Noor Almaani

Revertant Mosaicism in Recessive Dystrophic Epidermolysis Bullosa

TO THE EDITOR Revertant mosaicism refers to the presence of two genetically heterogeneous populations of cells as a result of spontaneous genetic correction during mitosis (Hall, 1988; Jonkman et al., 1997). This phenomenon has been reported in several inherited diseases, including severe combined immunodeficiency, Bloom’s syndrome, Fanconi’s anemia, X-linked Wiscott–Aldrich syndrome, Duchenne muscular dystrophy, and tyrosinemia type I (Hirschhorn, 2003). With regard to genodermatoses, cutaneous revertant mosaicism has been described in epidermolysis bullosa (EB) (Fine et al., 2008). Notably, in vivo reversion of mutations in LAMB3, COL17A1, and KRT14 has underscored cutaneous mosaicism in non-Herlitz junctional EB and EB simplex, respectively (Darling et al., 1999; Schuilenga-Hut et al., 2002; Smith et al., 2004; Pasmooij et al., 2005, 2007; Jonkman and Pasmooij, 2009). Of potential clinical interest, such genetic events may not be that rare— perhaps occurring in up to one-third of cases of non-Herlitz junctional EB (Jonkman and Pasmooij, 2009). Multiple corrective mechanisms have been proposed or observed, including back mutations, intragenic crossovers, mitotic gene conversions, and second-site mutations (Jonkman et al., 1997; Pasmooij et al., 2005; Frank and Happle, 2007). Indeed, several different corrective processes can occur in the same patient (Jonkman and Pasmooij, 2009). The implications for phenotype, however, depend on several factors, including the timing and extent of the revertant mosaicism. Here we report a further example of revertant mosaicism in a different sub-type of EB, with probable intragenic crossover in the COL7A1 gene leading to restoration of basement membrane collagen VII and anchoring fibrils in a patch of skin in an individual with recessive dystrophic EB. The proband is a 41-year-old Caucasian British man with severe generalized recessive dystrophic EB (Fine et al., 2008). He has mutilating scars with bilateral mitten deformities and a history of recurrent squamous cell carcinomas. His skin is prone to traumainduced blistering, although for as long as he can remember, two small patches of skin on his left wrist and right shin never seem to blister despite repeated trauma. Examination of these sites revealed areas approximately 8 5 cm that resembled the normal skin in appearance and texture (Figure 1a and b). To explain the phenotypic heterogeneity, and following ethics committee approval (St Thomas’ Hospital Ethics Committee: 07/H0802/104) and informed consent and in accordance with the Declaration of Helsinki principles, skin from the left wrist was investigated with biopsy specimens taken from both the blister-prone area (unreverted) and the normal-appearing skin (reverted). Immunolabeling for collagen VII (clone LH 7.2; Sigma-Aldrich, Poole, UK) and transmission electron microscopy were performed as described elsewhere (McGrath et al., 1993) and the results are illustrated in Figure 1c–e. Sequencing of peripheral leukocyte genomic DNA revealed that the patient is a compound heterozygote for two loss-of-function mutations in COL7A1, c.1732C4T (p.Arg578X) in exon 13 (maternal) and c.7786delG (p.Gly2593fsX4) in exon 104 (paternal) (Figure 2a). Both of these are recurrent mutations within the white British population (Mellerio et al., 1997). To explain the heterogeneous skin phenotype, genomic DNA and RNA were extracted from whole skin from both unreverted and reverted areas using standard kits (DNA Extraction Minikit and RNAEasy Minikit, Qiagen, Crawley, UK), as well as from cultured fibroblasts from both sites (TRIzol, Invitrogen, Paisley, UK) using the manufacturer’s protocols. The fibroblast cultures were performed using standard methods (Wong et al., 2008). cDNA was generated using commercial kits and protocols (IScript cDNA generation kit, Biorad, Hemel Hempstead, UK). Reverse transcriptase-PCR was performed across the sites of both mutations and the results are illustrated in Figure 2b. Real-time reverse transcriptase-PCR was also carried out to assess COL7A1 gene expression in the different skin and cell samples with primers (details available on request) specifically designed to amplify a 200-bp region of the COL7A1 30UTR (SyberGreen Master Mix, Applied Biosystems, Warrington, UK). Reverse transcriptasePCR for each cDNA sample was carried out in triplicate and the results are illustrated in Figure 2c. Collectively, these investigations indicated that the reverted skin expressed collagen VII and that anchoring fibrils were present. In addition, the reverse transcriptasePCR data demonstrated that there was expression of wild-type cDNA spanning the frameshift mutation in exon 104 and that COL7A1 gene expression levels were similar to those seen in the patient’s brother who was heterozygous for one mutant COL7A1 allele (Figure 2c). Moreover, the gene correction in the patient’s reverted skin appeared to have occurred in keratinocytes rather than fibroblasts. To explore the mechanism of this correction, we performed long-range sequencing of the patient’s reverted skin cDNA using LongAmp Taq DNA polymerase (New England Biolabs, Hitchin, UK). We first searched for polymorphisms to distinguish between maternal and paternal alleles and identified differences for a common PvuII polymorphism in exon

HB-EGF induces COL7A1 expression in keratinocytes and fibroblasts: Possible mechanism underlying allogeneic fibroblast therapy in recessive dystrophic epidermolysis bullosa

TO THE EDITOR Recessive dystrophic epidermolysis bullosa (RDEB) is a mechanobullous disease caused by mutations in the COL7A1 gene that encodes type VII collagen (C7) at the dermal–epidermal junction (DEJ) (Fine et al., 2008). The C7 protein is synthesized by both keratinocytes and fibroblasts (Stanley et al., 1985). We demonstrated previously that intradermal injections of allogeneic fibroblasts in RDEB can increase C7 expression at the DEJ, although that study did not disclose how long the benefits were sustained for or indicate a therapeutic mode of action (Wong et al., 2008). Allogeneic fibroblasts could not be detected 2 weeks after injection but, in some individuals, there was increased C7 protein at the DEJ for at least 3 months. To investigate this further, we injected 5 10 per cm normal control allogeneic fibroblasts (ICX-RHY, Vavelta, Intercytex; Manchester, UK) into one RDEB subject and took biopsies at days 7, 15, 30, 90, 180, 270, and 360. Each injection volume was 0.25 ml cm ; we also injected a similar volume of normal saline to adjacent skin (noting observations by Venugopal et al. (2010) that saline can also increase C7 in RDEB skin) and took biopsies at days 15 and 90. Clinicopathological details of the subject studied have been published previously (case 5; Wong et al., 2008). This patient is a compound heterozygote for the COL7A1 mutations c.2044C4T (p.Arg682X) and IVS87þ 4A4G. This particular donor splice-site mutation creates a leaky splice site, which leads to either in-frame skipping of exon 87 (69-bp) or wild-type sequence; this also allows for tracking of the mutant allele in skin biopsy complementary DNA. We assessed C7 immunolabeling at the DEJ (as described previously; Wong et al., 2008), COL7A1 gene expression by quantitative real-time RT-PCR (using a primer pair upstream of the splice-site mutation and another pair spanning the potential exon skip), and gene expression profiling using Sentrix Human-6 Whole Genome Expression Beadchips (Illumina, San Diego, CA); see Supplementary Materials and Methods online for full details. Quantification of immunofluorescence microscopy intensities for C7 labeling (see Wong et al., 2008 for Materials and Methods) is detailed in Supplementary Table 1 online. C7 labeling increased 15 days after fibroblast injection and was maintained for at least 270 days but returned to baseline levels by 360 days. For the saline control, at day 15, we noted a slight increase in labeling at the DEJ but a return to baseline by 90 days. COL7A1 gene expression, using both sets of primers (Figure 1a and b) showed that following fibroblast injection, there was a 420-fold increase at days 15 and 90, but this returned to baseline at day 180 and thereafter. Saline injection also resulted in a approximately 5-fold increase in patient skin COL7A1 gene expression at day 15, but levels at day 90 were similar to baseline (Figure 1a and b). For the COL7A1 primers spanning the splicesite mutation, at baseline, the ratio of wild-type to in-frame exon skip of exon 87 transcripts was approximately 2:3 (Figure 1b): this expression ratio persisted in all biopsy material. With regard to mechanism, no differences were noted in gene expression for cytokines already known to increase COL7A1 (full details are shown in Supplementary Tables 2–10 online). Of note, however, we observed a 43-fold increase in expression of the gene for heparin-binding epidermal growth factor-like growth factor (HB-EGF) (Iwamoto and Mekada, 2000), and quantitative real-time RT-PCR showed a similar temporal pattern to the COL7A1 quantitative real-time RTPCR data (linear correlation, r1⁄40.978; Po0.0001; compare HB-EGF data in Figure 1c with COL7A1 expression in Figure 1a and b). We also noted that gene expression profiles of FOS (linear correlation, r1⁄40.864; Po0.0006) and JUN (linear correlation, r1⁄40.945; Po0.0001) were also highly similar to the pattern of increased COL7A1 expression at the different time points (Supplementary Tables 2–10 online). JUN and FOS form the AP-1 transcription complex, which can bind to the COL7A1 promoter and enhance gene expression (Nakano et al., 2001). To investigate whether the upregulation of HB-EGF might be related to the increased expression of COL7A1, subconfluent-cultured keratinocytes and fibroblasts, both normal control and from two subjects with RDEB (this study individual and an unrelated subject with the COL7A1 mutations c.1732C4T (p.Arg578X) and c.7786delG (p.Gly2593fsX4), were treated with 100 ng ml 1 recombinant HB-EGF protein (R&D Systems,

A systematic review of the literature on the treatment of pityriasis rubra pilaris type 1 with TNF-antagonists

Background  Adult pityriasis rubra pilaris (PRP) type 1 is a rare chronic papulosquamous disorder with clinical and histological parallels with psoriasis. Treatment is challenging and recent case reports suggest a potential role for tumour necrosis factor (TNF) antagonists.

New Glycine Substitution Mutations in Type VII Collagen Underlying Epidermolysis Bullosa Pruriginosa but the Phenotype is not Explained by a Common Polymorphism in the Matrix Metalloproteinase-1 Gene Promoter

Epidermolysis bullosa (EB) pruriginosa is an unusual variant of dystrophic EB in which intense itching can lead to striking skin changes resembling acquired skin disorders such as nodular prurigo or hypertrophic lichen planus. The molecular pathology involves mutations in the COL7A1 gene, but the nature of the mutations is similar to those seen in other non-pruritic forms of dystrophic EB. The mechanism of the dramatic phenotypic differences is currently unknown. In this study we assessed the incidence of a common functional polymor-phism in the matrix metalloproteinase-1 gene promoter (1G or 2G at nucleotide -1607) in individuals with EB pruriginosa (n = 27) compared with non-itchy dominant dystrophic EB (n = 23), recessive dystrophic EB (n = 25) and normal controls (n = 50). The hypothesis is that the 2G allele, which was previously shown to increase matrix metalloproteinase-1 activity and lead to increased degradation of type VII collagen, could explain the phenotypic heterogeneity encountered in dominant forms of EB, particularly the itchy EB pruriginosa phenotype. The rationale is that increased type VII collagen degradation could trigger an inflammatory response leading to itchy skin characteristic of EB pruriginosa. All 27 individuals with EB pruriginosa were heterozygous for dominant-negative glycine substitution mutations in the COL7A1 gene, six of which have not been reported previously. The frequency of the 2G allele in these subjects (46.3%) was greater than in the controls (42.0%), but less than in non-itchy dominant dystrophic EB (52.2%) or recessive dystrophic EB (62.0%), indicating that variants of a common functional polymorphism in the matrix metalloproteinase-1 gene promoter do not account for the itchy skin phenotype. The pathophysiology of EB pruriginosa remains unexplained.

Autosomal dominant junctional epidermolysis bullosa

Background  Epidermolysis bullosa (EB) encompasses a heterogeneous group of inherited skin disorders associated with trauma‐induced blistering. The junctional forms of EB (JEB), Herlitz JEB, non‐Herlitz JEB and JEB associated with pyloric atresia have all been attributed to autosomal recessive inheritance. We describe a 7‐year‐old girl with defective dental enamel, trauma‐induced blistering and subsequent scarring. Her mother, a carrier of the mutation p.G627V in the collagen XVII gene (COL17A1) had evidence of hypoplastic dental enamel without skin blistering. Her grandmother had non‐Herlitz JEB as a result of a compound heterozygous mutation in COL17A1 (p.G627V and c.3514ins25).

Loss-of-Function FERMT1 Mutations in Kindler Syndrome Implicate a Role for Fermitin Family Homolog-1 in Integrin Activation

Kindler syndrome is an autosomal recessive disorder characterized by skin atrophy and blistering. It results from loss-of-function mutations in the FERMT1 gene encoding the focal adhesion protein, fermitin family homolog-1. How and why deficiency of fermitin family homolog-1 results in skin atrophy and blistering are unclear. In this study, we investigated the epidermal basement membrane and keratinocyte biology abnormalities in Kindler syndrome. We identified altered distribution of several basement membrane proteins, including types IV, VII, and XVII collagens and laminin-332 in Kindler syndrome skin. In addition, reduced immunolabeling intensity of epidermal cell markers such as beta1 and alpha6 integrins and cytokeratin 15 was noted. At the cellular level, there was loss of beta4 integrin immunolocalization and random distribution of laminin-332 in Kindler syndrome keratinocytes. Of note, active beta1 integrin was reduced but overexpression of fermitin family homolog-1 restored integrin activation and partially rescued the Kindler syndrome cellular phenotype. This study provides evidence that fermitin family homolog-1 is implicated in integrin activation and demonstrates that lack of this protein leads to pathological changes beyond focal adhesions, with disruption of several hemidesmosomal components and reduced expression of keratinocyte stem cell markers. These findings collectively provide novel data on the role of fermitin family homolog-1 in skin and further insight into the pathophysiology of Kindler syndrome.

The molecular skin pathology of familial primary localized cutaneous amyloidosis

Please cite this paper as: The molecular skin pathology of familial primary localized cutaneous amyloidosis. Experimental Dermatology 2010; 19: 416–423.

Revertant Mosaicism in Kindler Syndrome

TO THE EDITOR Revertant mosaicism (RM) is a genetic phenomenon that results in the spontaneous partial or complete correction of an affected phenotype (Jonkman et al., 1997). In the skin, RM has been reported in various subtypes of epidermolysis bullosa and ichthyosis with confetti (for review, see Lai-Cheong et al., 2011). The mechanisms underlying RM include back mutation, second-site mutation, intragenic crossover, and gene conversion (Jonkman et al., 2003), and indeed several different corrective processes can occur in the same patient (Jonkman and Pasmooij, 2009). Here, we investigate a case of RM in Kindler syndrome (KS; OMIM 173650), an autosomal recessive disorder characterized by skin fragility, atrophy, and photosensitivity

Identical Glycine Substitution Mutations in Type VII Collagen May Underlie Both Dominant and Recessive Forms of Dystrophic Epidermolysis Bullosa

Autosomal dominant and recessive forms of dystrophic epidermolysis bullosa (DEB) result from mutations in the type VII collagen gene (COL7A1). Although paradigms have emerged for genotype/phenotype correlation in DEB, some pathogenic mutations in COL7A1, notably glycine substitutions within the type VII collagen triple helix, may lead to diagnostic difficulties, since certain glycine substitutions can result in either dominant or recessive mutant alleles. Delineation of glycine substitution mutations into two discrete groups, however, is made difficult by observations that, for some particular glycine substitutions in type VII collagen, the same mutation can result in both dominant and recessive disease. In this report we describe four further glycine missense mutations: p.Gly1483Asp, p.Gly1770Ser, p.Gly2213Arg and p.Gly2369Ser, which can lead to either dominant or recessive DEB, and which result in a spectrum of clinical abnormalities. We also identify a further 30 new glycine substitution mutations that cause either dominant or recessive DEB, but not both. In screening the COL7A1 gene for mutations in individuals with DEB our data highlight that delineation of glycine substitutions in type VII collagen has important implications for genetic counselling.

Genitourinary Tract Involvement in Epidermolysis Bullosa

Involvement of the genitourinary tract has been described in many different types of epidermolysis bullosa (EB). Pathology may be broadly divided into problems resulting in obstruction, that may in turn lead to hydroureter or hydronephrosis, or disease primarily affecting the renal parenchyma. Left unrecognized and untreated, renal tract disease may lead to chronic renal failure, and consequent problems associated with providing renal replacement therapy. Management of the urogenital tract in EB should therefore focus on detecting symptoms suggestive of obstruction and regular monitoring to detect problems as early as possible.