Mass Spectrometric Characterization of the Surface-Associated 42 kDa Lipoprotein JlpA as a Glycosylated Antigen in Strains of Campylobacter jejuni Nichollas E. Scott,† Daniel R. Bogema,‡,§ Angela M. Connolly,† Linda Falconer, Steven P. Djordjevic,‡,# and Stuart J. Cordwell*,†,| School of Molecular and Microbial Biosciences, The University of Sydney, Australia 2006, NSW Department of Primary Industries, Elizabeth Macarthur Agricultural Institute, Camden, Australia 2570, School of Biological Sciences, University of Wollongong, Wollongong, Australia 2522, and Discipline of Pathology, School of Medical Sciences, The University of Sydney, Australia 2006 Received June 22, 2009
Campylobacter jejuni is the most common cause of bacterial gastroenteritis in the developed world. Immunoproteomics highlighted a 42-45 kDa antigen that comigrated on two-dimensional (2-DE) gels with the C. jejuni major outer membrane protein (MOMP). Predictive analysis revealed two candidates for the identity of the antigen, the most likely of which was the surface-associated lipoprotein, JlpA. Recombinant JlpA (rJlpA) reacted with patient sera, confirming that JlpA is antigenic. Polyclonal antibodies raised against rJlpA reacted against 3 JlpA mass variants from multiple C. jejuni. These variants differed by approximately 1.5 kDa, suggesting the presence of the N-linked C. jejuni glycan on two sites. Soybean agglutinin affinity and 2-DE purified 2 JlpA glycoforms (43.5 and 45 kDa). Their identities were confirmed using mass spectrometry following trypsin digest. Glycopeptides within JlpA variants were identified by proteinase-K digestion, graphite micropurification and MS-MS. Sites of glycosylation were confirmed as asparagines 107 and 146, both of which are flanked by the N-linked sequon. Sequence analysis confirmed that the N146 sequon is conserved in all C. jejuni genomes examined to date, while the N107 sequon is absent in the reference strain NCTC 11168. Western blotting confirmed the presence of only a single JlpA glycoform in both virulent (O) and avirulent (GS) isolates of NCTC 11168. MS analysis showed that JlpA exists as 3 discrete forms, unmodified, glycosylated at N146, and glycosylated at both N146/107, suggesting glycan addition at N146 is necessary for N107 glycosylation. Glycine extracts and Western blotting revealed that doubly glycosylated JlpA was the predominant form on the C. jejuni JHH1 surface; however, glycosylation is not required for antigenicity. This is the first study to identify N-linked glycosylation of a surface-exposed C. jejuni virulence factor and to show strain variation in glycosylation sites. Keywords: JlpA • Campylobacter jejuni • glycosylation • antigen • mass spectrometry • outer membrane proteins
Introduction Campylobacter jejuni is a microaerophilic, spiral-shaped and motile Gram-negative bacterium, that is the most frequent cause of food- and water-borne diarrheal illness worldwide,1 infecting approximately 1% of the population in the United States,2 and causing 223 000 human infections in Australia,3 * Corresponding Author: Dr. Stuart J. Cordwell, School of Molecular and Microbial Biosciences, Building GO8, The University of Sydney, Australia 2006. Phone: (+61-2) 9351-6050; fax, (+61-2) 9351-4726; e-mail, s.cordwell@ usyd.edu.au. † School of Molecular and Microbial Biosciences, The University of Sydney. ‡ Elizabeth Macarthur Agricultural Institute. § University of Wollongong. # Present address: Institute for Biotechnology of Infectious Diseases, P.O. Box 123, The University of Technology Sydney, Australia 2007. | School of Medical Sciences, The University of Sydney.
4654 Journal of Proteome Research 2009, 8, 4654–4664 Published on Web 08/18/2009
per annum. C. jejuni infections are often caused by consumption of contaminated poultry, although other sources are possible,4 and infection with as few as 500 cells has been documented.5 Symptoms range from mild, noninflammatory diarrhea, to severe abdominal cramps, vomiting and inflammation.6 C. jejuni has also been linked to the development of two chronic immune-mediated disorders, Guillain-Barre´ Syndrome7 and, more recently, immunoproliferative small intestine disease, an infection-induced lymphoma originating in the mucosal-associated lymphoid tissue.8 C. jejuni thus presents a significant burden on health systems worldwide, with the cost of infection and associated sequelae being estimated at $8 billion a year in the U.S.9 Strategies for limiting C. jejuni infections in the wider population are therefore of significant interest and include detection of C. jejuni within poultry flocks10,11 and prepared meats,12,13 limitation of colonization 10.1021/pr900544x CCC: $40.75
2009 American Chemical Society
research articles
Glycosylation of the JlpA Antigen in Campylobacter jejuni 14,15
using competitive strains, and reduction of bacterial load using antimicrobial agents.16,17 A major challenge for the development of infection control strategies is our limited understanding of C. jejuni virulence, particularly with respect to invasion and survival within the host. Despite the comprehensive sequencing of several representative genomes,18-22 the genes responsible for colonization and other virulence traits are largely to be elucidated. The development of disease relies on the organism adapting to the gut environment, adherence to intestinal epithelial cells, followed by internalization, invasion and toxin production leading to host cell death.23 Several adhesins have been identified, including the fibronectin-binding protein CadF,24 and the surface-associated lipoprotein, JlpA.25 JlpA promotes adherence to epithelial cells, potentially via interactions with surfaceexposed heat shock protein 90-R.26 Such adhesins are of interest because they not only facilitate the initial steps needed for disease, but may also be useful vaccine candidates. A unique molecular trait of C. jejuni is the ability to posttranslationally modify proteins by the N-linked addition of a 7 residue glycan (GalNAc-R1,4-GalNAc-R1,4-[Glcβ1,3]-GalNAcR1,4-GalNAc-R1,4-GalNAc-R1,3-Bac-β1, where Bac is bacillosamine [2,4-diacetamido-2,4,6 trideoxyglucopyranose]) at the consensus sequon D/E-X-N-X-S/T.27 The modification occurs exclusively on noncytoplasmic proteins and over 20 glycoproteins have been identified.27-30 The glycosylation apparatus is essential for virulence;31,32 however, mutagenesis studies on currently known glycoproteins are yet to reveal virulence factor/s that require glycosylation for mediating colonization or cell adhesion.28,29 A better understanding will only be possible when a comprehensive analysis of C. jejuni glycoproteins, across multiple virulent strains, has been performed. The development of novel vaccines is an appealing approach to limit C. jejuni in the human population.33-36 Immunological studies have shown that a protective humoral response is generated in response to C. jejuni that can provide protection against subsequent challenge.37 Campylobacter-specific immunoglobulin has been used to remove chronic infection in hypogammaglobulinemic patients.38 Examination of the immunological response to C. jejuni has been undertaken39-45 and multiple antigens are targeted. The distribution and reactivity of individual antigens confirm that few demonstrate seroconversion within all patients, and diverse antibody patterns exist.41,42 This absence of seroconversion occurs with multiple immunogens, including Omp1846 and the PEB antigens.47 The intensity of the response can also differ between patients, an effect observed for CadF where multiple patient sera react, but with clear differences in intensity.24 Several immunogens have been identified in C. jejuni; however, it is likely that more exist. Immunoproteomics has been used to identify proteins recognized by chicken maternal antibodies,48 while another study investigated the C. jejuni membrane proteome and found evidence for the presence of an unknown immunogen comigrating on two-dimensional electrophoresis (2-DE) gels with the major outer membrane protein (MOMP).49 Here we confirm the identity of this 42-45 kDa antigen as the surface-associated lipoprotein JlpA. We also show that JlpA is glycosylated at two sites: a pattern observed in all C. jejuni strains examined, with the exception of the reference strain NCTC 11168. Glycosylation correlates with surface localization, but is not required for recognition by the humoral response. JlpA is the first surface-exposed virulence
factor to be confirmed as a glycoprotein and our data highlight that glycosylation may vary between strains.
Materials and Methods Bacterial Strains and Growth. Bacterial strains used in this study are outlined in Supporting Information. C. jejuni were cultured on parallel Skirrow’s agar (Oxoid, U.K.) plates in a microaerophilic environment of 5% O2, 5% CO2 and 90% N2 at 37 °C for 48 h. Plates were flooded with 5 mL of sterile phosphate-buffered saline (PBS) and colonies removed with a cell scraper. Cells were washed 3 times in PBS and collected by centrifugation at 12 000g. Cells were lyophilized and stored at -80 °C until required.50,51 DNA Extraction. Chromosomal DNA was extracted from C. jejuni JHH1 using Instagene matrix (Bio-Rad, Hercules, CA) according to the manufacturer’s instructions and as per Burnett et al.52 Plasmid DNA was extracted from Escherichia coli as previously described.53 Construction of Recombinant JlpA. A polyhistidine fusion JlpA construct was created using chromosomal DNA extracts from C. jejuni JHH1. The full-length JlpA open reading frame was amplified by PCR using Pwo polymerase (Roche, Basel, Switzerland) and the primers JlpAF, 5′CACCATGAAAAAAGGTATTTTTCTCTCTATT3′, and JlpAR, 5′TTAAAATGACGCTCCGCCCATTA3′. The JlpA PCR product was isolated via gel excision using an Ultrafree-DA centrifugal filter unit (Millipore, Billerica, MA) and ligated into the pET100/D-TOPO vector (Invitrogen, Carlsbad, CA). TOP10 chemically competent cells were transformed according to the manufacturer and incubated at 37 °C on LB agar containing 100 µg/mL ampicillin (Sigma, St. Louis, MO). Plasmids were extracted using a QIAprep Spin Miniprep Kit (QIAGEN) and screened for correct orientation by PCR and confirmed by DNA sequencing. Protein Expression and Purification. Recombinant JlpA (rJlpA) was expressed as a hexahistidyl fusion protein and purified under denaturing conditions using immobilized metal affinity chromatography (IMAC) resin as previously described,54 with minor alteration. Briefly, chemically competent E. coli BL21Star (Invitrogen) was transformed with the JlpA-containing plasmid (as above), inoculated into LB broth containing 100 µg/mL ampicillin and cultured overnight at 37 °C. This was then used to inoculate a larger culture, which was grown to midlog phase (OD600 ) 0.5-0.8) before induction of recombinant expression with 1 mM IPTG for 3 h. Cells were collected at 10 000g for 30 min at 4 °C and lysed in 8 M urea buffer (8 M urea, 0.01 M Tris, 0.1 M NaH2PO4, pH 8.0) with gentle rocking for 1 h at room temperature. Cell debris was removed by centrifugation at 10 000g for 30 min and protein was bound to Profinity IMAC Ni-Charged Resin (Bio-Rad) at room temperature for 1 h. The solution was loaded onto a glass column, drained and washed twice with 8 M urea buffer (pH 6.3). Bound proteins were eluted with low pH urea buffer (pH 5.9 and pH 4.5). Proteins were analyzed for purity by SDS-PAGE and then dialyzed for 48 h with multiple buffer changes against PBS containing 0.1% SDS. Creation of Polyclonal JlpA Antiserum. Polyclonal serum was generated via primary and secondary intramuscular injections of antigen into New Zealand White rabbits at 2-week intervals. Recombinant JlpA (approximately 0.5 mg) was prepared for injection by mixing equal volumes of purified protein (approximately 300 µL) and Freund’s incomplete adjuvant (Sigma). Preimmune serum was collected and screened against JHH1 whole cell lysates to confirm no prior exposure to C. Journal of Proteome Research • Vol. 8, No. 10, 2009 4655
research articles jejuni. Serum was collected via cardiac bleeding. Reactivity to rJlpA was confirmed by Western blotting. Preparation of C. jejuni Protein Extracts. Two milligrams of freeze-dried bacteria was suspended in Laemmli loading buffer [24.8 mM Tris, 10 mM Glycerol, 0.5% (w/v) sodium dodecyl sulfate (SDS), 3.6 mM β-mercaptoethanol and 0.001% (w/v) of bromophenol blue (pH6.8)] and heated at 95 °C for 10 min. Insoluble material was removed by centrifugation at 20 000g, 4 °C for 15 min. The resulting supernatant was removed and used for SDS-PAGE. C. jejuni membrane protein-enriched fractions were isolated by the modified sodium carbonate precipitation method as outlined previously49 using a Bio-Rad ReadyPrep Protein Extraction Kit (Membrane II). The resulting supernatant was dried and stored at -20 °C until needed for 2-DE. Surface protein extractions were performed using the method of Logan and Trust.55 C. jejuni cultures were incubated on ice in glycine-HCl buffer [0.2 M glycine (pH 2.2)] and stirred gently for 15 min. The mixture was then centrifuged at 10 000g, 4 °C for 15 min, pH neutralized using NaOH and dialyzed against ultrapure water overnight using SnakeSkin pleated dialysis tubing with a 3500 Da molecular mass cutoff (Pierce, Rockford, IL). Dialyzed samples were then freeze-dried. Soybean Agglutinin Enrichment of C. jejuni Glycoproteins. Glycoprotein enrichment was performed according to Young et al.30 Freeze-dried glycine protein extracts were resuspended in Tris-buffered saline (TBS) [0.05 M Tris-HCl, 0.15 M NaCl (pH 7.5)] and centrifuged to remove insoluble material. Resuspended protein extracts were passed through lectin columns composed of 250 µL of soybean agglutinin (SBA) agarose slurry (Vector Laboratories, Burlingame, CA) in poly prep chromatography columns (Bio-Rad) previously equilibrated with 40 bed vol of TBS. The unbound fractions were collected and subjected to methanol protein precipitation,56 while the bound proteins were washed with 60 vol of TBS. Bound glycoproteins were eluted with 100 mM N-acetyl-D-galactosamine in TBS and dialyzed against Milli-Q water overnight using a Mini Dialysis Kit with a 1000 Da molecular mass cutoff (Amersham Biosciences, Buckinghamshire, U.K.) and then freeze-dried. SDS-PAGE. Freeze-dried and rJlpA samples were resuspended in Laemmli loading buffer and heated at 95 °C for 10 min, and insoluble material was removed by centrifugation at 20 000g, 4 °C for 15 min. Supernatants were loaded onto 16% polyacrylamide resolving gels with a 5% stacking gel and run in a Mini-PROTEAN 3 electrophoresis chamber. 2-DE. Freeze-dried proteins were resuspended in 2-DE sample buffer [5 M urea, 2 M thiourea, 0.1% carrier ampholytes, 2% (w/v) CHAPS, 2% (w/v) sulfobetaine 3-10, 2 mM tributylphosphine; Bio-Rad] and 250 µg was used to reswell precast 17 cm pH 4-7 immobilized pH gradient (IPG) strip gels (BioRad). Isoelectric focusing was performed using an IEF Cell (BioRad) for a total of 80 kVh. Proteins were reduced, alkylated and detergent-exchanged, and then separated on second-dimension SDS-PAGE gels as previously described.49 SDS-PAGE and 2-DE gels were fixed and double stained in Sypro Ruby and colloidal Coomassie Blue as described.49 Western Blotting. Proteins from unstained 2-DE gels were transferred to PVDF membrane using a Trans-Blot Semi-Dry Electrophoretic Transfer Cell (Bio-Rad) at 25 V for 2 h. Proteins from SDS-PAGE gels were transferred using a Criterion Wet Electrophoretic Transfer Cell (Bio-Rad) for 1 h at 400 mA. PVDF membranes were blocked overnight in 5% BSA (Sigma) and probed with either a 1/200 dilution of convalescent patient 4656
Journal of Proteome Research • Vol. 8, No. 10, 2009
Scott et al. serum derived 5 weeks post-C. jejuni infection or 1/1000 JlpAspecific rabbit antiserum. Proteins reactive to patient serum were detected using a 1/1000 dilution of goat-anti-human immunoglobulin (Millipore), followed by incubation in 3,3′diaminobenzidine tetrahydrochloride (DAB; Sigma). Proteins reacting against JlpA-specific rabbit antiserum were detected using a 1/1000 dilution of goat-anti-rabbit immunoglobulin (Millipore), followed by incubation in Supersignal West Pico Chemiluminescent substrate according to the manufacturer’s instructions (Pierce). The resulting blots were visualized using Hyperfilm ECL (GE LifeSciences, Amersham, U.K.) and a CP100 film processor (AGFA, Mortsel, Belgium). Images from stained 2-DE gels and Western blots were overlapped using the imaging program PD-Quest (Bio-Rad) to facilitate identification of antigenic proteins. Densitometry was also performed using PDQuest. Protein Identification by Mass Spectrometry. 2-DE isolated protein spots were processed as described.56 Briefly, spots were excised and destained [60:40 solution of 40 mM NH4HCO3 (pH 7.8)/100% acetonitrile] for 1 h at room temperature. The solution was removed and gel pieces were vacuum-dried for 1 h. Gel spots were rehydrated in 8 µL of trypsin solution [12 ng µL-1 of sequencing grade modified trypsin (Promega, Madison, WI) in 40 mM NH4HCO3] at 4 °C for 1 h. Excess trypsin was removed, and gel pieces were resuspended in 25 µL of 40 mM NH4HCO3 and incubated overnight at 37 °C. Peptides were concentrated and desalted using C18 Perfect Pure Tips (Eppendorf, Hamburg, Germany) and eluted directly onto a target plate with matrix (R-cyano-4-hydroxycinnamic acid) as described.49 Peptide mass maps were generated using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) and a Voyager DE-STR instrument (Applied Biosystems, Foster City, CA). Data from peptide mass maps were used to perform searches of the NCBI, SWISSPROT and TrEMBL databases, via the program MASCOT (www.matrixscience.com). JlpA peptide sequences were confirmed using MS-MS on an Applied Biosystems Q-STAR XL equipped with o-MALDI source. Following acquisition of peptide mapping data, individual parent ions were selected for MS-MS and the instrument switched into product ion mode. Collision energy was set between 60 and 100 depending on the m/z of the parent ion. Spectra were annotated in Analyst (Applied Biosystems) and searched against the NCBI database using the BLAST ‘search for short, nearly exact matches’ algorithm. Characterization of Glycopeptides from JlpA. Glycosylation sites were elucidated as described by Larsen et al.57 Tryptic peptides were further digested overnight using proteinase-K (400 ng in Milli-Q H2O, Sigma). Samples were then concentrated and desalted using Poros 20 R2 resin (Applied Biosystems) microcolumns, and the unbound flow-through applied to hand-held activated carbon/graphite microcolumns as described.58 Bound glycopeptides were eluted using 30% (v/v) acetonitrile and 0.2% (v/v) formic acid and dried. Glycopeptides were resuspended in 0.1% formic acid, loaded onto Econo 12 static tips (New Objective, Woburn, MA) and characterized by MS-MS on a Q-STAR XL instrument via electrospray ionization (ESI). Individual parent ions were selected for MS-MS and collision energy set between 60 and 100 depending on the m/z of the parent ion. Spectra were annotated in Analyst and manually matched to JlpA peptide sequences.
Glycosylation of the JlpA Antigen in Campylobacter jejuni
research articles
Figure 1. Antigenicity of recombinant JlpA (rJlpA). Left, total protein stain; right, Western blot against patient serum.
Results Sequence Prediction Identifies JlpA as a Possible Antigen. Previous work on the membrane subproteome of C. jejuni JHH1 suggested the presence of an unidentified antigen with molecular mass of 42-45 kDa and comigrating on 2-DE gels with the highly abundant MOMP.49 We interrogated the proteome of JHH1 to determine a list of candidate proteins using the following criteria: (i) pI between 4.7 and 5.2; (ii) predicted mass between 40 and 46 kDa; (iii) predicted solubility compatible with gel-based technology (
