- Email: [email protected]
Usefulness of immunoblotting using purified laminin 5 in the diagnosis of anti-laminin 5 cicatricial pemphigoid
Journal of Dermatological Science (2003) 33, 113 /119
www.elsevier.com/locate/jdermsci
Usefulness of immunoblotting using purified laminin 5 in the diagnosis of anti-laminin 5 cicatricial pemphigoid Yoshiko Hisamatsua, Toshio Nishiyamab, Satoshi Amanob, Chihiro Matsuic, Riza Ghohestanid, Takashi Hashimotoa,* a
Department of Dermatology, Kurume University School of Medicine, 67 Asahimachi, Kurume, Fukuoka 830-0011, Japan b Shiseido Life Science Research Center, Yokohama, Japan c Department of Dermatology, Toyama Medical and Pharmaceutical University, Toyama, Japan d Department of Dermatology, Thomas Jefferson Medical College, Philadelphia, PA, USA Received 25 March 2003; received in revised form 11 June 2003; accepted 16 June 2003
KEYWORDS Cicatricial pemphigoid; Immunoblotting; Laminin 5; Recombinant protein
Summary Background: Anti-laminin 5 cicatricial pemphigoid (CP) is a mucosaldominant subepithelial blistering disease characterized by IgG anti-basement membrane zone autoantibodies, that bind to dermal side of 1 M NaCl split skin and immunoprecipitate laminin 5. Laminin 5 is an epidermis-specific extracellular matrix consisting of a3, b3 and g2 subunits. Recent studies have suggested that autoantibodies of anti-laminin 5 CP recognize the G domains of a3 subunit. Objective: We examined the reactivity of anti-laminin 5 CP by immunoblotting using purified laminin 5 and recombinant proteins of a3 subunit. Method: We first examined the reactivity of anti-laminin 5 CP by immunoblotting using purified laminin 5. To further investigate the epitopes in the G domains of a3 subunit, we produced recombinant proteins of G1 /2, G1 /3, G2 /3, G3 /5 domains, that covered entire G domain, and examined the reactivity of anti-laminin 5 CP sera with these recombinant proteins by immunoblotting. Results: By immunoblotting using purified laminin 5, 7 of 21 anti-laminin 5 CP sera reacted with a3 subunit, while 8 sera reacted with b3 subunit and one serum reacted with g2 subunit. Two sera reacted with both a3 and b3 subunits, while seven sera did not show positive reactivity. This result indicates that the reactivity of antilaminin 5 CP sera is much more heterogeneous, although the previous studies suggested that most sera reacted with a3 subunit. However, in the studies using recombinant proteins of G domains of a3 subunit, none of the CP sera, including the sera reactive with a3 subunit in purified laminin 5, reacted with any recombinant proteins. The reason for this negative reactivity with the recombinant proteins is not clear. Conclusion: The immunoblotting using purified laminin 5 should be useful technique for the diagnosis of anti-laminin 5 CP, although the sensitivity was less than conventional immunoprecipitation analysis. – 2003 Japanese Society for Investigative Dermatology. Published by Elsevier Ireland Ltd. All rights reserved.
Abbreviations: CP, cicatricial pemphigoid; GST, glutathione-S -transferase. *Corresponding author. Tel.: /81-942-31-7571; fax: /81-942-34-2620. E-mail address: [email protected] (T. Hashimoto). 0923-1811/03/$30.00 – 2003 Japanese Society for Investigative Dermatology. Published by Elsevier Ireland Ltd. All rights reserved. doi:10.1016/S0923-1811(03)00158-0
114
1. Introduction Laminin 5 (epiligrin) is an epidermis-specific extracellular matrix consisting of disulfide-linked a3, b3 and g2 subunits [1 /5]. The a3 and g2 subunits exist as cell-associated precursors of 200 and 155 kDa, respectively; these precursors are rapidly processed following secretion to 165 and 105 kDa, respectively [1]. The b3 subunit exists as a 140 kDa polypeptide that does not undergo processing [1]. Laminin 5 binds to its receptor, a6 b4 integrin, and play an important role in the adhesion at the basement membrane zone of the epidermis. Anti-laminin 5 cicatricial pemphigoid (CP) is a mucosal-dominant subepithelial blistering disease characterized by IgG anti-basement membrane zone autoantibodies, that bind to the dermal side of 1 M NaCl split skin [6] and immunoprecipitate laminin 5 from extracts of human keratinocytes [7,8]. Immunoblotting studies using extracts of cultured keratinocytes revealed that the majority of anti-laminin 5 CP sera reacted with a3 subunit [9]. However, a few sera reacted with either b3 or g2 subunits [10,11]. In addition, using neonatal mouse animal model, it has been shown that antibodies to laminin 5 can reproduce skin lesion in the mice, thus indicating that the autoantibodies to laminin 5 in the CP patients should play a pathogenic role [12]. Furthermore, recent studies using bacterial expressed recombinant proteins have suggested that most CP sera recognized the G domains of a3 subunit [13]. In this study, we examined the reactivity of antilaminin 5 CP by immunoblotting using the purified laminin 5. To further investigate the epitopes on G domains of a3 subunit, we produced recombinant proteins of G1 /2, G1 /3, G2 /3, and G3 /5 domains, that cover entire G domain, and examined the reactivity of anti-laminin 5 CP sera by immunoblotting using these recombinant proteins.
2. Materials and methods 2.1. Human sera and antibodies Sera were obtained from 21 patients with antilaminin 5 CP (13 patients from Asia and eight patients from USA). All patients had subepithelial blistering and erosive lesions of mucous membranes. In addition, all the sera had circulating IgG autoantibodies that bound exclusively to the dermal side of 1 M NaCl split skin on indirect immunofluorescence, and immunoprecipitated laminin 5 from lysate of biosynthetically radiolabeled
Y. Hisamatsu et al.
human keratinocytes. Sera from 20 normal volunteers were also used as controls. Two polyclonal rabbit antibodies (designated as Lapin 1 and Lapin 2) raised against a recombinant protein of G2 /G3 domain of a3 subunit were also used. These polyclonal antibodies were shown to react strongly and exclusively with processed form of a3 subunit on immunoblotting using keratinocyte culture media. However, by immunofluorescence, these polyclonal antibodies reacted only faintly with the basement membrane zone in the normal skin sections and relatively weakly with dermal side of the 1 M NaCl split skin section. We also used commercially available anti-glutathione-S -transferase (GST) polyclonal antibody (Sigma, USA). All the sera and antibodies were stored at /30 8C or at 4 8C in the presence of 0.1% sodium azide.
2.2. Purification of laminin 5 Human keratinocytes were cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum. The laminin 5 was purified from the culture medium by using anti-laminin 5 affinity column, prepared by Sepharose gel conjugated with a monoclonal antibody against the laminin a3 chain. The detailed methods were described previously [3 /5].
2.3. Generation of bacterial expression constructs of G domains of a3 subunits of human laminin 5 Domain-specific GST-fusion proteins of G domains of a3 subunit of laminin 5 were generated, as reported previously [14,15]. Four cDNA fragments were amplified by polymerase chain reaction, using keratinocyte cDNA library as templates, as well as appropriate primers shown in Table 1. The keratinocyte cDNA library was prepared from mRNA obtained from the cultured squamous cell line, KU8. To prepare the recombinant protein of G1 /2 domain, we used the primer pair, 5-G1 /G2 and 3G1 /G2. To prepare the recombinant protein of G1 /3 domain, we used the primer pair, 5-G1 /G2 and 3-G1 /G3. To prepare the recombinant protein of G2 /3 domain, we used the primer pair, 5-G2 /G3 and 3-G1 /G3. To prepare the recombinant protein of G3 /5 domain, we used the primer pair, 5-G3 /G5 and 3-G3 /G5. In the 5? region of each primer, the restriction enzyme site for NotI was incorporated to facilitate to prepare construct. PCR products were purified by PCR purification kit (QIAGEN), digested with NotI , and ligated into
Immunoblotting for anti-laminin 5 cicatricial pemphigoid
115
Table 1 The list of primers used to amplify G-domains of a3 subunit of laminin 5 Name
Sequence
5-G1 /G2 3-G1 /G2 3-G1 /G3 5-G2 /G3 5-G3 /G5 3-G3 /G5
gccgcggccgc gccgcggccgc gccgcggccgc gccgcggccgc gccgcggccgc gccgcggccgc
Restriction enzyme ACATCTCTGTCCTTGTTTCTCCAAAGGCCC ACTTTTTGGTTACTCCCACAGTATCATTC AAGCATCTTGCCACACCTTCACGCTCC TGTAGAAGGAGGAAGGAAGAGTCAGAC TGCTCGGAAGACTGGAAGCTTGTGCGATCTGCC TGGTCAGGACAACCATTCAGACTGACAGGCC
the 3? end of GST gene in the bacterial expression vector pGEX-5X-3 (Pharmacia LKB Biotechnology, NJ), which had been digested by the same restriction enzymes and gel purified. For all fragments, PCR was performed on the condition: 94 8C 1 min, one cycle; 98 8C 20 s, 68 8C 15 min, 30 cycles; and 72 8C 10 min, one cycle. All the constructs were subjected to nucleotide sequence analysis by ABI PRISMƒ 310 Genetic Analyzer using 5?- and 3?-pGEX sequencing primers, to confirm the identity and fidelity of subcloning steps.
2.4. Preparation of GST-fusion proteins of G domains E. coli strain XL-1 Blue was transfected with plasmid constructs and grown in 25 ml LB medium at 37 8C overnight. After the overnight culture was added to 250 ml fresh medium, cells were grown at 37 8C for 3 h and IPTG was added to a final concentration of 1.0 mM to induce expression of the fusion protein. After an additional incubation at 37 8C for 3 h, cells were harvested by centrifugation. Pellets were re-suspended in 3-ml phosphate-buffered saline containing 1% Triton-X-100, mildly sonicated and clarified by centrifugation. All fusion proteins were not soluble in this buffer, the pellets were further extracted by resuspending and sonicating in 3 ml 2 M urea [16]. Subsequently, the pellets were resuspended in 1-ml Laemmli’s sample buffer, boiled for 5 min and centrifuged [16]. The supernatant was used as antigen sources for immunoblot analysis.
NotI NotI NotI NotI NotI NotI
at 4 8C with patients’ sera diluted at 1:20 in dilution buffer. Blots were then incubated with horseradish peroxidase-conjugated rabbit anti-human IgG antiserum (1:100, gamma-chain-specific) (Dako, Denmark) as a secondary antibody for 3 h at room temperature and color was developed with 4chloro-1-naphtol in the presence of H2O2. For immunoblotting of recombinant proteins, the four fusion recombinant proteins were separated by 9% SDS-polyacrylamide gels, and transferred onto nitrocellulose membrane. Because these proteins were not soluble in 1% Triton X-100 solution, and could not be purified by glutathione-sepharose 4B column, the patients’ sera were pre-absorbed with E. coli lysate to reduce background reactivity due to non-specific anti-E. coli antibodies. E. coli lysate was prepared as described below. Two hundred milliliter LB medium was added with 20 ml of overnight culture of XL-1 Blue transfected with pGEX without insert, cultured for 3 h, and induced by 1.0 mM IPTG. After the bacteria reached confluence, bacterial pellet was re-suspended in 5 ml of phosphate-buffered saline, subjected to three cycles of freezing/thawing, sonicated mildly, and kept at /80 8C until use. For pre-absorption, the sera diluted in dilution buffer were mixed with the same volume of E. coli lysate, agitated at room temperature for 1 h, and centrifuged at 10 min at 14 000 rpm. The supernatant was used as the first antibody. Following procedures were the same as those for the immunoblotting for purified laminin 5.
2.5. Immunoblot analysis For immunoblotting of purified laminin 5, the sample was directly solubilized in Laemmli’s sample buffer, applied on 6% SDS-polyacrylamide gels, and transferred onto nitrocellulose membrane. The blots were blocked with Tris-buffered saline with 3% skim milk (dilution buffer) for 1 h at room temperature. The blots were incubated overnight
3. Results
3.1. The protein structure of laminin 5 The protein structure of laminin 5 is depicted in Fig. 1.
116
Y. Hisamatsu et al.
3.2. The schematic diagram of the structure of a3 subunit of human laminin 5 and the four truncated recombinant GST-fusion recombinant proteins (Fig. 2) In Fig. 2, we depict an entire structure of a3 subunit of laminin 5 with tandem G1 /G5 domains in its C-terminal domain. We also show the positions and the numbers of residues of all the recombinant proteins; i.e. G1 /2, G1 /3, G2 /3 and G3 /5.
3.3. Circulating IgG antibodies in the antilaminin 5 CP patients’ sera reacted with various subunits of laminin 5 by immunoblot analysis using purified laminin 5 The purified laminin 5 was reduced and applied to 6% SDS-polyacrylamide gels. In this laminin 5 preparation, amido black staining showed three clear protein bands with molecular sizes of 165 kDa (processed form of a3 subunit), 140 (b3 subunit), and 105 kDa (processed form of g2 subunit) (Fig. 3). No protein bands of the 200 kDa unprocessed form of a3 subunit and the 155 kDa unprocessed form of g2 subunit were seen. A weak band seen at the position of about 145 kDa was considered to be an degradation product of a3 subunit, because the two rabbit anti-a3 subunit polyclonal antibodies identified clearly both the 165 kDa processed form of a3 subunit and the 145 kDa band (Fig. 3). Circulating IgG antibodies of 7 of 21 patients with anti-laminin 5 CP bound to the 165 kDa processed form of a3 subunit of laminin 5 (Fig. 3). Some of them also reacted with the 145-kDa proteins. Eight of 21 CP sera bound to the 140 kDa b3 subunit of laminin 5. Only one CP serum clearly bound to the 105 kDa processed form of g2 subunit of laminin 5. Two sera reacted with both a3 and b3 subunits. In contrast, seven sera showed no positive reactivity with any subunits. None of these three subunits of laminin 5 were not reacted by either 20 control normal human sera or normal rabbit serum (Fig. 3).
3.4. None of the anti-laminin 5 CP sera reacted with any truncated recombinant fusion proteins We produced the four truncated recombinant proteins of G domain of a3 subunit and examined the reactivity of the anti-laminin 5 CP sera with these proteins. We used all the 21 CP patients’ sera, which include seven sera reactive with a3 subunit of purified laminin 5, 10 normal control
Fig. 1 The protein structure of laminin 5. Laminin 5 is a heterotrimeric protein composed of disulfide-linked a3, b3, and g2 subunits. The a3 subunit exists as 200 kDa unprocessed and 165 kDa processed polypeptides. The b3 subunit is 140 kDa polypeptide that does not undergo processing. The g2 subunit of laminin 5 exists as 155 kDa unprocessed and 105 kDa processed polypeptides. The a3 chain has a large globular domain in the C-terminus, termed G domain. This C-terminal globular domain is further subdivided into five homologous subdomains; i.e. G1 /G5, each of which consists of 180 /200 amino acid residues. They contain binding sites for heparin and other extracellular matrix proteins, as well as cellular receptors including integrins. According to the recent studies, the proteolytic cleavage of the a3 chain occurs between the G3 and G4 domains.
sera, anti-GST polyclonal antibody and two anti-a3 subunit polyclonal antibodies. None of the 21 antilaminin 5 CP sera and ten normal sera showed any specific reactivity with these four recombinant proteins, while three polyclonal antibodies showed the specific reactivity (Fig. 4).
4. Discussion In the first part of this study, we performed immunoblotting using purified laminin 5. By this immunoblot analysis, seven sera of 21 anti-laminin
Immunoblotting for anti-laminin 5 cicatricial pemphigoid
Fig. 2 The schematic diagram of the structure of a3 subunit of human laminin 5 and the four truncated recombinant GST-fusion recombinant proteins. The upper panel depicts an entire structure of a3 subunit of laminin 5, in which there are five homologous G1 /G5 domains in its C-terminal domain. The middle panel indicates the positions and the numbers of residues of all the recombinant proteins used in this study; i.e. G1 /2, G1 /3, G2 /3, and G3 /5. The lower line indicates the size of cDNA.
5 CP sera reacted with a3 subunit, while 8 sera reacted with b3 subunit and one serum reacted with g2 subunit. However, seven sera did not show positive reactivity. Therefore, the immunoblotting of purified laminin 5 should be a useful method for the diagnosis of anti-laminin 5 CP, although conventional immunoprecipitation analysis using radiolabeled keratinocyte lysate is still more sensitive
117
method. Furthermore, the previous studies indicated that the major antigenic domain is a3 subunit [9]. However, the results shown in the present study indicate that the reactivity of anti-laminin 5 CP sera is more heterogeneous, and all the three subunits of laminin 5 can be recognized by CP patients’ sera. The reason why seven CP sera did not react with any of the subunit in this study is unclear. It may be conceivable that the titer of the antibodies in the sera were too low to be detected in our immunoblot analysis. More likely, there sera may contain antibodies against conformation-dependent epitopes, which can be detected by immunoprecipitation but not by immunoblotting. In the second part of this study, we prepared recombinant proteins. The previous studies using the same pGEX expression system showed that most anti-laminin 5 CP sera reacted with the recombinant proteins of G domains of a3 subunit [13]. However, unexpectedly, by the present immunoblotting study using the four recombinant proteins of G domains of a3 subunit, none of the recombinant proteins were reacted by any patients’ sera, including the sera reactive with a3 subunit in purified laminin 5. The reason for this discrepancy is at the present unknown. All of our four recombinant proteins were recognized by anti-GST polyclonal antibodies and the two polyclonal antibodies raised against G2 /G3
Fig. 3 The results of immunoblot analysis using purified laminin 5. Protein staining by amido black showed three major protein bands (lane 1). One of the polyclonal antibodies to G2 /G3 (Lapin 2) showed strong bands with 165 kDa processed form of a3 subunit and the 145 kDa degradation product (lane 2). The representative 11 anti-laminin 5 CP sera showed reactivity with a3, b3, and g2 subunits (lanes 3 /13), while normal sera did not show any positive reactivity (lane 14 /16). The positions of the 165 kDa processed form of a3 subunit, the 145 kDa degraded a3 subunit, 140 kDa b3, and the 105 kDa processed form of g2 subunits are indicated by arrows in the left margin.
118
Y. Hisamatsu et al.
Fig. 4 The results of immunoblot analysis using four truncated recombinant proteins of G domain of a3 subunit. Panels a /d showed the reactivity in immunoblot analyses for G1 /G2 domain, G1 /G3 domain, G2 /G3 domain and G3 / G5 domain, respectively. In each panel, lanes 1 and 7 show the reactivity of anti-GST polyclonal antibody, lanes 2 /6 show the reactivity of 5 representative anti-laminin 5 CP sera, which reacted with a3 subunit of purified laminin 5, lanes 8 /10 show the reactivity of normal control sera, and lanes 11 and 12 show the reactivity of two polyclonal antibodies to G2 /G3 domain (Lapin 1 and Lapin 2). Anti-GST polyclonal antibody and anti-G2 /G3 polyclonal antibodies reacted with each recombinant protein. However, none of the CP sera and normal sera showed positive reactivity with the recombinant proteins. An arrowhead in the left of each panel indicates the position of intact recombinant protein. In all the panels, the lane with the same number shows the reactivity of the same serum and same polyclonal antibodies. Polyclonal antibodies to G2 /G3 (lanes 11 and 12) produced a strong non-specific protein band. These bands are seen in the top of the lanes in panels a and c, and seen as the second bands in panels b and d.
domain, and showed expected sizes, indicating that the four recombinant proteins were produced properly. Therefore, the reason for the negative reactivity with the recombinant proteins is not clear. However, in the previous studies, the GST fusion proteins were solubilized and purified by glutathione-sepharose column [13], whereas our GST fusion proteins were not soluble and could not be purified in the present study. Therefore, it may be conceivable that some contaminated materials may interfere with the binding of the antibodies to recombinant proteins in our study. Alternatively, in the previous study [13], alkaline phosphatase
method was used to detect the protein bands, while we used peroxidase/4-chloro-1-naphthol method for detection. There may be some difference of sensitivity between the two systems.
Acknowledgements We are grateful to Professor Kim B. Yancey, Department of Dermatology, Wisconsin University, Milwaukee for generously providing us with eight CP sera, as well as important critical discussions. We also thank Miss Michiyo Noge, Miss Yuko Kawano
Immunoblotting for anti-laminin 5 cicatricial pemphigoid
and Miss Ayumi Suzuki for technical assistance and Miss Akiko Tanaka for secretarial work. This work was supported by Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture of Japan, a grant from the Ministry of Health and Welfare of Japan, and a grant for Open Research Center Project from the Ministry of Education.
[8]
[9]
[10]
References [1] Marinkovich MP, Lunstrum GP, Burgeson RE. The anchoring filament protein kalinin is synthesized and secreted as a high molecular weight precursor. J Biol Chem 1992;267:17900 /6. [2] Champliaud MF, Lunstrum GP, Rousselle P, Nishiyama T, Keene DR, Burgeson RE. Human amnion contains a novel laminin variant, laminin 7, which like laminin 6, covalently associates with laminin 5 to promote stable epithelialstromal attachment. J Cell Biol 1996;132:1189 /98. [3] Amano S, Scott IC, Takahara K, Koch M, Champliaud MF, Gerecke DR, et al. Bone morphogenetic protein 1 is an extracellular processing enzyme of the laminin 5 gamma 2 chain. J Biol Chem 2000;275:22728 /35. [4] Amano S, Nishiyama T, Burgeson RE. A specific and sensitive ELISA for laminin 5. J Immunol Methods 1999;224:161 /9. [5] Nishiyama T, Amano S, Tsunenaga M, Kadoya K, Takeda A, Adachi E, et al. The importance of laminin 5 in the dermalepidermal basement membrane. J Dermatol Sci 2000;25:S51 /9. [6] Lazarova Z, Yancey KB. Reactivity of autoantibodies from patients with defined subepidermal bullous diseases against 1 mol/l salt-split skin. Specificity, sensitivity, and practical considerations. J Am Acad Dermatol 1996;35:398 /403. [7] Domloge-Hultsch N, Gammon WR, Briggaman RA, Gil SG, Carter WG, Yancey KB. Epiligrin, the major human keratinocyte integrin ligand, is a target in both an acquired
[11]
[12]
[13]
[14]
[15]
[16]
119 autoimmune and an inherited subepidermal blistering skin disease. J Clin Invest 1992;90:1628 /33. Domloge-Hultsch N, Anhalt GJ, Gammon WR, Lazarova Z, Briggaman R, Welch M, et al. Antiepiligrin cicatricial pemphigoid. A subepithelial bullous disorder. Arch Dermatol 1994;130:1521 /9. Kirtschig G, Marinkovich MP, Burgeson RE, Yancy KB. Antibasement membrane autoantibodies in patients with antiepiligrin cicatricial pemphigoid bind the a subunit of laminin 5. J Invest Dermatol 1995;105:543 /8. Lazarova Z, Hsu R, Yee C, Yancey KB. Antiepiligrin cicatricial pemphigoid represents an autoimmune response to subunits present in laminin 5 (alpha3 beta3 gamma2). Br J Dermatol 1998;139:791 /7. Nousari HC, Rencic A, Hsu R, Yancey KB, Anhalt GJ. Antiepiligrin cicatricial pemphigoid with antibodies against the gamma2 subunit of laminin 5. Arch Dermatol 1999;135:173 /6. Lazarova Z, Hsu R, Briggaman RA, Yancey KB. Related Fab fragments directed against laminin 5 induce subepidermal blisters in neonatal mice. Clin Immunol 2000;95:26 /32. Lazarova Z, Yee C, Lazar J, Yancy KB. IgG autoantibodies in patients with anti-epiligrin cicatricial pemphigoid recognize the G domain of the laminin 5 a-subunit. Clin Immunol 2001;101:100 /5. Nagata Y, Karashima T, Watt FM, Salmhofer W, Kanzaki T, Hashimoto T. Paraneoplastic pemphigus sera react strongly with multiple epitopes on the various regions of envoplakin and periplakin, except for C-terminal homologous domain of periplakin. J Invest Dermatol 2001;116:556 /63. Hamada T, Nagata Y, Tomita M, Salmhofer W, Hashimoto T. Bullous pemphigoid sera react specifically with various domains of BP230, most frequently with C-terminal domain, by immunoblot analyses using bacterial recombinant proteins covering the entire molecule. Exp Dermatol 2001;10:256 /63. Tanaka M, Hashimoto T, Amagai M, Shimizu N, Ikeguchi N, Tsubata T, et al. Characterization of bullous pemphigoid antibodies by use of recombinant bullous pemphigoid antigen proteins. J Invest Dermatol 1991;97:725 /8.
www.elsevier.com/locate/jdermsci
Usefulness of immunoblotting using purified laminin 5 in the diagnosis of anti-laminin 5 cicatricial pemphigoid Yoshiko Hisamatsua, Toshio Nishiyamab, Satoshi Amanob, Chihiro Matsuic, Riza Ghohestanid, Takashi Hashimotoa,* a
Department of Dermatology, Kurume University School of Medicine, 67 Asahimachi, Kurume, Fukuoka 830-0011, Japan b Shiseido Life Science Research Center, Yokohama, Japan c Department of Dermatology, Toyama Medical and Pharmaceutical University, Toyama, Japan d Department of Dermatology, Thomas Jefferson Medical College, Philadelphia, PA, USA Received 25 March 2003; received in revised form 11 June 2003; accepted 16 June 2003
KEYWORDS Cicatricial pemphigoid; Immunoblotting; Laminin 5; Recombinant protein
Summary Background: Anti-laminin 5 cicatricial pemphigoid (CP) is a mucosaldominant subepithelial blistering disease characterized by IgG anti-basement membrane zone autoantibodies, that bind to dermal side of 1 M NaCl split skin and immunoprecipitate laminin 5. Laminin 5 is an epidermis-specific extracellular matrix consisting of a3, b3 and g2 subunits. Recent studies have suggested that autoantibodies of anti-laminin 5 CP recognize the G domains of a3 subunit. Objective: We examined the reactivity of anti-laminin 5 CP by immunoblotting using purified laminin 5 and recombinant proteins of a3 subunit. Method: We first examined the reactivity of anti-laminin 5 CP by immunoblotting using purified laminin 5. To further investigate the epitopes in the G domains of a3 subunit, we produced recombinant proteins of G1 /2, G1 /3, G2 /3, G3 /5 domains, that covered entire G domain, and examined the reactivity of anti-laminin 5 CP sera with these recombinant proteins by immunoblotting. Results: By immunoblotting using purified laminin 5, 7 of 21 anti-laminin 5 CP sera reacted with a3 subunit, while 8 sera reacted with b3 subunit and one serum reacted with g2 subunit. Two sera reacted with both a3 and b3 subunits, while seven sera did not show positive reactivity. This result indicates that the reactivity of antilaminin 5 CP sera is much more heterogeneous, although the previous studies suggested that most sera reacted with a3 subunit. However, in the studies using recombinant proteins of G domains of a3 subunit, none of the CP sera, including the sera reactive with a3 subunit in purified laminin 5, reacted with any recombinant proteins. The reason for this negative reactivity with the recombinant proteins is not clear. Conclusion: The immunoblotting using purified laminin 5 should be useful technique for the diagnosis of anti-laminin 5 CP, although the sensitivity was less than conventional immunoprecipitation analysis. – 2003 Japanese Society for Investigative Dermatology. Published by Elsevier Ireland Ltd. All rights reserved.
Abbreviations: CP, cicatricial pemphigoid; GST, glutathione-S -transferase. *Corresponding author. Tel.: /81-942-31-7571; fax: /81-942-34-2620. E-mail address: [email protected] (T. Hashimoto). 0923-1811/03/$30.00 – 2003 Japanese Society for Investigative Dermatology. Published by Elsevier Ireland Ltd. All rights reserved. doi:10.1016/S0923-1811(03)00158-0
114
1. Introduction Laminin 5 (epiligrin) is an epidermis-specific extracellular matrix consisting of disulfide-linked a3, b3 and g2 subunits [1 /5]. The a3 and g2 subunits exist as cell-associated precursors of 200 and 155 kDa, respectively; these precursors are rapidly processed following secretion to 165 and 105 kDa, respectively [1]. The b3 subunit exists as a 140 kDa polypeptide that does not undergo processing [1]. Laminin 5 binds to its receptor, a6 b4 integrin, and play an important role in the adhesion at the basement membrane zone of the epidermis. Anti-laminin 5 cicatricial pemphigoid (CP) is a mucosal-dominant subepithelial blistering disease characterized by IgG anti-basement membrane zone autoantibodies, that bind to the dermal side of 1 M NaCl split skin [6] and immunoprecipitate laminin 5 from extracts of human keratinocytes [7,8]. Immunoblotting studies using extracts of cultured keratinocytes revealed that the majority of anti-laminin 5 CP sera reacted with a3 subunit [9]. However, a few sera reacted with either b3 or g2 subunits [10,11]. In addition, using neonatal mouse animal model, it has been shown that antibodies to laminin 5 can reproduce skin lesion in the mice, thus indicating that the autoantibodies to laminin 5 in the CP patients should play a pathogenic role [12]. Furthermore, recent studies using bacterial expressed recombinant proteins have suggested that most CP sera recognized the G domains of a3 subunit [13]. In this study, we examined the reactivity of antilaminin 5 CP by immunoblotting using the purified laminin 5. To further investigate the epitopes on G domains of a3 subunit, we produced recombinant proteins of G1 /2, G1 /3, G2 /3, and G3 /5 domains, that cover entire G domain, and examined the reactivity of anti-laminin 5 CP sera by immunoblotting using these recombinant proteins.
2. Materials and methods 2.1. Human sera and antibodies Sera were obtained from 21 patients with antilaminin 5 CP (13 patients from Asia and eight patients from USA). All patients had subepithelial blistering and erosive lesions of mucous membranes. In addition, all the sera had circulating IgG autoantibodies that bound exclusively to the dermal side of 1 M NaCl split skin on indirect immunofluorescence, and immunoprecipitated laminin 5 from lysate of biosynthetically radiolabeled
Y. Hisamatsu et al.
human keratinocytes. Sera from 20 normal volunteers were also used as controls. Two polyclonal rabbit antibodies (designated as Lapin 1 and Lapin 2) raised against a recombinant protein of G2 /G3 domain of a3 subunit were also used. These polyclonal antibodies were shown to react strongly and exclusively with processed form of a3 subunit on immunoblotting using keratinocyte culture media. However, by immunofluorescence, these polyclonal antibodies reacted only faintly with the basement membrane zone in the normal skin sections and relatively weakly with dermal side of the 1 M NaCl split skin section. We also used commercially available anti-glutathione-S -transferase (GST) polyclonal antibody (Sigma, USA). All the sera and antibodies were stored at /30 8C or at 4 8C in the presence of 0.1% sodium azide.
2.2. Purification of laminin 5 Human keratinocytes were cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum. The laminin 5 was purified from the culture medium by using anti-laminin 5 affinity column, prepared by Sepharose gel conjugated with a monoclonal antibody against the laminin a3 chain. The detailed methods were described previously [3 /5].
2.3. Generation of bacterial expression constructs of G domains of a3 subunits of human laminin 5 Domain-specific GST-fusion proteins of G domains of a3 subunit of laminin 5 were generated, as reported previously [14,15]. Four cDNA fragments were amplified by polymerase chain reaction, using keratinocyte cDNA library as templates, as well as appropriate primers shown in Table 1. The keratinocyte cDNA library was prepared from mRNA obtained from the cultured squamous cell line, KU8. To prepare the recombinant protein of G1 /2 domain, we used the primer pair, 5-G1 /G2 and 3G1 /G2. To prepare the recombinant protein of G1 /3 domain, we used the primer pair, 5-G1 /G2 and 3-G1 /G3. To prepare the recombinant protein of G2 /3 domain, we used the primer pair, 5-G2 /G3 and 3-G1 /G3. To prepare the recombinant protein of G3 /5 domain, we used the primer pair, 5-G3 /G5 and 3-G3 /G5. In the 5? region of each primer, the restriction enzyme site for NotI was incorporated to facilitate to prepare construct. PCR products were purified by PCR purification kit (QIAGEN), digested with NotI , and ligated into
Immunoblotting for anti-laminin 5 cicatricial pemphigoid
115
Table 1 The list of primers used to amplify G-domains of a3 subunit of laminin 5 Name
Sequence
5-G1 /G2 3-G1 /G2 3-G1 /G3 5-G2 /G3 5-G3 /G5 3-G3 /G5
gccgcggccgc gccgcggccgc gccgcggccgc gccgcggccgc gccgcggccgc gccgcggccgc
Restriction enzyme ACATCTCTGTCCTTGTTTCTCCAAAGGCCC ACTTTTTGGTTACTCCCACAGTATCATTC AAGCATCTTGCCACACCTTCACGCTCC TGTAGAAGGAGGAAGGAAGAGTCAGAC TGCTCGGAAGACTGGAAGCTTGTGCGATCTGCC TGGTCAGGACAACCATTCAGACTGACAGGCC
the 3? end of GST gene in the bacterial expression vector pGEX-5X-3 (Pharmacia LKB Biotechnology, NJ), which had been digested by the same restriction enzymes and gel purified. For all fragments, PCR was performed on the condition: 94 8C 1 min, one cycle; 98 8C 20 s, 68 8C 15 min, 30 cycles; and 72 8C 10 min, one cycle. All the constructs were subjected to nucleotide sequence analysis by ABI PRISMƒ 310 Genetic Analyzer using 5?- and 3?-pGEX sequencing primers, to confirm the identity and fidelity of subcloning steps.
2.4. Preparation of GST-fusion proteins of G domains E. coli strain XL-1 Blue was transfected with plasmid constructs and grown in 25 ml LB medium at 37 8C overnight. After the overnight culture was added to 250 ml fresh medium, cells were grown at 37 8C for 3 h and IPTG was added to a final concentration of 1.0 mM to induce expression of the fusion protein. After an additional incubation at 37 8C for 3 h, cells were harvested by centrifugation. Pellets were re-suspended in 3-ml phosphate-buffered saline containing 1% Triton-X-100, mildly sonicated and clarified by centrifugation. All fusion proteins were not soluble in this buffer, the pellets were further extracted by resuspending and sonicating in 3 ml 2 M urea [16]. Subsequently, the pellets were resuspended in 1-ml Laemmli’s sample buffer, boiled for 5 min and centrifuged [16]. The supernatant was used as antigen sources for immunoblot analysis.
NotI NotI NotI NotI NotI NotI
at 4 8C with patients’ sera diluted at 1:20 in dilution buffer. Blots were then incubated with horseradish peroxidase-conjugated rabbit anti-human IgG antiserum (1:100, gamma-chain-specific) (Dako, Denmark) as a secondary antibody for 3 h at room temperature and color was developed with 4chloro-1-naphtol in the presence of H2O2. For immunoblotting of recombinant proteins, the four fusion recombinant proteins were separated by 9% SDS-polyacrylamide gels, and transferred onto nitrocellulose membrane. Because these proteins were not soluble in 1% Triton X-100 solution, and could not be purified by glutathione-sepharose 4B column, the patients’ sera were pre-absorbed with E. coli lysate to reduce background reactivity due to non-specific anti-E. coli antibodies. E. coli lysate was prepared as described below. Two hundred milliliter LB medium was added with 20 ml of overnight culture of XL-1 Blue transfected with pGEX without insert, cultured for 3 h, and induced by 1.0 mM IPTG. After the bacteria reached confluence, bacterial pellet was re-suspended in 5 ml of phosphate-buffered saline, subjected to three cycles of freezing/thawing, sonicated mildly, and kept at /80 8C until use. For pre-absorption, the sera diluted in dilution buffer were mixed with the same volume of E. coli lysate, agitated at room temperature for 1 h, and centrifuged at 10 min at 14 000 rpm. The supernatant was used as the first antibody. Following procedures were the same as those for the immunoblotting for purified laminin 5.
2.5. Immunoblot analysis For immunoblotting of purified laminin 5, the sample was directly solubilized in Laemmli’s sample buffer, applied on 6% SDS-polyacrylamide gels, and transferred onto nitrocellulose membrane. The blots were blocked with Tris-buffered saline with 3% skim milk (dilution buffer) for 1 h at room temperature. The blots were incubated overnight
3. Results
3.1. The protein structure of laminin 5 The protein structure of laminin 5 is depicted in Fig. 1.
116
Y. Hisamatsu et al.
3.2. The schematic diagram of the structure of a3 subunit of human laminin 5 and the four truncated recombinant GST-fusion recombinant proteins (Fig. 2) In Fig. 2, we depict an entire structure of a3 subunit of laminin 5 with tandem G1 /G5 domains in its C-terminal domain. We also show the positions and the numbers of residues of all the recombinant proteins; i.e. G1 /2, G1 /3, G2 /3 and G3 /5.
3.3. Circulating IgG antibodies in the antilaminin 5 CP patients’ sera reacted with various subunits of laminin 5 by immunoblot analysis using purified laminin 5 The purified laminin 5 was reduced and applied to 6% SDS-polyacrylamide gels. In this laminin 5 preparation, amido black staining showed three clear protein bands with molecular sizes of 165 kDa (processed form of a3 subunit), 140 (b3 subunit), and 105 kDa (processed form of g2 subunit) (Fig. 3). No protein bands of the 200 kDa unprocessed form of a3 subunit and the 155 kDa unprocessed form of g2 subunit were seen. A weak band seen at the position of about 145 kDa was considered to be an degradation product of a3 subunit, because the two rabbit anti-a3 subunit polyclonal antibodies identified clearly both the 165 kDa processed form of a3 subunit and the 145 kDa band (Fig. 3). Circulating IgG antibodies of 7 of 21 patients with anti-laminin 5 CP bound to the 165 kDa processed form of a3 subunit of laminin 5 (Fig. 3). Some of them also reacted with the 145-kDa proteins. Eight of 21 CP sera bound to the 140 kDa b3 subunit of laminin 5. Only one CP serum clearly bound to the 105 kDa processed form of g2 subunit of laminin 5. Two sera reacted with both a3 and b3 subunits. In contrast, seven sera showed no positive reactivity with any subunits. None of these three subunits of laminin 5 were not reacted by either 20 control normal human sera or normal rabbit serum (Fig. 3).
3.4. None of the anti-laminin 5 CP sera reacted with any truncated recombinant fusion proteins We produced the four truncated recombinant proteins of G domain of a3 subunit and examined the reactivity of the anti-laminin 5 CP sera with these proteins. We used all the 21 CP patients’ sera, which include seven sera reactive with a3 subunit of purified laminin 5, 10 normal control
Fig. 1 The protein structure of laminin 5. Laminin 5 is a heterotrimeric protein composed of disulfide-linked a3, b3, and g2 subunits. The a3 subunit exists as 200 kDa unprocessed and 165 kDa processed polypeptides. The b3 subunit is 140 kDa polypeptide that does not undergo processing. The g2 subunit of laminin 5 exists as 155 kDa unprocessed and 105 kDa processed polypeptides. The a3 chain has a large globular domain in the C-terminus, termed G domain. This C-terminal globular domain is further subdivided into five homologous subdomains; i.e. G1 /G5, each of which consists of 180 /200 amino acid residues. They contain binding sites for heparin and other extracellular matrix proteins, as well as cellular receptors including integrins. According to the recent studies, the proteolytic cleavage of the a3 chain occurs between the G3 and G4 domains.
sera, anti-GST polyclonal antibody and two anti-a3 subunit polyclonal antibodies. None of the 21 antilaminin 5 CP sera and ten normal sera showed any specific reactivity with these four recombinant proteins, while three polyclonal antibodies showed the specific reactivity (Fig. 4).
4. Discussion In the first part of this study, we performed immunoblotting using purified laminin 5. By this immunoblot analysis, seven sera of 21 anti-laminin
Immunoblotting for anti-laminin 5 cicatricial pemphigoid
Fig. 2 The schematic diagram of the structure of a3 subunit of human laminin 5 and the four truncated recombinant GST-fusion recombinant proteins. The upper panel depicts an entire structure of a3 subunit of laminin 5, in which there are five homologous G1 /G5 domains in its C-terminal domain. The middle panel indicates the positions and the numbers of residues of all the recombinant proteins used in this study; i.e. G1 /2, G1 /3, G2 /3, and G3 /5. The lower line indicates the size of cDNA.
5 CP sera reacted with a3 subunit, while 8 sera reacted with b3 subunit and one serum reacted with g2 subunit. However, seven sera did not show positive reactivity. Therefore, the immunoblotting of purified laminin 5 should be a useful method for the diagnosis of anti-laminin 5 CP, although conventional immunoprecipitation analysis using radiolabeled keratinocyte lysate is still more sensitive
117
method. Furthermore, the previous studies indicated that the major antigenic domain is a3 subunit [9]. However, the results shown in the present study indicate that the reactivity of anti-laminin 5 CP sera is more heterogeneous, and all the three subunits of laminin 5 can be recognized by CP patients’ sera. The reason why seven CP sera did not react with any of the subunit in this study is unclear. It may be conceivable that the titer of the antibodies in the sera were too low to be detected in our immunoblot analysis. More likely, there sera may contain antibodies against conformation-dependent epitopes, which can be detected by immunoprecipitation but not by immunoblotting. In the second part of this study, we prepared recombinant proteins. The previous studies using the same pGEX expression system showed that most anti-laminin 5 CP sera reacted with the recombinant proteins of G domains of a3 subunit [13]. However, unexpectedly, by the present immunoblotting study using the four recombinant proteins of G domains of a3 subunit, none of the recombinant proteins were reacted by any patients’ sera, including the sera reactive with a3 subunit in purified laminin 5. The reason for this discrepancy is at the present unknown. All of our four recombinant proteins were recognized by anti-GST polyclonal antibodies and the two polyclonal antibodies raised against G2 /G3
Fig. 3 The results of immunoblot analysis using purified laminin 5. Protein staining by amido black showed three major protein bands (lane 1). One of the polyclonal antibodies to G2 /G3 (Lapin 2) showed strong bands with 165 kDa processed form of a3 subunit and the 145 kDa degradation product (lane 2). The representative 11 anti-laminin 5 CP sera showed reactivity with a3, b3, and g2 subunits (lanes 3 /13), while normal sera did not show any positive reactivity (lane 14 /16). The positions of the 165 kDa processed form of a3 subunit, the 145 kDa degraded a3 subunit, 140 kDa b3, and the 105 kDa processed form of g2 subunits are indicated by arrows in the left margin.
118
Y. Hisamatsu et al.
Fig. 4 The results of immunoblot analysis using four truncated recombinant proteins of G domain of a3 subunit. Panels a /d showed the reactivity in immunoblot analyses for G1 /G2 domain, G1 /G3 domain, G2 /G3 domain and G3 / G5 domain, respectively. In each panel, lanes 1 and 7 show the reactivity of anti-GST polyclonal antibody, lanes 2 /6 show the reactivity of 5 representative anti-laminin 5 CP sera, which reacted with a3 subunit of purified laminin 5, lanes 8 /10 show the reactivity of normal control sera, and lanes 11 and 12 show the reactivity of two polyclonal antibodies to G2 /G3 domain (Lapin 1 and Lapin 2). Anti-GST polyclonal antibody and anti-G2 /G3 polyclonal antibodies reacted with each recombinant protein. However, none of the CP sera and normal sera showed positive reactivity with the recombinant proteins. An arrowhead in the left of each panel indicates the position of intact recombinant protein. In all the panels, the lane with the same number shows the reactivity of the same serum and same polyclonal antibodies. Polyclonal antibodies to G2 /G3 (lanes 11 and 12) produced a strong non-specific protein band. These bands are seen in the top of the lanes in panels a and c, and seen as the second bands in panels b and d.
domain, and showed expected sizes, indicating that the four recombinant proteins were produced properly. Therefore, the reason for the negative reactivity with the recombinant proteins is not clear. However, in the previous studies, the GST fusion proteins were solubilized and purified by glutathione-sepharose column [13], whereas our GST fusion proteins were not soluble and could not be purified in the present study. Therefore, it may be conceivable that some contaminated materials may interfere with the binding of the antibodies to recombinant proteins in our study. Alternatively, in the previous study [13], alkaline phosphatase
method was used to detect the protein bands, while we used peroxidase/4-chloro-1-naphthol method for detection. There may be some difference of sensitivity between the two systems.
Acknowledgements We are grateful to Professor Kim B. Yancey, Department of Dermatology, Wisconsin University, Milwaukee for generously providing us with eight CP sera, as well as important critical discussions. We also thank Miss Michiyo Noge, Miss Yuko Kawano
Immunoblotting for anti-laminin 5 cicatricial pemphigoid
and Miss Ayumi Suzuki for technical assistance and Miss Akiko Tanaka for secretarial work. This work was supported by Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture of Japan, a grant from the Ministry of Health and Welfare of Japan, and a grant for Open Research Center Project from the Ministry of Education.
[8]
[9]
[10]
References [1] Marinkovich MP, Lunstrum GP, Burgeson RE. The anchoring filament protein kalinin is synthesized and secreted as a high molecular weight precursor. J Biol Chem 1992;267:17900 /6. [2] Champliaud MF, Lunstrum GP, Rousselle P, Nishiyama T, Keene DR, Burgeson RE. Human amnion contains a novel laminin variant, laminin 7, which like laminin 6, covalently associates with laminin 5 to promote stable epithelialstromal attachment. J Cell Biol 1996;132:1189 /98. [3] Amano S, Scott IC, Takahara K, Koch M, Champliaud MF, Gerecke DR, et al. Bone morphogenetic protein 1 is an extracellular processing enzyme of the laminin 5 gamma 2 chain. J Biol Chem 2000;275:22728 /35. [4] Amano S, Nishiyama T, Burgeson RE. A specific and sensitive ELISA for laminin 5. J Immunol Methods 1999;224:161 /9. [5] Nishiyama T, Amano S, Tsunenaga M, Kadoya K, Takeda A, Adachi E, et al. The importance of laminin 5 in the dermalepidermal basement membrane. J Dermatol Sci 2000;25:S51 /9. [6] Lazarova Z, Yancey KB. Reactivity of autoantibodies from patients with defined subepidermal bullous diseases against 1 mol/l salt-split skin. Specificity, sensitivity, and practical considerations. J Am Acad Dermatol 1996;35:398 /403. [7] Domloge-Hultsch N, Gammon WR, Briggaman RA, Gil SG, Carter WG, Yancey KB. Epiligrin, the major human keratinocyte integrin ligand, is a target in both an acquired
[11]
[12]
[13]
[14]
[15]
[16]
119 autoimmune and an inherited subepidermal blistering skin disease. J Clin Invest 1992;90:1628 /33. Domloge-Hultsch N, Anhalt GJ, Gammon WR, Lazarova Z, Briggaman R, Welch M, et al. Antiepiligrin cicatricial pemphigoid. A subepithelial bullous disorder. Arch Dermatol 1994;130:1521 /9. Kirtschig G, Marinkovich MP, Burgeson RE, Yancy KB. Antibasement membrane autoantibodies in patients with antiepiligrin cicatricial pemphigoid bind the a subunit of laminin 5. J Invest Dermatol 1995;105:543 /8. Lazarova Z, Hsu R, Yee C, Yancey KB. Antiepiligrin cicatricial pemphigoid represents an autoimmune response to subunits present in laminin 5 (alpha3 beta3 gamma2). Br J Dermatol 1998;139:791 /7. Nousari HC, Rencic A, Hsu R, Yancey KB, Anhalt GJ. Antiepiligrin cicatricial pemphigoid with antibodies against the gamma2 subunit of laminin 5. Arch Dermatol 1999;135:173 /6. Lazarova Z, Hsu R, Briggaman RA, Yancey KB. Related Fab fragments directed against laminin 5 induce subepidermal blisters in neonatal mice. Clin Immunol 2000;95:26 /32. Lazarova Z, Yee C, Lazar J, Yancy KB. IgG autoantibodies in patients with anti-epiligrin cicatricial pemphigoid recognize the G domain of the laminin 5 a-subunit. Clin Immunol 2001;101:100 /5. Nagata Y, Karashima T, Watt FM, Salmhofer W, Kanzaki T, Hashimoto T. Paraneoplastic pemphigus sera react strongly with multiple epitopes on the various regions of envoplakin and periplakin, except for C-terminal homologous domain of periplakin. J Invest Dermatol 2001;116:556 /63. Hamada T, Nagata Y, Tomita M, Salmhofer W, Hashimoto T. Bullous pemphigoid sera react specifically with various domains of BP230, most frequently with C-terminal domain, by immunoblot analyses using bacterial recombinant proteins covering the entire molecule. Exp Dermatol 2001;10:256 /63. Tanaka M, Hashimoto T, Amagai M, Shimizu N, Ikeguchi N, Tsubata T, et al. Characterization of bullous pemphigoid antibodies by use of recombinant bullous pemphigoid antigen proteins. J Invest Dermatol 1991;97:725 /8.