- Email: [email protected]
Localizing antibody-defined immunoreactivity in Porphyromonas gingivalis HtpG recognized by human serum utilizing selective protein expression
Journal of Immunological Methods 285 (2004) 165 – 170 www.elsevier.com/locate/jim
Localizing antibody-defined immunoreactivity in Porphyromonas gingivalis HtpG recognized by human serum utilizing selective protein expression Domenica G. Sweier *, Charles E. Shelburne, Jemiah Cameron, Dennis E. Lopatin Department of Biologic and Materials Sciences, School of Dentistry, The University of Michigan, 1011 North University Avenue, Campus Box 1078, Ann Arbor, MI 48109-1078, USA Received 6 August 2003; received in revised form 19 November 2003; accepted 24 November 2003
Abstract Earlier studies suggested that specific immunoreactive domains of the prokaryotic homologue of Hsp90, HtpG, might contribute to the virulence of the periodontal pathogen, Porphyromonas gingivalis (Pg) [J. Periodontol. 70 (1999) 1185]. Since serum antibodies to this protein appeared to be associated with oral health, we developed a rapid epitope-mapping system that could be tailored to detect antibodies against specific immunoreactive regions of the Pg HtpG protein. This paper describes the use of Caulobacter crescentus (Cc) and the creation of a Cc RsaA fusion protein library that defined specific regions of the Pg HtpG protein. The fusion protein library was used to identify immunoreactive regions in the Pg HtpG dominant in patient and control sera. The development of methods to rapidly localize dominant immunoreactive regions in protein antigens may prove useful for the development of screening tests, vaccines and therapeutics in periodontal and other infectious diseases. D 2004 Elsevier B.V. All rights reserved. Keywords: Epitope; Immunoreactivity; Fusion proteins; Peptide libraries
1. Introduction Identification of immunodominant regions of an antigen is often a critical step in determining its biological activity or virulence mechanism. The process of identifying the unit of an antigen that is recognized by an antibody or a T-cell receptor is referred to as ‘‘epitope mapping.’’ While simple chemical molecules (haptens) can act as epitopes * Corresponding author. Tel.: +1-734-764-4414; fax: +1-734764-2425. E-mail address: [email protected] (D.G. Sweier). 0022-1759/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.jim.2003.11.009
as can nucleic acids and carbohydrate, we generally refer to an epitope as a sequence of amino acids on a protein antigen specifically recognized by the immune system (Mahler et al., 2003). Epitope mapping is generally accomplished by the use of combinatorial libraries. Library methods generally fit into one of five categories (Liu et al., 2003): (i) biological, (ii) spatially addressable, (iii) synthetic, requiring deconvolution, (iv) one-bead one-compound (OBOC) and (v) synthetic, using affinity chromatography. Porphyromonas gingivalis has been implicated as a causative organism in periodontal disease (Griffen
166
D.G. Sweier et al. / Journal of Immunological Methods 285 (2004) 165–170
et al., 1998). Since both clinical studies and in vitro studies suggested that anti-Pg HtpG antibodies were protective, our goal was to identify the HtpG regions immunodominant in healthy subjects, but not by those who developed periodontal disease (Lopatin et al., 1999). We chose to use a biological method of generating a combinatorial library of peptides due to ease of use, flexibility and cost. Additionally, this method allowed us to isolate, using PCR technology, specific segments of interest along the length of the protein. This paper describes the use of Caulobacter crescentus (Cc) and the creation of a Cc RsaA fusion protein library, which defines specific regions of the HtpG protein, as a tool in elucidating the antibody-defined immunoreactive regions in the P. gingivalis HtpG protein (Bingle et al., 1997a,b, 2000; Umelo-Njaka et al., 2001).
2. Materials and methods 2.1. Bacterial strains and cultivation P. gingivalis (ATCC 33277) was grown under anaerobic conditions at 37 jC. The cultures were maintained on blood agar plates (BAPs: enriched trypticase soy agar + 5% defibrinated sheep’s blood) or anaerobic medium 1 plates (Remel). Suspension cultures were grown in brain heart infusion (BHI) broth (BBL) or mycoplasma broth base (BBL) supplemented with 5 Ag ml 1 menadione and 5 Ag ml 1 hemin. Transfected C. crescentus was grown at 30 jC on or in peptone yeast extract (PYE) solid or liquid medium containing 2 Ag ml 1 chloramphenicol. For expression of the fusion protein, the Caulobacter were grown at 30 jC in liquid M11 Expression Medium (Invitrogen) containing 2 Ag ml 1 chloramphenicol. Transfection-competent Caulobacter were purchased from the manufacturer (Invitrogen). 2.2. DNA isolation DNA was isolated from bacterial cells using the WizardR Genomic DNA Isolation Kit (Promega) per manufacturer’s instructions. The genomic DNA was stored at 4 jC until used.
2.3. Peptide construction P. gingivalis htpG (Lopatin et al., 2000) was used in the construction of 19 consecutive sequences of 108 bp each, along the length of the htpG gene from 5V to 3V, hereafter designated #1 –#19. A pair of nondegenerative primer sequences, 18 –20 bp in length, was designed and synthesized for each segment (Invitrogen). Amplicons were generated with Taq DNA polymerase (PCR SuperMix, Invitrogen) using P. gingivalis ATCC 33277 DNA as a template by PCR using the oligonucleotide primer pairs specific for each segment. The amplicon size was confirmed by resolution on a 1.5% agarose gel, and the concentration was estimated using the A260 measurement. Each amplicon was subsequently utilized for fusion protein construction by the Caulobacter Expression System (PureProk, Invitrogen) per manufacturer’s instructions. Briefly, each amplicon was cloned into the TA cloning vector, pCX-TOPOR (Invitrogen), and the insert sequence, frame and orientation were confirmed by double-strand sequencing at the University of Michigan Biomedical Research Core Facilities. This process was repeated for each of the 19 segments representative of the Pg htpG. The final products consisted of plasmids each with the potential of constitutively expressing a fusion protein consisting of a fragment of the Caulobacter RsaA protein and a Pg peptide representative of a specific segment along the Pg HtpG. The Caulobacter cells were transfected with the plasmids, and the fusion proteins were expressed and purified using the manufacturer’s protocols. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDSPAGE) and Western immunoblotting were completed to confirm the presence of a fusion protein using rabbit anti-Pg rHtpG antiserum (rHtpG is recombinant HtpG) (Sweier et al., 2003) and rabbit anti-Cc RsaA antiserum (Invitrogen). 2.4. Western immunoblotting Western immunoblotting was completed as described elsewhere (Lopatin et al., 2000). Briefly, bacterial extracts were prepared by TCA (trichloroacetic acid) precipitation. A protein assay (Bio-Rad Laboratories) was used to determine protein concentration of the extracts. The extracts were boiled in
D.G. Sweier et al. / Journal of Immunological Methods 285 (2004) 165–170
NuPage LDS Sample Buffer (Novex), resolved by SDS-PAGE and electrophoretically transferred to nitrocellulose membranes (Novex). The blots were probed with polyclonal rabbit (Rb) anti-Pg rHtpG (Sweier et al., 2003) or polyclonal Rb anti-Cc RsaA (Invitrogen) antibodies and detected with horseradish peroxidase-conjugated goat anti-rabbit immunoglobulin G antibody. Detection of immunoreactive bands was performed by chemiluminescence (ECL, Amersham) and exposure to X-ray film (Kodak X-Omat AR).
167
al., 1993, 1994). Briefly, 20 serum samples were chosen, 10 representative of periodontal disease and 10 representative of health based on the gingival index, which is a clinical measure of periodontal disease (Loe, 1967). These are the same samples that were used to define the positive relationship between anti-Hsp90 antibody titer, periodontal health and P. gingivalis colonization (Lopatin et al., 1999). The sera were clarified by centrifugation at 10,000 g for 10 min and stored at 20 jC until assayed. 2.7. Statistical and in silico analyses
2.5. ELISA A total of 100 Al of diluted Cc RsaA or one of the 19 Pg HtpG peptide-Cc RsaA fusion proteins (5 Ag ml 1) in carbonate coating buffer (15 mM Na2CO3, 35 mM NaHCO3, pH 9.5) was added to each well of 96-well flat-bottom microELISA plates (Immunlon 1, Dynatech Laboratories) and allowed to adsorb for 30 min at room temperature. The contents of the wells were aspirated, and the plates were blocked with 0.1% non-fat dry milk (NFDM) in Tris-buffered saline (TBS) for 1 h. After washing the plates with TBS, 100 Al of 1:20 human serum diluted with 0.1% NFDM in TBS were added to each well and allowed to incubate for 1 h. After washing the wells with TBS, 100 Al of diluted (1:100 with 0.1% NFDM in TBS) rabbit anti-IgG, IgA or IgM, conjugated to fluorescein isothiocyante (FITC, Kirkegarrd and Perry Laboratories) were added. After a 1-h incubation, the plates were washed with TBS and read using the appropriate excitation, 485 nm, and emission, 535 nm wavelengths, with a GENios microplate reader (Tecan). A total of 20 serum samples were assayed, 10 representative of periodontal disease and 10 representative of health. Serum samples were tested in triplicate, had corresponding control C. crescentus RsaA readings subtracted, and averaged.
Data were generated and reported by the GENios microplate reader (Tecan) using Magellan v3.0 software (Tecan) and were analyzed using Excel v10 and SPSS v11.0. SPSS v11.0 was used to accomplish ANOVA, and a mixed model analysis of disease states, immunoglobulins and fusion proteins, including pair-wise comparisons.
3. Results 3.1. Generation of C. crescentus RsaA and Pg HtpG peptide fusion proteins Fig. 1 displays Western immunoblots comparing the reactivity of the Cc RsaA – Pg HtpG peptide fusion proteins with both the Rb anti-Pg rHtpG (Sweier et
2.6. Human serum The human serum samples used in these experiments consisted of archived samples obtained from an earlier study as approved by the University of Michigan Institutional Review Board (Lopatin et al., 1999). A complete description of the clinical status of the serum donors is given in earlier studies (Bagramian et
Fig. 1. Immunoreactivity of engineered fusion proteins. Panel A: Rb anti-Pg rHtpG, 1:100,000. Panel B: Rb anti-Cc RsaA, 1:5,000. Lane 1, Pg rHtpG; lane 2, Cc RsaA; lane 3, Pg HtpG peptide #5-Cc RsaA; lane 4, Pg HtpG peptide #16-Cc RsaA. There was no detectable reactivity between Pg rHtpG and Rb anti-Cc RsaA, or between Cc RsaA and Rb anti-Pg rHtpG (data not shown). Proteins were loaded at 300 ng total protein except for Pg rHtpG 75 ng and Cc RsaA 100 ng.
168
D.G. Sweier et al. / Journal of Immunological Methods 285 (2004) 165–170
al., 2003), Panel A, and the Rb anti-Cc RsaA (Invitrogen), Panel B, antisera. Reactivities of Pg rHtpG and Cc RsaA proteins are shown for comparison in Fig. 1, lanes 1 and 2, respectively, in both Panels A and B. The control Pg rHtpG and Cc RsaA proteins migrate at the expected 78 and 33 kDa, respectively. The fusion proteins consisting of Cc RsaA and a Pg HtpG peptide migrate at the expected f 38 kDa. The difference is indicative of an insert of f 4– 5 kDa, consistent with the size of the engineered Pg HtpG peptide (108 bp f 36 aa f 4– 5 kDa). Fusion proteins #5 and #16 were chosen as examples to avoid unnecessary repetition of data. All of the peptides, 13 at the time of this writing, elicited anti-RsaA – Pg HtpG peptide IgA, IgG and IgM titers. Note that each fusion protein reacts with both the Rb anti-Cc RsaA and the
Rb anti-Pg rHtpG antisera. There was no evidence of cross-reactivity between the control Cc RsaA protein and the Rb anti-Pg rHtpG antibody or between the Pg rHtpG protein and the Rb anti-Cc RsaA antibody (data not shown). 3.2. Serum reactivity of Pg HtpG peptides in fusion proteins Fig. 2 consists of boxplot graphs comparing the relative amounts of serum IgA, IgG and IgM to two fusion proteins, Pg HtpG peptide #5-Cc RsaA and Pg HtpG peptide #16-Cc RsaA. Note the differences in peptide reactivity in health and disease states. In a mixed model analysis including pair-wise comparisons, IgA, IgG and IgM levels were not significantly different from one another. The fusion proteins, however, did display significant differences from one another, many down to p < 0.005, in reactivity to the human serum samples. In fact, human anti-Pg HtpG peptide #5-Cc RsaA is statistically different than human anti-Pg HtpG peptide #16-Cc RsaA, p < 0.005, in the elicited response from human serum samples.
4. Discussion
Fig. 2. Immunoglobulin levels to engineered fusion proteins. Boxplots exhibiting fusion protein reactivity against human serum from healthy and periodontally diseased donors. Each box represents 10 donors of either health or disease. The boxes represent data between the 25th and the 75th percentiles. The whiskers extend to the minimum and maximum values, excluding outliers. The median value, 50th percentile, is indicated by the horizontal line within each box.
P. gingivalis has been implicated as a causative organism in periodontal disease (Griffen et al., 1998). We began our study of the P. gingivalis HtpG by cloning the gene, htpG, and expressing a recombinant protein, Pg rHtpG, which we used to generate polyclonal antibodies in rabbit (Lopatin et al., 2000; Sweier et al., 2003). We continued our studies into the possible role played by P. gingivalis HtpG in the pathogenesis of periodontal disease by generating an htpG disruption mutant. We used this mutant to evaluate the HtpG role in growth, adherence and invasion of P. gingivalis (Sweier et al., 2003). Additionally, clinical and in vitro studies suggested that anti-Pg HtpG antibodies were protective (Lopatin et al., 1999 and unpublished data). In our continuing study of the P. gingivalis HtpG, we wanted to determine if there were specific immunodominant regions recognized by healthy individuals but not by those susceptible to P. gingivalis-associated periodontal disease.
D.G. Sweier et al. / Journal of Immunological Methods 285 (2004) 165–170
Many common epitope mapping systems use peptide libraries generated by techniques that can be technically difficult, time-consuming and even prohibitively expensive. Selective peptide expression using fusion protein technology offers a fast, easy and economical way to produce a library of peptides to be used experimentally. The Caulobacter cloning and protein expression system represents such a system. Using this system, we were able to be very selective in the precise coding regions we were interested in expressing as fusion proteins. This was possible by using PCR technology to specify the regions to be cloned and expressed. Double-strand sequencing of the cloned region once in an expression plasmid gave us added certainty that we had restricted our cloning region to a specific linear segment of the Pg htpG. We were successful in expressing proteins with sensitive and specific reactivity to both the Rb anti-Cc RsaA and Rb anti-Pg rHtpG antisera as shown in the immunoblots in Fig. 1. Note that bands reactive on Western immunoblotting represent immunoreactivity resulting from the denaturing conditions of SDSPAGE. Therefore, only linear immunoreactive regions become evident. This, by nature, excludes any regions that may exhibit immunoreactivity dependent on the three-dimensional conformation of the protein, i.e. the native state of the protein. The immunoassays, in contrast, do not denature the proteins and the proteins may be closer to their native state, theoretically. However, they are fusion proteins generated in a non-native host so it is not possible to accurately predict what three-dimensional conformations they display. We were able to map the immunoreactive regions along the Pg HtpG protein, a putative virulence determinant in the Pg organism, an established periodontal disease pathogen. By segmenting the HtpG protein, we were able to express fusion proteins representative of the Pg HtpG linearly along the length of the protein from N- to C-terminus. Fig. 2 shows two representative Cc RsaA fusion proteins created with Pg HtpG peptides #5 and #16. The variability of the data is typical for ELISA antiimmunoglobulin detection using serum (Doty et al., 1982). Since antisera to a single protein are, by its very nature, polyclonal, multiple epitopes are recognized or partially recognized. However, due to genetic restrictions, not every epitope is immuno-
169
genic in every individual. By limiting the length of the fusion protein inserts, one can minimize the number of epitopes recognized by any single fusion protein. Once regions of interest are identified, inserts that span the overlap between peptides, or new inserts that subdivide initial inserts can be readily synthesized to further elucidate the actual epitopes. This paper describes the adaptation of selective protein expression, by use of a commercial fusion protein expression system, to the mapping of antibody-defined immunoreactive regions on protein antigens. The system is fast, easy, flexible and economical. It provides a level of specificity and sensitivity useful for many applications involving antibody-defined epitope mapping of protein antigens. Acknowledgements This research was supported by PHS grants DE000423 and DE11117. References Bagramian, R.A., Farghaly, M.M., Lopatin, D., Sowers, M., Syed, S.A., Pomerville, J.L., 1993. Periodontal disease in an Amish population. J. Clin. Periodontol. 20, 269 – 272. Bagramian, R.A., Farghaly, M.M., Lopatin, D., Sowers, M., Syed, S.A., Pomerville, J.L., 1994. A comparison of periodontal disease among rural Amish and non-Amish adults. J. Clin. Periodontol. 21, 386 – 390. Bingle, W.H., Nomellini, J.F., Smit, J., 1997a. Cell-surface display of a Pseudomonas aeruginosa strain K pilin peptide within the paracrystalline S-layer of Caulobacter crescentus. Mol. Microbiol. 26, 277 – 288. Bingle, W.H., Nomellini, J.F., Smit, J., 1997b. Linker mutagenesis of the Caulobacter crescentus S-layer protein: toward a definition of an N-terminal anchoring region and a C-terminal secretion signal and the potential for heterologous protein secretion. J. Bacteriol. 179, 601 – 611. Bingle, W.H., Nomellini, J.F., Smit, J., 2000. Secretion of the Caulobacter crescentus S-layer protein: further localization of the Cterminal secretion signal and its use for secretion of recombinant proteins. J. Bacteriol. 182, 3298 – 3301. Doty, S.L., Lopatin, D.E., Syed, S.A., Smith, F.N., 1982. Humoral immune response to oral microorganisms in periodontitis. Infect. Immun. 37, 499 – 505. Griffen, A.L., Becker, M.R., Lyons, S.R., Moeschberger, M.L., Leys, E.J., 1998. Prevalence of Porphyromonas gingivalis and periodontal health status. J. Clin. Microbiol. 36, 3239 – 3242. Liu, R., Enstrom, A.M., Lam, K.S., 2003. Combinatorial peptide
170
D.G. Sweier et al. / Journal of Immunological Methods 285 (2004) 165–170
library methods for immunobiology research. Exp. Hematol. 31, 11 – 30. Loe, H., 1967. The gingival index, the plaque index and the retention index systems. J. Periodontol. 38, 610 – 616 (Suppl.). Lopatin, D.E., Shelburne, C.E., Van Poperin, N., Kowalski, C.J., Bagramian, R.A., 1999. Humoral immunity to stress proteins and periodontal disease. J. Periodontol. 70, 1185 – 1193. Lopatin, D.E., Combs, A., Sweier, D.G., Fenno, J.C., Dhamija, S., 2000. Characterization of heat-inducible expression and cloning of HtpG (Hsp90 homologue) of Porphyromonas gingivalis. Infect. Immun. 68, 1980 – 1987.
Mahler, M., Bluthner, M., Pollard, K.M., 2003. Advances in B-cell epitope analysis of autoantigens in connective tissue diseases. Clin. Immunol. 107, 65 – 79. Sweier, D.G., Combs, A., Shelburne, C.E., Fenno, J.C., Lopatin, D.E., 2003. Construction and characterization of a Porphyromonas gingivalis htpG disruption mutant. FEMS Microbiol. Lett. 225, 101 – 106. Umelo-Njaka, E., Nomellini, J.F., Yim, H., Smit, J., 2001. Development of small high-copy-number plasmid vectors for gene expression in Caulobacter crescentus. Plasmid 46, 37 – 46.
Localizing antibody-defined immunoreactivity in Porphyromonas gingivalis HtpG recognized by human serum utilizing selective protein expression Domenica G. Sweier *, Charles E. Shelburne, Jemiah Cameron, Dennis E. Lopatin Department of Biologic and Materials Sciences, School of Dentistry, The University of Michigan, 1011 North University Avenue, Campus Box 1078, Ann Arbor, MI 48109-1078, USA Received 6 August 2003; received in revised form 19 November 2003; accepted 24 November 2003
Abstract Earlier studies suggested that specific immunoreactive domains of the prokaryotic homologue of Hsp90, HtpG, might contribute to the virulence of the periodontal pathogen, Porphyromonas gingivalis (Pg) [J. Periodontol. 70 (1999) 1185]. Since serum antibodies to this protein appeared to be associated with oral health, we developed a rapid epitope-mapping system that could be tailored to detect antibodies against specific immunoreactive regions of the Pg HtpG protein. This paper describes the use of Caulobacter crescentus (Cc) and the creation of a Cc RsaA fusion protein library that defined specific regions of the Pg HtpG protein. The fusion protein library was used to identify immunoreactive regions in the Pg HtpG dominant in patient and control sera. The development of methods to rapidly localize dominant immunoreactive regions in protein antigens may prove useful for the development of screening tests, vaccines and therapeutics in periodontal and other infectious diseases. D 2004 Elsevier B.V. All rights reserved. Keywords: Epitope; Immunoreactivity; Fusion proteins; Peptide libraries
1. Introduction Identification of immunodominant regions of an antigen is often a critical step in determining its biological activity or virulence mechanism. The process of identifying the unit of an antigen that is recognized by an antibody or a T-cell receptor is referred to as ‘‘epitope mapping.’’ While simple chemical molecules (haptens) can act as epitopes * Corresponding author. Tel.: +1-734-764-4414; fax: +1-734764-2425. E-mail address: [email protected] (D.G. Sweier). 0022-1759/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.jim.2003.11.009
as can nucleic acids and carbohydrate, we generally refer to an epitope as a sequence of amino acids on a protein antigen specifically recognized by the immune system (Mahler et al., 2003). Epitope mapping is generally accomplished by the use of combinatorial libraries. Library methods generally fit into one of five categories (Liu et al., 2003): (i) biological, (ii) spatially addressable, (iii) synthetic, requiring deconvolution, (iv) one-bead one-compound (OBOC) and (v) synthetic, using affinity chromatography. Porphyromonas gingivalis has been implicated as a causative organism in periodontal disease (Griffen
166
D.G. Sweier et al. / Journal of Immunological Methods 285 (2004) 165–170
et al., 1998). Since both clinical studies and in vitro studies suggested that anti-Pg HtpG antibodies were protective, our goal was to identify the HtpG regions immunodominant in healthy subjects, but not by those who developed periodontal disease (Lopatin et al., 1999). We chose to use a biological method of generating a combinatorial library of peptides due to ease of use, flexibility and cost. Additionally, this method allowed us to isolate, using PCR technology, specific segments of interest along the length of the protein. This paper describes the use of Caulobacter crescentus (Cc) and the creation of a Cc RsaA fusion protein library, which defines specific regions of the HtpG protein, as a tool in elucidating the antibody-defined immunoreactive regions in the P. gingivalis HtpG protein (Bingle et al., 1997a,b, 2000; Umelo-Njaka et al., 2001).
2. Materials and methods 2.1. Bacterial strains and cultivation P. gingivalis (ATCC 33277) was grown under anaerobic conditions at 37 jC. The cultures were maintained on blood agar plates (BAPs: enriched trypticase soy agar + 5% defibrinated sheep’s blood) or anaerobic medium 1 plates (Remel). Suspension cultures were grown in brain heart infusion (BHI) broth (BBL) or mycoplasma broth base (BBL) supplemented with 5 Ag ml 1 menadione and 5 Ag ml 1 hemin. Transfected C. crescentus was grown at 30 jC on or in peptone yeast extract (PYE) solid or liquid medium containing 2 Ag ml 1 chloramphenicol. For expression of the fusion protein, the Caulobacter were grown at 30 jC in liquid M11 Expression Medium (Invitrogen) containing 2 Ag ml 1 chloramphenicol. Transfection-competent Caulobacter were purchased from the manufacturer (Invitrogen). 2.2. DNA isolation DNA was isolated from bacterial cells using the WizardR Genomic DNA Isolation Kit (Promega) per manufacturer’s instructions. The genomic DNA was stored at 4 jC until used.
2.3. Peptide construction P. gingivalis htpG (Lopatin et al., 2000) was used in the construction of 19 consecutive sequences of 108 bp each, along the length of the htpG gene from 5V to 3V, hereafter designated #1 –#19. A pair of nondegenerative primer sequences, 18 –20 bp in length, was designed and synthesized for each segment (Invitrogen). Amplicons were generated with Taq DNA polymerase (PCR SuperMix, Invitrogen) using P. gingivalis ATCC 33277 DNA as a template by PCR using the oligonucleotide primer pairs specific for each segment. The amplicon size was confirmed by resolution on a 1.5% agarose gel, and the concentration was estimated using the A260 measurement. Each amplicon was subsequently utilized for fusion protein construction by the Caulobacter Expression System (PureProk, Invitrogen) per manufacturer’s instructions. Briefly, each amplicon was cloned into the TA cloning vector, pCX-TOPOR (Invitrogen), and the insert sequence, frame and orientation were confirmed by double-strand sequencing at the University of Michigan Biomedical Research Core Facilities. This process was repeated for each of the 19 segments representative of the Pg htpG. The final products consisted of plasmids each with the potential of constitutively expressing a fusion protein consisting of a fragment of the Caulobacter RsaA protein and a Pg peptide representative of a specific segment along the Pg HtpG. The Caulobacter cells were transfected with the plasmids, and the fusion proteins were expressed and purified using the manufacturer’s protocols. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDSPAGE) and Western immunoblotting were completed to confirm the presence of a fusion protein using rabbit anti-Pg rHtpG antiserum (rHtpG is recombinant HtpG) (Sweier et al., 2003) and rabbit anti-Cc RsaA antiserum (Invitrogen). 2.4. Western immunoblotting Western immunoblotting was completed as described elsewhere (Lopatin et al., 2000). Briefly, bacterial extracts were prepared by TCA (trichloroacetic acid) precipitation. A protein assay (Bio-Rad Laboratories) was used to determine protein concentration of the extracts. The extracts were boiled in
D.G. Sweier et al. / Journal of Immunological Methods 285 (2004) 165–170
NuPage LDS Sample Buffer (Novex), resolved by SDS-PAGE and electrophoretically transferred to nitrocellulose membranes (Novex). The blots were probed with polyclonal rabbit (Rb) anti-Pg rHtpG (Sweier et al., 2003) or polyclonal Rb anti-Cc RsaA (Invitrogen) antibodies and detected with horseradish peroxidase-conjugated goat anti-rabbit immunoglobulin G antibody. Detection of immunoreactive bands was performed by chemiluminescence (ECL, Amersham) and exposure to X-ray film (Kodak X-Omat AR).
167
al., 1993, 1994). Briefly, 20 serum samples were chosen, 10 representative of periodontal disease and 10 representative of health based on the gingival index, which is a clinical measure of periodontal disease (Loe, 1967). These are the same samples that were used to define the positive relationship between anti-Hsp90 antibody titer, periodontal health and P. gingivalis colonization (Lopatin et al., 1999). The sera were clarified by centrifugation at 10,000 g for 10 min and stored at 20 jC until assayed. 2.7. Statistical and in silico analyses
2.5. ELISA A total of 100 Al of diluted Cc RsaA or one of the 19 Pg HtpG peptide-Cc RsaA fusion proteins (5 Ag ml 1) in carbonate coating buffer (15 mM Na2CO3, 35 mM NaHCO3, pH 9.5) was added to each well of 96-well flat-bottom microELISA plates (Immunlon 1, Dynatech Laboratories) and allowed to adsorb for 30 min at room temperature. The contents of the wells were aspirated, and the plates were blocked with 0.1% non-fat dry milk (NFDM) in Tris-buffered saline (TBS) for 1 h. After washing the plates with TBS, 100 Al of 1:20 human serum diluted with 0.1% NFDM in TBS were added to each well and allowed to incubate for 1 h. After washing the wells with TBS, 100 Al of diluted (1:100 with 0.1% NFDM in TBS) rabbit anti-IgG, IgA or IgM, conjugated to fluorescein isothiocyante (FITC, Kirkegarrd and Perry Laboratories) were added. After a 1-h incubation, the plates were washed with TBS and read using the appropriate excitation, 485 nm, and emission, 535 nm wavelengths, with a GENios microplate reader (Tecan). A total of 20 serum samples were assayed, 10 representative of periodontal disease and 10 representative of health. Serum samples were tested in triplicate, had corresponding control C. crescentus RsaA readings subtracted, and averaged.
Data were generated and reported by the GENios microplate reader (Tecan) using Magellan v3.0 software (Tecan) and were analyzed using Excel v10 and SPSS v11.0. SPSS v11.0 was used to accomplish ANOVA, and a mixed model analysis of disease states, immunoglobulins and fusion proteins, including pair-wise comparisons.
3. Results 3.1. Generation of C. crescentus RsaA and Pg HtpG peptide fusion proteins Fig. 1 displays Western immunoblots comparing the reactivity of the Cc RsaA – Pg HtpG peptide fusion proteins with both the Rb anti-Pg rHtpG (Sweier et
2.6. Human serum The human serum samples used in these experiments consisted of archived samples obtained from an earlier study as approved by the University of Michigan Institutional Review Board (Lopatin et al., 1999). A complete description of the clinical status of the serum donors is given in earlier studies (Bagramian et
Fig. 1. Immunoreactivity of engineered fusion proteins. Panel A: Rb anti-Pg rHtpG, 1:100,000. Panel B: Rb anti-Cc RsaA, 1:5,000. Lane 1, Pg rHtpG; lane 2, Cc RsaA; lane 3, Pg HtpG peptide #5-Cc RsaA; lane 4, Pg HtpG peptide #16-Cc RsaA. There was no detectable reactivity between Pg rHtpG and Rb anti-Cc RsaA, or between Cc RsaA and Rb anti-Pg rHtpG (data not shown). Proteins were loaded at 300 ng total protein except for Pg rHtpG 75 ng and Cc RsaA 100 ng.
168
D.G. Sweier et al. / Journal of Immunological Methods 285 (2004) 165–170
al., 2003), Panel A, and the Rb anti-Cc RsaA (Invitrogen), Panel B, antisera. Reactivities of Pg rHtpG and Cc RsaA proteins are shown for comparison in Fig. 1, lanes 1 and 2, respectively, in both Panels A and B. The control Pg rHtpG and Cc RsaA proteins migrate at the expected 78 and 33 kDa, respectively. The fusion proteins consisting of Cc RsaA and a Pg HtpG peptide migrate at the expected f 38 kDa. The difference is indicative of an insert of f 4– 5 kDa, consistent with the size of the engineered Pg HtpG peptide (108 bp f 36 aa f 4– 5 kDa). Fusion proteins #5 and #16 were chosen as examples to avoid unnecessary repetition of data. All of the peptides, 13 at the time of this writing, elicited anti-RsaA – Pg HtpG peptide IgA, IgG and IgM titers. Note that each fusion protein reacts with both the Rb anti-Cc RsaA and the
Rb anti-Pg rHtpG antisera. There was no evidence of cross-reactivity between the control Cc RsaA protein and the Rb anti-Pg rHtpG antibody or between the Pg rHtpG protein and the Rb anti-Cc RsaA antibody (data not shown). 3.2. Serum reactivity of Pg HtpG peptides in fusion proteins Fig. 2 consists of boxplot graphs comparing the relative amounts of serum IgA, IgG and IgM to two fusion proteins, Pg HtpG peptide #5-Cc RsaA and Pg HtpG peptide #16-Cc RsaA. Note the differences in peptide reactivity in health and disease states. In a mixed model analysis including pair-wise comparisons, IgA, IgG and IgM levels were not significantly different from one another. The fusion proteins, however, did display significant differences from one another, many down to p < 0.005, in reactivity to the human serum samples. In fact, human anti-Pg HtpG peptide #5-Cc RsaA is statistically different than human anti-Pg HtpG peptide #16-Cc RsaA, p < 0.005, in the elicited response from human serum samples.
4. Discussion
Fig. 2. Immunoglobulin levels to engineered fusion proteins. Boxplots exhibiting fusion protein reactivity against human serum from healthy and periodontally diseased donors. Each box represents 10 donors of either health or disease. The boxes represent data between the 25th and the 75th percentiles. The whiskers extend to the minimum and maximum values, excluding outliers. The median value, 50th percentile, is indicated by the horizontal line within each box.
P. gingivalis has been implicated as a causative organism in periodontal disease (Griffen et al., 1998). We began our study of the P. gingivalis HtpG by cloning the gene, htpG, and expressing a recombinant protein, Pg rHtpG, which we used to generate polyclonal antibodies in rabbit (Lopatin et al., 2000; Sweier et al., 2003). We continued our studies into the possible role played by P. gingivalis HtpG in the pathogenesis of periodontal disease by generating an htpG disruption mutant. We used this mutant to evaluate the HtpG role in growth, adherence and invasion of P. gingivalis (Sweier et al., 2003). Additionally, clinical and in vitro studies suggested that anti-Pg HtpG antibodies were protective (Lopatin et al., 1999 and unpublished data). In our continuing study of the P. gingivalis HtpG, we wanted to determine if there were specific immunodominant regions recognized by healthy individuals but not by those susceptible to P. gingivalis-associated periodontal disease.
D.G. Sweier et al. / Journal of Immunological Methods 285 (2004) 165–170
Many common epitope mapping systems use peptide libraries generated by techniques that can be technically difficult, time-consuming and even prohibitively expensive. Selective peptide expression using fusion protein technology offers a fast, easy and economical way to produce a library of peptides to be used experimentally. The Caulobacter cloning and protein expression system represents such a system. Using this system, we were able to be very selective in the precise coding regions we were interested in expressing as fusion proteins. This was possible by using PCR technology to specify the regions to be cloned and expressed. Double-strand sequencing of the cloned region once in an expression plasmid gave us added certainty that we had restricted our cloning region to a specific linear segment of the Pg htpG. We were successful in expressing proteins with sensitive and specific reactivity to both the Rb anti-Cc RsaA and Rb anti-Pg rHtpG antisera as shown in the immunoblots in Fig. 1. Note that bands reactive on Western immunoblotting represent immunoreactivity resulting from the denaturing conditions of SDSPAGE. Therefore, only linear immunoreactive regions become evident. This, by nature, excludes any regions that may exhibit immunoreactivity dependent on the three-dimensional conformation of the protein, i.e. the native state of the protein. The immunoassays, in contrast, do not denature the proteins and the proteins may be closer to their native state, theoretically. However, they are fusion proteins generated in a non-native host so it is not possible to accurately predict what three-dimensional conformations they display. We were able to map the immunoreactive regions along the Pg HtpG protein, a putative virulence determinant in the Pg organism, an established periodontal disease pathogen. By segmenting the HtpG protein, we were able to express fusion proteins representative of the Pg HtpG linearly along the length of the protein from N- to C-terminus. Fig. 2 shows two representative Cc RsaA fusion proteins created with Pg HtpG peptides #5 and #16. The variability of the data is typical for ELISA antiimmunoglobulin detection using serum (Doty et al., 1982). Since antisera to a single protein are, by its very nature, polyclonal, multiple epitopes are recognized or partially recognized. However, due to genetic restrictions, not every epitope is immuno-
169
genic in every individual. By limiting the length of the fusion protein inserts, one can minimize the number of epitopes recognized by any single fusion protein. Once regions of interest are identified, inserts that span the overlap between peptides, or new inserts that subdivide initial inserts can be readily synthesized to further elucidate the actual epitopes. This paper describes the adaptation of selective protein expression, by use of a commercial fusion protein expression system, to the mapping of antibody-defined immunoreactive regions on protein antigens. The system is fast, easy, flexible and economical. It provides a level of specificity and sensitivity useful for many applications involving antibody-defined epitope mapping of protein antigens. Acknowledgements This research was supported by PHS grants DE000423 and DE11117. References Bagramian, R.A., Farghaly, M.M., Lopatin, D., Sowers, M., Syed, S.A., Pomerville, J.L., 1993. Periodontal disease in an Amish population. J. Clin. Periodontol. 20, 269 – 272. Bagramian, R.A., Farghaly, M.M., Lopatin, D., Sowers, M., Syed, S.A., Pomerville, J.L., 1994. A comparison of periodontal disease among rural Amish and non-Amish adults. J. Clin. Periodontol. 21, 386 – 390. Bingle, W.H., Nomellini, J.F., Smit, J., 1997a. Cell-surface display of a Pseudomonas aeruginosa strain K pilin peptide within the paracrystalline S-layer of Caulobacter crescentus. Mol. Microbiol. 26, 277 – 288. Bingle, W.H., Nomellini, J.F., Smit, J., 1997b. Linker mutagenesis of the Caulobacter crescentus S-layer protein: toward a definition of an N-terminal anchoring region and a C-terminal secretion signal and the potential for heterologous protein secretion. J. Bacteriol. 179, 601 – 611. Bingle, W.H., Nomellini, J.F., Smit, J., 2000. Secretion of the Caulobacter crescentus S-layer protein: further localization of the Cterminal secretion signal and its use for secretion of recombinant proteins. J. Bacteriol. 182, 3298 – 3301. Doty, S.L., Lopatin, D.E., Syed, S.A., Smith, F.N., 1982. Humoral immune response to oral microorganisms in periodontitis. Infect. Immun. 37, 499 – 505. Griffen, A.L., Becker, M.R., Lyons, S.R., Moeschberger, M.L., Leys, E.J., 1998. Prevalence of Porphyromonas gingivalis and periodontal health status. J. Clin. Microbiol. 36, 3239 – 3242. Liu, R., Enstrom, A.M., Lam, K.S., 2003. Combinatorial peptide
170
D.G. Sweier et al. / Journal of Immunological Methods 285 (2004) 165–170
library methods for immunobiology research. Exp. Hematol. 31, 11 – 30. Loe, H., 1967. The gingival index, the plaque index and the retention index systems. J. Periodontol. 38, 610 – 616 (Suppl.). Lopatin, D.E., Shelburne, C.E., Van Poperin, N., Kowalski, C.J., Bagramian, R.A., 1999. Humoral immunity to stress proteins and periodontal disease. J. Periodontol. 70, 1185 – 1193. Lopatin, D.E., Combs, A., Sweier, D.G., Fenno, J.C., Dhamija, S., 2000. Characterization of heat-inducible expression and cloning of HtpG (Hsp90 homologue) of Porphyromonas gingivalis. Infect. Immun. 68, 1980 – 1987.
Mahler, M., Bluthner, M., Pollard, K.M., 2003. Advances in B-cell epitope analysis of autoantigens in connective tissue diseases. Clin. Immunol. 107, 65 – 79. Sweier, D.G., Combs, A., Shelburne, C.E., Fenno, J.C., Lopatin, D.E., 2003. Construction and characterization of a Porphyromonas gingivalis htpG disruption mutant. FEMS Microbiol. Lett. 225, 101 – 106. Umelo-Njaka, E., Nomellini, J.F., Yim, H., Smit, J., 2001. Development of small high-copy-number plasmid vectors for gene expression in Caulobacter crescentus. Plasmid 46, 37 – 46.