[Skip to Content]
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
Case Report/Case Series
December 2014

High Anti–Desmoglein 3 Antibody ELISA Index and Negative Indirect Immunofluorescence Result in a Patient With Pemphigus Vulgaris in RemissionEvaluation of the Antibody Profile by Newly Developed Methods

Author Affiliations
  • 1Department of Dermatology and Allergology, Juntendo University Graduate School of Medicine, Tokyo, Japan
  • 2Department of Dermatology, Keio University School of Medicine, Tokyo, Japan
  • 3Department of Dermatology, Hamamatsu University School of Medicine, Hamamatsu, Japan
  • 4Department of Dermatology, Okayama University Graduate School of Medicine, Dentistry Pharmaceutical Sciences, Okayama, Japan
JAMA Dermatol. 2014;150(12):1327-1330. doi:10.1001/jamadermatol.2014.411

Importance  Pemphigus vulgaris (PV) is a disease that features blistering of the skin and mucous membranes caused by autoantibodies directed against desmoglein 3 (Dsg3) and/or desmoglein 1 (Dsg1). Indirect immunofluorescence (IIF) and enzyme-linked immunosorbent assay (ELISA) are 2 methods that are widely used to measure Dsg3 or Dsg1 antibody titers in PV. Although the titers of these autoantibodies are generally correlated with disease activity, some patients with a high ELISA index do not have severe symptoms. We encountered a patient with PV in remission, who had a high anti-Dsg3 antibody ELISA index while the IIF result was negative.

Observations  The anti-Dsg3 antibodies of our patient mainly recognized Ca2+-dependent conformational epitopes and targeted mature Dsg3 protein. We report this case focusing on the discrepancy between ELISA and IIF findings, as well as on the specific characteristics of the patient’s autoantibodies evaluated by newly developed methods.

Conclusions and Relevance  This case emphasizes that a discrepancy between disease activity, the ELISA index for Dsg3, and/or IIF findings can occur in PV. Further research on similar patients will be required to elucidate the pathogenic mechanisms in patients with PV who have nonpathogenic antibodies and show a discrepancy between ELISA and IIF.


It is known that some patients with pemphigus vulgaris (PV) with a high enzyme-linked immunosorbent assay (ELISA) index do not necessarily have severe symptoms. The autoantibodies of such patients could display interesting characteristics, and many studies of nonpathogenic autoantibodies have been performed to understand more about the pathogenesis of PV. We encountered a patient with PV in remission who had a high ELISA index for anti-Dsg3 antibody while the indirect immunofluorescence (IIF) finding was negative, and we report this case focusing on the autoantibody profile.

Report of a Case

In November 2001, a woman in her 70s presented with oral erosions. Histopathological examination of biopsy specimens obtained from the oral and esophageal mucosa by her previous physician revealed epidermal blisters and acantholysis. These findings led to a diagnosis of PV, and treatment was started with prednisolone (40 mg/d). Although oral steroid therapy was continued, the patient’s blisters and erosions gradually became widespread. In August 2004, she was referred to us because of difficulty with oral intake.

At the first visit, physical examination revealed blisters and erosions all over her body, including on the oral mucosa. She had no relevant medical history. Laboratory test results revealed that the complete blood cell count and biochemical profile were within normal limits, but the serum anti-Dsg3 and anti-Dsg1 antibody indexes (determined by ELISA) were 2020 and 120, respectively. Examination of biopsy specimens of the oral mucosa again revealed epidermal blisters and suprabasal acantholysis. Direct immunofluorescence examination of the oral and esophageal mucosa demonstrated intracellular deposits of IgG and C3 in the epidermis. Indirect immunofluorescence also revealed intercellular staining of the epidermis for IgG (Figure 1A).

Figure 1.
Indirect Immunofluorescence Findings
Indirect Immunofluorescence Findings

A, Indirect Immunofluorescence showed interkeratinocytes staining at the onset in 2004. B, Indirect immunofluorescence with normal human skin as substrates performed during remission in 2009 showed negative results. C, Indirect Immunofluorescence during remission with guinea pig esophagus as substrates also showed negative results.

The patient was already taking prednisolone (35 mg/d) on admission. She was treated with betamethasone at a dose of 4 mg/d, and double filtration plasmapheresis (DFPP) was commenced. Her skin lesions improved rapidly with DFPP, which was continued until clinical remission was achieved in October 2004 after 6 sessions. Thereafter, azathioprine was added at 100 mg/d. Remission was maintained when betamethasone was switched to prednisolone and while prednisolone was tapered until it was ceased in October 2009. Azathioprine therapy was also stopped in December 2009.

The ELISA index for anti-Dsg3 antibody decreased when remission was achieved, but increased again in 2006 and has remained in the range of 2000 to 3000 since then (Figure 2). In contrast, ELISA for anti-Dsg1 antibody has been negative since remission was achieved. Findings from IIF have generally been negative while the patient was in remission from 2009 to 2012 (Figure 1B and C), but a positive IIF result was detected at the onset of 2004 (Figure 1A). Indirect immunofluorescence was not only performed with normal human skin but also with guinea pig esophagus as a substrate because the latter is known to be a more sensitive substrate than the former.13 However, the results were the same with both substrates.

Figure 2.
Treatment and Course
Treatment and Course

The enzyme-linked immunosorbent assay (ELISA) index for anti–desmoglein 3 (anti-Dsg3) antibody (Ab) decreased when the patient achieved remission, but increased again in 2006. DFPP indicates double filtration plasmapheresis.

We (K.K., Y.A., and K.I.) performed an EDTA-treated ELISA of the patient’s serum sample. We devised a conformational ELISA index by subtracting the EDTA-ELISA index from the conventional ELISA index, which was considered to be a better indicator of disease activity than the conventional ELISA index.4 Although a slightly lower value was obtained compared with the conventional ELISA index, the titer was still elevated to approximately 2000 (Table 1). This result suggested that most of the patient’s antibodies were directed against Ca2+-dependent epitopes.

Table 1.  
The Results of Conformational ELISA Index
The Results of Conformational ELISA Index

We (K.K., Y.A., and K.I.) evaluated the effect of PV-IgG purified from this patient on cell-cell adhesion of DJM-1 cells by the previously reported method.4 The patient’s PV-IgG did not cause a decrease of cell-cell adhesion, suggesting that her autoantibodies were probably nonpathogenic.

We (J.Y.) analyzed the patient’s serum antibodies to determine whether most of them recognized precursor Dsg3 (preDsg3) molecules. Recombinant proteins of precursor and mature Dsg3 (matDsg3) using Chinese hamster ovary cells as previously reported.5 The optical density obtained by ELISA targeting matDsg3 was similar to that for preDsg3 (Table 2). After adsorption of matDsg3, the ELISA for preDsg3 became negative. In contrast, adsorption of preDsg3 did not result in a negative ELISA result for matDsg3, although it reduced the optical density to some extent. These results suggested that most of the anti-Dsg3 antibodies of our patient recognized matDsg3, while some recognized both preDsg3 and matDsg3, and there were few or none recognizing preDsg3 alone. Immunoprecipitation of the patient’s IgG targeting preDsg3 and matDsg3 also confirmed that the antibodies recognized both preDsg3 and matDsg3, supporting the ELISA findings (see eFigure in the Supplement).

Table 2.  
The Results of ELISA for preDsg3 and matDsg3
The Results of ELISA for preDsg3 and matDsg3

Our patient remained in remission for approximately 11 years, including more than 6 years without any treatment. Despite being in remission, she persistently had a high ELISA index for anti-Dsg3 antibody, although the IIF result was negative. The positive IIF result at the onset of PV contains some possibilities. The result at the onset is consistent with the blistering lesions that began on the oral mucosa. It suggests that the anti-Dsg3 antibodies at the onset were pathogenic, but they changed to nonpathogenic antibodies during the course, which was not detected with IIF. Another possibility would be that the presence of anti-Dsg1 antibodies in the early period of the disease mainly attributed to the positive IIF result because the IIF result was negative after anti-Dsg1 antibodies became undetectable in the remission phase. This suggests that the negative IIF result during remission was associated with the profile of our patient’s anti-Dsg3 antibodies. The present case emphasizes that a discrepancy between disease activity, the ELISA index for Dsg3, and/or IIF findings can occur in PV. What is the antibody profile that underlies this discrepancy?

Previous studies have shown that pathogenic monoclonal anti-Dsg3 antibodies (mAbs) predominantly recognize Ca2+-dependent conformational epitopes.6,7 It was also reported that pathogenic mAbs from patients with PV or mice preferentially bind to epitopes on matDsg, whereas nonpathogenic mAbs either only bind to preDsg or bind with both preDsg and matDsg.8,9 Moreover, Kamiya et al4 reported that the EDTA-treated ELISA revealed a decrease of anti-Dsg3 antibodies targeting Ca2+-dependent epitopes in the inactive phase of PV.

We initially predicted that the autoantibodies of our patient would mainly recognize non–Ca2+-dependent epitopes and would mainly bind to preDsg3 because antibodies targeting preDsg3 synthesized in the endoplasmic reticulum might be negative by IIF.9 However, we unexpectedly found evidence that most of the antibodies in our patient recognized Ca2+-dependent conformational epitopes and targeted matDsg3. This is consistent with a previous report that the antibodies of patients with PV predominantly recognize epitopes at the N-terminus of Dsg throughout the course of the disease.10

There have been some previous reports of similar patients with a high ELISA index and negative IIF result. In patients with rheumatoid arthritis taking thiol compounds, autoantibodies targeting nonconformational epitopes of either Dsg1 or Dsg3 have been detected by ELISA without blistering skin lesions, while the IIF result remained negative.11 The authors mentioned that the discrepancy was possibly related to a difference of peptide structure between the recombinant Dsg used for the ELISA and native human Dsg in epidermal cryosections, and these patients never developed features of PV. In addition, Li et al12 reported that epitope spread occurs in patients with endemic pemphigus foliaceus during remission and relapse. Serum obtained from some patients during remission predominantly reacted with EC5, and the IIF result was negative despite the ELISA result being positive. Therefore, they hypothesized that the EC5 domain may be sequestered in vivo because of its proximity to the cell membrane, which may prevent it from binding to circulating antibodies.

It is known that a combination of mAbs targeting different epitopes of Dsg3 shows stronger pathogenicity than a single mAb.13 Yamagami et al14 characterized Dsg-specific mAbs from patients with pemphigus and reported that specific amino acids in the heavy-chain complementarity-determining region 3 (HCDR3) are critical for pathogenicity, although antibody binding can be uncoupled from pathogenicity. In their study, replacement of a specific amino acid abolished pathogenicity and also decreased the IIF titer in some cases, although the IIF result did not become negative.14

It is also possible that a point mutation of an antigenic epitope or antibody abolished pathogenicity in the present patient and affected the IIF response, but the actual reason for the discrepancy between ELISA and IIF findings remains uncertain at present. It seems that only the result of the cell dissociation study was correlated with the pathogenicity of anti-Dsg3 antibodies in our patient.


The results of the analysis on the patient’s serum sample emphasizes that a discrepancy between disease activity, the ELISA index for Dsg3, and/or IIF findings can occur in PV. Further research into other cases like our patient will be required to elucidate the mechanisms operating in patients with PV with nonpathogenic antibodies who show a discrepancy between ELISA and IIF results.

Back to top
Article Information

Accepted for Publication: February 18, 2014.

Corresponding Author: Shigaku Ikeda, MD, PhD, Department of Dermatology and Allergology, Juntendo University Graduate School of Medicine, 2-1-1 Hongo, Bunkyo-ku, Tokyo 113-8421 (ikeda@juntendo.ac.jp).

Published Online: July 30, 2014. doi:10.1001/jamadermatol.2014.411.

Author Contributions: Drs Nakahara and Ikeda had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Nakahara, Takagi, Ikeda.

Acquisition, analysis, or interpretation of data: Nakahara, Yamagami, Kamiya, Aoyama, Iwatsuki.

Drafting of the manuscript: Nakahara, Ikeda.

Critical revision of the manuscript for important intellectual content: All authors.

Obtained funding: Yamagami, Ikeda.

Administrative, technical, or material support: Nakahara, Kamiya, Aoyama, Iwatsuki, Ikeda.

Study supervision: Takagi, Ikeda.

Conflict of Interest Disclosures: None reported.

Funding/Support: This work was supported partially by “Research on Measures for Intractable Diseases” Project matching fund subsidy (H23-028) from Ministry Health, Labor, and Welfare, Japan (Drs Takagi, Kamiya, Aoyama, Iwatsuki, and Ikeda).

Role of the Sponsor: The Ministry Health, Labor, and Welfare had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Additional Contributions: Takatoshi Kuhara, DVM, PhD, Atopy Research Center, Juntendo University Graduate School of Medicine, provided assistance for indirect immunofluorescence with guinea pig esophagus. No financial compensation was provided.

Schmidt  E, Dähnrich  C, Rosemann  A,  et al.  Novel ELISA systems for antibodies to desmoglein 1 and 3: correlation of disease activity with serum autoantibody levels in individual pemphigus patients. Exp Dermatol. 2010;19(5):458-463.PubMedArticle
Ng  PP, Thng  ST, Mohamed  K, Tan  SH.  Comparison of desmoglein ELISA and indirect immunofluorescence using two substrates (monkey oesophagus and normal human skin) in the diagnosis of pemphigus. Australas J Dermatol. 2005;46(4):239-241.PubMedArticle
Zagorodniuk  I, Weltfriend  S, Shtruminger  L,  et al.  A comparison of anti-desmoglein antibodies and indirect immunofluorescence in the serodiagnosis of pemphigus vulgaris. Int J Dermatol. 2005;44(7):541-544.PubMedArticle
Kamiya  K, Aoyama  Y, Shirafuji  Y,  et al.  Detection of antibodies against the non-calcium-dependent epitopes of desmoglein 3 in pemphigus vulgaris and their pathogenic significance. Br J Dermatol. 2012;167(2):252-261.PubMedArticle
Toumi  A, Saleh  MA, Yamagami  J,  et al.  Autoimmune reactivity against precursor form of desmoglein 1 in healthy Tunisians in the area of endemic pemphigus foliaceus. J Dermatol Sci. 2013;70(1):19-25.PubMedArticle
Anzai  H, Fujii  Y, Nishifuji  K,  et al.  Conformational epitope mapping of antibodies against desmoglein 3 in experimental murine pemphigus vulgaris. J Dermatol Sci. 2004;35(2):133-142.PubMedArticle
Tsunoda  K, Ota  T, Aoki  M,  et al.  Induction of pemphigus phenotype by a mouse monoclonal antibody against the amino-terminal adhesive interface of desmoglein 3. J Immunol. 2003;170(4):2170-2178.PubMedArticle
Yokouchi  M, Saleh  MA, Kuroda  K,  et al.  Pathogenic epitopes of autoantibodies in pemphigus reside in the amino-terminal adhesive region of desmogleins which are unmasked by proteolytic processing of prosequence. J Invest Dermatol. 2009;129(9):2156-2166.PubMedArticle
Sharma  PM, Choi  EJ, Kuroda  K, Hachiya  T, Ishii  K, Payne  AS.  Pathogenic anti-desmoglein MAbs show variable ELISA activity because of preferential binding of mature versus proprotein isoforms of desmoglein 3. J Invest Dermatol. 2009;129(9):2309-2312.PubMedArticle
Ohyama  B, Nishifuji  K, Chan  PT,  et al.  Epitope spreading is rarely found in pemphigus vulgaris by large-scale longitudinal study using desmoglein 2-based swapped molecules. J Invest Dermatol. 2012;132(4):1158-1168.PubMedArticle
Yamamoto  T, Takata-Michigami  M, Hisamatsu  Y,  et al.  A prospective analysis of anti-desmoglein antibody profiles in patients with rheumatoid arthritis treated with thiol compounds. J Dermatol Sci. 2010;59(3):170-175.PubMedArticle
Li  N, Aoki  V, Hans-Filho  G, Rivitti  EA, Diaz  LA.  The role of intramolecular epitope spreading in the pathogenesis of endemic pemphigus foliaceus (fogo selvagem). J Exp Med. 2003;197(11):1501-1510.PubMedArticle
Müller  R, Svoboda  V, Wenzel  E, Müller  HH, Hertl  M.  IgG against extracellular subdomains of desmoglein 3 relates to clinical phenotype of pemphigus vulgaris. Exp Dermatol. 2008;17(1):35-43.PubMed
Yamagami  J, Payne  AS, Kacir  S, Ishii  K, Siegel  DL, Stanley  JR.  Homologous region of antibody H-chain CDR-3 in patients with pemphigus cause pathogenicity. J Clin Invest. 2010;120(11):4111-4117.PubMedArticle