
	Version PDF

Anti-basement membrane zone antibodies in elderly patients with pruritic disorders and diabetes mellitus

European Journal of Dermatology. Volume 18, Number 5, 534-8, September-October 2008, Investigative report

DOI : 10.1684/ejd.2008.0483

Summary

Author(s) : Hana Jedlickova, Jana Racovska, Andrea Niedermeier, Josef Feit, Michael Hertl , Department of Dermatovenereology, Masaryk University, Brno, Czech Republic St. Anna University Hospital, Pekarska 53, 656 91 Brno, Czech Republic, Department of Dermatology and Allergology, Philipps University, Marburg, Germany, Department of Pathology, Masaryk University, Brno, Czech Republic.

Summary : Anti-basement membrane zone (anti-BMZ) antibodies are detectable in a low percentage of elderly subjects without clinical signs of bullous pemphigoid (BP). BP may initially mimic other pruritic dermatoses and may be more common in patients with diabetes mellitus (DM), since DM is frequently associated with pruritic disorders. The aim of the present study was to analyse a possible association of BP and DM and to detect subclinical BP among elderly patients with pruritic dermatoses. Ninety elderly patients (78.6 ± 4.7 years) treated for dermatologic conditions were divided into four groups: I. DM+/pruritus+, II. DM–/pruritus+, III. DM+/pruritus–, and IV. DM–/pruritus–. Patients’ sera were tested by indirect immunofluorescence (IIF) on monkey oesophagus\; positive or dubious results were further evaluated by ELISA with human recombinant BP180 and BP230 proteins and purified laminin 5. Positive results were found in 1 of 21 (4.8%) patients in group I, 6/31 (19.3%) patients in group II, 1/18 (5.5%) patients in group III, 3/20 (15%) patients in group IV. In the whole cohort positive anti-BMZ antibodies of linear or basal cell cytoplasmic and membrane pattern were found in 11 cases (12.2%). ELISA was positive in 11/29 (37.9%) tested sera for at least one antigen (BP180, BP230 and laminin 5 ELISA was positive in 7, 5, and 2 sera, respectively). Positive IIF corresponded with positive ELISA in 6/11 (54.5%) cases (ELISA with BP180, BP230, laminin 5 was positive in 5, 3, and 1 serum, respectively). Thus, by IIF, a significant portion of elderly patients had anti-BMZ antibodies and these findings were confirmed by ELISA. There was no statistically significant difference in the presence of anti-BMZ antibodies among the groups I-IV. Thus, the association of anti-BMZ antibodies with age overrules the potential association with DM and/or pruritus.

Keywords : anti-basement membrane zone antibody, bullous pemphigoid, diabetes mellitus, ELISA, indirect immunofluorescence

Pictures

ARTICLE

Auteur(s) : Hana Jedlickova¹, Jana Racovska¹, Andrea Niedermeier², Josef Feit³, Michael Hertl²

¹Department of Dermatovenereology, Masaryk University, Brno, Czech Republic St. Anna University Hospital, Pekarska 53, 656 91 Brno, Czech Republic
²Department of Dermatology and Allergology, Philipps University, Marburg, Germany
³Department of Pathology, Masaryk University, Brno, Czech Republic

accepté le 13 Mai 2008

Anti-basement membrane zone (BMZ) antibodies are a hallmark of distinct subepidermal autoimmune disorders including bullous pemphigoid (BP), mucous membrane pemphigoid (MMP), and epidermolysis bullosa acquisita (EBA). BP is the most common autoimmune bullous disorder in adults and affects mainly the elderly exceeding 70 years of age. Initially BP may resemble fixed urticaria, prurigo, erythema multiforme, nummular eczema and scabies [1, 2]. Pruritus is common. In some of these patients, the disease may never progress to full-blown BP with blisters [3]. Typically, BP is associated with IgG antibodies reactive with two components of the dermoepidermal BMZ, BP180 and BP230 [4]. It is noteworthy that anti BMZ antibodies can occasionally be detected in other dermatoses and also in healthy individuals.

In our own experience, approximately one third of patients with BP suffer from diabetes mellitus (DM); this association may be accidental [5]. The overall prevalence of non-insulin dependent DM is estimated at 2%, and in populations over 60 years of age at 10% [6]. Epidemiological data for older populations are sparse. In the study of Warram and Krolewski the total prevalence of DM in the U.S. white population for the age category 60-74 was 17.3% and did not increase in individuals 75 years and older [7]. A potential association of BP and DM was previously suggested. Rosina et al. and Chuang et al. found a significantly higher association of BP with DM compared to controls (32% to 9%, and 23% to 3.6%, respectively [8, 9]). A disturbed glucose metabolism may lead to changes in structural and functional properties of extracellular matrix proteins by a non-enzymatic glycation process; this is particularly the case for collagens which are exposed to glycation [10].

Pruritus is a common complaint in DM. Its frequency has been reported from 3 to 40% of patients, with infectious diseases, eczematous diseases, metabolic disturbances, autonomic and sensitive neuropathies as possible causes [11-14]. Some skin changes in diabetics can mimic BP. Prurigo-like lesions are quite common and bullosis diabeticorum is a complication of insulin-dependent DM characterised by the formation of subepidermal blisters. Previous studies of bullosis diabeticorum described anchoring fibril impairment and cleavage within the lamina lucida [15].

In the present study, anti-BMZ antibodies were investigated in elderly subjects with DM and with/without pruritus to analyse the aforementioned potential association of BP with DM. This task seems relevant since a direct association of anti-BMZ antibodies and DM would help to identify a potential trigger factor which, apart from age, may increase the risk for loss of tolerance against components of the dermoepidermal BMZ.

Materials and methods

Patients

A total of 90 patients (78.6 ± 4.7 years) were recruited between August 2003 and October 2005 at the Ist Department of Dermatovenereology of the St. Anna University Hospital, Brno. The patients were divided into four groups:

I. 21 diabetic patients with pruritic skin changes.
II. 31 non-diabetic patients with pruritic skin changes.
III. 18 diabetic patients without pruritic dermatoses.
IV. 20 non-diabetic patients with non-pruritic dermatoses.

All the diabetic patients suffered from non insulin-dependent DM. The mean duration of DM was 12.3 years in group I and 6.3 years in the group III. Pruritic dermatoses included nummular eczema, scabies, prurigo, chronic urticaria, contact eczema, seborrhoeic dermatitis, pruritus sine materia, psoriasis, and drug eruptions. Patients in the control groups III and IV suffered from leg ulcers, psoriasis, parapsoriasis, herpes zoster, erysipelas, basal cell carcinoma, tinea corporis, and seborrhoeic dermatitis.

Serological analyses

Indirect immunofluorescence (IIF) of the patients’ sera was performed on monkey oesophagus (The Binding Site, UK) using FITC conjugated swine anti human polyvalent IgG, IgA, IgM antibodies (Sevapharma, Czech Republic) at a dilution of 1:5 and 1:10. Positive sera were tested with FITC conjugated rabbit anti human IgG/IgA (DakoCytomation, Denmark) on monkey oesophagus. Sera with positive basal cell cytoplasmic and membrane IIF were retested on human tissue sections (samples of healthy skin obtained by dermatosurgery). Sera with positive or questionable reactivity were further tested by ELISA with recombinant BP180 and BP230 proteins produced in a baculovirus expression system and purified laminin 5 (α3, β3 and γ2 chains), as recently described [16-18]. The following recombinant proteins were employed: the entire extracellular portion of BP180 - BP180ex (amino acid residues (aa) 486-1497); BP180-N1, containing the NC16A domain and the collagenous domain Col15 (aa 490-812); BP180-N2 containing the NC16A domain (aa 467-567), BP180-M (aa 809-1106); BP180-C1 (aa 1048-1465); BP180-C2 (aa 1352-1465); BP230-C1 (aa 1881-2649); BP230-N (aa 1-1307). The presence of intercellular substance antibodies (ICS) and anti-nuclear antibodies (ANA) was also examined, when evaluating IIF on monkey oesophagus.

Statistical analysis

To test the relations among the groups the X² test of good correlation was used.

Results

Detection of anti-BMZ autoantibodies by IIF

We evaluated as positive IIF findings either the presence of a linear labeling along the basement membrane zone (BMZ) or a staining of the cell membrane and peripheral cytoplasm of basal epithelial cells (BC). Positive results were found in 1/21 (4.8%) patients in group I (DM+/pruritus+), in 6/31 (19.4%) patients in group II (DM–/pruritus+), in 1/18 (5.5%) patients in group III (DM+/pruritus–), and in 3/20 (15%) patients in group IV (DM–/pruritus–) (table 1). Linear positivity along the BMZ was detected only in 3 cases (figure 1A). Eight sera labelled the peripheral cytoplasm and the cell membrane of BC (figure 1B-D) (tables 1 and 2). This type of IF was considered positive, as discussed in more detail later on. The sera with positive anti BC IF were retested on human skin. On this substrate, only cytoplasmic positivity of BC was detected.
Table 1 Immunoserological findings in the studied patient groups

	IIF: BM	IIF: BC	IIF: BM+BC	ELISA: No of positive sera for at least one antigen from tested samples	ELISA BP180	ELISA BP230	ELISA LN5
I. DM+ /pruritus+(n-21)	0	1	1	4/5	2	1	1
II. DM–/ pruritus+(n-31)	2	4	6	5/11	3	3	1
III. DM+/pruritus–(n-18)	0	1	1	0/7	0	0	0
IV. DM–/ pruritus–(n-20)	1	2	3	2/6	2	1	1
Total	3	8	11	11/29	7	5	3

Table 2 Characteristics of patients with anti-basement membrane zone antibodies by IIF and corresponding ELISA results

Patient n. / group	BC	BM	Ig	ELISA	DM	Diagnosis
17/I	+		IgG	BP180-M BP180-C2	+	Paraneoplastic exanthema
1/II		+	IgG	BP230-C, BP230-N, BP180-M, BP180-N2, BP180-C2 Laminin 5	–	Scabies
5/II		+	IgG	neg	–	Bullous drug eruption
12/II	+		IgG	neg	–	Bullous scarring disease
25/II	+		IgG	neg	–	Nummular eczema
26/II	+		IgG	BP180ex BP180-M BP180-C2	–	Scabies
28/II	+		IgG	BP230-N	–	Nummular eczema
18/III	+		IgA	neg	+	Ulcus cruris
4/IV		+	IgG	neg	–	Psoriasis
5/IV	+		IgG	BP180-C2, laminin 5	–	Parapsoriasis
18/IV	+		IgG	BP230-N, BP180-ex, BP180-C1, BP180-N2, BP180-M, BP180-C2	–	Seborrhoeic dermatitis

Detection of anti-BMZ autoantibodies by ELISA

A total of 29 sera were tested by ELISA with recombinant human BP180, BP230 and purified laminin 5. Patients with positive (11 cases) or dubious IIF results were chosen for this assay. Positive ELISA values were found in 11/29 cases (37.9%). Most of the sera showed IgG reactivity to one or two epitopes of BP180, BP230 or laminin. Three sera reacted both with BP180 and BP230 and/or laminin (table 2). Of the three sera with linear positivity, ELISA confirmed reactivity in a case of scabies; five of 8 sera with IIF BC positivity were reactive by ELISA. Overall, detection of ELISA reactivity corresponded with positive IIF in six cases (54.5%).

Findings of traces of linear BMZ labeling and of positive ICS in the lower epithelial cell layers (15 cases) were considered as dubious IIF. Three cases with a clinical picture possibly suggestive of subclinical BP (chronic urticaria, prurigo and scabies) were also tested by ELISA, despite negative IIF. Of 18 patients with dubious IIF results or suspicious clinical pictures, a positive ELISA was detected in 5 cases. In the whole group of sera tested by ELISA, recombinant BP180 was recognized by IgG in the majority of the cases (7 sera). Noteworthy, IgG reactivity against the immunodominant BP180-NC16A domain (BP180-N2) was only found in two sera, two other sera reacted with BP180-N1 (NC16A and Col15 domain). In the entire cohort of 90 patients, anti-ICS antibodies and ANA (granular pattern) were found in 24 and 27 patients (26.7% and 30%), respectively. Their high incidence was surprising, though probably non-specific. IgG deposits with an ICS pattern can be found in IIF at low titres due to the cross reactions with anti-blood group antigen antibodies, and low titre ANA are common in the elderly [19, 20].

Statistical analysis

The difference among patients with and without DM in BM or BC positivity in IIF was not statistically significant (X² = 3.530; df = 1; p = 0.060). The difference among patients with and without pruritus in BM or BC positivity in IIF was also not statistically significant (X² = 0.179; df = 1; p = 0.673). There was no significant difference in ANA occurrence in diabetic and non-diabetic patients (X² = 2.600; df = 1; p = 0.107). In addition, there was no significant difference in the occurrence of ICS in diabetic and non-diabetic patients (X² = 2.983; df = 1; p = 0.084).

Discussion

Rieckhoff-Cantoni and co-workers detected anti-BMZ antibodies by immunoblotting in 13 of 97 (13%) patients with different disorders including eczema, lichen planus, prurigo and pruritus sine materia [21]. Davies et al. found high titre serum antibodies against the dermoepidermal BMZ in 2 eczema patients using normal human skin, monkey oesophagus, guinea pig lip and patient’s skin as a substrate [22]. This finding may be an epiphenomenon, since inflammatory processes of the skin may lead to tissue destruction and loss of tolerance against components of the BMZ. Moreover, elderly patients may be more prone to loss of immunological tolerance against self antigens of the BMZ leading to BP. Hachisuka et al. investigated anti BMZ IgG reactivity by IIF on guinea pig oesophagus in healthy elderly subjects. Anti-BMZ IgG were detected in 6 of 32 (19%) subjects, whereas human salt split epidermis was positive only in one case. Immunoblot analysis of the patients’ sera with human epidermal extract was negative, while 4 sera were reactive with BP230 using guinea pig epidermal extracts [23].

In the present study, elderly patients had a rather high prevalence of anti-BMZ antibodies by IIF (12.2%), with no statistically significant differences among the patients with or without DM and with or without pruritus. We would have expected a higher frequency of anti-BMZ antibodies in diabetic patients but this hypothesis was not confirmed. A possible explanation is offered by a study of Nakagawa et al., who showed that non-enzymatic glycation of antibodies in vitro impaired antigen-antibody binding depending on glucose concentration [24]. Thus hyperglycaemia might cause a false decrease in the titre of autoantibodies.

A cytoplasmic and membrane staining pattern of basal epithelial cells was seen more frequently than the linear pattern. IgG reactivity with cytoplasm and cell membrane of BC by IIF with various substrates is sometimes described, but is generally considered as non-specific. Some authors have reported an association of this particular binding pattern with drug eruptions [25], burns [26] or oral lichen planus [27]. However, Nishikawa et al. detected this reactivity pattern in BP sera by complement IIF on human skin substrate [28]. Previous studies showed that BP180 is not only expressed in hemidesmosomes located along the basal cell membrane, but also distributed on the lateral and apical plasma membrane of basal cells as a pool [29]. Staining of human epidermis with monoclonal anti-BP180 antibodies demonstrated BP180 distribution on the entire surface of basal cells [29]. Pathogenic antibodies in BP mainly target the NC16A domain located within the extracellular portion of BP180 [30]. Several groups have shown that more than 50% of BP sera tested by ELISA react with at least one epitope other than NC16A, located both in the intra- or extra-cellular portion of BP180 [31, 32]. BP230 is a cytoplasmic protein connecting the cytoskeleton with components of the hemidesmosomal adhesion complex. Even though anti-BP230 IgG antibodies can be detected in 80-90% of BP sera, current evidence suggests that BP230 is presumably of secondary importance in the pathogenesis of BP. Laminin 5 is an extracellular glycoprotein of epithelial cell basement membranes, but it is also expressed in the cytoplasm of neoplastic epithelial cells as an early event [33]. In this study, 8 sera stained by IIF the peripheral cytoplasm and membrane of basal epithelial cells: 5 of these sera were reactive with BP180, BP230 and/or laminin 5 by ELISA, while the antigen specificity of the 3 remaining sera could not be further defined. Thus, the observed cytoplasmic and membrane staining pattern of basal epithelial cells most likely reflects reactivity with BP230, BP180, laminin 5 and other not yet identified components of the dermo-epidermal BMZ.

Conclusion

Our findings suggest that the incidence of anti-BMZ autoantibodies is not associated with diabetes mellitus or chronic pruritic disorders. The basal cell cytoplasmic and membrane staining pattern detected by IIF should be further defined as we believe that it is related to BP antibodies.

Acknowledgments

We gratefully acknowledge the excellent technical help by M. Urbankova, M. Rozkydalova and E. Podstawa and the statistical evaluation by J. Jarkovský and D. Nemethova (Institute of biostatistics and analyses, Masaryk University). This study was supported by a grant of the Czech Ministry of Health IGA NG 7353-3 (to H.J.) and by grants from the Wilhelm-Sander-Stiftung (2004.120.1 to M.H.) and from the Deutsche Forschungsgemeinschaft (He1602/8-1; 8-2 to M.H.). Conflict of interest: None.

References

1 Bornhövd E, Partscht K, Flaig MJ, Messer G. Bullöse skabiesbefall getriggertes bullöses pemphigoid. Hautarzt 2001; 52: 56-61.

2 Gengoux P, Lachapelle JM. Pemphigoid presenting as atypical excoriated prurigo. Regarding 11 cases. Dermatology 1997; 194: 392-4.

3 Laffitte E, Borradori L. Bullous pemphigoid. Challenging situations. In: Hertl M, ed. Autoimmune diseases of the skin, 1st ed. Wien NewYork: Springer, 2005: 71-93.

4 Yancey KB. The pathophysiology of autoimmune blistering diseases. J Clin Invest 2005; 115: 825-8.

5 Jedlickova H, Vlasin Z. Bullous pemphigoid and internal diseases. Abstracts of the 11^th Congress of the EADV Prague, JEADV suppl 1 2002; 16: 24.

6 Vohradníková O, Perusicová J. Prehled soucasné problematiky diabetického syndromu. In: Skin manifestations of diabetes mellitus. 1st ed. Prague: Maxdorf, 1996: 15-9.

7 Warram JH, Krolewski AS. Epidemiology of Diabetes mellitus. In: Kahn CR, Weir GC, King GL, Moses AC, Smith RJ, Jacobson AM, eds. Joslin´s Diabetes mellitus, 14th ed. Lippincott, Williams & Wilkins, 2004: 115-27.

8 Rosina P, Chieregato C, D´Onghia FS. Bullous pemphigoid and diabetes mellitus. Acta Derm Venereol 1996; 76: 497-8.

9 Chuang TY, Korkij W, Soltani K, Clayman J, Cook J. Increased frequency of diabetes mellitus in patients with bullous pemphigoid: a case-control study. J Am Acad Dermatol 1984; 11: 1099-102.

10 Turk Z, Mišur I, Turk N, Benko B. Rat tissue collagen modified by advanced glycation: correlation with duration of diabetes and glycemic control. Clin Chem Lab Med 1999; 37: 813-20.

11 Al-Mutairi N, Zaki A, Sharma AK, Al-Sheltawi M. Cutaneous manifestations of diabetes mellitus. Study from Farwaniya hospital, Kuwait. Med Princ Pract 2006; 15: 427-30.

12 Bhat YJ, Gupta V, Kudyar RP. Cutaneous manifestations of diabetes mellitus. Int J Diab Dev Ctries 2006; 26: 152-5.

13 Neilly JB, Martin A, Simpson N, MacCuish AC. Pruritus in diabetes mellitus: investigation of prevalence and correlation with diabetes control. Diabetes Care 1986; 9: 273-5.

14 Wohlrab J, Wohlrab D, Meiss F. Skin diseases in diabetes mellitus. J Dtsch Dermatol Ges 2007; 5: 37-53.

15 Toonstra J. Bullosis diabeticorum. Report of a case with a review of the literature. J Am Acad Dermatol 1985; 13: 799-805.

16 Hofmann SC, Tamm K, Hertl M, Borradori L. Diagnostic value of an enzyme-linked immunosorbent assay using BP180 recombinant proteins in elderly patients with pruritic skin disorders. Br J Dermatol 2003; 149: 910-2.

17 Thoma-Uszynski S, Uter W, Schwietzke S, Hofmann SC, Hunziker T, Bernard P, Treudler R, Zouboulis CC, Schuler G, Borradori L, Hertl M. BP230- and BP180-specific auto-antibodies in bullous pemphigoid. J Invest Dermatol 2004; 122: 1413-22.

18 Bekou V, Thoma-Uszynski S, Wendler O, Uter W, Schwietzke S, Hunziker T, Zouboulis CC, Schuler G, Sorokin L, Hertl M. Detection of laminin 5-specific auto-antibodies in mucous membrane and bullous pemphigoid sera by ELISA. J Invest Dermatol 2005; 124: 732-40.

19 Andersen-Ranberg K, Høier-Madsen M, Wiik A, Jeune B, Hegedüs L. High prevalence of autoantibodies among Danish centenarians. Clin Exp Immunol 2004; 138: 158-63.

20 Nilsson B-O, Skogh T, Ernerudh J, Johansson B, Löfgren S, Wikby A, Dahle C. Antinuclear antibodies in the oldest-old women and men. J Autoimmun 2006; 27: 281-8.

21 Rieckhoff-Cantoni L, Bernard P, Didierjean L, Imhof K, Kinloch-de-Loes S, Saurat J-H. Frequency of bullous pemphigoid-like antibodies as detected by western immunoblot analysis in pruritic dermatoses. Arch Dermatol 1992; 128: 791-4.

22 Davies MG, Pembroke AC, Bhogal B, Davies E, Vergani D. Circulating basement membrane zone antibodies in eczema. Clin Exp Dermatol 1984; 9: 36-41.

23 Hachisuka H, Kurose K, Karashima T, Mori O, Maeyama Y. Serum from normal elderly individuals contains anti-basement membrane zone antibodies. Arch Dermatol 1996; 132: 1201-5.

24 Nakagawa A, Kobayashi N, Yamashina Y, Nakatani C, Muramatsu T, Mori T, Shirai T. Effect of nonenzymatic glycosylation on the titers of circulating autoantibodies in pemphigus and pemphigoid. J Dermatol 1998; 25: 710-5.

25 van Joost T. Incidence of circulating antibodies reactive with basal cells of skin in drug reactions. Acta Derm Venereol 1974; 54: 183-7.

26 Ablin RJ, Holder IA. Antibodies to basal cells. J Invest Dermatol 1981; 76: 523.

27 Lin SC, Sun A, Wu YC, Chiang CP. Presence of anti-basal cell antibodies in oral lichen planus. J Am Acad Dematol 1992; 26: 943-7.

28 Nishikawa T, Kurihara S, Harada T, Hatano H. Binding of bullous pemphigoid antibodies to basal cells. J Invest Dermatol 1980; 74: 389-91.

29 Kitajima Y, Nojiri M, Yamada T, Hirako Y, Owaribe K. Internalization of the 180 kDa bullous pemphigoid antigen as immune complexes in basal keratinocytes: an important early event in blister formation in bullous pemphigoid. Br J Dermatol 1998; 138: 71-6.

30 Borradori L, Sonnenberg A. Structure and function of hemidesmosomes: more than simple adhesion complexes. J Invest Dermatol 1999; 112: 411-8.

31 Hofmann S, Thoma-Uszynski S, Hunziker T, Bernard P, Koebnick C, Stauber A, Schuler G, Borradori L, Hertl M. Severity and phenotype of bullous pemphigoid relate to autoantibody profile against the NH2- and COOH-terminal regions of the BP180 ectodomain. J Invest Dermatol 2002; 119: 1065-73.

32 Di Zenzo G, Grosso F, Terracina M, Mariotti F, De Pita O, Owaribe K, Mastrogiacomo A, Sera F, Borradori L, Zambruno G. Characterization of the anti-BP180 autoantibody reactivity profile and epitope mapping in bullous pemphigoid patients. J Invest Dermatol 2004; 122: 103-10.

33 Kainulainen T, Autio-Harmainen H, Oikarinen A, Salo S, Tryggvason K, Salo T. Altered distribution and synthesis of laminin-5 (kalinin) in oral lichen planus, epithelial dysplasias and squamous cell carcinomas. Br J Dermatol 1997; 136: 331-6.