Content uploaded by Tadashi Terui
Author content
All content in this area was uploaded by Tadashi Terui on Jan 01, 2022
Content may be subject to copyright.
CONCISE COMMUNICATION
Linear immunoglobulin A/G bullous dermatosis associated
with ulcerative colitis
Asuka ONOE,
1
Daisuke MATSUURA,
1
Tadashi TERUI,
2
Norito ISHII,
3,4
Takashi HASHIMOTO,
4
Toyoko OCHIAI
1
1
Department of Dermatology, Nihon University Hospital,
2
Division of Dermatological Science, Department of Dermatology, Nihon
University School of Medicine, Tokyo,
3
Department of Dermatology, Kurume University School of Medicine,
4
Kurume University
Institute of Cutaneous Cell Biology, Kurume, Japan
ABSTRACT
Linear immunoglobulin (Ig)A/G bullous dermatosis (LAGBD) is an autoimmune bullous disease characterized by
formation of subepidermal blisters and linear deposition of IgA and IgG antibodies along the basement membrane
zone (BMZ). The association between linear IgA bullous dermatosis and ulcerative colitis (UC) is well recognized,
but reports of UC-associated LAGBD are lacking. We have reported a 24-year-old man suffering from LAGBD
associated with UC, which occurred before exacerbations of skin rash. A skin biopsy indicated a subepidermal
blister with an infiltration of primarily neutrophils and eosinophils in the dermis. Direct immunofluorescence (IF)
studies showed a linear deposition of IgA, IgG and C3c. Indirect IF of human skin revealed IgA and IgG anti-BMZ
autoantibodies. Indirect IF of 1 M NaCl-split human skin demonstrated reactivity of IgA and IgG antibodies at the
epidermal side. Immunoblotting showed that IgG antibodies reacted to the BP180 NC16a domain and 120-kDa lin-
ear IgA dermatosis-1, and enzyme-linked immunoassay detected IgG anti-BP230 antibodies. Administration of
prednisolone and diaminodiphenyl sulfone (DDS) via the p.o. route improved skin lesions and bowel conditions.
These results suggest that the bowel inflammation observed in UC may have a causative effect of initiation of the
immune response to the skin and development of the bullous skin lesions in LAGBD. A combination of DDS and
corticosteroid could be a recommended therapeutic option for patients with LAGBD with UC.
Key words: BP180 NC16a domain, immunoblotting, linear immunoglobulin A dermatosis-1, linear
immunoglobulin A/G bullous dermatosis, ulcerative colitis.
INTRODUCTION
Linear immunoglobulin (Ig)A/G bullous dermatosis (LAGBD) is
an autoimmune bullous disease characterized by formation of
subepidermal blisters and linear deposition of IgA and IgG anti-
bodies along the basement membrane zone (BMZ).
1
Ulcerative
colitis (UC) is a clinical form of inflammatory bowel disease.
The association between linear IgA bullous dermatosis (LABD)
and UC is well recognized, but reports of LAGBD associated
with UC are lacking. Here, we describe a patient with LAGBD
who had suffered from UC. We also discuss the implications of
our observed findings.
CASE REPORT
A 24-year-old Japanese man presented in December 2015 with
an 8-month history of worsening blisters on the trunk and
limbs. He had suffered from UC for 10 months before being
referred to us. He was treated with mesalazine (2.4 g, p.o.) and
prednisolone (10 mg, p.o.) once daily. Physical examination at
first visit to our outpatient clinic revealed pigmented plaques
with erosions on the trunk (Fig. 1a), and erosive erythematous
plaques with raised edges and annularly arranged vesiculobul-
lae on the upper extremities (Fig. 1b). Mucous membranes
were not involved. Laboratory examination disclosed a white
blood cell count of 5200/lL with a differential count of 67.0%
neutrophils and 11.4% eosinophils. The hemoglobin level was
15.2 g/dL. Erythrocyte sedimentation rate was 14 mm. The
level of C-reactive protein was 1.31 mg/dL. An IgG chemilumi-
nescence enzyme immunoassay of BP180 non-collagenous
16a (NC16a) domain (MBL, Nagoya, Japan) was positive
(92.3 U/mL; normal, <9.0 U/mL). An IgG enzyme-linked
immunosorbent assay (ELISA) of BP230 was weakly positive
(index, 17.54; cut-off, 9.0). A skin biopsy specimen of an ery-
thematous plaque revealed a subepidermal blister with an
inflammatory infiltrate composed predominantly of neutrophils
and eosinophils in the superficial dermis, and with small neu-
trophil aggregates and microabscesses in the papillary dermis
(Fig. 2). Direct immunofluorescence (IF) evaluation showed lin-
ear deposition of IgA (Fig. 3a), IgG and C3c along the BMZ.
Correspondence: Daisuke Matsuura, M.D., Department of Dermatology, Nihon University Hospital, 1-6 Kanda Surugadai, Chiyoda-ku, Tokyo
101-8309, Japan. Email: matsuura.daisuke@nihon-u.ac.jp
Received 20 February 2017; accepted 5 May 2017.
1295©2017 Japanese Dermatological Association
doi: 10.1111/1346-8138.13934 Journal of Dermatology 2017; 44: 1295–1298
Indirect IF of normal human skin showed IgA and IgG antibod-
ies at titers of 1:10 and 1:40, respectively. In indirect IF using
1 M NaCl-split-normal human skin, both IgA (Fig. 3b) and IgG
(Fig. 3c) autoantibodies reacted at the epidermal side at titers
of 1:40 for both Ig classes. Immunoblot analyses of IgA and
IgG class antibodies were undertaken using normal human epi-
dermal extracts, recombinant proteins of the BP180 NC16a
and C-terminal domains, concentrated culture supernatant of
HaCaT cells, normal human dermal extracts and purified
human laminin-332 (Fig. 3d–i). The patient’s IgA antibodies did
not show any positive reactivity. IgG antibodies reacted with
the BP180 NC16a domain recombinant protein and the
120-kDa linear IgA dermatosis-1 (LAD-1) in the supernatants of
cultured HaCaT cells (Fig. 3e,g). Immunoblotting analyses of
normal human epidermal extract, recombinant protein of the
BP180 C-terminal domain, normal human dermal extracts and
purified human laminin-332 were negative. Based on these
findings, we diagnosed the patient as having LAGBD associ-
ated with UC. Administration of prednisolone (30 mg, p.o.,
once daily) improved his skin lesions and bowel conditions. As
the prednisolone dose was tapered to 18 mg once daily, the
cutaneous lesions relapsed. Once the diagnosis had been
confirmed, diaminodiphenyl sulfone (DDS; 50 mg once daily)
was added. After initiation of DDS treatment, skin and bowel
conditions were controlled progressively with prednisolone
(11 mg).
DISCUSSION
We report an unusual case of LAGBD with UC that occurred
before exacerbations of skin rash. Histopathology indicated a
subepidermal blister with an infiltration primarily of neutrophils
and eosinophils in the dermis. Direct IF showed linear deposi-
tion of IgA, IgG and C3c. Indirect IF revealed both IgA and IgG
anti-BMZ antibodies. Indirect IF of 1 M NaCl-split human skin
demonstrated that IgA and IgG autoantibodies bound to the
epidermal side.
Recently, immunoblot studies have indicated that LAGBD
sera show circulating IgA and IgG antibodies reactive with vari-
ous autoantigens: BP230,
2–4
BP180 NC16a domain,
4
97/120-
kDa LAD-1,
5,6
BP180 C-terminal domains,
4,5
laminin-332,
4,7,8
290-kDa type-VII collagen
4
and 200-kDa laminin-c1.
4,7
Our
(a)
(b)
Figure 1. (a) Clinical features of skin lesions formed on the
chest and the upper extremities. (b) Clinical features of erosive
erythematous plaques with raised edges and annularly
arranged vesiculobullae on the upper extremities.
(a)
(b)
Figure 2. (a,b) Histopathology of a skin biopsy specimen
taken from an erythematous plaque (hematoxylin–eosin, origi-
nal magnifications: [a] 9200; [b] 9400).
1296 ©2017 Japanese Dermatological Association
A. Onoe et al.
patient showed that IgG antibodies reacted with the BP180
NC16a domain and 120-kDa LAD-1 in immunoblotting analy-
ses, and IgG antibodies to BP230 in ELISA. IgA antibodies did
not react with any antigens in various antigen sources used in
immunoblotting analyses. The reason for this lack of IgA reac-
tivity is yet to be clarified. However, a target antigen for IgA
antibodies might have been conformational epitopes, which
were broken during the immunoblot procedure.
Linear IgA/IgG bullous dermatosis is associated with sys-
temic diseases such as malignant B-cell lymphoma,
2
Vogt–
Koyanagi–Harada disease (VKHD)
3
and interstitial pneumonia.
9
This is the first report of LAGBD associated with UC,
although the associations between other immunobullous skin
diseases and UC have been reported. LABD associated with
UC is now well known.
10,11
Paige et al.
10
studied 70 adult
patients with LABD in the UK and found that five of them
(7.1%) had UC (UC prevalence in the UK was 0.05%). They
stated that UC developed before LABD in all patients. Some
reports have shown total resolution of LABD after procto-
colectomy.
12
Cases of bullous pemphigoid with UC have
been reported, in which a diagnosis of UC was made before
development of skin lesions.
13
In our patient, UC occurred
2 months before exacerbations of skin rash, and combination
therapy of prednisolone and DDS was effective for both
bowel and skin conditions. This evidence suggests that the
bowel inflammation observed in UC may have a causative
effect on initiation of the immune response to cutaneous tis-
sues and development of immunobullous skin disease.
Although the mechanism for production of IgA and IgG
autoantibodies in our patient is unknown, we speculate that
abnormal permeability of the colonic mucosa in UC may
increase antigenic stimulation to colonic epithelial antigens
(a)
(d) (e) (f) (g) (h) (i)
(b) (c)
Figure 3. (a) Direct immunofluorescence for immunoglobulin (Ig)A (original magnification 9200). Indirect immunofluorescence using
1 M NaCl-split-normal human skin showed (b) IgA (original magnification 9400) and (c) IgG autoantibodies (original magnification
9400) reacting at the epidermal side at titers of 1:40 for the Ig classes. Immunoblotting analyses using (d) normal human epidermal
extract, (e) BP180 NC16a domain recombinant protein (RP), (f) BP180 C-terminal domain (RP), (g) concentrated culture supernatant
of HaCaT cells, (h) normal human dermal extract and (i) purified human laminin-332. IgG antibodies in our patient’s serum reacted
very weakly with BP180 NC16a RP and 120-KDa LAD-1 in the HaCaT cell culture supernatants. BP, bullous pemphigoid; EBA, epi-
dermolysis bullosa acquisita; MMP, mucous membrane pemphigoid; LAD, linear IgA bullous dermatosis; PNP, paraneoplastic pem-
phigus; p200, anti-p200 pemphigoid; PV, pemphigus vulgaris.
1297©2017 Japanese Dermatological Association
LAGBD associated with ulcerative colitis
and produce cross-reactive antibodies to cutaneous BMZ
antigens.
To the contrary, there are reports convincing to our specula-
tion. Kakugawa et al.
9
reported a case of interstitial pneumonia
associated with LAGBD that developed before lung disease.
They stated that direct IF of a lung biopsy specimen revealed
deposition of IgA, IgG and C3 along the epithelial cell mem-
brane and basement membrane of bronchioles and alveoli.
Yanagihara et al.
3
reported a case of concurrent LAGBD and
VKHD. They stated that the onset of VKHD was earlier than
that of LAGBD, and that the clinical time-course of both dis-
eases was not correlated perfectly.
Patients with LAGBD have responded well to DDS ther-
apy
2,4–6,8
in the same manner as that for LABD treatment.
11
DDS has efficacy against skin diseases characterized by neu-
trophil-rich infiltrates. Suda et al.
14
revealed that the mecha-
nism of the anti-inflammatory action of DDS is inhibition of the
calcium-dependent functions of neutrophils, such as release of
tissue-damaging oxidants and proteases in the affected skin.
As it was suggested that bowel inflammation might have initi-
ated the immune response to cutaneous antigens in our
patient, both UC and LAGBD should be controlled simultane-
ously. Therefore, a combination of DDS and corticosteroid
agents, which was effective both skin and bowel conditions,
could be a recommended therapeutic option for patients with
concurrent LAGBD and UC.
CONFLICT OF INTEREST: None declared.
REFERENCES
1 Zone JJ, Pazderka Smith E, Powell D, Taylor TB, Smith JB, Meyer
LJ. Antigenic specificity of an tibodies from patients with linear
basement memb rane deposition of IgA. Dermatology 1994; 189:
64–66.
2 Morita K, Fujisawa A, Yagi Y, et al. Immunoglobulin A anti-BP230
autoantibodies in linear immunoglobulin A/immunoglobulin G bul-
lous dermatosis. J Dermatol 2011; 38: 1030–1032.
3 Yanagihara S, Mizuno N, Naruse A, Tateishi C, Tsuruta D, Ishii M.
Linear immunoglobulin A/immunoglobulin G bullous dermatosis
associated with Vogt-Koyanagi-Harada disease. J Dermatol 2011;
38: 798–801.
4 Sakaguchi M, Bito T, Oda Y, et al. Three cases of linear IgA/IgG bul-
lous dermatosis showing IgA and IgG reactivity with multiple antigens,
particularly laminin-332. JAMA Dermatol 2013; 149: 1308–1313.
5 Shimizu S, Natsuga K, Shinkuma S, Yasui C, Tsuchiya K, Shimizu
H. Localized linear IgA/IgG bullous dermatosis. Acta Derm Venereol
2010; 90: 621–624.
6 Izaki S, Ito K, Ishii N, Hashimoto T, Fujita H, Terui T. Infantile linear
IgA/IgG bullous dermatosis. Eur J Dermatol 2016; 26:96–98.
7 Li X, Tsuchisaka A, Qian H, et al. Linear IgA/IgG bullous dermatosis
reacts with multiple laminins and integrins. Eur J Dermatol 2015; 25:
418–423.
8 Izaki S, Mitsuya J, Okada T, Koga H, Hashimoto T, Terui T. A case
of linear IgA/IgG bullous dermatosis with anti-laminin-332 autoanti-
bodies. Acta Derm Venereol 2015; 95: 359–360.
9 Kakugawa T, Tomimura S, Hayashi T, et al. Interstitial pneumonia
associated with linear immunoglobulin A/immunoglobulin G bullous
dermatosis. Respiration 2013; 86: 347–351.
10 Paige DG, Leonard JN, Wojnarowska F, Fry L. Linear IgA disease
and ulcerative colitis. Br J Dermatol 1997; 136: 779–782.
11 De Simone C, Guerriero C, Pellicano R. Linear IgA disease and
ulcerative colitis. Eur J Dermatol 1998; 8:48–50.
12 Watchorn RE, Ma S, Gulmann C, Keogan M, O’Kane M. Linear IgA
disease associated with ulcerative colitis: the role of surgery. Clin
Exp Dermatol 2014; 39: 327–329.
13 Sotiriou MC, Foo CW, Scholes CT, Zone JJ. Immunobullous disease
and ulcerative colitis: a case series of six patients. Br J Dermatol
2015; 173: 792–796.
14 Suda T, Suzuki Y, Matsui T, et al. Dapsone suppresses human neu-
trophil superoxide production and elastase release in a calcium-
dependent manner. Br J Dermatol 2005; 152: 887–895.
1298 ©2017 Japanese Dermatological Association
A. Onoe et al.