Report
Immunohistochemical localization of basement membrane laminin 5 and collagen IV in adult linear IgA disease Moetaz El-Domyati, MD, Hossam Abdel-Wahab, MD, and Hesham Ahmad, MD
Department of Dermatology, Faculty of Medicine, Al-Minya University, Al-Minya, Egypt Correspondence Moetaz El-Domyati, MD Department of Dermatology Al-Minya University 2 Obour Buildings Salah Salem St. Apt. 53, Nasr City 11371 Cairo, Egypt E-mails: [email protected] or [email protected] or [email protected] Conflicts of interest: None.
Abstract Background Linear immunoglobulin A disease (LAD), also known as linear IgA bullous dermatosis, is an autoimmune disorder characterized by subepidermal bullae caused by IgA autoantibodies directed against several antigens located in the basement membrane zone of the skin. Laminin 5 (laminin-332) is considered a key component of the lamina lucida/ lamina densa interface, which provides stable attachment of the epidermis to the dermis. Meanwhile, collagen IV is a major component of the lamina densa. Laminin 5 and collagen IV bind to the cell membrane and induce cytoskeletal rearrangements, which contribute to the basement membrane’s final mat-like structure. The study aimed to evaluate the immunohistochemical staining of laminin 5 and collagen IV and to identify the site of blister formation in formalin-fixed, paraffin-embedded skin biopsies from adults with LAD. Methods Skin biopsies from 20 adult patients with LAD were subjected to routine hematoxylin–eosin as well as immunohistochemical staining of collagen IV and laminin 5. Results Linear staining was positive on the floor of the blister for laminin 5 in 65% and collagen IV in 90% of biopsies denoting that the site of separation in most cases of LAD is above the lamina densa. Conclusions The use of laminin 5 and collagen IV immunohistochemistry can be considered as an adjuvant diagnostic tool and may aid in the identification of the site of blister formation in routine skin biopsies in adults with LAD.
Introduction
922
Linear IgA disease (LAD), also known as linear IgA bullous dermatosis, is a pruritic, vesiculobullous autoimmune disorder characterized by linear deposition of IgA autoantibodies directed against several different antigens in the basement membrane zone (BMZ) and subepidermal bullae that resemble skin lesions of bullous pemphigoid (BP) or dermatitis herpetiformis (DH).1,2 For a long time, the nosology of LAD remained controversial. It has been previously reported as a mixed bullous disease, a form of DH, a type of BP, or a distinct disease entity based on its immunopathologic findings.3 Two main clinical syndromes are distinguished, chronic bullous disease of childhood (CBDC) beginning in childhood and adult LAD beginning any time during adult life. They differ in their age of presentation and to some extent as regards their clinical signs, but there is much overlap, and the immunopathology and immunogenetics are common to both diseases.2 The clinical features of LAD are variable, and patients can present with findings suggestive of DH as well as subepidermal tense bullae that are often indistinguishable from BP. However, the vesiculobullous lesions often International Journal of Dermatology 2015, 54, 922–928
appear in a herpetiform arrangement on erythematous and/or normal-appearing skin.4,5 These tense bullae or vesicles vary in size and frequently tend to show annular or polycyclic arrangement due to the coalescence of lesions. Blisters commonly occur in a string of pearls or cluster of jewels pattern.1 Fully developed lesions show subepidermal bullae with mononuclear cells, and neutrophil and eosinophil infiltration. The infiltrate is superficial, perivascular and interstitially located, and visible at the base and edges of the bulla. In most cases, neutrophils prevail over eosinophils in the infiltrate. Neutrophils are usually found along the BMZ, which shows vacuolar degeneration. Neutrophils are also seen in the epidermis and in some dermal papillae, in particular, adjacent to bullae. They may form papillary neutrophilic abscesses as in DH.6 Meanwhile, in old LAD bullae, eosinophils may prevail over neutrophils in dermal papillae and subepidermal bullae. A BP bulla is difficult to differentiate from LAD in these situations.7 Perilesional skin or mucosa direct immunofluorescence (DIF) reveals linear IgA deposits along the BMZ, whereas circulating IgA antibodies to basement membrane proteins are demonstrated by indirect immunofluorescence.6 ª 2015 The International Society of Dermatology
El-Domyati et al.
Laminin 5 and collagen IV in linear IgA disease
In LAD, specific antibody binding has been found to various antigens on Western immunoblot. The most wellcharacterized is a 97 kDa antigen found in a modified epidermal extract (LAD97).8 This antigen is recognized by adult and childhood LAD sera that bind to the epidermal side of BMZ split skin.9 This antigen is identical to a portion of the extracellular domain of the 180 kDa antigen recognized by circulating IgG antibodies in BP (BPAg2).10 Prost et al.11 described the localization of IgA deposits in the majority of patients studied in a mirror image pattern on each side of the lamina densa, i.e., the lamina lucida and sublamina densa. Other studies have found IgA deposition in the lamina lucida, lamina densa, sublamina densa, in the hemidesmosome, or on the basal surface of keratinocytes. Indirect immunogold electron microscope using LAD sera with high titers of antibodies specific for LAD97 has demonstrated binding of these antibodies in the lamina lucida.12,13 Moreover, sera of some patients with LAD with high titers of antibodies specific for collagen VII also shows binding to the anchoring fibrils.14,15 Epidermal basement membrane contains a scaffolding of two network polymers, laminin isoforms and type IV collagen, in which diverse matrix glycoproteins such as nidogen, perlecan, fibulins, and integrins act as stabilizing bridges. Interaction of various molecules of BMZ (basal keratinocyte cell membrane, integrines, laminin 5, laminin 1, laminin 6, entactin, nidogen, type IV collagen, type VII collagen, fibronectin, type I collagen) give the major BMZ function, which is the adherence of epidermis to dermis16 (Fig. 1). Laminin 5, recently designated as laminin-332, is a heterodimer composed of a3, b3, and c2 chains.16 Previously, it was named as epiligrin, nicein, or BM-600 and is
Basal KC
basement membrane zone proteins and proposed site of blister formation in some subepidermal bullous diseases. BP, bullous pemphigoid; BPAG-1, bullous pemphigoid antigen 1 (230 kDa); BPAG-2, bullous pemphigoid antigen 2 (180 kDa); BSLE, bullous systemic lupus erythematosus; EBA, epidermolysis bullosa acquisita; HD, hemidesmosome; KC, keratinocyte; LAD, linear immunoglobulin A disease; PM, plasma membrane ª 2015 The International Society of Dermatology
considered one of the major lamina lucida components. It is one of the key components of the lamina lucida/lamina densa interface, which provides stable attachment of the epidermis to the dermis. Meanwhile, collagen IV is a major structural component of all basement membranes, including cutaneous BMZ and is found around blood vessels. Type IV collagen, a major component of the lamina densa, has a honeycomb or reticular pattern in contrast to the fibrillar pattern of other collagen types (Fig. 1). It has numerous interruptions in its collagenous repeating segment (Gly-X-Y), which confer flexibility, allowing it to form a super structure for incorporation of other basement membrane components.17 Laminin and collagen IV isotypes bind to the cell membrane through integrin or dystroglycan receptors and induce cytoskeletal rearrangements, which contribute to the basement membrane’s final mat-like structure. Immunopathology utilizing optical microscopy can contribute to the differentiation between several autoimmune bullous dermatoses by recognizing certain specific macromolecular components of the dermoepidermal junction.18 Therefore, immunohistochemical staining of laminin 5 and collagen IV could be helpful to identify the site of cleavage in subepidermal bullous disorders. It can be also a useful diagnostic tool in BP cases when DIF is not performed.19 Type IV collagen mapping of BMZ helps to distinguish BP from epidermolysis bullosa acquisita (EBA) in children, since in BP it is on the base of the blister (cleavage in lamina lucida), whereas in EBA it is on the roof.19,20 The aim of the present study is to evaluate the immunohistochemical staining of laminin 5 and collagen IV and to identify the site of blister formation in formalin-fixed, paraffin-embedded skin biopsies from adults with LAD.
KC intermediate fibres
HD
Figure 1 Schematic illustration of
BPAG1
Plectin
PM
BPAG2 Lamina lucida
6 4 Integrin
Site of blister BP ? LAD
Anchoring filaments
Lamina densa
Sublamina densa region
Report
Laminin 5 Collagen IV
Collagen I, III
Collagen VII Anchoring fibrils
EBA BSLE
International Journal of Dermatology 2015, 54, 922–928
923
924
Report
Laminin 5 and collagen IV in linear IgA disease
Materials and methods Twenty adult patients with LAD, who fulfilled the clinical picture and showed linear IgA BMZ deposits by DIF, were the subject of this retrospective study over the last 15 years. However, IgG was present in 25%, IgM in 25%, and C3 in 50% of specimens as recorded in patients’ clinical files. The diagnostic criteria were the characteristic clinical picture, histopathologic features, and exclusive or predominant presence of a homogeneous linear pattern of IgA by DIF at the BMZ. When all that was unclear, a therapeutic test with dapsone was considered. Any case that was not diagnosed using the above criteria was excluded from the study.
Skin biopsies Early blisters were evaluated by a single skin biopsy from every patient. Biopsies were fixed in 10% formalin then embedded in paraffin and sectioned into 5 lm sections to be used for routine hematoxylin–eosin as well as immunohistochemical staining of collagen IV and laminin 5. Immunohistochemical staining An immunoperoxidase technique was used for the demonstration of type IV collagen and laminin 5 epitopes. Type IV collagen Before staining, antigen retrieval was carried out by heating tissue sections with pepsin (catalog no. S 2002; Dako, Carpinteria, CA, USA) diluted in 0.2 N HCl, for 10 minutes at 37 °C. The primary antibody used was mouse antihuman type IV collagen (catalog no. 10760; ICN Biomedicals Inc., Costa Mesa, CA, USA). These were used in a dilution of 1 : 500 and were incubated with the tissue sections overnight at 4 °C in a humidified chamber. Laminin 5 Before staining, antigen retrieval was carried out by heating tissue sections with protease XXIV (catalog no. BK-1405-15; InnoGenex, San Ramon, CA, USA) for one minute at room temperature. The primary antibodies used were mouse antihuman laminin 5 (catalog no. L1225-35; US Biological, Swampscott, MA, USA) that recognize gamma 2 chain unique for laminin 5. These were used in a dilution of 1 : 200 and were incubated with the tissue sections overnight at 4 °C in a humidified chamber.
Results This retrospective study was conducted on 20 patients from attendants of Department of Dermatology Outpatient Clinic (Al-Minya University Hospital, Al-Minya, Egypt). Their ages ranged between 37 and 66 years with a mean SD of 53.1 8.58 years. The age at diagnosis International Journal of Dermatology 2015, 54, 922–928
El-Domyati et al.
ranged between 36 and 65 years with a mean SD of 52.4 8.44 years. The age at onset of the disease ranged between 35 and 62 years with a mean SD of 50.2 7.99 years. The duration of the disease at time of diagnosis ranged between 0.5 and 5 years with a mean SD of 1.58 1.35 years. The sex distribution shows that 60% were males (12 patients) and 40% were females (eight patients). The course of the disease showed remissions and exacerbations in 40% of patients. Pruritus was a variable feature; it was severe (25%), moderate (40%), mild (30%), or even entirely absent (5%). The lesions were formed of vesicles (100%) and bullae (75%). The contents of the bullae were clear serous or hemorrhagic fluid. The vesicles and bullae arose on an apparently normal or erythematous base. Iris-like lesions were present in 20% of cases. The vesicles and bullae were noticed to form the characteristic string of pearls (15%) or cluster of jewels sign (30%) in some cases. However, annular or polycyclic lesions (40%) were more apparent (Fig. 2). The trunk was the most commonly affected site (90%). Other sites of predilection were the upper and lower limbs (65%), inner thighs (50%), and genitalia (25%). Mucosal affection was absent in all patients. Histopathologically, a subepidermal separation was observed in all cases (100%). However, a classic histopathological DH picture with neutrophilic predominance and microabscess formation was seen in 40% of cases, while 35% had shown the characteristic picture of BP with eosinophilic predominance and no microabscess (Fig. 3). However, multilocular bullae were also regularly observed in 70%, and fibrin was only noticed in 40% of biopsies. The remaining 25% revealed a non-diagnostic histopathological picture intermediate between DH and BP. The mononuclear cell infiltrate was evidently increased (60%). Meanwhile marked epidermal changes (15%), basal cell vacuolization (50%), and rete tip infiltrating inflammatory cells (10%) were seen although not regularly observed. In the group of patients with a mixed DH–BP-like picture, the prominent feature was a rich cellular infiltrate in the papillary dermis and within the bulla cavity. The inflammatory cells were scattered throughout the papillae or arranged in a linear, closely aggregated manner immediately beneath the epidermis or at the base of the bulla. These cells were mostly neutrophils with eosinophils and mononuclear cells. Laminin 5 staining showed characteristic linear brownish deposits on the floor of the bullae in 13 (65%) biopsies, while collagen IV staining showed brownish linear deposits on the floor of the bullae in 18 (90%) biopsies. Collagen IV characteristic staining around the blood vessels was evident in all biopsies (Fig. 4). In the rest of the specimens, the basement membrane showed negative ª 2015 The International Society of Dermatology
El-Domyati et al.
Laminin 5 and collagen IV in linear IgA disease
(a)
(c)
(b)
(d)
Figure 2 Vesicles and bullae in different sites with polycyclic arrangement (a–d) with the characteristic string of pearls sign of linear immunoglobulin A disease (c,d)
staining at the site of the separation and bulla, so it was considered destroyed in cases showing no staining pattern at the site of separation (dermoepidermal junction). There was no relationship encountered between laminin 5 and collagen IV staining with other immunoglobulin (IgG, IgM) or complement (C3) deposits recorded. Discussion Linear IgA bullous dermatosis and CBDC are rare immune-mediated blistering skin diseases defined by the presence of homogeneous linear deposits of IgA at the cutaneous basement membrane and are considered a different presentation of the same disease process.21,22 These different clinical presentations appear to result from the IgA binding to different antigens.23 We studied 20 patients with adult LAD diagnosed over a period of 15 years. These cases have been mainly ª 2015 The International Society of Dermatology
Report
selected based on their clinical features beside its unique immunopathology consisting of an exclusive or predominant homogeneous linear deposition of IgA along the cutaneous BMZ by DIF. However, other subepidermal blistering diseases were sometimes excluded by the constellation of findings, including the characteristic clinical picture, histopathologic features, DIF, and therapeutic test with dapsone. Clinically, it is worth mentioning that mucous membrane involvement was not encountered in any of our Egyptian cases, whether adults in the present study or children with CBDC.24 On the contrary, other studies reported painful erosions and ulcerations produced by the rupture of blisters or vesicles that are very hard to see due to the continuous trauma and may appear everywhere in the oral cavity.25,26 They sometimes may also assume the character of a scarring lesion,27 desquamative gingivitis,28 or erosive cheilitis. Even though very rare, mucosal lesions may be the only clinical feature presented by a patient with LAD29 and may precede cutaneous lesions.26 Routine histopathology showed perivascular infiltrate in the entire study group, while rete tip infiltrate was present in 10% of the study group. Neutrophilic infiltrate was the rule in all biopsies. Meanwhile, mononuclear cells and eosinophils are present in 60% of biopsies, which agree with classical findings of LAD.23 Laminins are a family of high molecular weight extracellular matrix proteins, deposited in the basement membranes, also involved in cellular adhesion, growth, and differentiation.30,31 There are at least 15 types of laminin isoforms, resulting from a variety of combinations of a, b, and c chains.16,31 In skin, laminin 5 (laminin-332) is localized to the interface of the lamina lucida and the lamina densa, where it is thought to form a critical link between hemidesmosome–anchoring filament complexes, the lamina densa, and underlying anchoring fibrils17 (Fig. 1). Collagen IV is a unique macromolecule in different basement membranes that contains genetically distinct but structurally homologous a-chains. Six different a-chains have been identified so far. The existence of three major networks, namely a1/a2-containing, a3/a4/ a5-containing, and a1/a2/a5/a6-containing networks, has been established. The chain composition is determined by the non-collagenous NC1 domains, and the a-chains are linked to each other by covalent interactions through these domains. In the skin, the a1/a2-containing collagen IV network dominates within the dermal–epidermal junction, but the a1/a2/a5/a6-containing network is also likely to be present.32 Collagen IV serves as a key matrix within the lamina densa and has been related to several different diseases of basement membranes.17 International Journal of Dermatology 2015, 54, 922–928
925
926
Report
Laminin 5 and collagen IV in linear IgA disease
(a)
El-Domyati et al.
(c)
(b)
(d)
(a)
(c)
(b)
(d)
Figure 4 Linear laminin 5 staining on the floor of the bulla (a,b); and collagen IV staining on the floor of bulla and around blood vessels (c,d) (immunohistochemical staining; a, c 9 200, b,d 9 400) International Journal of Dermatology 2015, 54, 922–928
Figure 3 Subepidermal multilocular blister in a dermatitis herpetiformislike picture with microabscess, neutrophils, nuclear dust, and papillary edema (a) and base of the bulla shows numerous neutrophils and nuclear bands (b). A bullous pemphigoid-like picture with mixed inflammatory infiltrate (c), mononuclear cells and numerous eosinophils (d) (hematoxylin and eosin; a,c 9 100, b,d 9 400)
Linear immunoglobulin A disease is a heterogeneous disorder manifested by the number of target antigens (epithelial, dermal, or both sites) involved in the BMZ space (multiple distribution of immune deposits in the lamina lucida, lamina/sublamina densa, or mixed). This heterogeneity and the multiple target antigens involved in its pathogenesis can occasionally make the diagnosis challenging, particularly because LAD shows many overlapping clinical and immunohistopathologic features with other diseases such as BP and DH.1 El-Domyati et al.24 reported that laminin 5 was positive in 70% and collagen IV in 95% of the studied cases of CBDC on the floor of the bulla. This inspired us to study the presence of laminin 5 and collagen IV in adults with LAD. In the present study, laminin 5 (lamina lucida/ lamina densa interface) was positive on the floor of the blister in 65% of cases and collagen IV (lamina densa) in 90% of skin biopsies. Consequently, the early site of cleavage seemed to be present mainly between lamina densa, when not destroyed, and the basal cell plasma membrane, that is to say in the lamina lucida. Hence, the present study may also add further evidence that adult LAD could be considered the adulthood form or absolute counterpart of CBDC. This pilot study shows the significance of testing for both laminin 5 and collagen IV, which could be added to the investigative tests for LAD. This easy procedure could localize the site of blister formation in routine formalinfixed, paraffin-embedded skin biopsies in adults with LAD, especially in cases where the more sophisticated procedures, such as DIF and electron microscopy, were not performed. ª 2015 The International Society of Dermatology
El-Domyati et al.
Thus, we could consider this histopathologic and immunohistochemical feature of patients with LAD as a helpful adjunct to other parameters for differentiation among the group of subepidermal bullous diseases. Hence it could differentiate cases of EBA, especially IgA type,33,34 and bullous systemic lupus erythematosus,35,36 with positive staining on the roof rather than the floor of the blister as in LAD. In conclusion, the use of laminin 5 and collagen IV immunohistochemistry can be considered as an adjuvant diagnostic tool and may help identify the site of blister formation in formalin-fixed, paraffin-embedded skin biopsies in LAD, but it could not replace the reference method of DIF by which this group could precisely be defined.
Laminin 5 and collagen IV in linear IgA disease
13
14
15
16
17
References 1 Fortuna G, Marinkovich MP. Linear immunoglobulin A bullous dermatosis. Clin Dermatol 2012; 30: 38–50. 2 Wojnarowska F, Venning VA. Immunobullous diseases. In: Burns T, Breathnachs S, Cox N, Griffiths C, eds. Rook’s Textbook of Dermatology, 8th edn. Oxford: Blackwell Publishing Ltd, 2010: 40.45–40.51. 3 Chorzelski TP, Jablonska S. Diagnostic significance of immunofluorescent pattern in dermatitis herpetiformis. Int J Dermatol 1975; 14: 429–436. 4 Mobacken H, Kastrup W, Ljundhall K, et al. Linear IgA dermatosis: a study of ten adult patients. Acta Derm Venereol 1983; 63: 123–128. 5 Leonard JN, Haffenden GP, Ring NP, et al. Linear IgA disease in adults. Br J Dermatol 1982; 107: 301–316. 6 Rados J. Autoimmune blistering diseases: histologic meaning. Clin Dermatol 2011; 29: 377–388. 7 Verdolini R, Cerio R. Autoimmune subepidermal bullous skin disease: the impact of recent findings for the dermatopathologist. Virchows Arch 2003; 443: 184–193. 8 Zone JJ, Taylor TB, Kadunce DP, et al. Identification of the cutaneous basement membrane zone antigen and isolation of antibody in linear immunoglobulin A bullous dermatosis. J Clin Invest 1990; 85: 812–820. 9 Zone JJ, Taylor TB, Kadunce DP, et al. IgA antibodies in chronic bullous disease of childhood react with a 97 kDa basement membrane zone protein. J Invest Dermatol 1996; 106: 1277–1280. 10 Zone JJ, Taylor TB, Meyer LJ, et al. The 97 kDa linear IgA bullous disease antigen is identical to a portion of the extracellular domain of the 180 kDa bullous pemphigoid antigen, BPAg2. J Invest Dermatol 1998; 110: 207–210. 11 Prost C, De Leca AC, Combemale P, et al. Diagnosis of adult linear IgA dermatosis by immunoelectron microscopy in 16 patients with linear IgA deposits. J Invest Dermatol 1989; 92: 39–45. 12 Haftek M, Xone JJ, Taylor TB, et al. Immunogold localization of the 97 kD antigen of linear IgA bullous ª 2015 The International Society of Dermatology
18
19
20
21
22
23
24
25
26
27
28
Report
dermatosis (LAD) detected with patients sera. J Invest Dermatol 1994; 103: 656–659. Ishiko A, Shimizu H, Masunaga T, et al. 97 kDa linear IgA bullous dermatosis (LAD) antigen localizes to the lamina lucida of the epidermal basement membrane. J Invest Dermatol 1996; 106: 739–743. Hashimoto T, Ishiko A, Shimizu H, et al. A case of linear IgA bullous dermatosis with IgA anti-type VII Collagen autoantibodies. Br J Dermatol 1996; 134: 336–339. Caux F, Kirtschig G, Lemarchand-Venencie F, et al. IgA-epidermolysis bullosa acquisita in a child resulting in blindness. Br J Dermatol 1997; 137: 270–275. Aumailley M, Bruckner-Tuderman L, Carter WG, et al. A simplified laminin nomenclature. Matrix Biol 2005; 24: 326–332. Yancey KB, Allen DM. The biology of the basement membrane zone. In: Bolognia JL, Jorizzo JL, Rapini RP, eds. Dermatology. Mosby: St. Louis: 2008: 403–415. Karpouzis A, Vamvassakis E, Stavrianeas N, et al. Ultrastructural immunocytochemistry of autoimmune bullous diseases. Australas J Dermatol 2002; 43: 113–119. Fisler RE, Saeb M, Liang MG, et al. Childhood bullous pemphigoid: a clinicopathologic study and review of the literature. Am J Dermatopathol 2003; 25: 183–189. Nemeth AJ, Klein AD, Gould EW, et al. Childhood bullous pemphigoid. Clinical and immunologic features, treatment, and prognosis. Arch Dermatol 1991; 127: 378–386. Zone JJ, Pazderka SE, Powell D, et al. Antigenic specificity of antibodies from patients with linear basement membrane deposition of IgA. Dermatology 1994; 189 (Suppl.): 64–66. Dmochowski M, Hashimoto T, Bhogal BS, et al. Immunoblotting studies of linear IgA disease. J Dermatol Sci 1993; 6: 194–200. Rao CL, Hall RP III. Linear immunoglobulin A and chronic bullous disease of childhood. In: Wolff K, Goldsmith LA, Katz SI, et al., eds. Fitzpatrick Dermatology in General Medicine, 7th edn. New York: McGraw-Hill Medical Publisher, 2008: 485–490. El-Domyati M, Abd-El-Raheem T, Ahmed H, et al. Laminin 5 and collagen IV in the diagnosis of linear IgA bullous dermatosis. Egypt J Dermatol Androl 2009; 29: 81–90. Eguia del Valle A, Aguirre Urizar JM, Martinez Sahuquillo A. Oral manifestations caused by the linear IgA disease. Med Oral 2004; 9: 39–44. Chan LS, Regezi JA, Cooper KD. Oral manifestations of linear IgA disease. J Am Acad Dermatol 1990; 22: 362–365. Edwards S, Wojnarowska F, Armstrong LM. Chronic bullous disease of childhood with oral mucosal scarring. Clin Exp Dermatol 1991; 16: 41–43. ORegan E, Bane A, Flint S, et al. Linear IgA disease presenting as desquamative gingivitis: a pattern poorly recognized in medicine. Arch Otolaryngol Head Neck Surg 2004; 130: 469–472.
International Journal of Dermatology 2015, 54, 922–928
927
928
Report
Laminin 5 and collagen IV in linear IgA disease
29 del Valle AE, Martinez-Sahuquillo A, Padron JR, et al. Two cases of linear IgA disease with clinical manifestations limited to the gingiva. J Periodontol 2003; 74: 879–882. 30 Colognato H, Yurchenco PD. Form and function: the laminin family of heterotrimers. Dev Dyn 2000; 218: 213–234. 31 Tzu J, Marinkovich MP. Bridging structure with function: structural, regulatory, and developmental role of laminins. Int J Biochem Cell Biol 2008; 40: 199–214. 32 Myllyharju J, Kivirikko KI. Collagens, modifying enzymes and their mutations in humans, flies and worms. Trends Genet 2004; 20: 33–43.
International Journal of Dermatology 2015, 54, 922–928
El-Domyati et al.
33 Vodegel RM, de Jong MC, Pas HH, et al. IgAmediated epidermolysis bullosa acquisita: two cases and review of the literature. J Am Acad Dermatol 2002; 47: 919–925. 34 Buijsrogge JJ, de Jong MC, Meijer HJ, et al. Inflammatory epidermolysis bullosa acquisita with coexistent IgA antibodies to plectin. Clin Exp Dermatol 2005; 30: 531–534. 35 Yell JA, Wojnarowska F. Bullous skin disease in lupus erythematosus. Lupus 1997; 6: 112–121. 36 Vassileva S. Bullous systemic lupus erythematosus. Clin Dermatol 2004; 22: 129–138.
ª 2015 The International Society of Dermatology
Immunohistochemical localization of basement membrane laminin 5 and collagen IV in adult linear IgA disease Moetaz El-Domyati, MD, Hossam Abdel-Wahab, MD, and Hesham Ahmad, MD
Department of Dermatology, Faculty of Medicine, Al-Minya University, Al-Minya, Egypt Correspondence Moetaz El-Domyati, MD Department of Dermatology Al-Minya University 2 Obour Buildings Salah Salem St. Apt. 53, Nasr City 11371 Cairo, Egypt E-mails: [email protected] or [email protected] or [email protected] Conflicts of interest: None.
Abstract Background Linear immunoglobulin A disease (LAD), also known as linear IgA bullous dermatosis, is an autoimmune disorder characterized by subepidermal bullae caused by IgA autoantibodies directed against several antigens located in the basement membrane zone of the skin. Laminin 5 (laminin-332) is considered a key component of the lamina lucida/ lamina densa interface, which provides stable attachment of the epidermis to the dermis. Meanwhile, collagen IV is a major component of the lamina densa. Laminin 5 and collagen IV bind to the cell membrane and induce cytoskeletal rearrangements, which contribute to the basement membrane’s final mat-like structure. The study aimed to evaluate the immunohistochemical staining of laminin 5 and collagen IV and to identify the site of blister formation in formalin-fixed, paraffin-embedded skin biopsies from adults with LAD. Methods Skin biopsies from 20 adult patients with LAD were subjected to routine hematoxylin–eosin as well as immunohistochemical staining of collagen IV and laminin 5. Results Linear staining was positive on the floor of the blister for laminin 5 in 65% and collagen IV in 90% of biopsies denoting that the site of separation in most cases of LAD is above the lamina densa. Conclusions The use of laminin 5 and collagen IV immunohistochemistry can be considered as an adjuvant diagnostic tool and may aid in the identification of the site of blister formation in routine skin biopsies in adults with LAD.
Introduction
922
Linear IgA disease (LAD), also known as linear IgA bullous dermatosis, is a pruritic, vesiculobullous autoimmune disorder characterized by linear deposition of IgA autoantibodies directed against several different antigens in the basement membrane zone (BMZ) and subepidermal bullae that resemble skin lesions of bullous pemphigoid (BP) or dermatitis herpetiformis (DH).1,2 For a long time, the nosology of LAD remained controversial. It has been previously reported as a mixed bullous disease, a form of DH, a type of BP, or a distinct disease entity based on its immunopathologic findings.3 Two main clinical syndromes are distinguished, chronic bullous disease of childhood (CBDC) beginning in childhood and adult LAD beginning any time during adult life. They differ in their age of presentation and to some extent as regards their clinical signs, but there is much overlap, and the immunopathology and immunogenetics are common to both diseases.2 The clinical features of LAD are variable, and patients can present with findings suggestive of DH as well as subepidermal tense bullae that are often indistinguishable from BP. However, the vesiculobullous lesions often International Journal of Dermatology 2015, 54, 922–928
appear in a herpetiform arrangement on erythematous and/or normal-appearing skin.4,5 These tense bullae or vesicles vary in size and frequently tend to show annular or polycyclic arrangement due to the coalescence of lesions. Blisters commonly occur in a string of pearls or cluster of jewels pattern.1 Fully developed lesions show subepidermal bullae with mononuclear cells, and neutrophil and eosinophil infiltration. The infiltrate is superficial, perivascular and interstitially located, and visible at the base and edges of the bulla. In most cases, neutrophils prevail over eosinophils in the infiltrate. Neutrophils are usually found along the BMZ, which shows vacuolar degeneration. Neutrophils are also seen in the epidermis and in some dermal papillae, in particular, adjacent to bullae. They may form papillary neutrophilic abscesses as in DH.6 Meanwhile, in old LAD bullae, eosinophils may prevail over neutrophils in dermal papillae and subepidermal bullae. A BP bulla is difficult to differentiate from LAD in these situations.7 Perilesional skin or mucosa direct immunofluorescence (DIF) reveals linear IgA deposits along the BMZ, whereas circulating IgA antibodies to basement membrane proteins are demonstrated by indirect immunofluorescence.6 ª 2015 The International Society of Dermatology
El-Domyati et al.
Laminin 5 and collagen IV in linear IgA disease
In LAD, specific antibody binding has been found to various antigens on Western immunoblot. The most wellcharacterized is a 97 kDa antigen found in a modified epidermal extract (LAD97).8 This antigen is recognized by adult and childhood LAD sera that bind to the epidermal side of BMZ split skin.9 This antigen is identical to a portion of the extracellular domain of the 180 kDa antigen recognized by circulating IgG antibodies in BP (BPAg2).10 Prost et al.11 described the localization of IgA deposits in the majority of patients studied in a mirror image pattern on each side of the lamina densa, i.e., the lamina lucida and sublamina densa. Other studies have found IgA deposition in the lamina lucida, lamina densa, sublamina densa, in the hemidesmosome, or on the basal surface of keratinocytes. Indirect immunogold electron microscope using LAD sera with high titers of antibodies specific for LAD97 has demonstrated binding of these antibodies in the lamina lucida.12,13 Moreover, sera of some patients with LAD with high titers of antibodies specific for collagen VII also shows binding to the anchoring fibrils.14,15 Epidermal basement membrane contains a scaffolding of two network polymers, laminin isoforms and type IV collagen, in which diverse matrix glycoproteins such as nidogen, perlecan, fibulins, and integrins act as stabilizing bridges. Interaction of various molecules of BMZ (basal keratinocyte cell membrane, integrines, laminin 5, laminin 1, laminin 6, entactin, nidogen, type IV collagen, type VII collagen, fibronectin, type I collagen) give the major BMZ function, which is the adherence of epidermis to dermis16 (Fig. 1). Laminin 5, recently designated as laminin-332, is a heterodimer composed of a3, b3, and c2 chains.16 Previously, it was named as epiligrin, nicein, or BM-600 and is
Basal KC
basement membrane zone proteins and proposed site of blister formation in some subepidermal bullous diseases. BP, bullous pemphigoid; BPAG-1, bullous pemphigoid antigen 1 (230 kDa); BPAG-2, bullous pemphigoid antigen 2 (180 kDa); BSLE, bullous systemic lupus erythematosus; EBA, epidermolysis bullosa acquisita; HD, hemidesmosome; KC, keratinocyte; LAD, linear immunoglobulin A disease; PM, plasma membrane ª 2015 The International Society of Dermatology
considered one of the major lamina lucida components. It is one of the key components of the lamina lucida/lamina densa interface, which provides stable attachment of the epidermis to the dermis. Meanwhile, collagen IV is a major structural component of all basement membranes, including cutaneous BMZ and is found around blood vessels. Type IV collagen, a major component of the lamina densa, has a honeycomb or reticular pattern in contrast to the fibrillar pattern of other collagen types (Fig. 1). It has numerous interruptions in its collagenous repeating segment (Gly-X-Y), which confer flexibility, allowing it to form a super structure for incorporation of other basement membrane components.17 Laminin and collagen IV isotypes bind to the cell membrane through integrin or dystroglycan receptors and induce cytoskeletal rearrangements, which contribute to the basement membrane’s final mat-like structure. Immunopathology utilizing optical microscopy can contribute to the differentiation between several autoimmune bullous dermatoses by recognizing certain specific macromolecular components of the dermoepidermal junction.18 Therefore, immunohistochemical staining of laminin 5 and collagen IV could be helpful to identify the site of cleavage in subepidermal bullous disorders. It can be also a useful diagnostic tool in BP cases when DIF is not performed.19 Type IV collagen mapping of BMZ helps to distinguish BP from epidermolysis bullosa acquisita (EBA) in children, since in BP it is on the base of the blister (cleavage in lamina lucida), whereas in EBA it is on the roof.19,20 The aim of the present study is to evaluate the immunohistochemical staining of laminin 5 and collagen IV and to identify the site of blister formation in formalin-fixed, paraffin-embedded skin biopsies from adults with LAD.
KC intermediate fibres
HD
Figure 1 Schematic illustration of
BPAG1
Plectin
PM
BPAG2 Lamina lucida
6 4 Integrin
Site of blister BP ? LAD
Anchoring filaments
Lamina densa
Sublamina densa region
Report
Laminin 5 Collagen IV
Collagen I, III
Collagen VII Anchoring fibrils
EBA BSLE
International Journal of Dermatology 2015, 54, 922–928
923
924
Report
Laminin 5 and collagen IV in linear IgA disease
Materials and methods Twenty adult patients with LAD, who fulfilled the clinical picture and showed linear IgA BMZ deposits by DIF, were the subject of this retrospective study over the last 15 years. However, IgG was present in 25%, IgM in 25%, and C3 in 50% of specimens as recorded in patients’ clinical files. The diagnostic criteria were the characteristic clinical picture, histopathologic features, and exclusive or predominant presence of a homogeneous linear pattern of IgA by DIF at the BMZ. When all that was unclear, a therapeutic test with dapsone was considered. Any case that was not diagnosed using the above criteria was excluded from the study.
Skin biopsies Early blisters were evaluated by a single skin biopsy from every patient. Biopsies were fixed in 10% formalin then embedded in paraffin and sectioned into 5 lm sections to be used for routine hematoxylin–eosin as well as immunohistochemical staining of collagen IV and laminin 5. Immunohistochemical staining An immunoperoxidase technique was used for the demonstration of type IV collagen and laminin 5 epitopes. Type IV collagen Before staining, antigen retrieval was carried out by heating tissue sections with pepsin (catalog no. S 2002; Dako, Carpinteria, CA, USA) diluted in 0.2 N HCl, for 10 minutes at 37 °C. The primary antibody used was mouse antihuman type IV collagen (catalog no. 10760; ICN Biomedicals Inc., Costa Mesa, CA, USA). These were used in a dilution of 1 : 500 and were incubated with the tissue sections overnight at 4 °C in a humidified chamber. Laminin 5 Before staining, antigen retrieval was carried out by heating tissue sections with protease XXIV (catalog no. BK-1405-15; InnoGenex, San Ramon, CA, USA) for one minute at room temperature. The primary antibodies used were mouse antihuman laminin 5 (catalog no. L1225-35; US Biological, Swampscott, MA, USA) that recognize gamma 2 chain unique for laminin 5. These were used in a dilution of 1 : 200 and were incubated with the tissue sections overnight at 4 °C in a humidified chamber.
Results This retrospective study was conducted on 20 patients from attendants of Department of Dermatology Outpatient Clinic (Al-Minya University Hospital, Al-Minya, Egypt). Their ages ranged between 37 and 66 years with a mean SD of 53.1 8.58 years. The age at diagnosis International Journal of Dermatology 2015, 54, 922–928
El-Domyati et al.
ranged between 36 and 65 years with a mean SD of 52.4 8.44 years. The age at onset of the disease ranged between 35 and 62 years with a mean SD of 50.2 7.99 years. The duration of the disease at time of diagnosis ranged between 0.5 and 5 years with a mean SD of 1.58 1.35 years. The sex distribution shows that 60% were males (12 patients) and 40% were females (eight patients). The course of the disease showed remissions and exacerbations in 40% of patients. Pruritus was a variable feature; it was severe (25%), moderate (40%), mild (30%), or even entirely absent (5%). The lesions were formed of vesicles (100%) and bullae (75%). The contents of the bullae were clear serous or hemorrhagic fluid. The vesicles and bullae arose on an apparently normal or erythematous base. Iris-like lesions were present in 20% of cases. The vesicles and bullae were noticed to form the characteristic string of pearls (15%) or cluster of jewels sign (30%) in some cases. However, annular or polycyclic lesions (40%) were more apparent (Fig. 2). The trunk was the most commonly affected site (90%). Other sites of predilection were the upper and lower limbs (65%), inner thighs (50%), and genitalia (25%). Mucosal affection was absent in all patients. Histopathologically, a subepidermal separation was observed in all cases (100%). However, a classic histopathological DH picture with neutrophilic predominance and microabscess formation was seen in 40% of cases, while 35% had shown the characteristic picture of BP with eosinophilic predominance and no microabscess (Fig. 3). However, multilocular bullae were also regularly observed in 70%, and fibrin was only noticed in 40% of biopsies. The remaining 25% revealed a non-diagnostic histopathological picture intermediate between DH and BP. The mononuclear cell infiltrate was evidently increased (60%). Meanwhile marked epidermal changes (15%), basal cell vacuolization (50%), and rete tip infiltrating inflammatory cells (10%) were seen although not regularly observed. In the group of patients with a mixed DH–BP-like picture, the prominent feature was a rich cellular infiltrate in the papillary dermis and within the bulla cavity. The inflammatory cells were scattered throughout the papillae or arranged in a linear, closely aggregated manner immediately beneath the epidermis or at the base of the bulla. These cells were mostly neutrophils with eosinophils and mononuclear cells. Laminin 5 staining showed characteristic linear brownish deposits on the floor of the bullae in 13 (65%) biopsies, while collagen IV staining showed brownish linear deposits on the floor of the bullae in 18 (90%) biopsies. Collagen IV characteristic staining around the blood vessels was evident in all biopsies (Fig. 4). In the rest of the specimens, the basement membrane showed negative ª 2015 The International Society of Dermatology
El-Domyati et al.
Laminin 5 and collagen IV in linear IgA disease
(a)
(c)
(b)
(d)
Figure 2 Vesicles and bullae in different sites with polycyclic arrangement (a–d) with the characteristic string of pearls sign of linear immunoglobulin A disease (c,d)
staining at the site of the separation and bulla, so it was considered destroyed in cases showing no staining pattern at the site of separation (dermoepidermal junction). There was no relationship encountered between laminin 5 and collagen IV staining with other immunoglobulin (IgG, IgM) or complement (C3) deposits recorded. Discussion Linear IgA bullous dermatosis and CBDC are rare immune-mediated blistering skin diseases defined by the presence of homogeneous linear deposits of IgA at the cutaneous basement membrane and are considered a different presentation of the same disease process.21,22 These different clinical presentations appear to result from the IgA binding to different antigens.23 We studied 20 patients with adult LAD diagnosed over a period of 15 years. These cases have been mainly ª 2015 The International Society of Dermatology
Report
selected based on their clinical features beside its unique immunopathology consisting of an exclusive or predominant homogeneous linear deposition of IgA along the cutaneous BMZ by DIF. However, other subepidermal blistering diseases were sometimes excluded by the constellation of findings, including the characteristic clinical picture, histopathologic features, DIF, and therapeutic test with dapsone. Clinically, it is worth mentioning that mucous membrane involvement was not encountered in any of our Egyptian cases, whether adults in the present study or children with CBDC.24 On the contrary, other studies reported painful erosions and ulcerations produced by the rupture of blisters or vesicles that are very hard to see due to the continuous trauma and may appear everywhere in the oral cavity.25,26 They sometimes may also assume the character of a scarring lesion,27 desquamative gingivitis,28 or erosive cheilitis. Even though very rare, mucosal lesions may be the only clinical feature presented by a patient with LAD29 and may precede cutaneous lesions.26 Routine histopathology showed perivascular infiltrate in the entire study group, while rete tip infiltrate was present in 10% of the study group. Neutrophilic infiltrate was the rule in all biopsies. Meanwhile, mononuclear cells and eosinophils are present in 60% of biopsies, which agree with classical findings of LAD.23 Laminins are a family of high molecular weight extracellular matrix proteins, deposited in the basement membranes, also involved in cellular adhesion, growth, and differentiation.30,31 There are at least 15 types of laminin isoforms, resulting from a variety of combinations of a, b, and c chains.16,31 In skin, laminin 5 (laminin-332) is localized to the interface of the lamina lucida and the lamina densa, where it is thought to form a critical link between hemidesmosome–anchoring filament complexes, the lamina densa, and underlying anchoring fibrils17 (Fig. 1). Collagen IV is a unique macromolecule in different basement membranes that contains genetically distinct but structurally homologous a-chains. Six different a-chains have been identified so far. The existence of three major networks, namely a1/a2-containing, a3/a4/ a5-containing, and a1/a2/a5/a6-containing networks, has been established. The chain composition is determined by the non-collagenous NC1 domains, and the a-chains are linked to each other by covalent interactions through these domains. In the skin, the a1/a2-containing collagen IV network dominates within the dermal–epidermal junction, but the a1/a2/a5/a6-containing network is also likely to be present.32 Collagen IV serves as a key matrix within the lamina densa and has been related to several different diseases of basement membranes.17 International Journal of Dermatology 2015, 54, 922–928
925
926
Report
Laminin 5 and collagen IV in linear IgA disease
(a)
El-Domyati et al.
(c)
(b)
(d)
(a)
(c)
(b)
(d)
Figure 4 Linear laminin 5 staining on the floor of the bulla (a,b); and collagen IV staining on the floor of bulla and around blood vessels (c,d) (immunohistochemical staining; a, c 9 200, b,d 9 400) International Journal of Dermatology 2015, 54, 922–928
Figure 3 Subepidermal multilocular blister in a dermatitis herpetiformislike picture with microabscess, neutrophils, nuclear dust, and papillary edema (a) and base of the bulla shows numerous neutrophils and nuclear bands (b). A bullous pemphigoid-like picture with mixed inflammatory infiltrate (c), mononuclear cells and numerous eosinophils (d) (hematoxylin and eosin; a,c 9 100, b,d 9 400)
Linear immunoglobulin A disease is a heterogeneous disorder manifested by the number of target antigens (epithelial, dermal, or both sites) involved in the BMZ space (multiple distribution of immune deposits in the lamina lucida, lamina/sublamina densa, or mixed). This heterogeneity and the multiple target antigens involved in its pathogenesis can occasionally make the diagnosis challenging, particularly because LAD shows many overlapping clinical and immunohistopathologic features with other diseases such as BP and DH.1 El-Domyati et al.24 reported that laminin 5 was positive in 70% and collagen IV in 95% of the studied cases of CBDC on the floor of the bulla. This inspired us to study the presence of laminin 5 and collagen IV in adults with LAD. In the present study, laminin 5 (lamina lucida/ lamina densa interface) was positive on the floor of the blister in 65% of cases and collagen IV (lamina densa) in 90% of skin biopsies. Consequently, the early site of cleavage seemed to be present mainly between lamina densa, when not destroyed, and the basal cell plasma membrane, that is to say in the lamina lucida. Hence, the present study may also add further evidence that adult LAD could be considered the adulthood form or absolute counterpart of CBDC. This pilot study shows the significance of testing for both laminin 5 and collagen IV, which could be added to the investigative tests for LAD. This easy procedure could localize the site of blister formation in routine formalinfixed, paraffin-embedded skin biopsies in adults with LAD, especially in cases where the more sophisticated procedures, such as DIF and electron microscopy, were not performed. ª 2015 The International Society of Dermatology
El-Domyati et al.
Thus, we could consider this histopathologic and immunohistochemical feature of patients with LAD as a helpful adjunct to other parameters for differentiation among the group of subepidermal bullous diseases. Hence it could differentiate cases of EBA, especially IgA type,33,34 and bullous systemic lupus erythematosus,35,36 with positive staining on the roof rather than the floor of the blister as in LAD. In conclusion, the use of laminin 5 and collagen IV immunohistochemistry can be considered as an adjuvant diagnostic tool and may help identify the site of blister formation in formalin-fixed, paraffin-embedded skin biopsies in LAD, but it could not replace the reference method of DIF by which this group could precisely be defined.
Laminin 5 and collagen IV in linear IgA disease
13
14
15
16
17
References 1 Fortuna G, Marinkovich MP. Linear immunoglobulin A bullous dermatosis. Clin Dermatol 2012; 30: 38–50. 2 Wojnarowska F, Venning VA. Immunobullous diseases. In: Burns T, Breathnachs S, Cox N, Griffiths C, eds. Rook’s Textbook of Dermatology, 8th edn. Oxford: Blackwell Publishing Ltd, 2010: 40.45–40.51. 3 Chorzelski TP, Jablonska S. Diagnostic significance of immunofluorescent pattern in dermatitis herpetiformis. Int J Dermatol 1975; 14: 429–436. 4 Mobacken H, Kastrup W, Ljundhall K, et al. Linear IgA dermatosis: a study of ten adult patients. Acta Derm Venereol 1983; 63: 123–128. 5 Leonard JN, Haffenden GP, Ring NP, et al. Linear IgA disease in adults. Br J Dermatol 1982; 107: 301–316. 6 Rados J. Autoimmune blistering diseases: histologic meaning. Clin Dermatol 2011; 29: 377–388. 7 Verdolini R, Cerio R. Autoimmune subepidermal bullous skin disease: the impact of recent findings for the dermatopathologist. Virchows Arch 2003; 443: 184–193. 8 Zone JJ, Taylor TB, Kadunce DP, et al. Identification of the cutaneous basement membrane zone antigen and isolation of antibody in linear immunoglobulin A bullous dermatosis. J Clin Invest 1990; 85: 812–820. 9 Zone JJ, Taylor TB, Kadunce DP, et al. IgA antibodies in chronic bullous disease of childhood react with a 97 kDa basement membrane zone protein. J Invest Dermatol 1996; 106: 1277–1280. 10 Zone JJ, Taylor TB, Meyer LJ, et al. The 97 kDa linear IgA bullous disease antigen is identical to a portion of the extracellular domain of the 180 kDa bullous pemphigoid antigen, BPAg2. J Invest Dermatol 1998; 110: 207–210. 11 Prost C, De Leca AC, Combemale P, et al. Diagnosis of adult linear IgA dermatosis by immunoelectron microscopy in 16 patients with linear IgA deposits. J Invest Dermatol 1989; 92: 39–45. 12 Haftek M, Xone JJ, Taylor TB, et al. Immunogold localization of the 97 kD antigen of linear IgA bullous ª 2015 The International Society of Dermatology
18
19
20
21
22
23
24
25
26
27
28
Report
dermatosis (LAD) detected with patients sera. J Invest Dermatol 1994; 103: 656–659. Ishiko A, Shimizu H, Masunaga T, et al. 97 kDa linear IgA bullous dermatosis (LAD) antigen localizes to the lamina lucida of the epidermal basement membrane. J Invest Dermatol 1996; 106: 739–743. Hashimoto T, Ishiko A, Shimizu H, et al. A case of linear IgA bullous dermatosis with IgA anti-type VII Collagen autoantibodies. Br J Dermatol 1996; 134: 336–339. Caux F, Kirtschig G, Lemarchand-Venencie F, et al. IgA-epidermolysis bullosa acquisita in a child resulting in blindness. Br J Dermatol 1997; 137: 270–275. Aumailley M, Bruckner-Tuderman L, Carter WG, et al. A simplified laminin nomenclature. Matrix Biol 2005; 24: 326–332. Yancey KB, Allen DM. The biology of the basement membrane zone. In: Bolognia JL, Jorizzo JL, Rapini RP, eds. Dermatology. Mosby: St. Louis: 2008: 403–415. Karpouzis A, Vamvassakis E, Stavrianeas N, et al. Ultrastructural immunocytochemistry of autoimmune bullous diseases. Australas J Dermatol 2002; 43: 113–119. Fisler RE, Saeb M, Liang MG, et al. Childhood bullous pemphigoid: a clinicopathologic study and review of the literature. Am J Dermatopathol 2003; 25: 183–189. Nemeth AJ, Klein AD, Gould EW, et al. Childhood bullous pemphigoid. Clinical and immunologic features, treatment, and prognosis. Arch Dermatol 1991; 127: 378–386. Zone JJ, Pazderka SE, Powell D, et al. Antigenic specificity of antibodies from patients with linear basement membrane deposition of IgA. Dermatology 1994; 189 (Suppl.): 64–66. Dmochowski M, Hashimoto T, Bhogal BS, et al. Immunoblotting studies of linear IgA disease. J Dermatol Sci 1993; 6: 194–200. Rao CL, Hall RP III. Linear immunoglobulin A and chronic bullous disease of childhood. In: Wolff K, Goldsmith LA, Katz SI, et al., eds. Fitzpatrick Dermatology in General Medicine, 7th edn. New York: McGraw-Hill Medical Publisher, 2008: 485–490. El-Domyati M, Abd-El-Raheem T, Ahmed H, et al. Laminin 5 and collagen IV in the diagnosis of linear IgA bullous dermatosis. Egypt J Dermatol Androl 2009; 29: 81–90. Eguia del Valle A, Aguirre Urizar JM, Martinez Sahuquillo A. Oral manifestations caused by the linear IgA disease. Med Oral 2004; 9: 39–44. Chan LS, Regezi JA, Cooper KD. Oral manifestations of linear IgA disease. J Am Acad Dermatol 1990; 22: 362–365. Edwards S, Wojnarowska F, Armstrong LM. Chronic bullous disease of childhood with oral mucosal scarring. Clin Exp Dermatol 1991; 16: 41–43. ORegan E, Bane A, Flint S, et al. Linear IgA disease presenting as desquamative gingivitis: a pattern poorly recognized in medicine. Arch Otolaryngol Head Neck Surg 2004; 130: 469–472.
International Journal of Dermatology 2015, 54, 922–928
927
928
Report
Laminin 5 and collagen IV in linear IgA disease
29 del Valle AE, Martinez-Sahuquillo A, Padron JR, et al. Two cases of linear IgA disease with clinical manifestations limited to the gingiva. J Periodontol 2003; 74: 879–882. 30 Colognato H, Yurchenco PD. Form and function: the laminin family of heterotrimers. Dev Dyn 2000; 218: 213–234. 31 Tzu J, Marinkovich MP. Bridging structure with function: structural, regulatory, and developmental role of laminins. Int J Biochem Cell Biol 2008; 40: 199–214. 32 Myllyharju J, Kivirikko KI. Collagens, modifying enzymes and their mutations in humans, flies and worms. Trends Genet 2004; 20: 33–43.
International Journal of Dermatology 2015, 54, 922–928
El-Domyati et al.
33 Vodegel RM, de Jong MC, Pas HH, et al. IgAmediated epidermolysis bullosa acquisita: two cases and review of the literature. J Am Acad Dermatol 2002; 47: 919–925. 34 Buijsrogge JJ, de Jong MC, Meijer HJ, et al. Inflammatory epidermolysis bullosa acquisita with coexistent IgA antibodies to plectin. Clin Exp Dermatol 2005; 30: 531–534. 35 Yell JA, Wojnarowska F. Bullous skin disease in lupus erythematosus. Lupus 1997; 6: 112–121. 36 Vassileva S. Bullous systemic lupus erythematosus. Clin Dermatol 2004; 22: 129–138.
ª 2015 The International Society of Dermatology
