| Sign In to gain access to subscriptions and/or personal tools. |
Autoimmunity to deltaNp63alpha in Chronic Ulcerative Stomatitis
1 Department of Oral and Maxillofacial Pathology, School of Dental Medicine, Tufts University, DHS-646A, One Kneeland Street, Boston, MA 02111-1527, USA; Correspondence: * corresponding author, lynn.solomon{at}tufts.edu
Chronic ulcerative stomatitis (CUS) is a recently described mucocutaneous condition in which patients experience chronic, painful, ulcerative lesions of the oral mucosa. CUS is diagnosed by immunofluorescence studies that demonstrate antinuclear antibodies. These autoantibodies are specific for a protein, deltaNp63alpha, which is normally expressed in basal cell nuclei of stratified squamous epithelia. The purpose of this study was to characterize the autoimmune response in CUS. Protein antigens were produced by in vitro transcription/translation of polymerase chain-reaction (PCR)-amplified cDNAs. We used immunoblotting and immunoprecipitation experiments with serum from CUS patients to examine the (1) antibody isotype, (2) immunogenic functional domains of the deltaNp63alpha antigen, and (3) cross-reactivity with homologous p53, p73, and p63 proteins. Results demonstrate CUS patient antibodies to deltaNp63alpha, and 52% of cases have circulating IgA isotype antibodies. The N-terminal and DNA-binding domains are the immunodominant regions, and antibody cross-reactivity with p53, p63, and p73 isoforms is limited.
Key Words: p63 • chronic ulcerative stomatitis • autoimmunity
Chronic ulcerative stomatitis (CUS) is a painful, debilitating condition that primarily affects the oral mucosa with chronic, exacerbating, and remitting ulcerations. Since CUS is a recently described condition, there are limited numbers of reported cases. The average age of CUS patients is 60 years, and women are most commonly affected (Solomon et al., 2003). CUS is one of several immunologically mediated conditions that affect the oral cavity, e.g., pemphigoid, pemphigus vulgaris, and lichen planus (Scully and Porter, 1997). Generally, these conditions are successfully managed with corticosteroid therapy (Gonzalez-Moles and Scully, 2005a), and treatment failures often result from incorrect diagnosis (Gonzalez-Moles and Scully, 2005b). In CUS, corticosteroids are less effective than other therapeutics, such as hydroxychloroquine (Chorzelski et al., 1998), underscoring the importance of accurate diagnosis. The histopathologic findings are non-specific, although suggestive features include atrophic, parakeratinized, stratified squamous epithelium, lichenoid inflammatory cell infiltrates, basal cell degeneration, and cytoid bodies (Solomon et al., 2003). Immunofluorescence studies have revealed the presence of autoantibodies with a stratified epithelial specific-antinuclear antibody (SES-ANA) pattern. These autoantibodies target an antigen, deltaNp63alpha, which is a nuclear protein normally present in basal and parabasal cells of stratified squamous epithelia (Lee et al., 1999). DeltaNp63alpha is a member of a family of nuclear transcription factors, including p63, p73, and the p53 tumor suppressor gene, which share considerable sequence homology (Choi et al., 2002). The homology extends to protein structures, which are characterized by arrangement in functional domains. The N-terminus of domain 1 is generated from the transcriptional start codon and has a transactivating (TA) function. Domain 2 is a DNA-binding domain, while domain 3 is an oligomerization domain. Domain 4 includes the C-terminus. Additional complexity is introduced in p63 and p73 proteins by the production of several isoforms. When an internal transcriptional start site is used, an N-terminal truncated isoform of domain 1 is created, indicated in writing as "deltaN". Alternate mRNA splicing produces C-termini of various lengths; the specific isoforms are indicated by the addition of a Greek letter suffix, e.g., alpha, beta, gamma. Currently, the convention is to refer to the p63/p73 family members by using the appropriate (TA or deltaN) prefix, and a Greek letter suffix, to identify specific isoforms. Investigators in several fields independently described the antigen in CUS, deltaNp63alpha; thus, it is found in the literature under several different names (Schmale and Bamberger, 1997; Augustin et al., 1998; Osada et al., 1998; Senoo et al., 1998; Trink et al., 1998; Yang et al., 1998; Lee et al., 1999; Hibi et al., 2000). The p63 gene is located on chromosome 3q27-29 (Yang et al., 1998). The molecular mass of deltaNp63alpha is approximately 70 kDa, and the cDNA has a 1761-bp open reading frame [GenBank accession #AF091627, National Center for Biotechnology Information (NCBI), http://www.ncbi.nlm.nih.gov/]. Various isoforms of p63 are expressed in a tissue-specific manner (Dellavalle et al., 2001). The present study was undertaken to characterize the autoimmune response in CUS and to provide a potential diagnostic rationale for the use of these autoantibodies. Molecular methods were applied in our study to examine the autoantibody isotype, immunogenic functional domains of the deltaNp63alpha antigen, and cross-reactivity with homologous p53, p73, and p63 proteins.
Serum Samples Serum samples from 21 CUS patients, diagnosed by clinical, histologic, and immunofluorescence findings, were obtained as generous gifts from Dr. Stephania Jablonska (Katedra i Klinika Dermatologiczna, Warsaw, Poland) and from IMMCO Diagnostics. Dr. Jablonska is supportive of this investigation and gave permission to publish the results. Control sera from IMMCO Diagnostics included 16 samples from patients with dermatologic or rheumatic clinical conditions. All sera were existing diagnostic samples and thus exempt from human subjects review, as provided for research involving "the collection or study of existing diagnostic specimens, if the information is recorded by the investigator in a manner that the subjects cannot be identified, directly or through identifiers linked to the subjects". Additional information is available at http://www.hhs.gov/ohrp/humansubjects/guidance/45cfr46.htm#46.101, the US Department of Health & Human Services Web site.
Primer Design and PCR Amplification of deltaNp63alpha and Domains Primer sets were designed for PCR amplification of functional domains: domain 1F (5'-GGATCCTAATACGACTCACTAT AGGGAACAGCTAACATGTTGTACCTGGAA-3') and domain 1R (5'-TTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTCAC TTTTTCTTACAGCTAACATGTTGTACCTGGAA-3'); domain 2F (5'-GGATCCTAATACGACTCACTATAGGGAACAG CTAACATGCCGCACAGTTTCGACGTGTCCT-3') and domain 2R (5'-TTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTCACTTT GTACTGTCCGAAACTTGCTG-3'); domain 3F (5'-GGATCCTAATACGACTCACTATAGGGAACAGCTAACATG CCGCACAGTTTCGACGTGTCCT-3') and domain 3R (5'-TTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTCACTTTGTACTG TCCGAAACTTGCTG-3'; domain 4F (5'-GGATCCTAATA CGACTCACTATAGGGAACAGCTAACATGACAGCTAACAT GTTGTACCTGGAA-3') and domain 4R (5'-TTTTTTTTTTTTT TTTTTTTTTTTTTTTTTTCACTCCCCCTCCTCTTTGATGC-3'). Incorporation of transcription and poly A translational signaling sequences, e.g., the oligo dT stretch (Beckler et al., 2000), resulted in a primer length longer than usual. Gel-purified primers (Integrated DNA Technologies, Coralville, IA, USA) and AccuPrimeTM Taq DNA polymerase (Invitrogen, Carlsbad, CA, USA) were used in a GeneAmp® thermal-cycler (PE Applied Biosystems, Foster City, CA, USA) as per manufacturers’ instructions under standard conditions (Saiki, 1990). Combination domains—1&2, 2&3, 3&4, 1,2&3, and 2,3&4—were created with appropriate primers.
In vitro Transcription/Translation (IVTT)
Western Immunoblot
Immunoprecipitation
Cross-reactivity of CUS Patient Antibodies with p53, p63, and p73 Isoforms Proteins were subjected to SDS-PAGE, electrotransferred to nitrocellulose membranes, and incubated with streptavidin/alkaline phosphatase and substrate to demonstrate successful IVTT. For detection of immunologic reactivity, duplicate membranes were immunoblotted with human sera (1:50 dilution). After being washed, membranes were incubated with alkaline phosphatase-conjugated donkey anti-human IgG secondary antibody (Jackson Laboratories) and subsequently with Western Blue® Substrate for Alkaline Phosphatase (Promega).
Western Immunoblot All of the CUS patient sera were immunologically reactive with deltaNp63alpha protein (Fig., A  ). However, donkey anti-human IgG (H+L) secondary antibody also reacts with IgM and IgA, which share light chains; thus we cannot conclude that the patient antibodies are exclusively IgG. Immunoblotting with IgA isotype-specific secondary antibody showed that 52% of the sera were positive, and 48% were negative for IgA antibodies to deltaNp63alpha (Fig., B). None of the control sera was immunologically reactive with deltaNp63alpha (Fig., C). Positive results were identified based on visual detection of a reactive band.
Immunoblotting of functional domains showed that all CUS sera tested recognized common deltaNp63alpha domains, although individual sera varied. The results are summarized (Table 1 ). Functional domain combinations served to confirm results of the individual domains. Domain 1 is a small peptide and did not express well in IVTT; thus it was unable to be tested individually. However, since 5 of 6 sera tested recognized combination domains 1&2, and none recognized domain 2, it is reasonable to assume that domain 1 must be the antigenic epitope. These results show that the antigenic epitopes of linearized peptides are domains 1 and 4, the N- and C-termini, respectively.
Immunoprecipitation Our experiments showed that all CUS patient sera tested had antibodies that immunoprecipitated folded domains of deltaNp63alpha protein. However, Protein L was not isotype-specific; thus we cannot conclude that these antibodies are exclusively IgG. Visual detection of a reactive band identified positive results. The results summary (Table 2 ) shows that all CUS sera strongly recognized domain 2 and combination domain 1&2 in folded conformations. Only 1 of 8 sera showed weak positive recognition of domain 3. Domain 4 was strongly recognized by 2 sera, and 5 sera produced a weak band. Of 8 samples tested, 3 recognized combination domains 3&4 with a strong band, and 3 produced a weak band. Domain 1 was unable to be tested individually. These results indicate that folded domain 2 is the most strongly antigenic epitope. Folded domain 4 is also antigenic, although not as strongly or frequently.
Cross-reactivity of CUS Patient Antibodies with p53, p63, and p73 Isoforms Homologous deltaNp63alpha proteins—specifically p53, TAp73alpha, TAp73-gamma, TAp73delta, deltaNp73alpha, deltaNp73beta, deltaNp73gamma, and TAp63alpha—were produced with IVTT. Visual detection of a reactive band identified positive results. The results summary (Table 3 ) shows that all CUS sera tested were immunologically reactive with the p63 isoform TAp63alpha. Two sera recognized additional isoforms, one recognizing TAp73alpha and delta-Np73alpha, while the other reacted with the TAp73gamma isoform. One serum was extremely cross-reactive, recognizing all p73 and p63 isoforms tested, although it did not recognize p53. The secondary antibody used, donkey anti-human IgG (H+L), also reacts with IgM and IgA, which share light chains; thus we cannot conclude that the patient antibodies are exclusively IgG. Ten duplicate blots were created for controls; all control sera were negative for immunologic reactivity with any of the isoforms (data not shown).
In our studies, we used in vitro systems to produce deltaNp63alpha, discrete functional domains, and homologous proteins. Immunoblotting and immunoprecipitation experiments with CUS patient sera determined the (1) presence of autoantibodies, (2) immunogenic functional domains, and (3) cross-reactivity with homologous p53, p73, and p63 proteins. In vitro systems employed to produce proteins used rabbit reticulocyte lysates, because they produce appropriate mammalian-type post-translational protein modification and folding. The IVTT-produced proteins were biotinylated to aid detection, and it is possible that addition of this small molecule may have affected epitope reactivity. Thus, our results should be confirmed with an independent method to determine immunogenic epitopes conclusively. Immunological reactivity to deltaNp63alpha was shown with all sera examined by immunoblotting. The presence of IgA antibodies in CUS patient sera has not previously been studied, and 52% of samples had circulating IgA immunoreactive with deltaNp63alpha. In mucous membrane pemphigoid, circulating IgA is clinically significant; patients with dual circulating IgG and IgA responses have more severe disease (Setterfield et al., 1998). Future studies to determine if CUS patients with circulating IgA have disease clinically more severe would be useful. Linear epitopes are detected by immunoblot, and conformational epitopes are detected by immunoprecipitation. Immunoblotting confirmed antibodies to linearized epitopes of functional domains 1 and 4. In contrast, immunoprecipitation showed that the most strongly immunogenic region was a conformational epitope of functional domain 2. It is not unexpected that multiple epitopes are recognized, since the normal immune response is polyclonal. The tissue milieu in CUS is inflammatory, and proteins are degraded; thus, linear sequences, in addition to conformational epitopes, may become immunogenic. This work is the first to examine cross-reactivity of CUS patient antibodies with p53 family isoforms. DeltaNp63alpha exhibits various homologies to other p53, p63, and p73 proteins. p53 and p63 share 60% and 37% identical residues in functional domains 2 and 3, respectively (Barbieri and Pietenpol, 2006). Amino acid sequence comparison between p63alpha and p73alpha shows 85%, 60%, and 50% homology between functional domains 2, 3, and 4, respectively (Yang et al., 2002). All CUS sera tested cross-reacted with TAp63alpha, which is not surprising, since there is only a small N-terminal difference between TA- and deltaN- p63alpha. Most sera exhibited limited cross-reactivity to p63 and 73 isoforms, while none cross-reacted with p53. DeltaNp63alpha is the p63 isoform preferentially expressed in stratified squamous epithelia, where it is essential for differentiation (Koster et al., 2004) and the maintenance of progenitor cell populations to sustain development, morphogenesis, and the proliferative potential (Ratovitski et al., 2001; Marchbank et al., 2003; Barbieri and Pietenpol, 2006). Homology between p53 family members allows for binding to consensus DNA sequences (Yang et al., 2002), and deltaN isoforms exert dominant-negative effects over p53, p63, and p73 activities (Murray-Zmijewski et al., 2006). Another p63 function is maintenance of epithelial integrity via regulation of desmosomal adhesion complex stability (Ihrie et al., 2005). While CUS patient antibodies to deltaNp63alpha are demonstrated within basal cell nuclei in vivo, their role in pathogenesis is unknown. Several studies support the intracellular and intranuclear penetration of antibodies, although not without controversy (Alarcon-Segovia et al., 1996; Reichlin, 1998; Rivadeneyra-Espinoza and Ruiz-Arguelles, 2006). It is interesting to speculate that CUS autoantibodies to the DNA-binding domain may interfere with the ability of deltaNp63alpha to oppose apoptosis and maintain epithelial integrity. This may be especially significant in the oral cavity, where epithelial turnover is high, and the site is frequently exposed to minor trauma. These results have importance when one considers that, currently, the gold standard in diagnosis of immunologically mediated conditions is immunofluorescence. The facilities to perform these studies are limited, and the tests are costly. As a result, many oral ulcerative conditions are treated empirically, without a diagnosis; thus, the true incidence and prevalence of many of these conditions, including CUS, are unknown. Future development of a simple immunoassay may help to establish the true incidence and prevalence of CUS and allow for meaningful comparisons of treatment efficacies.
This paper is based in part on a thesis submitted to the faculty of the Graduate School of the University at Buffalo in partial fulfillment of the requirements for the degree of Master of Science in the Oral Sciences. The financial support of IMMCO Diagnostics, Inc., Buffalo, NY, is gratefully acknowledged. Received for publication March 6, 2006. Revision received February 20, 2007. Accepted for publication April 15, 2007.
Journal of Dental Research, Vol. 86, No. 9,
826-831 (2007)
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
-chain-specific (cat. #109-056-011), or Goat Anti-Mouse IgG (H+L) AffiniPure (cat. #115-055-062), all from the same source (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA, USA). Western Blue® Substrate for Alkaline Phosphatase (Promega) was used to develop color. 
Np63