C];[337G4T] p.[L41P];[G113C]
Dorsum of the foot, no cleavage
Peeling and superficial blisters mainly on dorsal, but also on volar aspects of hands and feet; peeling on knees
6
7 Years, Spanish
11 Months
c.[337G4T];[337G4T] p.[G113C];[G113C]
Foot, intracorneal blister
Peeling and superficial blisters on volar and dorsal aspects of hands and feet; occasionally crusty erosions on knees and elbows; aggravation upon contact with water
7
11 Years, German Not known
c.[337G4T];[337G4T] p.[G113C];[G113C]
Foot, intracorneal blister
Not available
8
12 Years, German 3 Weeks
c.[337G4T];[640delC] p.[G113C];[L214CfsX15]
Not known, no cleavage
Large, superficial blisters on ankles; mild blistering and peeling on extremities
9
20 Years, Hungarian
Since birth
c.[337G4T];[337G4T] p.[G113C];[G113C]
Foot, intracorneal blister
Mild peeling on volar and dorsal aspects of hands and feet
10
27 Years, German Not known
c.[337G4T];[337G4T] p.[G113C];[G113C]
Palm, intracorneal Mild peeling on volar and dorsal aspects of hands and blister feet
11
33 Years, German Not known
c.[337G4T];[337G4T] p.[G113C];[G113C]
Lower arm, no cleavage
No.
8 months and 12 years) and presented with superficial blistering and peeling on volar and dorsal aspects of hands and feet (Figure 1a–c). It is noteworthy that patient 5 reported to have had erosions on knees after friction, and patient 8 had large, superficial blisters on the ankles (Figure 1c) and peeling on the legs, arms, and elbows. The three oldest patients (patients 9–11 in Table 1), 20, 27, and 33 years of age, only had peeling and were very mildly affected. In patients 1–4, 7, 8, and 11, epidermolysis bullosa simplex was initially suspected, and in patients 2, 3, and 7 keratin (KRT) 5 and 14 genes were analyzed and mutations excluded. We disclosed four previously unreported TGM5 mutations: c.2T4C, p.M1T in exon 1 (in patient 2), c.122T4C, p.L41P in exon 2 (in patient 5), c.640delC, p.L214CfsX15, in exon 5 (in patients 3 and 8), and c.1811_1815delinsTCCT TCA, p.S604IfsX9 in exon 11 (in patient 3; Figure 1b–d, Table 1). The first mutation affects the start codon and interferes with the initiation of translation, probably causing an N-terminal truncated TGase 5. This presumption is supported by the prediction of additional initiation sites in the complementary DNA sequence (http://atgpr.dbcls.jp/) and by the positive TGase 5 signal present in the skin of the patient (see below). The pathogenic role of the variation p.L41P is suggested by its absence in 100 control individuals
Clinical features
Mild peeling on volar and dorsal aspects of feet
and by the prediction to be ‘‘possibly damaging’’ (Polyphen, http://genetics.bwh.harvard.edu/pph/). The deletion c.640delC causes a frameshift and premature termination of translation, p.L214CfsX15, in the catalytic core domain. The last mutation, an insertion/deletion, leads to a frameshift and premature stop codon formation in the b-barrel 1 domain, p.S604IfsX9, probably leading to a C-terminal truncation of TGase 5. However, no material was available to verify whether frameshift mutations led to mRNA decay and/or truncated proteins. The recurrent mutation p.G113C was identified in seven patients in a homozygous state and in two patients in a heterozygous state (Table 1). Intriguingly, this mutation was also detected in a heterozygous state in 3 of 100 control individuals, together with the single-nucleotide polymorphism p.T109M in all patients and controls. The parents of patients 3, 5, 6, and 8 were available for analysis, and were heterozygous carriers of the mutations of their offspring. Impact of TGM5 mutations in keratinocytes
As data on TGase 5 functions mainly rely on experiments using recombinant protein (Candi et al., 2001, 2002; Pietroni et al., 2008), we used APSS as a genetic model to gain insight into the physiological functions of this enzyme in www.jidonline.org 2423
M Pigors et al. TGM5 Mutations Impact Epidermal Differentiation
a
b
2-Year-old
p.[L41P];[G113C]
T
390 G
T 53
260 G 53
G
G
T
C
395 G
C 53
265 G 53
G
260 A
A
C
c.337G>T, p.G113C G 53
G 53
12-Year-old p.[G113C];[L214CfsX15]
Control
Control G
c
7-Year-old
p.[G113C];[G113C]
T 53
C
265 C
T
Control A
C
C
G
C
G 44
115 C 44
T
c.122T>C, p.L41P A 56
A 44
C 44
C 35
C 35
250 T
G
C
G
G
120 C 29
G 29
G 44
c.640delC, p.L214CfsX15
135 G 53
C
A 46
C 56
C 53
T 26
T 20
G 29
d M1T L41P G113C
L214CfsX15 W255R
β-Sandwich
Catalytic core
1
S604IfsX9 K445N β-Barrel 1
β-Barrel 2 720
Figure 1. Clinical presentation of acral peeling skin syndrome and TGM5 mutations. (a–c) Patients presented with peeling and superficial blisters (arrows) mainly on volar and dorsal aspects of hands and feet. Note large erosions after blisters over the ankles in the 12-year-old patient who was wearing high-ankle leather shoes during summer. The lower panel shows chromatograms of controls and patients’ partial sequences of exon 3, 2, and 5 of TGM5 indicating the recurrent variation p.G113C, as well as the mutations p.L41P and p.L214CfsX15. (d) Schematic representation of transglutaminase 5 (TGase 5) and its domains (Grenard et al., 2001). The four, to our knowledge previously unreported, mutations are indicated in red. Both missense mutations, p.M1T and p.L41P, are located in the b-sandwich domain. The deletion/insertion mutations, p.L214CfsX15 and p.S604Ifsx9, are found in the catalytic core domain or b-barrel 1 domain of TGase 5, respectively. Previously reported mutations are shown in black. TGase 5 domain structures were designed with the DOG 2.0 software (Ren et al., 2009).
keratinocytes. We analyzed primary keratinocytes derived from two patients homozygous for the mutation p.G113C (designated as G113C-keratinocytes). Although cell culture is not an ideal model to study epidermal differentiation, treatment of keratinocytes with Ca2 þ for 7 days is known to induce expression of epidermal transglutaminases and main differentiation proteins (Pillai et al., 1990). These were analyzed at both mRNA and protein levels. Interestingly, the TGM5 mRNA levels were markedly increased (up to 6-fold) in G113C keratinocytes, as compared with normal keratinocytes (Figure 2a). In contrast, the expression of the genes for other transglutaminases present in the upper epidermal layers, transglutaminase 1 (TGM1) and 3 (TGM3), were comparable to the control. mRNA levels of TGase 5 substrates (Candi et al., 2005), LOR (4-fold), and IVL (3-fold), as well as KRT1 (up to 10-fold), were 2424 Journal of Investigative Dermatology (2012), Volume 132
upregulated, whereas that of KRT5, a marker of basal keratinocytes, remained unchanged (Figure 2a). Further, these results were validated at the protein level. TGase 5 is a highly insoluble protein present in several forms, including a monomeric form of 84 kDa and an autocatalytic, cross-linked form of higher molecular mass (Pietroni et al., 2008). As monomeric TGase 5 is only present to a low extent in differentiating keratinocytes in vitro (Candi et al., 2002), only a faint band was detected in all samples. In contrast, the autocatalytic, inactive form of TGase 5 was enhanced 2-fold in G113C keratinocytes (Figure 2b), presumably accounting for the mRNA upregulation. In agreement with the quantitative real-time PCR data, the level of transglutaminase 1 protein (TGase 1) was not changed to compensate for TGase 5 (Figure 2b). More importantly, the expression of IVL was increased 2-fold and that of CDSN, one of the most crucial
M Pigors et al.
a
Normalized fold expression
TGM5 Mutations Impact Epidermal Differentiation
Gene expression 10 8 6 4 2 0 TGM5
TGM1
TGM3
LOR
Control keratinocytes
b Control
TGase 5 G113C-KCs
Control
IVL
KRT1
KRT5
G113C-keratinocytes
TGase 1 G113C-KCs
Corneodesmosin Control G113C-KCs
250 95
72
55 95
Actin Involucrin G113C-KCs Control
72 130 55
36
Actin
Actin
Actin
Figure 2. Molecular characterization of acral peeling skin syndrome keratinocytes. (a) Keratinocytes from two patients homozygous for p.G113C (G113C-keratinocytes) and a control were treated for 7 days with Ca2 þ and used for RNA extraction for quantitative real-time PCR (a) and lysates for immunoblotting (b). Upregulation of the TGM5 transcript was observed in the G113C keratinocytes, whereas levels of TGM1 and TGM3 were comparable to that of control cells. The differentiation markers loricrin (LOR), involucrin (IVL), and keratin (KRT1) were strongly elevated in the patients’ cells. Keratin 5 (KRT5) remained unchanged. (b) Immunoblotting of control and G113C keratinocytes (G113C-KCs) lysates is shown. Transglutaminase 5 (TGase 5) monomeric form (lower arrow, 84 kDa) showed similar levels in control and patients’ lysates, whereas the autocatalytic form of TGase 5 (upper arrow) was markedly increased in the patients’ samples (2-fold). TGase 1 expression remained unchanged in patients’ cells. In contrast, involucrin and corneodesmosin were increased in the patients as compared with the control. Antibodies to b-actin were used to control loading. The numbers on the left side of each blot represent molecular weights in kilodaltons.
cell–cell adhesion molecules in the corneocytes, was increased up to 6-fold in G113C keratinocytes after 7 days of Ca2 þ stimulation (Figure 2b). Molecular characterization of APSS skin
Next, we validated the findings in situ by indirect immunofluorescence staining of the skin of patients 2, 5, and 10, all of them bearing different mutation constellations. All patients and control samples showed positive signals for TGase 5 confined to the cell membranes, suggesting that the enzymatic activity, rather than the expression or localization, is altered (Figure 3). To investigate this further, we performed an in situ TGase activity assay. Although this test does not discriminate among the activities of the epidermal TGases, it is known that TGase 1 is active at both pH 8.4 and 7.4, and the optimal pH for TGase 3 and 5 is considered to be B8.4 (Raghunath et al., 1998; Candi et al., 2001). We found a diminution of TGase activity at pH 8.4 in the skin of all patients, which was more pronounced in the patient
homozygous for p.M1T. At pH 7.4, TGase activity was preserved in the skin of all three patients (Figure 3). At both pH conditions, negative controls showed no fluorescent signal at the cell periphery. In agreement with the in vitro data, the signals for CDSN and LOR were increased in all patients, independent of the mutations. LOR, in particular, accumulated in the granular layer and shifted to the upper stratum corneum in the skin of all three patients, which showed a 2-fold increase of the intensity of the LOR signal, in contrast to the controls whose expression was lower and restricted to the granular layer. In addition, staining for KRT 10 appeared slightly increased in the patients’ skin. The abnormal signal within the stratum corneum is most probably due to false-positive antibody reaction. KRT 5/6 remained unchanged (Figure 4). In addition, transmission electron microscopy was performed to unravel more subtle morphological changes in the APSS skin with the homozygous mutation p.G113C. In particular, the granular cells contained stellate keratohyalin www.jidonline.org 2425
M Pigors et al. TGM5 Mutations Impact Epidermal Differentiation
p.[M1T];[M1T] -Sole-
p.[L41];[G113C] -Foot-
Control -Palm-
p.[G113C;[G113C] -Palm-
TGase activity pH 7.4
TGase activity pH 8.4
TGase 5
Control -Sole-
Figure 3. Transglutaminase 5 (TGase 5) expression and TGase activity assay in acral peeling skin syndrome skin. Indirect immunofluorescence showed TGase 5 expression confined to the cell membrane of keratinocytes in all three patients, predominantly in the granular layer of the epidermis. A TGase activity assay was performed at pH 8.4 and 7.4. The activity at pH 8.4 was reduced at the cell periphery in the upper spinous and granular layers of the skin of all patients, especially in the patient homozygous for p.M1T. At pH 7.4, TGase activity in patients’s skin was preserved. TGase 2 is active at the level of the basement membrane zone and dermis. Bars ¼ 50 mm.
Corneodesmosin
Control -Sole-
p.[M1T];[M1T] -Sole-
p.[L41P];[G113C] -Foot-
Control -Palm-
p.[G113C;[G113C] -Palm-
*
*
Keratin 5/6
Keratin 10
Loricrin
*
*
Figure 4. Indirect immunofluorescence staining of control and patients’ skin. In patients’ skin, expression of corneodesmosin, loricrin, and keratin 10 was increased. Particularly, loricrin showed strikingly elevated levels in the patients, and, in contrast to the control skin, it was also present in the cornified layer. The antibody to loricrin gave a weak signal at the basal layer of the epidermis, which is likely to represent unspecific staining. Keratin 5/6 was not changed. Skin cleavage occurred within the stratum corneum (asterisk). Bars ¼ 50 mm.
2426 Journal of Investigative Dermatology (2012), Volume 132
M Pigors et al. TGM5 Mutations Impact Epidermal Differentiation
a
b g
i
t
n
c
d *
t
t t
t d v
v
d d
d
Figure 5. Transmission electron microscope analysis of acral peeling skin syndrome skin. (a) A detailed electron micrograph of a granular keratinocyte is shown with stellate keratohyalin granules (g), which were inhomogenously electron-dense and less electron-dense tonofilaments (t), and nuclear material (n). Original magnification 5,000. Bar ¼ 1,000 nm. (b) The membranes of the cells in the lower cornified layer were normal, and the intercellular dense bodies (i) were homogenous. Original magnification 10,000. Bar ¼ 500 nm. (c) Cleavage (asterisk) was present between the cornified cells, and the cytoplasm next to the split showed loosened and disrupted tonofilament bundles (t), but the cell membrane and cornified envelope appeared normal (v). Original magnification 5,000. Bar ¼ 1,000 nm. (d) Desmosomes (d) were intact between degenerated cells with loosened tonofilaments (t). Original magnification 10,000. Bar ¼ 500 nm.
granules with abnormal varying electron-dense and lessdense areas, in addition to homogenously dense areas of tonofibrils (Figure 5a). The keratinization of the lower horny cells, as well as the intercellular dense bodies, appeared unchanged (Figure 5b). Cleavage was observed between the cornified cells, and the cytoplasm next to the split showed loosened and disrupted tonofilament bundles (Figure 5c). Desmosomes were normal and intercellular debris was not found. The cell membrane and cornified envelope were also intact (Figure 5d). DISCUSSION Here, we describe 11 patients with APSS and four previously unreported TGM5 mutations, and explore their impact on epidermal differentiation. Notably, we identified the first truncating mutations in TGM5, although the majority of the patients harbored the common mutation p.G113C in either a heterozygous or homozygous state. For the first time, we also disclosed p.G113C in patients from Eastern and Southern Europe, proving it to be a recurrent mutation of possibly ancestral origin disseminated in Europe (Cassidy et al., 2005; Kiritsi et al., 2010; van der Velden et al., 2012). But more importantly, we identified this variant in a heterozygous state in 3 of 100 control individuals, suggesting that the carrier frequency and, consequently, APSS may actually be more widespread than anticipated. As the phenotype is mild, it could remain undiagnosed or misdiagnosed in most cases.
Patients in this study had mild features, but were referred to us during episodes of aggravation induced by heat and/or mechanical trauma. On the basis of experiments using recombinant enzyme, TGase 5 has been predicted to be a major contributor to epidermal differentiation, especially to the formation and assembly of the cornified envelope (Candi et al., 2002). In vivo, however, APSS is associated with a limited and mild clinical picture, which is in contrast to the severe phenotype of lamellar ichthyosis caused by TGM1 mutations. This suggests that TGase 1 and 3 are sufficient for a relatively stable cornified envelope, but environmental insults determine the disease manifestations. Although TGase 1 and 3 expression levels are not increased, common TGase substrates, IVL and LOR, are upregulated. IVL is an early differentiation marker that associates, together with envoplakin–periplakin heterotetramers, with the cell membrane of the keratinocytes. The crosslinking of these proteins at this stage is primarily performed by TGase 1 in the upper spinous layer (Ishida-Yamamoto et al., 1996; Steinert and Marekov, 1997; Kalinin et al., 2001). High levels of IVL in APSS, and the fact that tissue separation occurs between the granular and cornified layers or higher, suggest that early processes of cornified envelope formation are established properly. LOR, on the other hand, constitutes the main component of the cornified envelope (Candi et al., 2005) and is expressed during late stages of cornification to reinforce the already existing scaffold of cornified envelope precursor proteins (Yoneda and Steinert, 1993; Ishida-Yamamoto et al., 1996; Kalinin et al., 2001). Before the cross-linking process, LOR is sequestered in granules, in which it is predicted to complex with profilaggrin, and possibly also with components of the desmosomes (Yoneda et al., 1992, 2012; Ishida-Yamamoto et al., 1996). When TGase 5 activity is missing, incorporation of LOR into the cornified envelope is impaired and it accumulates in the granular and cornified layer, which may also explain the presence of abnormal keratohyalin granules in the granular cells. Similar to LOR, cross-linking of other components of the corneocytes will be destabilized by shearing forces and exposure to water. The ultrastructural findings are very discreet and similar to those reported previously (Hashimoto et al., 2000; Cassidy et al., 2005). The level of cleavage in APSS with TGM5 mutations is unclear (Cassidy et al., 2005; Kharfi et al., 2009), but in our case it was observed between the corneocytes. Further, we hypothesize that owing to the subtle changes in stability of the cornified envelope in the horny cells KRTs are upregulated. In this context, the focal disruption of tonofilament bundles appears very interesting and remains to be elucidated. Similarly, high levels of CDSN should stabilize cell–cell contacts as a compensatory mechanism for the fragile cornified envelope. In conclusion, our results indicate that TGM5 mutations induce upregulation of genes for proteins involved in epidermal differentiation, including KRTs 1 and 10, IVL, and LOR. As assembly of the cornified cell envelope can be adequately established in the early stages, presumably by the www.jidonline.org 2427
M Pigors et al. TGM5 Mutations Impact Epidermal Differentiation
activity of other TGases, it seems that in patients with APSS only the terminal formation of the cornified envelope, mainly consisting of LOR, is disturbed. Consequently, APSS patients show relatively mild skin fragility at sites of high mechanical stress. In other body regions that are less burdened, we assume that other TGases are sufficient to establish a stable cornified envelope that can resist the mechanical forces of the environment. MATERIALS AND METHODS Patients and clinical samples The diagnosis was based on clinical assessment and morphological analysis of skin biopsies. After written informed consent, skin biopsies and EDTA blood were obtained from the patients, and whenever possible EDTA blood from their parents for mutation analysis was also obtained. The project was approved by the ethics committee of the University of Freiburg and was conducted according to the Declaration of Helsinki Principles.
Mutation detection Genomic DNA was extracted from EDTA blood and cultured keratinocytes using the QiAmp DNA Mini Kit, as well as from cryosections using the QiAmp DNA FFPE Tissue Kit (both from Qiagen, Hilden, Germany). Amplification of all 15 TGM5 exons and exon/intron boundaries was performed as described earlier (Cassidy et al., 2005). All PCR products were submitted to automated nucleotide sequencing in an ABI 3130XL genetic analyzer using Big Dye Terminator Chemistry (Applied Biosystems, Darmstadt, Germany). DNA sequences were compared with the reference sequence from NCBI Entrez Nucleotide database (transcript NM_201631.2 and NG_016124.1) using the Mutation SurveyorTM DNA variant analysis software (version 2.61 Softgenetics, State College, PA). Mutations were confirmed by resequencing. Mutation verification in 100 unrelated control individuals was carried out by sequencing as well.
Cell culture Primary keratinocytes from patients’ and control skin were cultured in serum-free keratinocyte medium supplemented with 25 mg ml1 bovine pituitary extract and 0.2 ng ml1 recombinant epidermal growth factor (Invitrogen, Darmstadt, Germany) using standard methods (Herz et al., 2006). Upon reaching 80–90% confluency, the cells were treated with 1.2 mM CaCl2 for 7 days.
RNA extraction and quantitative real-time PCR Total RNA was isolated from keratinocytes using the RNeasy Plus Mini Kit (Qiagen), transcribed into complementary DNA (Fermentas, St Leon-Rot, Germany), and subjected to quantitative real-time PCR using the iQTM SYBRw Green Supermix and Biorad CFX96 RealTime PCR Detection System (both from Bio-Rad, Munich, Germany). The data were analyzed using the BioRad CFX Manager Software (version 1.5). Expression levels were calculated relative to those of hypoxanthine phosphoribosyltransferase (HPRT1), 18s RNA, and plakoglobin. Efficiencies were determined for each marker and were shown to be close to the efficiency of the normalizing marker. Relative expression was determined as 2DDCT. The control sample was set to 1, and expression levels in the patient cells were indicated as fold change compared with the control. 2428 Journal of Investigative Dermatology (2012), Volume 132
Protein extraction and immunoblot analysis Keratinocytes were lysed in a buffer containing 25 mM Tris-HCl, pH 8.0, 100 mM NaCl, 1 mM EDTA, 1 mM dithiothreitol, 1% NP-40, 1% Triton X-100, 1 mM protease inhibitor cocktail set III (Merck, Darmstadt, Germany), and 10 mM Pefabloc SC (Merck) for 1 hour with agitation at 4 1C, and clarified by centrifugation at 20,000 g for 20 minutes (Pietroni et al., 2008). Immunoblot analysis was performed as described elsewhere (Has et al., 2009) using primary antibodies to TGase 5, TGase 1 (both Santa Cruz, Heidelberg, Germany), IVL (clone SY5; Sigma, Munich, Germany), CDSN (Santa Cruz), and b-actin (clone AC; Sigma) to control loading. The signals obtained for IVL and CDSN from western blot analysis were quantified with the ImageJ gel analysis program (version 1.38, http://rsbweb.nih.gov/ij/) and normalized to b-actin.
Indirect immunofluorescence staining Indirect immunofluorescence staining of the patients’ and sitematched control skin was performed on 5-mm cryosections, which were air-dried and incubated with primary antibodies at room temperature overnight. Primary antibodies to the following proteins were used: TGase 5, CDSN (both Santa Cruz), LOR (Abcam, Cambridge, UK), KRT 10 (clone LH2; Santa Cruz), and KRT 5/6 (clone D5/16 B4; Dako, Glostrup, Denmark). The secondary antibodies were Alexa-488 anti-mouse IgG, Alexa-488 anti-goat IgG, or Alexa-488 anti-rabbit IgG (Invitrogen). Nuclei were stained with 40 ,6-diamidino-2-phenylindole (Millipore, Temecula, CA). Stained sections were observed with an Axiophot fluorescence microscope (Carl Zeiss, Jena, Germany). Images were captured using the Zeiss internal software. To quantify LOR expression, the fluorescence signal of four skin sections of each patient and matching control was quantified using ImageJ after background subtraction.
Transglutaminase activity assay The TGase activity assay was performed as described elsewhere (Raghunath et al., 1998; Cheng et al., 2010). In brief, 5-mm cryosections from patients’ and site-matched control skin were blocked with 1% BSA in 0.1 M Tris/HCl, pH 8.4 or 7.4, for 30 minutes at room temperature, and then incubated with the substrate buffer (at the respective pH) containing either 5 mM CaCl2 or 20 mM EDTA (for the negative control) and biotinylated-X-cadaverine (AnaSpec, San Jose, CA) for 2 hours at room temperature. The reaction was stopped by incubating the sections with PBS/10 mM EDTA for 5 minutes. The sections were washed two times with PBS and then incubated with Streptavidin-conjugated Alexa-488 (Invitrogen) for 1 hour, and washed three times in PBS. Sections were visualized with an Axiophot fluorescence microscope (Carl Zeiss). Images were captured using the Zeiss internal software.
Transmission electron microscopy For electron microscopy, small pieces of tissue were fixed in 3% glutaraldehyde and postfixed in 1% osmium tetroxide according to routine procedures. Ultrathin sections were contrasted with uranyl acetate and lead citrate before being studied in a Jeol JEM 100SX electron microscope. CONFLICT OF INTEREST The authors state no conflict of interest.
M Pigors et al. TGM5 Mutations Impact Epidermal Differentiation
ACKNOWLEDGMENTS We thank the patients and their families who participated in this study, as well as Vera Morand, Margit Schubert, and Ka¨the Thoma for expert technical assistance. We acknowledge Dr Emilie Leclerc and Dr Nathalie Jonca (CHU Purpan, Toulouse) for excellent scientific support. This work was supported by the Network Epidermolysis Bullosa Grant from the Federal Ministry for Education and Research (BMBF) and by the Excellence Initiative of the German Federal Governments and the Freiburg Institute for Advanced Studies FRIAS, School of Life Sciences–LifeNet.
REFERENCES Bowden PE (2011) Peeling skin syndrome: genetic defects in late terminal differentiation of the epidermis. J Invest Dermatol 131:561–4 Candi E, Oddi S, Paradisi A et al. (2002) Expression of transglutaminase 5 in normal and pathologic human epidermis. J Invest Dermatol 119: 670–7 Candi E, Oddi S, Terrinoni A et al. (2001) Transglutaminase 5 cross-links loricrin, involucrin, and small proline-rich proteins in vitro. J Biol Chem 276:35014–23 Candi E, Schmidt R, Melino G (2005) The cornified envelope: a model of cell death in the skin. Nat Rev Mol Cell Biol 6:328–40 Cassidy AJ, van Steensel MA, Steijlen PM et al. (2005) A homozygous missense mutation in TGM5 abolishes epidermal transglutaminase 5 activity and causes acral peeling skin syndrome. Am J Hum Genet 77:909–17 Cheng X, Jin J, Hu L et al. (2010) TRP channel regulates EGFR signaling in hair morphogenesis and skin barrier formation. Cell 141:331–43 Grenard P, Bates MK, Aeschlimann D (2001) Evolution of transglutaminase genes: identification of a transglutaminase gene cluster on human chromosome 15q15. Structure of the gene encoding transglutaminase X and a novel gene family member, transglutaminase Z. J Biol Chem 276:33066–78 Has C, Herz C, Zimina E et al. (2009) Kindlin-1 is required for RhoGTPasemediated lamellipodia formation in keratinocytes. Am J Pathol 175:1442–52 Hashimoto K, Hamzavi I, Tanaka K et al. (2000) Acral peeling skin syndrome. J Am Acad Dermatol 43:1112–9 Herz C, Aumailley M, Schulte C et al. (2006) Kindlin-1 is a phosphoprotein involved in regulation of polarity, proliferation, and motility of epidermal keratinocytes. J Biol Chem 281:36082–90
Kharfi M, El Fekih N, Ammar D et al. (2009) A missense mutation in TGM5 causes acral peeling skin syndrome in a Tunisian family. J Invest Dermatol 129:2512–5 Kiritsi D, Cosgarea I, Franzke CW et al. (2010) Acral peeling skin syndrome with TGM5 gene mutations may resemble epidermolysis bullosa simplex in young individuals. J Invest Dermatol 130:1741–6 Oji V, Eckl KM, Aufenvenne K et al. (2010a) Loss of corneodesmosin leads to severe skin barrier defect, pruritus, and atopy: unraveling the peeling skin disease. Am J Hum Genet 87:274–81 Oji V, Tadini G, Akiyama M et al. (2010b) Revised nomenclature and classification of inherited ichthyoses: results of the First Ichthyosis Consensus Conference in Soreze 2009. J Am Acad Dermatol 63:607–41 Pavlovic S, Krunic AL, Bulj TK et al. (2011) Acral peeling skin syndrome: a clinically and genetically heterogeneous disorder. Pediatr Dermatol 29:258–63 Pietroni V, Di Giorgi S, Paradisi A et al. (2008) Inactive and highly active, proteolytically processed transglutaminase-5 in epithelial cells. J Invest Dermatol 128:2760–6 Pillai S, Bikle DD, Mancianti ML et al. (1990) Calcium regulation of growth and differentiation of normal human keratinocytes: modulation of differentiation competence by stages of growth and extracellular calcium. J Cell Physiol 143:294–302 Proksch E, Brandner JM, Jensen JM (2008) The skin: an indispensable barrier. Exp Dermatol 17:1063–72 Raghunath M, Hennies HC, Velten F et al. (1998) A novel in situ method for the detection of deficient transglutaminase activity in the skin. Arch Dermatol Res 290:621–7 Ren J, Wen L, Gao X et al. (2009) DOG 1.0: illustrator of protein domain structures. Cell Res 19:271–3 Shwayder T, Conn S, Lowe L (1997) Acral peeling skin syndrome. Arch Dermatol 133:535–6 Steinert PM, Marekov LN (1997) Direct evidence that involucrin is a major early isopeptide cross-linked component of the keratinocyte cornified cell envelope. J Biol Chem 272:2021–30 van der Velden JJ, Jonkman MF, McLean WH et al. (2012) A recurrent mutation in the TGM5 gene in European patients with acral peeling skin syndrome. J Dermatol Sci 65:74–6 Yoneda K, Hohl D, McBride OW et al. (1992) The human loricrin gene. J Biol Chem 267:18060–6
Ishida-Yamamoto A, Eady RA, Watt FM et al. (1996) Immunoelectron microscopic analysis of cornified cell envelope formation in normal and psoriatic epidermis. J Histochem Cytochem 44:167–75
Yoneda K, Nakagawa T, Lawrence OT et al. (2012) Interaction of the Profilaggrin N-terminal domain with loricrin in human cultured keratinocytes and epidermis. J Invest Dermatol 132:1206–14
Kalinin A, Marekov LN, Steinert PM (2001) Assembly of the epidermal cornified cell envelope. J Cell Sci 114:3069–70
Yoneda K, Steinert PM (1993) Overexpression of human loricrin in transgenic mice produces a normal phenotype. Proc Natl Acad Sci USA 90:10754–8
www.jidonline.org 2429
Copyright of Journal of Investigative Dermatology is the property of Nature Publishing Group and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.