E-cadherin Is an Additional Immunological Target for Pemphigus Autoantibodies

ORIGINAL ARTICLE

E-cadherin Is an Additional Immunological Target for Pemphigus Autoantibodies Flor Evangelista1, David A. Dasher1, Luis A. Diaz1, Phillip S. Prisayanh1 and Ning Li1 Pemphigus foliaceus (PF) and pemphigus vulgaris (PV) are autoimmune blistering diseases characterized by autoantibodies against desmoglein (Dsg)1 and Dsg3, respectively. The role of classical cadherins as immunological targets of pemphigus autoantibodies is unknown. In this study, we tested the reactivity of sera from patients with PF, Fogo Selvagem (FS), and PV by immunoprecipitation coupled with immunoblotting (IP–IB) and ELISA techniques using a baculovirus-expressed ectodomain of E-cadherin. By IP–IB, anti-E-cadherin reactivity was detected in all tested sera of PF (n ¼ 13) and FS (n ¼ 15) patients, and in 79% of mucocutaneoustype PV patients (n ¼ 33), but in none of the mucosal-type PV patients (n ¼ 7). By ELISA, anti-E-cadherin IgG was detected in most pemphigus sera that produced strong E-cadherin bands by IP–IB. The immunoreactivity of PF/FS sera with E-cadherin was also demonstrated by IP–IB using human epidermal extracts. However, immunofluorescence staining of A431DE cells (E-cadherin positive, Dsg1 negative) with pemphigus sera showed negative results. Immunoadsorption and competitive ELISA analysis suggest that most of the anti-E-cadherin antibodies cross-react with Dsg1, whereas others may represent independent antibodies that do not cross-react with Dsg1. The functional relevance of these anti-E-cadherin IgG autoantibodies detected in these pemphigus sera remains to be defined. Journal of Investigative Dermatology (2008) 128, 1710–1718; doi:10.1038/sj.jid.5701260; published online 24 January 2008

INTRODUCTION Adherens junctions and desmosomes are two intercellular junctions in the epidermis that allow keratinocytes to adhere to one another and maintain the integrity of the epithelium. Classical cadherins (located in adherens junctions) and desmosomal cadherins, sharing high sequence homology in their ectodomain, play a particularly important role in the dynamic regulation of intercellular adhesion (Patel et al., 2003; Getsios et al., 2004; Gumbiner, 2005 [20]). E-cadherin and P-cadherin are classical cadherins expressed in the epidermis. E-cadherin is expressed in all layers of the epidermis, whereas P-cadherin is limited to the basal cell layer (Hirai et al., 1989). The desmosomal cadherins consist of four desmogleins (Dsg1–Dsg4) and three desmocollins (Dsc1–Dsc3), all encoded by separate genes. The expression of different members of the desmosomal cadherins is tissuedependent and spatial-regulated (Getsios et al., 2004). Dsg1 1

Department of Dermatology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA Correspondence: Dr Ning Li and Dr Luis A. Diaz, Department of Dermatology, University of North Carolina at Chapel Hill, 3100 Thurston Building, CB no. 7287, Chapel Hill, North Carolina 27599, USA. E-mails: [email protected] and [email protected] Abbreviations: BP, bullous pemphigoid; Dsc, desmocollin; Dsg, desmoglein; FS, Fogo Selvagem; His, histidine; IB, immunoblotting; IF, immunofluorescence; IP, immunoprecipitation; mcPV, mucocutaneous-type pemphigus vulgaris; mPV, mucosal-type pemphigus vulgaris; PF, pemphigus foliaceus; PV, pemphigus vulgaris; rDsg1, recombinant Dsg1 Received 4 January 2007; revised 15 November 2007; accepted 20 November 2007; published online 24 January 2008

1710 Journal of Investigative Dermatology (2008), Volume 128

and Dsc1 are the major desmosomal cadherins in the skin where they are expressed throughout the epidermis but most prominently in the upper layers. Expression of Dsg3 and Dsc3 is predominant in the lower epidermis and decreases gradually toward the upper layers. Dsg1/Dsg3 and Dsc1/ Dsc3 are mostly restricted to the stratified epithelia, whereas Dsg2 and Dsc2 are expressed ubiquitously in most desmosome-containing tissues. The expression of Dsg2 in the epidermis is low and mostly confined to the basal cell layer. Dsg4 is a newly identified member of the Dsg family expressed in the suprabasal layers of the epidermis (Kljuic et al., 2003; Whittock and Bower, 2003). Pemphigus is a group of autoimmune skin blistering diseases characterized by pathogenic autoantibodies against Dsg1 and Dsg3 with resultant loss of epithelial cell–cell adhesion, known as acantholysis, leading to blister and bulla formation (Stanley et al., 1984; Amagai et al., 1991). Pemphigus vulgaris (PV) and pemphigus foliaceus (PF) are the two classical forms of pemphigus (Lever, 1953). Fogo Selvagem (FS) is an endemic form of PF found in certain regions of Brazil where environmental factors may trigger disease initiation and progression (Diaz et al., 1989). Blisters and erosions on the skin and mucous membranes with suprabasilar acantholysis seen histologically characterize PV, whereas PF and FS affect only the skin and produce subcorneal acantholysis (Lever, 1953). It is well established that the sera of PF and FS patients possess anti-Dsg1 IgG autoantibodies (Stanley et al., 1984, 1986), whereas the sera of PV patients contain anti-Dsg3 IgG autoantibodies (Amagai et al., 1991). On the basis of the clinical and the anti-Dsg & 2008 The Society for Investigative Dermatology

F Evangelista et al. E-cadherin Autoantibodies in Pemphigus Sera

autoantibody profile, PV is further divided into two subsets: the mucosal type (mPV) and the mucocutaneous type (mcPV) (Ding et al., 1997). Patients with mcPV exhibit mucosal and skin disease clinically, with autoantibodies against Dsg1 and Dsg3 detected by serological evaluation; mPV patients, however, lack cutaneous disease and possess solely anti-Dsg3 autoantibodies (Ding et al., 1997; Amagai et al., 1999; Harman et al., 2001; Jamora et al., 2003). Passive transfer of total IgG and affinity-purified anti-Dsg1 IgG or anti-Dsg3 IgG from sera of patients with PF/FS or PV to neonatal mice induces skin blisters in these animals that mirror the key clinical and histological findings of each corresponding human disease (Anhalt et al., 1982; Roscoe et al., 1985; Amagai et al., 1992, 1995; Ding et al., 1999). Anti-Dsg3 autoantibodies are detected in the sera of a small number of PF/FS patients (o7%) (Arteaga et al., 2002). Other desmosomal cadherins such as Dsc1–Dsc3 are recognized by IgA antibodies from some clinical variants of IgA pemphigus but not by typical PV or PF IgG (Hisamatsu et al., 2004). Several additional antigens have also been reported to react with some PV or PF sera. These include Dsg4 (Kljuic et al., 2003; Nagasaka et al., 2004), pemphaxin (Nguyen et al., 2000a), acetylcholine receptor (Nguyen et al., 2000b), the a-chain of the high-affinity IgE receptor (Fiebiger et al., 1998), and thyroid peroxidase (Pitoia et al., 2005). In this study, we have evaluated the immunoreactivity of PF, FS, and PV sera with baculovirus-expressed recombinant E-cadherin. We found that many sera from patients with PF, FS, and mcPV reacted with E-cadherin. RESULTS Construction and baculovirus expression of the ectodomains of E-cadherin, Dsg1, and Dsg3

The recombinant baculovirus encoding the entire ectodomains of human Dsg1 and Dsg3, each containing a C-terminal histidine (His)-tag, had previously been constructed and expressed as soluble proteins (Ding et al., 1997). For this study, the recombinant baculovirus encoding the entire ectodomain of human E-cadherin fused with a C-terminal His-tag was generated. The three recombinant proteins were produced as secreted forms in High Five cells infected by baculovirus. Figure 1a depicts the immunoblot analysis of the culture supernatants using a mouse monoclonal anti-His antibody showing the presence of E-cadherin, Dsg1, and Dsg3. The expression of E-cadherin was confirmed by immunoblot analysis using a goat polyclonal antibody to human E-cadherin (R&D Systems, Minneapolis, MN) (Figure 1b, lane 1). E-cadherin ectodomain is recognized by PF and FS sera

Serum samples from PF (n ¼ 13) and FS (n ¼ 15) patients were tested by immunoprecipitation coupled with immunoblotting (IP–IB) for their immunoreactivity with the E-cadherin ectodomain. All PF (13/13, 100%) and FS (15/15, 100%) sera immunoprecipitated E-cadherin producing strong bands by the majority of sera and weak bands by others. These results are shown in Table 1 and Figure 2. As expected, anti-Dsg1 autoantibodies were detected in all sera from PF and FS patients with active disease. Anti-Dsg3 autoantibodies

1

2

3

1

2

3

Figure 1. Baculovirus expression of the ectodomain of human E-cadherin, Dsg1, and Dsg3. IB analysis of the High Five culture supernatants containing the secreted recombinant proteins E-cadherin (lane 1), Dsg1 (lane 2), and Dsg3 (lane 3). Proteins were electrophoresed on 10% SDS-PAGE, transferred to a nitrocellulose membrane, and probed with antibodies to the (a) His-tag or (b) human E-cadherin.

Table 1. Prevalence of antibodies against E-cadherin, Dsg1, and Dsg3 in pemphigus patients and controls by IP-IB IP positive Number of patients

AntiE-cadherin

Anti-Dsg1 Anti-Dsg3

PF

13

13 (100%)

13 (100%)

0 (0%)

FS

15

15 (100%)

15 (100%)

1 (7%)

Pemphigus group

mPV

7

0 (0%)

0 (0%)

7 (100%)

mcPV

33

26 (79%)

28 (85%)

33 (100%)

BP

20

0 (0%)

3 (15%)1

0 (%)

SLE

3

0 (0%)

0 (0%)

0 (0%)

EBA

2

0 (0%)

0 (0%)

0 (0%)

Sezary syndrome

1

0 (0%)

0 (0%)

0 (0%)

Normal subjects

19

0 (0%)

0 (0%)

0 (0%)

Non-pemphigus group

BP, bullous pemphigoid; Dsg, desmoglein; EBA, epidermolysis bullosa acquisita; FS, Fogo Selvagem; IB, immunoblotting; IP, immunoprecipitation; mcPV, mucocutaneous-type pemphigus vulgaris; mPV, mucosaltype pemphigus vulgaris; PF, pemphigus foliaceus; SLE, systemic lupus erythematosus. 1 One of the three BP sera was also positive by E-cadherin ELISA and Dsg1 ELISA (see also Table 2). The other two BP samples showed no antiE-cadherin autoantibodies by ELISA and IP.

were detected in only 1 out of 15 FS sera (7%) tested by IP–IB, a finding that confirms a previous report (Arteaga et al., 2002). PF sera did not react with Dsg3 by IP–IB. The immunoreactivity of PF and FS sera with the E-cadherin ectodomain was also tested by an E-cadherin ELISA (Table 2). Using the cutoff point of 30, as determined by receiver-operating characteristic (ROC) analysis (sensitivity: 79% and specificity: 90%), we found positive tests in 46% of PF sera (n ¼ 13) and 79% of FS sera (n ¼ 89). The majority of PF and FS sera exhibiting moderate-to-strong E-cadherin bands by IP were also positive by ELISA (15 out of www.jidonline.org 1711

F Evangelista et al. E-cadherin Autoantibodies in Pemphigus Sera

Table 3. Correlation between E-cadherin ELISA and E-cadherin IP

E-cadherin

PF/FS

Dsg1

mcPV

Strong Weak Negative Strong Weak Negative IP (n=17) IP (n=11) IP (n=0) IP (n=12) IP (n=14) IP (n=7)

Dsg3 + – 1 2 3 4 5 6 7 8 9 10 11 12 PF

FS

Figure 2. Anti-E-cadherin IgG autoantibodies detected in PF and FS patients. Representative IP–IB results from six PF (lanes 1–6) and six FS (lanes 7–12) sera are shown. The E-cadherin- and Dsg1-recombinant proteins were immunoprecipitated by all of the PF and FS sera, but none immunoprecipitated Dsg3. Well-characterized serum samples from a PF or PV patient were used as positive controls ( þ ); normal human serum was used as a negative control (). One PF serum (lane 3) and another FS (lane 7) produced a weak E-cadherin band.

ELISA (+) 15 (88%) 3 (27%)

0

8 (67%)

ELISA ()

0

4 (33%) 13 (93%) 7 (100%)

2 (12%) 8 (73%)

1 (7%)

0

FS, Fogo Selvagem; IP, immunoprecipitation; mcPV, mucocutaneous-type pemphigus vulgaris; PF, pemphigus foliaceus.

E-cadherin Dsg1 Dsg3 + – 1 2 3 4 5 6

Table 2. Prevalence of antibodies against E-cadherin by ELISA Group of patients

Number of patients

ELISA-positive anti-E-cadherin

Pemphigus group PF

13

6 (46%)

FS

89

70 (79%)

mPV

7

0(0%)

mcPV

33

9 (27%)

BP

20

1 (5%)1

SLE

3

0 (0%)

EBA

2

0 (0%)

Sezary syndrome

1

0 (0%)

Normal subjects

60

6 (10%)

Non-pemphigus group

BP, bullous pemphigoid; EBA, epidermolysis bullosa acquisita; FS, Fogo Selvagem; mcPV, mucocutaneous-type pemphigus vulgaris; mPV, mucosaltype pemphigus vulgaris; PF, pemphigus foliaceus; SLE, systemic lupus erythematosus. 1 This BP serum was also positive for Dsg1 autoantibodies by IP.

17 cases). Sera that produced weak IP bands were often negative by ELISA (8 out of 11 cases) (Table 3). These results may suggest that certain E-cadherin epitopes producing weak bands by IP–IB were undetectable by ELISA. The ectodomain of E-cadherin is recognized by sera from mcPV, but not mPV, patients

A total of 40 PV sera from patients with active disease were included in this study. Seven of these patients had mPV and 33 had mcPV. As summarized in Table 1, anti-Dsg3 autoantibodies were detected in the sera of all PV patients (40 out of 40, 100%). Anti-Dsg1 autoantibodies were present in 28 out of 33 mcPV sera (85%) and were absent in the sera 1712 Journal of Investigative Dermatology (2008), Volume 128

mPV

+ – 1 2 3 4 5 6 7 8 9 10 mcPV

Figure 3. Anti-E-cadherin IgG autoantibodies detected in PV patients. Representative IP–IB results from (a) mPV and (b) mcPV sera are shown. All PV sera reacted with Dsg3. E-cadherin was not recognized by mPV sera but was bound by the majority of mcPV sera. A clear correlation between the reaction with E-cadherin and Dsg1 was observed. The reaction varies from strong reaction (lanes 1–3), moderate reaction (lanes 4, 8, and 9), weak reaction (lane 5) to no reaction (lanes 6, 7, 10). ( þ ) Positive control; (), negative control.

of 7 mPV patients (0%). Sera from 26 out of 33 mcPV patients immunoprecipitated E-cadherin (79%). In contrast, the sera from seven mPV patients did not bind E-cadherin. Representative IP–IB results for the two types of PV are shown in Figure 3a and b, respectively. The E-cadherin ELISA analysis of the 40 samples of PV sera is shown in Table 2. Using this assay, seven mPV sera produced negative results. In contrast, 9 out of 33 mcPV sera tested (27%) exhibited positive results. The low percentage of positive E-cadherin sera by ELISA (27%, Table 2) compared with the IP results (85%, see Table 1) seem large owing to the sensitivity and nature of the two assays. As summarized in Table 3, the majority of mcPV sera that showed moderate-to-strong E-cadherin bands by IP produced positive E-cadherin ELISA tests (8 out of 12, 67%). In contrast, 13 out of 14 mcPV sera (93%) that showed weak E-cadherin binding by IP produced negative E-cadherin ELISA results. Of note, 4 out of 12 mcPV sera that showed strong (n ¼ 1) or moderate (n ¼ 3) E-cadherin bands by IP produced negative ELISA results. This discrepancy may be explained by differences in epitope availability in the setting of IP, a true liquid-phase reaction, versus ELISA, a semiliquidphase assay. Reactivity of E-cadherin ectodomain with sera from clinically normal controls and patients with other diseases

A set of sera from patients with bullous pemphigoid (BP) (n ¼ 20), epidermolysis bullosa acquisita (n ¼ 2), lupus (n ¼ 3), Sezary syndrome (n ¼ 1), and normal subjects (n ¼ 19) were also tested by IP–IB for the presence of

F Evangelista et al. E-cadherin Autoantibodies in Pemphigus Sera

autoantibodies against Dsg1, Dsg3, and E-cadherin (Table 1). A representative IP–IB is shown in Figure 4. None of the 45 non-pemphigus sera reacted with E-cadherin or Dsg3 by IP. Three of the BP sera tested (n ¼ 20) reacted weakly with Dsg1 using this assay (15%). One of the three Dsg1-positive BP sera also reacted with E-cadherin by ELISA (Table 2). We and others have reported that approximately 10% of BP sera recognized Dsg1 (Ishii et al., 1997; Warren et al., 2000). The same panel of non-pemphigus patients’ sera was also tested by E-cadherin ELISA, and the results are shown in Table 2. Using this assay, 6 out of 60 normal human sera (10%) and 1 out of 20 BP sera (5%) produced positive results (readings above 30). The six normal human sera that produced positive E-cadherin ELISA scores were negative by IP. The only E-cadherin ELISA-positive BP serum (n ¼ 20) was negative by IP for E-cadherin but positive for Dsg1. This sample was a typical BP serum that stained the epidermal side of salt-split skin at a titer of 1:640 but did not bind the epidermal intercellular spaces as shown by indirect immunofluorescence (IF) techniques. Correlation between anti-E-cadherin and anti-Dsg1 autoantibodies tested by IP–IB

As shown in Table 4, a strong correlation was found between the presence of anti-Dsg1 and anti-E-cadherin antibodies in the sera of PF/FS (n ¼ 28) and mcPV patients (n ¼ 33). As shown in this table, all Dsg1-positive PF/FS sera possess anti-E-cadherin autoantibodies. On the other hand, five mcPV sera and seven mPV sera with no detectable antiDsg1 autoantibodies showed negative E-cadherin reactivity. Only 2 out of 28 Dsg1-positive mcPV sera (7%) lacked antiE-cadherin autoantibodies by IP–IB. Immunoadsorption studies

The possible cross-reactivity between anti-E-cadherin and anti-Dsg1 was tested in four pemphigus sera by liquid-phase immunoadsorption followed by IP–IB. As shown in Figure 5a, preincubation of sera from an FS and a PF patient with control culture supernatant (containing no recombinant Dsg1 (rDsg1)) did not affect the reactivity of E-cadherin (lanes 1 and 3). However, preincubation of these sera with a culture supernatant containing rDsg1 eliminated the E-cadherin reactivity (lanes 2 and 4). Similarly, preincubation of sera from two mcPV patients (Figure 5b) with a culture supernatant containing no recombinant proteins (lane 1) or rDsg3 (lane 3) did not abolish the E-cadherin reactivity. In contrast, preincubation with rDsg1 inhibited the E-cadherin reactivity (lane 2). These results indicated that at least in the four pemphigus sera tested, anti-E-cadherin antibodies crossreacted with Dsg1. To further determine whether all anti-Dsg1 antibodies cross-reacted with E-cadherin, we performed a competitive inhibition of the ELISA. As shown in Figure 6a, when eight pemphigus sera (three PF and five FS) were preincubated with E-cadherin and then tested by Dsg1-ELISA, there was minimal inhibition of the binding (5–9%) of the anti-Dsg1 reactivity in three sera (PF-3, FS-2, and FS-4) and a moderate inhibition (22–38%) in five sera (PF-2, PF-4, FS-3, FS-5, and FS-6). These

E-cadherin Dsg1 Dsg3 + – 1 2 3 4 5

6 7 8 9 10 11 12

Lupus

BP sera

Figure 4. IP–IB analysis of sera from patients with other diseases. Representative results from sera of other diseases are shown: systemic lupus erythematosus (lanes 1–3), epidermolysis bullosa acquisita (lane 4), Sezary syndrome (lane 5), and BP (lanes 6–12). A weak positive reaction with two BP sera to Dsg1 was detected (lanes 6 and 10). None of these sera reacted with E-cadherin or Dsg3. ( þ ) Positive control; () negative control.

Table 4. Correlation between antibodies against E-cadherin and Dsg1 in pemphigus patients by IP-IB mcPV

PF/FS Anti-Dsg1 (+) (n=28)

Anti-Dsg1 (+) (n=28)

Anti-Dsg1 () (n=5)

Anti-E-cadherin (+)

28 (100%)

26 (93%)

0 (0%)

Anti-E-cadherin ()

0 (0%)

2 (7%)

5/5 (100%)

Dsg, desmoglein; FS, Fogo Selvagem; IB, immunoblotting; IP, immunoprecipitation; mcPV, mucocutaneous-type pemphigus vulgaris; PF, pemphigus foliaceus.

results indicate that the majority of anti-Dsg1 autoantibodies in these sera do not or minimally cross-react with E-cadherin. Figure 6b shows the reverse experiment, that is, preincubation of the same set of sera with Dsg1 followed by examination of the adsorbed sera by the E-cadherin ELISA. As shown in this figure, preadsorption of these sera with Dsg1 reduced E-cadherin ELISA reactivity dramatically in three sera (77–85% inhibition) (PF-2, PF-3, and PF-4), moderately in three sera (47–64% inhibition) (FS-2, FS-3, and FS-6), and minimally in two sera (7–12% inhibition) (FS-4 and FS-5). These data indicate that not all anti-E-cadherin antibodies in PF/FS sera cross-react with Dsg1. Immunofluorescence staining of A431 by pemphigus sera

To test whether anti-E-cadherin autoantibodies present in pemphigus sera recognize E-cadherin on cells, we performed IF staining on A431D cells (E-cadherin negative) and A431DE cells (E-cadherin positive). These two cell lines were generated from the A431 cell line at Dr Margaret Wheelock’s laboratory (Lewis et al., 1997). The A431 cell line was originated from a human epithelial carcinoma and is known to express E-cadherin and Dsg3, but not Dsg1 (Ishii et al., 2005). We tested three pemphigus sera (two PF and one FS) and none produced a consistent IF staining (data not shown). PF sera recognize E-cadherin from normal human epidermal extract

To demonstrate that anti-E-cadherin autoantibodies from pemphigus sera also recognize native E-cadherin, we immunoprecipitated this antigen from a human epidermal www.jidonline.org 1713

F Evangelista et al. E-cadherin Autoantibodies in Pemphigus Sera

1

2

3

4

FS-1

PF-1

1 1

2

3

1

mcPV-1

2

3

mcPV-2

Figure 5. Elimination of E-cadherin IP reactivity in pemphigus serum by preincubation with rDsg1. Sera samples from (a) FS and PF patients or (b) mcPV patients were preincubated with a culture supernatant containing rDsg1, rDsg3, or no recombinant protein and then IP with E-cadherin. The precipitations were subjected to IB probed with a goat antiserum against E-cadherin. As shown, preincubation with a culture supernatant containing no recombinant protein did not affect the reactivity of E-cadherin (a, lanes 1 and 3; b, lane 1). However, preincubation with a culture supernatant containing rDsg1 eliminated or greatly reduced the E-cadherin reactivity (a, lanes 2 and 4; b, lane 2). In contrast, preincubation with a culture supernatant containing rDsg3 did not affect the reactivity of E-cadherin (b, lane 3).

Inhibition Reactivity left Dsg1 ELISA (using serum preincubated with E-cadherin)

% Inhibition (black)

a 100 90 80 70 60 50 40 30 20 10 0

PF-2

PF-4

FS-2 FS-3 Serum

FS-4

FS-5

FS-6

Inhibition Reactivity left E-cadherin ELISA (serum preincubated with Dsg1)

b % Inhibition (black)

PF-3

100 90 80 70 60 50 40 30 20 10 0 PF-2

PF-3

PF-4

FS-2 FS-3 Serum

FS-4

FS-5

FS-6

Figure 6. Competitive inhibition of the ELISA. A set of eight pemphigus sera were (a) preincubated with purified recombinant E-cadherin and tested for Dsg1 binding or (b) preincubated with purified rDsg1 and tested for E-cadherin binding by ELISA. The black section of each bar shows the % inhibition.

1714 Journal of Investigative Dermatology (2008), Volume 128

2

3

4

5

6

Figure 7. IP of normal human epidermal extracts with PF/FS sera. Epidermal extracts from neonatal foreskins were immunoprecipitated with a normal human serum (lane 1), PF sera (lanes 2–5), or an FS serum (lane 6) and subjected to IB using a mouse mAb to human E-cadherin intracellular domain. As shown, all the five patients’ sera precipitated the 120-kDa E-cadherin (lanes 2–6). The normal human serum did not precipitate E-cadherin (lane 1).

extract with five pemphigus sera (four PF and one FS). The immunoprecipitates were analyzed by IB using a mouse mAb to human E-cadherin. As shown in Figure 7, all the five pemphigus sera tested immunoprecipitated a 120-kDa protein, which reacted with the mAb to E-cadherin (lanes 2–6). In contrast, a normal human serum failed to precipitate E-cadherin (lane 1). DISCUSSION Experimental evidence accumulated over the last several years supports the notion that the primary autoantibodies in PF and PV are anti-Dsg1 and anti-Dsg3 IgG autoantibodies, respectively. These autoantibodies are pathogenic (Anhalt et al., 1982; Roscoe et al., 1985; Amagai et al., 1992, 1995; Ding et al., 1999; Arteaga et al., 2002) and recognize epitopes located on the EC1–2 domains of these desmosomal cadherins (Sekiguchi et al., 2001; Li et al., 2003). Binding of these pathogenic autoantibodies is calcium dependent and conformation sensitive (Eyre and Stanley, 1987; Kowalczyk et al., 1995). In this study, we provide in vitro evidence that a classical cadherin, E-cadherin, is a possible additional immunological target for pemphigus autoantibodies. Using IP–IB analysis, the ectodomain of E-cadherin was recognized by IgG from the sera of most PF, FS, and mcPV pemphigus patients tested (Figures 2 and 3, Table 1). In contrast, sera from normal controls or patients with other diseases such as BP, epidermolysis bullosa acquisita, lupus, and Sezary syndrome did not recognize E-cadherin (Figure 4). Interestingly, anti-E-cadherin antibodies were detected only in pemphigus patients with cutaneous disease (PF, FS, and mcPV). Patients with mPV lacked anti-E-cadherin antibodies by IP–IB analysis. The presence of anti-E-cadherin antibodies in pemphigus sera was also tested by ELISA (Table 2). The majority of sera that produced strong to moderate E-cadherin bands by IP–IB showed positive reaction by ELISA, whereas sera that reacted weakly with E-cadherin by IP–IB were generally negative by ELISA (Table 3). These results indicate that the IP–IB assay is more sensitive to detect anti-E-cadherin antibodies than the ELISA. The immunoreactivity of pemphigus sera with native E-cadherin was further demonstrated by IP–IB using normal

F Evangelista et al. E-cadherin Autoantibodies in Pemphigus Sera

human epidermal extracts (Figure 7). However, IF staining of A431DE cells (E-cadherin positive but Dsg1 negative) using three PF/FS sera produced unconvincing results (data not shown). Two possible explanations may account for the negative IF staining: (1) the epitopes recognized by pemphigus anti-E-cadherin antibodies are masked in vivo, and (2) the IF assay was not sensitive enough to detect the low levels of autoantibodies in the sera of these patients. The presence of anti-E-cadherin antibodies was correlated with the presence of anti-Dsg1 autoantibodies in pemphigus patients (Table 4). This observation suggests the possible existence of a population of autoantibodies capable of recognizing epitopes present on both antigens because of the great degree of homology of these molecules. Indeed, the IP reactivity of four pemphigus sera (two PF and two PV) with E-cadherin was abolished upon adsorption with Dsg1 but not with Dsg3 (Figure 5). Competitive ELISA analysis (Figure 6), using another set of pemphigus sera (n ¼ 8), suggests the existence of at least three IgG autoantibody populations in these sera. The first appears to encompass a large population of anti-Dsg1 antibodies that do not cross-react with E-cadherin (that is, they are not removed by E-cadherin adsorption). The second comprises a small population of antiE-cadherin antibodies that do not cross-react with Dsg1 (that is, they are not removed by Dsg1 adsorption). Finally, the third comprises a small population of cross-reactive antibodies that recognize both Dsg1 and E-cadherin. Ongoing epitope mapping studies are aimed at verifying these observations. Once epitope-specific anti-E-cadherin autoantibodies are available, it would be possible to explore their pathogenicity by passive transfer studies in the experimental mouse model of pemphigus (Anhalt et al., 1982) The potential pathogenic role of the anti-E-cadherin autoantibodies in pemphigus sera is as yet unknown. However, several observations indicate that E-cadherin may be involved in the pathogenesis of pemphigus acantholysis. It has been shown that the expression of E-cadherin is markedly reduced on acantholytic keratinocytes (Furukawa et al., 1994) and increased in the serum of pemphigus patients (Matsuyoshi et al., 1995). The addition of PV IgG to keratinocyte cultures increases phosphorylation of E-cadherin in addition to other desmosomal and adherens junctional proteins such as Dsg3, b-catenin, and plakoglobin (Aoyama et al., 1999; Nguyen et al., 2004), as well as p38MAPK and heat-shock protein 27 (Berkowitz et al., 2005), a process quite likely linked to the development of acantholysis. It is also noteworthy that E-cadherin functions not only in adhesion but also in desmosome assembly and cell signaling (Lewis et al., 1997; Goodwin and Yap, 2004; Yin and Green, 2004). Antibodies blocking E-cadherin function are known to affect desmosome assembly (Lewis et al., 1994) and promote cell apoptosis (Fouquet et al., 2004), processes thought to play roles in pemphigus acantholysis (Kitajima, 2002; Wang et al., 2004; Calkins et al., 2006; Li et al., 2006). Another intriguing observation is that upon binding of pemphigus autoantibodies to the epidermal cell surface, widening of the intercellular spaces in regions between the desmosomal junctions, where E-cadherin is the major adhesive molecule

between keratinocytes (Horiguchi et al., 1994), precedes desmosome splitting (Hashimoto and Lever, 1970; Takahashi et al., 1985; Futamura et al., 1989; Horiguchi et al., 1994). It is plausible that autoantibodies in PF/FS and mcPV may target adhesion molecules in both desmosomes and adherens junctions. An orchestrated dysfunction of both Dsg1mediated and E-cadherin-mediated adhesion may contribute to pemphigus epidermal injury. This possibility needs to be tested in future studies. In conclusion, these studies provide in vitro evidence that the majority of PF, FS, and mcPV sera possess IgG reactivity to E-cadherin ectodomain. These findings underscore the complexity and heterogeneity of the humoral response in pemphigus autoimmunity and raise an open question of whether this anti-E-cadherin reactivity in pemphigus sera is relevant to disease. MATERIALS AND METHODS Sources of human sera Sera from patients with PF (n ¼ 13), FS (n ¼ 89), PV (n ¼ 40), BP (n ¼ 20), epidermolysis bullosa acquisita (n ¼ 2), lupus (n ¼ 3), and Sezary syndrome (n ¼ 1), as well as sera from clinically normal individuals (n ¼ 60), from the blood bank of the University of North Carolina (UNC) were tested. Patients with PV were further divided into two groups: those with mPV (n ¼ 7) and those with mcPV (n ¼ 33). Diagnoses had previously been confirmed using established clinical, histological, and serological criteria. This study was approved by the Institutional Review Board at UNC. All clinical investigation was conducted according to the Declaration of Helsinki Principles. Participants gave their written informed consent.

Subcloning of the E-cadherin ectodomain into a baculovirus expression vector cDNA encoding the entire ectodomain of human E-cadherin was amplified by PCR using a template of the full-length cDNA clone constructed in a eukaryotic expression vector pcDNA3 (a generous gift from Dr Barry M Gumbiner, Department of Cell Biology, University of Virginia). The following pair of PCR primers was used for amplification: forward primer: 50 -cgcgcggccgccgccatgggcccttgga gccgc-30 ; reverse primer: 50 -gccgcatgcctacgcgtgatggtgatgatggtgcgcttg caatcctgcttcgacag-30 . For cloning purposes, restriction enzyme site NotI or SphI was included in the forward and reverse primers, respectively. A stop codon, as well as codons encoding a six-His tag, was included in the reverse primer. The amplified PCR product was digested with restriction enzymes NotI and SphI and then subcloned into the baculovirus expression vector pFastBac1 (Life Technologies, Gaithersburg, MD). The sequence of the subcloned E-cadherin was confirmed by DNA sequencing. The coding region of the construct begins with the endogenous initial Met followed by a putative signal sequence to ensure the secretion of the expressed E-cadherin ectodomain. An in-frame His-tag added to the C-terminus downstream of the EC5 domain was used to detect the recombinant E-cadherin by IP using an mAb against the His-tag.

Generation of E-cadherin recombinant baculovirus The recombinant baculovirus encoding human E-cadherin ectodomain was generated using the Bac-To-Bac baculovirus system (Life Technologies), according to the procedure provided by the www.jidonline.org 1715

F Evangelista et al. E-cadherin Autoantibodies in Pemphigus Sera

manufacturer. Briefly, the recombinant pFastBac1/E-Cad plasmid was transformed into DH10Bac containing a bacmid (a baculovirus shuttle vector) and a helper plasmid. Recombinant bacmid DNA was then isolated from the positive clones and used to transfect Sf9 insect cells cultured in Sf900-II serum-free medium (Life Technologies). Culture supernatant containing recombinant baculovirus particles was harvested after 72-hour transfection and used to generate a high titer of viral stock by Sf9 cell infections.

Production of baculovirus-expressed E-cadherin, Dsg1, and Dsg3 Soluble ectodomains of E-cadherin, Dsg1, and Dsg3 were produced in High Five insect cells by infection with high-titer recombinant baculovirus stocks of E-cadherin, Dsg1, and Dsg3 (Ding et al., 1997). Optimal infection conditions for each recombinant protein were determined by time-course studies with various multiplicities of infections. Cell culture supernatants were harvested and dialyzed against Tris-buffered saline (TBS; pH 7.2) containing 5 mM Ca2 þ . Supernatants were stored at 80 1C until used for IP.

Immunoprecipitation coupled with immunoblotting Aliquots of culture supernatants containing the baculovirus-expressed E-cadherin, Dsg1, or Dsg3 were incubated with 2 ml of serum utilizing a previously described procedure (Li et al., 2003). The precipitation was then subjected to SDS-PAGE and immunoblotted with an anti-His mAb coupled to horseradish peroxidase (QIAGEN, Valencia, CA). The reacted protein was detected using enhanced chemiluminescence reaction (ECL detection; Amersham Biosciences, Piscataway, NJ). Well-characterized sera were used for positive and negative controls for each assay.

Immunoadsorption studies Liquid-phase immunoadsorption studies were carried out using sera that produced optimal bands by IP–IB. Four sera were tested (one FS, one PF, and two mcPV). Aliquots of FS and PF sera were preincubated with culture supernatants containing rDsg1 (B500 ng of protein) and then tested by IP–IB against recombinant E-cadherin. Aliquots of mcPV sera were incubated with either rDsg1 (B500 ng of protein) or rDsg3 (B500 ng of protein) and then tested by IP–IB against E-cadherin. Samples of FS, PF, and mcPV sera incubated with culture supernatant lacking recombinant protein were included as controls.

Establishment of ELISA for E-cadherin The baculovirus-expressed E-cadherin ectodomain was purified by nickel affinity chromatography, according to the procedure described by Ding et al. (1999), and used for the ELISA. An initial chessboard titration experiment was carried out to determine the optimal amounts of E-cadherin to coat the ELISA plate and the dilutions of serum and secondary antibody conjugates to be used (Figure S1). The immunomicrotiter plate (Costar, Cambridge, MA) was coated with purified E-cadherin (200 ng per well) at 4 1C overnight. After washing five times with TBS containing 3.7 mM Ca2 þ and 0.05% Tween-20 (TBS/Ca2 þ /T-20), the plate was blocked with 1% BSA in TBS/Ca2 þ /T-20 at room temperature for 1 hour. The plate was then washed five times and incubated with duplicate samples of diluted serum (1:200) for 1 hour at room temperature. Following wash, the plate was incubated with a 1:1,500 dilution of horseradish peroxidase-conjugated goat anti-human IgG (Bio-Rad, 1716 Journal of Investigative Dermatology (2008), Volume 128

Hercules, CA) for 1 hour. The color development was achieved with the peroxidase substrate o-phenylenediamine. The index value was defined according to Diaz et al. (2004). ROC analysis was performed for the ELISA index values generated from a set of normal serum (n ¼ 60) and FS serum (n ¼ 89) samples (Table S1; S2). An ROC curve (Figure S2) was generated to display the sensitivity and specificity of the ELISA. A cutoff of 30 produced a sensitivity of 79% and a specificity of 90%. Index values above 30 were considered positive throughout this investigation.

Competitive inhibition of ELISA To study the possible cross-reactivity between anti-E-cadherin and anti-Dsg1 autoantibodies, we tested a set of eight pemphigus sera (three PF and five FS) using two ELISA competitive inhibition assays. For the E-cadherin competitive assay, diluted serum was preincubated with 2 mg of purified recombinant E-cadherin in 100 ml of TBS/ Ca2 þ /T-2/1% BSA at room temperature for 4 hours. The adsorbed serum was then tested by Dsg1 ELISA as previously described (Hilario-Vargas et al., 2006). For the Dsg1 competitive inhibition assay, diluted serum was preincubated with 2 mg of purified rDsg1 and then tested by the E-cadherin ELISA. Sera preincubated with buffer alone were included as non-absorbed controls. The competitive inhibition of the ELISA binding was calculated for each serum using the following formula: (% of inhibition) ¼ (1OD of adsorbed serum/OD of non-adsorbed serum)  100.

Cell culture and immunofluorescence staining A431D and A431DE cell lines were kindly provided by Dr Margaret Wheelock (University of Nebraska Medical Center, Omaha, NE). Cells were cultured at 37 1C in a 5% CO2 atmosphere in DMEM/10% fetal bovine serum containing 200 mg ml1 G418 as described (Lewis et al., 1997). Cells were grown to confluence on glass coverslips and used for IF staining using procedures described by Lewis et al. (1997) and Amagai et al. (2000). Three pemphigus serum (two PF and one FS) were tested at various dilutions (1:20 to 1:100) in duplicates. A normal human serum and a commercial goat anti-human E-cadherin ectodomain serum (R&D Systems) at 1:100 dilution were included as a negative and a positive control, respectively.

IP–IB of human epidermal extract with pemphigus sera Normal human epidermal sheets were obtained from neonatal foreskins by heating the skin specimens at 56 1C for 30 seconds in phosphate-buffered saline. The epidermal sheets were homogenized and sonicated in 1% Nonidet P-40/TBS/Ca2 þ containing 2 mM phenylmethylsulfonyl fluoride and a cocktail of protease inhibitors. After centrifuge, the supernatant was diluted in TBS/Ca2 þ and subjected to IP using patients’ serum (four PF, one FS) and a normal human serum as described (Li et al., 2003). The immunoprecipitate was then subjected to SDS-PAGE and IB using a mouse mAb to human E-cadherin intracellular domain (BD Biosciences, San Jose, CA). The reacted protein was detected using enhanced chemiluminescence reaction (ECL detection; Amersham Biosciences). CONFLICT OF INTEREST The authors state no conflict of interest.

ACKNOWLEDGMENTS This work was supported in part by US Public Health Service National Institutes of Health (NIH) Grants AR30281, AR32599, and AR07369 awarded

F Evangelista et al. E-cadherin Autoantibodies in Pemphigus Sera

to LA Diaz, and AR052109 and AR053313 awarded to N Li. We are grateful to Dr Barry M Gumbiner (Department of Cell Biology, University of Virginia) for providing us with the original eukaryotic expression cDNA clone encoding the full-length human E-cadherin, Dr Margaret J Wheelock (University of Nebraska Medical Center) for providing us with A431D and A431DE cell lines. We also thank Dr Bahjat F Qaqish (Department of Biostatistics, University of North Carolina) for ROC curve analysis of the established E-cadherin ELISA.

Fiebiger E, Hammerschmid F, Stingl G, Maurer D (1998) Anti-FcepsilonRIalpha autoantibodies in autoimmune-mediated disorders. Identification of a structure–function relationship. J Clin Invest 101:243–51

SUPPLEMENTARY MATERIAL

Futamura S, Martins C, Rivitti EA, Labib RS, Diaz LA, Anhalt GJ (1989) Ultrastructural studies of acantholysis induced in vivo by passive transfer of IgG from endemic pemphigus foliaceus (Fogo Selvagem). J Invest Dermatol 93:480–5

Table S1. IgG anti-E-cadherin ELISA index values for ROC determination. Table S2. ROC determination for IgG anti-E-cadherin autoantibodies in FS sera and normal human sera. Figure S1. ELISA chessboard titration.

Fouquet S, Lugo-Martinez VH, Faussat AM, Renaud F, Cardot P, Chambaz J (2004) Early loss of E-cadherin from cell–cell contacts is involved in the onset of anoikis in enterocytes. J Biol Chem 279:43061–9 Furukawa F, Takigawa M, Matsuyoshi N, Shirahama S, Wakita H, Fujita M et al. (1994) Cadherins in cutaneous biology. J Dermatol 21:802–13

Getsios S, Huen AC, Green KJ (2004) Working out the strength and flexibility of desmosomes. Nat Rev Mol Cell Biol 5:271–81

Figure S2. ROC curve of IgG anti-E-cadherin ELISA index values of normal human sera (n ¼ 60) and FS sera (n ¼ 89).

Goodwin M, Yap AS (2004) Classical cadherin adhesion molecules: coordinating cell adhesion, signaling and the cytoskeleton. J Mol Histol 35:839–44

REFERENCES

Gumbiner BM (2005) Regulation of cadherin-mediated adhesion in morphogenesis. Nat Rev Mol Cell Biol 6:622–34

Amagai M, Hashimoto T, Green KJ, Shimizu N, Nishikawa T (1995) Antigenspecific immunoadsorption of pathogenic autoantibodies in pemphigus foliaceus. J Invest Dermatol 104:895–901 Amagai M, Karpati S, Prussick R, Klaus-Koytun V, Stanley JR (1992) Autoantibodies against the amino-terminal cadherin-like binding domain of pemphigus vulgaris antigen are pathogenic. J Clin Invest 90:919–26

Harman KE, Seed PT, Gratian MJ, Bhogal BS, Challacobe SJ, Black MM (2001) The severity of cutaneous and oral pemphigus is related to desmoglein 1 and 3 antibody levels. Br J Dermatol 144:775–80 Hashimoto K, Lever WF (1970) An ultrastructural study of cell junctions in pemphigus vulgaris. Arch Dermatol 101:287–98

Amagai M, Klaus-Kovtun V, Stanley JR (1991) Autoantibodies against a novel epithelial cadherin in pemphigus vulgaris, a disease of cell adhesion. Cell 67:869–77

Hilario-Vargas J, Dasher DA, Li N, Aoki V, Hans-Filho G, dos Santos V et al. (2006) Prevalence of anti-desmoglein 3 antibodies in endemic regions of Fogo Selvagem in Brazil. J Invest Dermatol 126:2044–8

Amagai M, Tsunoda K, Suzuki H, Nishifuji K, Koyasu S, Nishikawa T (2000) Use of autoantigen-knockout mice in developing an active autoimmune disease model for pemphigus. J Clin Invest 105:625–31

Hirai Y, Nose A, Kobayashi S, Takeichi M (1989) Expression and role of E- and P-cadherin molecules in embryonic histogenesis. Development 105: 271–7

Amagai M, Tsunoda K, Zillikens D, Nagai T, Nishikawa T (1999) The clinical phenotype of pemphigus is defined by the anti-desmoglein autoantibody profile. J Am Acad Dermatol 40:67–70

Hisamatsu Y, Amagai M, Garrod DR, Kanzaki T, Hashimoto T (2004) The detection of IgG and IgA autoantibodies to desmocollins 1–3 by enzymelinked immunosorbent assays using baculovirus-expressed proteins, in atypical pemphigus but not in typical pemphigus. Br J Dermatol 151: 73–83

Anhalt GJ, Labib RS, Voorhees JJ, Beals TF, Diaz LA (1982) Induction of pemphigus in neonatal mice by passive transfer of IgG from patients with the disease. N Engl J Med 306:1189–96 Aoyama Y, Owada MK, Kitajima Y (1999) A pathogenic autoantibody, pemphigus vulgaris-IgG, induces phosphorylation of desmoglein 3, and its dissociation from plakoglobin in cultured keratinocytes. Eur J Immunol 29:2233–40

Horiguchi Y, Furukawa F, Fujita M, Imamura S (1994) Ultrastructural localization of E-cadherin cell adhesion molecule on the cytoplasmic membrane of keratinocytes in vivo and in vitro. J Histochem Cytochem 42:1333–40

Arteaga LA, Prisayanh PS, Warren SJP, Liu Z, Diaz LA, Lin MS (2002) A subset of pemphigus foliaceus patients exhibits pathogenic autoantibodies against desmoglein-1 and desmoglein-3. J Invest Dermatol 118:806–11

Ishii K, Amagai M, Hall RP, Hashimoto T, Takayanagi A, Gamou S et al. (1997) Characterization of autoantibodies in pemphigus using antigenspecific enzyme-linked immunosorbent assays with baculovirusexpressed recombinant desmogleins. J Immunol 159:2010–7

Berkowitz P, Hu P, Liu Z, Diaz LA, Enghild JJ, Chua MP et al. (2005) Desmosome signaling. inhibition of p38MAPK prevents pemphigus vulgaris IgG-induced cytoskeleton reorganization. J Biol Chem 280: 23778–84

Ishii K, Harada R, Matsuo I, Shirakata Y, Hashimoto K, Amagai M (2005) In vitro keratinocyte dissociation assay for evaluation of the pathogenicity of anti-desmoglein 3 IgG autoantibodies in pemphigus vulgaris. J Invest Dermatol 124:939–46

Calkins CC, Setzer SV, Jennings JM, Summers S, Tsunoda K, Amagai M et al. (2006) Desmoglein endocytosis and desmosome disassembly are coordinated responses to pemphigus autoantibodies. J Biol Chem 281:7623–34

Jamora M, Jiao JD, Bystryn JC (2003) Antibodies to desmoglein 1 and 3, and the clinical phenotype of pemphigus vulgaris. J Am Acad Dermatol 48:976–7

Diaz LA, Arteaga LA, Hilario-Vargas J, Valenzuela JG, Li N, Warren S et al. (2004) Anti-desmoglein-1 antibodies in onchocerciasis, leishmaniasis and Chagas disease suggest a possible etiological link to Fogo Selvagem. J Invest Dermatol 123:1045–51

Kitajima Y (2002) Mechanisms of desmosome assembly and disassembly. Clin Exp Dermatol 27:684–90

Diaz LA, Sampaio SAP, Rivitti EA, Martins CR, Cunha PR, Lombardi C et al. (1989) Endemic pemphigus foliaceus (Fogo Selvagem): II. Current and historic epidemiologic studies. J Invest Dermatol 92:4–12 Ding X, Aoki V, Mascaro JM Jr, Lopez-Swiderski A, Diaz LA, Fairley JA (1997) Mucosal and mucocutaneous (generalized) pemphigus vulgaris show distinct autoantibody profiles. J Invest Dermatol 109:592–6

Kljuic A, Bazzi H, Sundberg JP, Martinez-Mir A, O’Shaughnessy R, Mahomney MG et al. (2003) Desmoglein 4 in hair follicle differentiation and epidermal adhesion. Evidence from inherited hypotrichosis and acquired pemphigus vulgaris. Cell 113:249–60 Kowalczyk AP, Anderson JE, Borgwardt JE, Hashimoto T, Stanley JR, Green KJ (1995) Pemphigus sera recognize conformationally sensitive epitopes in the amino-terminal region of desmoglein-1. J Invest Dermatol 105:147–52 Lever WF (1953) Pemphigus. Medicine 32:1–123

Ding X, Diaz LA, Fairley JA, Giudice GJ, Liu Z (1999) The anti-desmoglein 1 autoantibodies in pemphigus vulgaris sera are pathogenic. J Invest Dermatol 112:739–43

Lewis JE, Jensen PJ, Wheelock MJ (1994) Cadherin function is required for human keratinocytes to assemble desmosomes and stratify in response to calcium. J Invest Dermatol 102:870–7

Eyre RW, Stanley JR (1987) Human autoantibodies against a desmosomal protein complex with a calcium-sensitive epitope are characteristic of pemphigus foliaceus patients. J Exp Med 165:1719–24

Lewis JE, Wahl JK III, Sass KM, Jensen PJ, Johnson KR, Wheelock MJ (1997) Cross-talk between adherens junctions and desmosomes depends on plakoglobin. J Cell Biol 136:919–34

www.jidonline.org 1717

F Evangelista et al. E-cadherin Autoantibodies in Pemphigus Sera

Li

N, Aoki V, Hans-Filho G, Rivitti EA, Diaz LA (2003) The role of intramolecular epitope spreading in the pathogenesis of endemic pemphigus foliaceus (Fogo Selvagem). J Exp Med 197: 1501–10

Li N, Zhao M, Liu Z, Diaz LA (2006) Pathogenic role of apoptosis in experimental pemphigus. J Invest Dermatol 126:29 (abstr.) Matsuyoshi N, Tanaka T, Toda K, Okamoto H, Furukawa F, Imamura S (1995) Soluble E-cadherin: a novel cutaneous disease marker. Br J Dermatol 132:745–9 Nagasaka T, Nishifuji K, Ota T, Whittock NV, Amagai M (2004) Defining the pathogenic involvement of desmoglein 4 in pemphigus and staphylococcal scalded skin syndrome. J Clin Invest 114:1484–92 Nguyen VT, Arredondo J, Chernyavsky AI, Kitajima Y, Pittelkow M, Grando SA (2004) Pemphigus vulgaris IgG and methylprednisolone exhibit reciprocal effects on keratinocytes. J Biol Chem 279: 2135–46 Nguyen VT, Ndoye A, Grando SA (2000a) Pemphigus vulgaris antibody identifies pemphaxin. A novel keratinocyte annexin-like molecule binding acetylcholine. J Biol Chem 275:29466–76 Nguyen VT, Ndoye A, Grando SA (2000b) Novel human alpha9 acetylcholine receptor regulating keratinocyte adhesion is targeted by pemphigus vulgaris autoimmunity. Am J Pathol 157:1377–91 Patel SD, Chen CP, Bahna F, Honig B, Shapiro L (2003) Cadherin-mediated cell–cell adhesion: sticking together as a family. Curr Opin Struct Biol 13:690–8 Pitoia F, Moncet D, Glorio R, Graciela Diaz A, Rodriguez-Costa G, Carbia S et al. (2005) Prevalence of thyroid autoimmunity in patients with pemphigus vulgaris. Medicina (Buenos Aires) 65:307–10

1718 Journal of Investigative Dermatology (2008), Volume 128

Roscoe JT, Diaz LA, Sampaio SAP, Castro RM, Labib RS, Takahashi Y et al. (1985) Brazilian pemphigus foliaceus autoantibodies are pathogenic to BALB/C mice by passive transfer. J Invest Dermatol 85:538–41 Sekiguchi M, Futei Y, Fujii Y, Iwasaki T, Nishikawa T, Amagai M (2001) Dominant autoimmune epitopes recognized by pemphigus antibodies map to the N-terminal adhesive region of desmogleins. J Immunol 167:5439–48 Stanley JR, Klaus-Kovtun V, Sampaio SAP (1986) Antigenic specificity of Fogo Selvagem autoantibodies is similar to North American pemphigus foliaceus and distinct from pemphigus vulgaris autoantibodies. J Invest Dermatol 87:197–801 Stanley JR, Koulu L, Thivolet C (1984) Distinction between epidermal antigens binding pemphigus vulgaris and pemphigus foliaceus autoantibodies. J Clin Invest 74:313–20 Takahashi Y, Patel HP, Labib RS, Diaz LA, Anhalt GJ (1985) Experimentally induced pemphigus vulgaris in neonatal BALB/c mice: a time-course study of clinical, immunologic, ultrastructural, and cytochemical changes. J Invest Dermatol 84:41–6 Wang X, Bregegere F, Frusic-Zlotkin M, Feinmesser M, Michel B, Milner Y (2004) Possible apoptotic mechanism in epidermal cell acantholysis induced by pemphigus vulgaris autoimmunoglobulins. Apoptosis 9:131–43 Warren SJ, Lin MS, Giudice GJ, Hoffman RG, Hans-Flho G, Aoki V et al. (2000) The prevalence of antibodies against desmoglein 1 in endemic pemphigus foliaceus in Brazil. Cooperative Group on Fogo Selvagem Research. N Engl J Med 343:23–30 Whittock NV, Bower C (2003) Genetic evidence for a novel human desmosomal cadherin, desmoglein 4. J Invest Dermato 120:523–30 Yin T, Green KJ (2004) Regulation of desmosome assembly in adhesion. Semin Cell Dev Biol 15:665–77