- Email: [email protected]
52-kDa subunit protein
FEMS Immunology and Medical Microbiology 24 (1999) 221^225
Protection against Helicobacter pylori infection by intestinal immunisation with a 50/52-kDa subunit protein Margaret L. Dunkley *, Simon J. Harris, Richard J. McCoy, Melissa J. Musicka, Fiona M. Eyers, Leone G. Beagley, Paul J. Lumley, Kenneth W. Beagley, Robert L. Clancy The Australian Institute of Mucosal Immunology, Cortecs plc, P.O. Box 418, Newcastle, NSW 2300, Australia Received 25 November 1998 ; accepted 18 February 1999
Abstract A mouse model of Helicobacter pylori infection was used to evaluate the vaccine antigen potential of the citrate synthase homologue protein purified from the H. pylori NCTC 11637 strain. Mice were immunised with the protein by intra-Peyer's patch immunisation. This route gives maximal intestinal immunisation and was used to screen oral vaccine candidate antigens without the added complication of simultaneously testing oral delivery systems. Two weeks post-immunisation mice were infected with Sydney strain H. pylori and 4 weeks after infection the mice were killed and the level of H. pylori infection in the stomach determined. Pre-immunisation with the 50/52-kDa protein led to a 84^91% reduction in H. pylori infection compared to unimmunised controls. ß 1999 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Helicobacter pylori; Citrate synthase antigen; Helicobacter pylori vaccine; Animal model
1. Introduction Helicobacter pylori infection has been established as a predominant cause of gastric disease [1] and eradication of H. pylori infection has been linked to resolution of duodenal ulcer [2]. Eradication therapies which commonly employ a combination of proton pump inhibitors and antibiotics have shown considerable success in the past; however, this approach is becoming less successful due to the * Corresponding author. Tel.: +61 (249) 291885; Fax: +61 (249) 292591; E-mail: [email protected]
increasing antibiotic resistance of H. pylori strains. This has led to the necessity to develop e¡ective prophylactic and therapeutic vaccines against H. pylori to target `at risk' groups and infected individuals respectively. Both prophylactic and therapeutic immunisation have been demonstrated in animal infection models. Prophylactic oral immunisation was initially demonstrated against H. felis infection in mice [3,4] and later against H. pylori infection in mice. H. pylori lysates, urease heat shock protein HsPA, and the cytotoxin VacA have all been demonstrated as e¡ective vaccine antigens in mouse infection models, and other H. pylori proteins are currently being tested for e¤cacy as vaccine antigens.
0928-8244 / 99 / $20.00 ß 1999 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII: S 0 9 2 8 - 8 2 4 4 ( 9 9 ) 0 0 0 3 0 - 9
FEMSIM 1021 6-5-99
222
M.L. Dunkley et al. / FEMS Immunology and Medical Microbiology 24 (1999) 221^225
E¡ective therapy is considered to be complete eradication of the bacteria and an e¡ective vaccine must be able to accomplish this either alone or in conjunction with other treatments. It has been established that a combination of vaccination and antibiotic therapy can successfully eliminate an H. pylori infection in mice under conditions where immunisation alone is unsuccessful in completely eradicating the infection [5]. The aim of the present study was to evaluate the H. pylori 50/52-kDa citrate synthase homologue protein as a vaccine antigen using a mouse model of H. pylori infection. Immunisation was via direct intraPeyer's patch injection of vaccine to obtain a maximal intestinal immunisation.
2. Materials and methods 2.1. Preparation of the 50/52-kDa protein The protein was prepared from a sonicate supernatant of H. pylori strain NCTC 11637 using a modi¢cation of a procedure used for the preparation of a crude reactive antigen fraction that was developed for use in a diagnostic assay for detection of H. pylori infection in man [6]. The H. pylori was grown on chocolate agar plates at 37³C for 3 days then harvested, washed and resuspended in phosphatebu¡ered saline (PBS), pH 7.2. This preparation was subjected to sonication using a Sanyo Soniprep 150 ultrasonic disintegrator with a 9.5-mm probe. The sonic amplitude level was set at 6 Wm and the machine was operated using 25 cycles of 30 s on and 60 s o¡ regulated by an MSE process timer. The sonicated preparation was centrifuged at 10 000Ug for 10 min and the supernatant ¢ltered through 0.45and 0.22-Wm ¢lters. The sonicate supernatant was partially puri¢ed by anion-exchange FPLC on a Mono Q HR 10/10 column (Pharmacia Biotech Ltd, Uppsala, Sweden) using 0.05 M Tris bu¡er pH 8.2 and a two-step gradient of Tris bu¡er containing 0.24 M NaCl and 1.0 M NaCl. Fractions containing the 50/52-kDa protein were pooled, concentrated, and subjected to gel ¢ltration FPLC on a Superose 6 column (Pharmacia Biotech Ltd, Uppsala, Sweden). Elution was with a 0.05 M Tris bu¡er (pH 7.2). Fractions containing the 50/52-kDa protein
were pooled, and further puri¢cation was obtained by low-pressure liquid chromatography on DEAESepharose CL6B (Pharmacia Biotech Ltd, Uppsala, Sweden) and ceramic hydroxyapatite (Bio-Rad Laboratories, Sydney, Australia). This separation comprised loading the pooled fractions from the Superose 6 column onto the 2.5-ml DEAE-Sephadex CL6B column equilibrated with 0.05 M Tris bu¡er (pH 7.4), washing with this bu¡er, then eluting with sequential step gradients of 0.05 M Tris bu¡er (pH 7.4) supplemented with 0.05 and 0.1 M NaCl. The ¢nal eluted material was loaded onto a 2.5-ml column of ceramic hydroxyapatite equilibrated with 0.005 M sodium phosphate bu¡er then eluted with 0.025 M sodium phosphate (pH 7.4) followed by a linear gradient of 0.025^0.2 M sodium phosphate (pH 7.4). Fractions containing the puri¢ed subunit protein were pooled, dialysed against PBS (pH 7.2) and stored at 370³C until required. The protein concentration was determined using the BCA protein assay kit (Pierce, Rockford, IL, USA) and an assay for urease activity was performed as previously described [6]. The protein was analysed by discontinuous SDS-PAGE (5% stacking gel and 12% slab gel), under reducing and non-reducing conditions and by native PAGE (8^25% gel gradient). Puri¢ed 50/52kDa subunit protein was transferred to polyvinylidene di£uoride (PVDF) membrane (Bio-Rad, Sydney, Australia), stained with amido black (Sigma, St. Louis, MO, USA), then destained and subunit bands excised. N-terminal amino acid sequencing of these bands was performed by Newcastle Protein at The University of Newcastle, Newcastle, NSW, Australia. 2.2. Immunisation of mice Female, speci¢c pathogen-free C57BL/6 mice were obtained from the Central Animal House at the University of Newcastle, NSW, Australia. Animal experiments were performed with the approval of the Animal Care and Ethics Committee of The University of Newcastle and mice were housed ¢ve per cage in isolator cages. Mice were immunised by the intraPeyer's patch (IPP) route to test the e¤cacy of the antigen as a vaccine candidate as this immunisation route has been shown to give a maximal intestinal immunisation [7,8] and is therefore useful for screen-
FEMSIM 1021 6-5-99
M.L. Dunkley et al. / FEMS Immunology and Medical Microbiology 24 (1999) 221^225
ing proteins which can be used in oral vaccines. Antigen (at 0.5 mg protein ml31 ) was contained in an homogenate of equal quantities of PBS and Freund's incomplete adjuvant. For IPP immunisation each mouse was anaesthetised by intraperitoneal injection of 200 Wl of a ketamine (Parnell Laboratories, Australia), xylazine (Bayer) mixture made by mixing 10 ml of ketamine (100 Wg ml31 and 1 ml of xylazine (100 Wg ml31 , the abdomen shaved and swabbed with 70% alcohol and a midline incision made in the skin and muscle layers to expose the intestine. Visible Peyer's patches were located along the length of the intestine and approximately 3 Wl of homogenate injected directly under the serosa of each Peyer's patch. The muscle and skin layers were sutured and the mouse kept warm until recovery from anaesthesia. For each experiment, 10 mice were immunised and another 10 mice left untreated as the unimmunised controls. 2.3. Infection of mice with H. pylori Mice were infected 2 weeks after immunisation. H. pylori Sydney strain 1 (SS1) was obtained from Prof. A. Lee, The University of New South Wales, Sydney, Australia. This strain of H. pylori has been shown to successfully colonise the stomachs of C57BL/6 mice [9]. The H. pylori was grown on chocolate agar plates for 3 days in a microaerophilic 37³C incubator and harvested into PBS. The concentration of H. pylori was determined from the optical density reading at 405 nm and a regression curve relating optical density to H. pylori concentration. Mice were infected, by gavage, on three successive days with a 100 Wl volume containing approximately 108 H. pylori, and actual concentration of live H. pylori was determined by culture of serial 10-fold dilutions of the live H. pylori preparation on chocolate agar for 3 days. The actual dose of live H.
223
pylori was therefore calculated retrospectively. The doses on the three successive days were: experiment 1: 4.5U108 , 4.25U108 , 4.25U108 ; and experiment 2: 1.5U108 , 1.0U108 , 8.75U108 . This is a total of 109 for experiment 1 and 1.125U109 for experiment 2. 2.4. Sample collection Four weeks after infection the mice were killed by intraperitoneal pentobarbital overdose and the stomachs removed. The stomachs were cut in half longitudinally and one half was homogenised in 1 ml of PBS and aliquots of serial dilutions plated out on chocolate agar plates and cultured for 3 days. Colonies were counted to determine the number of colony-forming units (CFU) of H. pylori in the stomach of each mouse. The mean þ S.E.M. was calculated for each group. Only mice showing detectable infection were included in this calculation as the infection did not establish in all mice. This absence of infection occurred in mice in both the unimmunised and immunised groups.
3. Results 3.1. Puri¢cation of the 50/52-kDa protein from H. pylori NCTC 11637 Approximately 0.2% of the sonicate supernatant protein was recoverable as puri¢ed 50/52-kDa protein. Analysis by SDS-PAGE indicated that the protein puri¢ed from the NCTC H. pylori exists as a discrete 50-kDa subunit protein under reducing conditions (Fig. 1) and as a doublet with molecular masses of 50 and 52 kDa under non-reducing conditions (not shown). In the native form the protein exists as a hexamer with an approximate molecular mass of 300^340 kDa (not shown). N-terminal se-
Table 1 Recovery of live H. pylori SS1 from unimmunised mice and mice pre-immunised with H. pylori NCTC 11637-derived 50/52-kDa protein Experiment
Number infected
Mouse group
Live bacteria per stomach (104 CFU)
1
10/12 8/9 9/10 11/11
Unimmunised (10) Immunised (8) Unimmunised (9) Immunised (11)
36 þ 7 3.2 þ 1.7 40 þ 9 6.6 þ 3.6
2
FEMSIM 1021 6-5-99
H. pylori cleared (% of unimmunised group) 91 84
224
M.L. Dunkley et al. / FEMS Immunology and Medical Microbiology 24 (1999) 221^225
ni¢cant (P 6 0.05) bacterial clearance (91% reduction in CFU) was observed in the immunised group compared to the unimmunised group. In experiment 2, signi¢cant (P 6 0.01) bacterial clearance (84% reduction in CFU) was observed in the immunised group compared to the unimmunised group.
4. Discussion
Fig. 1. SDS-PAGE analysis of a 50/52-kDa subunit protein puri¢ed from H. pylori NCTC 11637. Samples were analysed by discontinuous SDS-PAGE (5% stacking gel and 12% slab gel) under reducing conditions. Lanes : A, molecular mass markers, (numbers refer to size in kDa) ; B: the puri¢ed protein.
quence data on the reduced 50-kDa subunit yielded the amino acid sequence SVTLINNENNERYYFET. Both the 50- and 52-kDa subunits have identical sequences up to the 10th residue. This protein was identi¢ed by BLAST analysis, using the SwissProt database, as the H. pylori citrate synthase homologue. There was no detectable urease activity in the puri¢ed protein preparation. 3.2. Recovery of live H. pylori SS1 from mice
The quest for an e¡ective H. pylori vaccine has resulted in the puri¢cation and testing of many H. pylori proteins for e¤cacy as vaccine antigens. The 50/52-kDa protein identi¢ed as the citrate synthase homologue is a likely vaccine antigen as it is present in signi¢cant quantities in the H. pylori strain NCTC 11637 and is easily puri¢ed. The protein was demonstrated to be a potent vaccine antigen when used to pre-immunise mice by IPP injection. This suggests that this protein will be an appropriate antigen for inclusion in a human oral vaccine using an appropriate delivery system. The IPP route of immunisation is useful for screening antigens as potential oral vaccine candidates as it is independent of the e¤cacy of various oral delivery systems. E¡ective antigens identi¢ed in this way can then be incorporated into various delivery systems. Soluble proteins are not usually immunogenic when delivered directly to the gut lumen as, unlike particulate antigens, they are not readily taken up by Peyer's patch M cells [10,11]. Appropriate delivery systems are necessary to deliver the antigen to the Peyer's patches which are the intestinal immune induction site. The 50/52kDa protein is currently being tested in various oral delivery systems.
Acknowledgements Ms. Sarah Cooper is thanked for assistance with H. pylori preparations.
References
The pooled data from two experiments in which mice were pre-immunised IPP with the 50/52-kDa protein are shown in Table 1. In experiment 1, sig-
[1] Marshall, B.J., Armstrong, J.A., McGechie, D.B. and Glancy, R.J. (1985) Attempts to ful¢ll Koch's postulates for pyloric campylobacter. Med. J. Aust. 142, 436^439.
FEMSIM 1021 6-5-99
M.L. Dunkley et al. / FEMS Immunology and Medical Microbiology 24 (1999) 221^225 [2] Hentschel, E., Brandstatter, G., Dragostics, B., Hirschl, A.M., Neme, H. and Schutzek, K. et al. (1993) E¡ect of ranitidine and amoxycillin plus metronidazole on the eradication of Helicobacter pylori and the recurrence of duodenal ulcer. New Engl. J. Med. 328, 308^312. [3] Chen, M., Lee, A. and Hazell, S. (1992) Immunization against gastric Helicobacter infection in a mouse Helicobacter felis model. Lancet 339, 1120^1121. [4] Czinn, S.J. and Nedrud, J.G. (1991) Oral immunisation against Helicobacter pylori. Infect. Immun. 59, 2359^2363. [5] Blanchard, T.G., Nedrud, J.G., Zagorski, S.J. and Czinn, S.J. (1997) Antimicrobial therapy as an adjunct to oral immunization signi¢cantly enhances the resolution of gastritis. Gut 41, (Suppl. 1) A63. [6] Cripps, A.W., McShane, L., Beagley, L.G., Clancy, R.L., Beagley, K.W., Dunkley, M.L., Smith, C.J., Cresswell, M. and Tyreman, D.R. (1997) Identi¢cation and characterization of an antigen preparation of Helicobacter pylori, containing novel proteins and its use in an ELISA assay for the detection of Helicobater pylori infection. In: Mucosal Solutions ^ Advances in Mucosal Immunology (Husband, A., Beagley, K.,
[7]
[8]
[9]
[10]
[11]
225
Clancy, R., Collins, A., Cripps, A. and Emery, D., Eds.), Vol. 2. pp. 127^143. University of Sydney Press, Sydney. Dunkley, M.L. and Husband, A.J. (1986) The induction and migration of antigen-speci¢c helper cells for IgA responses in the intestine. Immunology 57, 379^385. Cripps, A.W., Dunkley, M.L. and Clancy, R.L. (1994) Mucosal and systemic immunizations with killed Pseudomonas aeruginosa protect against acute respiratory infection in rats. Infect. Immun. 62, 1427^1436. Lee, A., O'Rourke, J., Ungria, M.C.D., Robertson, B., Daskalopoulos, G. and Dixon, M.F. (1997) A standardized mouse model of Helicobacter pylori infection : Introducing the Sydney strain. Gastroenterology 112, 1386^1397. O' Hagen, D.T., Palin, K.J., Davis, S.S. and Cho, M.J. (1987) Intestinal absorption of proteins and macromolecules and the immunological response. CRC Crit. Rev. Ther. Drug Carrier Syst. 4, 197^220. Childers, N.K., Denys, F.R., McGee, N.F. and Michalek, S.M. (1990) Ultrastructural study of liposome uptake by M cells of rat Peyer's patch : An oral vaccine system for delivery of puri¢ed antigen. Reg. Immunol. 3, 8^16.
FEMSIM 1021 6-5-99
Protection against Helicobacter pylori infection by intestinal immunisation with a 50/52-kDa subunit protein Margaret L. Dunkley *, Simon J. Harris, Richard J. McCoy, Melissa J. Musicka, Fiona M. Eyers, Leone G. Beagley, Paul J. Lumley, Kenneth W. Beagley, Robert L. Clancy The Australian Institute of Mucosal Immunology, Cortecs plc, P.O. Box 418, Newcastle, NSW 2300, Australia Received 25 November 1998 ; accepted 18 February 1999
Abstract A mouse model of Helicobacter pylori infection was used to evaluate the vaccine antigen potential of the citrate synthase homologue protein purified from the H. pylori NCTC 11637 strain. Mice were immunised with the protein by intra-Peyer's patch immunisation. This route gives maximal intestinal immunisation and was used to screen oral vaccine candidate antigens without the added complication of simultaneously testing oral delivery systems. Two weeks post-immunisation mice were infected with Sydney strain H. pylori and 4 weeks after infection the mice were killed and the level of H. pylori infection in the stomach determined. Pre-immunisation with the 50/52-kDa protein led to a 84^91% reduction in H. pylori infection compared to unimmunised controls. ß 1999 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Helicobacter pylori; Citrate synthase antigen; Helicobacter pylori vaccine; Animal model
1. Introduction Helicobacter pylori infection has been established as a predominant cause of gastric disease [1] and eradication of H. pylori infection has been linked to resolution of duodenal ulcer [2]. Eradication therapies which commonly employ a combination of proton pump inhibitors and antibiotics have shown considerable success in the past; however, this approach is becoming less successful due to the * Corresponding author. Tel.: +61 (249) 291885; Fax: +61 (249) 292591; E-mail: [email protected]
increasing antibiotic resistance of H. pylori strains. This has led to the necessity to develop e¡ective prophylactic and therapeutic vaccines against H. pylori to target `at risk' groups and infected individuals respectively. Both prophylactic and therapeutic immunisation have been demonstrated in animal infection models. Prophylactic oral immunisation was initially demonstrated against H. felis infection in mice [3,4] and later against H. pylori infection in mice. H. pylori lysates, urease heat shock protein HsPA, and the cytotoxin VacA have all been demonstrated as e¡ective vaccine antigens in mouse infection models, and other H. pylori proteins are currently being tested for e¤cacy as vaccine antigens.
0928-8244 / 99 / $20.00 ß 1999 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII: S 0 9 2 8 - 8 2 4 4 ( 9 9 ) 0 0 0 3 0 - 9
FEMSIM 1021 6-5-99
222
M.L. Dunkley et al. / FEMS Immunology and Medical Microbiology 24 (1999) 221^225
E¡ective therapy is considered to be complete eradication of the bacteria and an e¡ective vaccine must be able to accomplish this either alone or in conjunction with other treatments. It has been established that a combination of vaccination and antibiotic therapy can successfully eliminate an H. pylori infection in mice under conditions where immunisation alone is unsuccessful in completely eradicating the infection [5]. The aim of the present study was to evaluate the H. pylori 50/52-kDa citrate synthase homologue protein as a vaccine antigen using a mouse model of H. pylori infection. Immunisation was via direct intraPeyer's patch injection of vaccine to obtain a maximal intestinal immunisation.
2. Materials and methods 2.1. Preparation of the 50/52-kDa protein The protein was prepared from a sonicate supernatant of H. pylori strain NCTC 11637 using a modi¢cation of a procedure used for the preparation of a crude reactive antigen fraction that was developed for use in a diagnostic assay for detection of H. pylori infection in man [6]. The H. pylori was grown on chocolate agar plates at 37³C for 3 days then harvested, washed and resuspended in phosphatebu¡ered saline (PBS), pH 7.2. This preparation was subjected to sonication using a Sanyo Soniprep 150 ultrasonic disintegrator with a 9.5-mm probe. The sonic amplitude level was set at 6 Wm and the machine was operated using 25 cycles of 30 s on and 60 s o¡ regulated by an MSE process timer. The sonicated preparation was centrifuged at 10 000Ug for 10 min and the supernatant ¢ltered through 0.45and 0.22-Wm ¢lters. The sonicate supernatant was partially puri¢ed by anion-exchange FPLC on a Mono Q HR 10/10 column (Pharmacia Biotech Ltd, Uppsala, Sweden) using 0.05 M Tris bu¡er pH 8.2 and a two-step gradient of Tris bu¡er containing 0.24 M NaCl and 1.0 M NaCl. Fractions containing the 50/52-kDa protein were pooled, concentrated, and subjected to gel ¢ltration FPLC on a Superose 6 column (Pharmacia Biotech Ltd, Uppsala, Sweden). Elution was with a 0.05 M Tris bu¡er (pH 7.2). Fractions containing the 50/52-kDa protein
were pooled, and further puri¢cation was obtained by low-pressure liquid chromatography on DEAESepharose CL6B (Pharmacia Biotech Ltd, Uppsala, Sweden) and ceramic hydroxyapatite (Bio-Rad Laboratories, Sydney, Australia). This separation comprised loading the pooled fractions from the Superose 6 column onto the 2.5-ml DEAE-Sephadex CL6B column equilibrated with 0.05 M Tris bu¡er (pH 7.4), washing with this bu¡er, then eluting with sequential step gradients of 0.05 M Tris bu¡er (pH 7.4) supplemented with 0.05 and 0.1 M NaCl. The ¢nal eluted material was loaded onto a 2.5-ml column of ceramic hydroxyapatite equilibrated with 0.005 M sodium phosphate bu¡er then eluted with 0.025 M sodium phosphate (pH 7.4) followed by a linear gradient of 0.025^0.2 M sodium phosphate (pH 7.4). Fractions containing the puri¢ed subunit protein were pooled, dialysed against PBS (pH 7.2) and stored at 370³C until required. The protein concentration was determined using the BCA protein assay kit (Pierce, Rockford, IL, USA) and an assay for urease activity was performed as previously described [6]. The protein was analysed by discontinuous SDS-PAGE (5% stacking gel and 12% slab gel), under reducing and non-reducing conditions and by native PAGE (8^25% gel gradient). Puri¢ed 50/52kDa subunit protein was transferred to polyvinylidene di£uoride (PVDF) membrane (Bio-Rad, Sydney, Australia), stained with amido black (Sigma, St. Louis, MO, USA), then destained and subunit bands excised. N-terminal amino acid sequencing of these bands was performed by Newcastle Protein at The University of Newcastle, Newcastle, NSW, Australia. 2.2. Immunisation of mice Female, speci¢c pathogen-free C57BL/6 mice were obtained from the Central Animal House at the University of Newcastle, NSW, Australia. Animal experiments were performed with the approval of the Animal Care and Ethics Committee of The University of Newcastle and mice were housed ¢ve per cage in isolator cages. Mice were immunised by the intraPeyer's patch (IPP) route to test the e¤cacy of the antigen as a vaccine candidate as this immunisation route has been shown to give a maximal intestinal immunisation [7,8] and is therefore useful for screen-
FEMSIM 1021 6-5-99
M.L. Dunkley et al. / FEMS Immunology and Medical Microbiology 24 (1999) 221^225
ing proteins which can be used in oral vaccines. Antigen (at 0.5 mg protein ml31 ) was contained in an homogenate of equal quantities of PBS and Freund's incomplete adjuvant. For IPP immunisation each mouse was anaesthetised by intraperitoneal injection of 200 Wl of a ketamine (Parnell Laboratories, Australia), xylazine (Bayer) mixture made by mixing 10 ml of ketamine (100 Wg ml31 and 1 ml of xylazine (100 Wg ml31 , the abdomen shaved and swabbed with 70% alcohol and a midline incision made in the skin and muscle layers to expose the intestine. Visible Peyer's patches were located along the length of the intestine and approximately 3 Wl of homogenate injected directly under the serosa of each Peyer's patch. The muscle and skin layers were sutured and the mouse kept warm until recovery from anaesthesia. For each experiment, 10 mice were immunised and another 10 mice left untreated as the unimmunised controls. 2.3. Infection of mice with H. pylori Mice were infected 2 weeks after immunisation. H. pylori Sydney strain 1 (SS1) was obtained from Prof. A. Lee, The University of New South Wales, Sydney, Australia. This strain of H. pylori has been shown to successfully colonise the stomachs of C57BL/6 mice [9]. The H. pylori was grown on chocolate agar plates for 3 days in a microaerophilic 37³C incubator and harvested into PBS. The concentration of H. pylori was determined from the optical density reading at 405 nm and a regression curve relating optical density to H. pylori concentration. Mice were infected, by gavage, on three successive days with a 100 Wl volume containing approximately 108 H. pylori, and actual concentration of live H. pylori was determined by culture of serial 10-fold dilutions of the live H. pylori preparation on chocolate agar for 3 days. The actual dose of live H.
223
pylori was therefore calculated retrospectively. The doses on the three successive days were: experiment 1: 4.5U108 , 4.25U108 , 4.25U108 ; and experiment 2: 1.5U108 , 1.0U108 , 8.75U108 . This is a total of 109 for experiment 1 and 1.125U109 for experiment 2. 2.4. Sample collection Four weeks after infection the mice were killed by intraperitoneal pentobarbital overdose and the stomachs removed. The stomachs were cut in half longitudinally and one half was homogenised in 1 ml of PBS and aliquots of serial dilutions plated out on chocolate agar plates and cultured for 3 days. Colonies were counted to determine the number of colony-forming units (CFU) of H. pylori in the stomach of each mouse. The mean þ S.E.M. was calculated for each group. Only mice showing detectable infection were included in this calculation as the infection did not establish in all mice. This absence of infection occurred in mice in both the unimmunised and immunised groups.
3. Results 3.1. Puri¢cation of the 50/52-kDa protein from H. pylori NCTC 11637 Approximately 0.2% of the sonicate supernatant protein was recoverable as puri¢ed 50/52-kDa protein. Analysis by SDS-PAGE indicated that the protein puri¢ed from the NCTC H. pylori exists as a discrete 50-kDa subunit protein under reducing conditions (Fig. 1) and as a doublet with molecular masses of 50 and 52 kDa under non-reducing conditions (not shown). In the native form the protein exists as a hexamer with an approximate molecular mass of 300^340 kDa (not shown). N-terminal se-
Table 1 Recovery of live H. pylori SS1 from unimmunised mice and mice pre-immunised with H. pylori NCTC 11637-derived 50/52-kDa protein Experiment
Number infected
Mouse group
Live bacteria per stomach (104 CFU)
1
10/12 8/9 9/10 11/11
Unimmunised (10) Immunised (8) Unimmunised (9) Immunised (11)
36 þ 7 3.2 þ 1.7 40 þ 9 6.6 þ 3.6
2
FEMSIM 1021 6-5-99
H. pylori cleared (% of unimmunised group) 91 84
224
M.L. Dunkley et al. / FEMS Immunology and Medical Microbiology 24 (1999) 221^225
ni¢cant (P 6 0.05) bacterial clearance (91% reduction in CFU) was observed in the immunised group compared to the unimmunised group. In experiment 2, signi¢cant (P 6 0.01) bacterial clearance (84% reduction in CFU) was observed in the immunised group compared to the unimmunised group.
4. Discussion
Fig. 1. SDS-PAGE analysis of a 50/52-kDa subunit protein puri¢ed from H. pylori NCTC 11637. Samples were analysed by discontinuous SDS-PAGE (5% stacking gel and 12% slab gel) under reducing conditions. Lanes : A, molecular mass markers, (numbers refer to size in kDa) ; B: the puri¢ed protein.
quence data on the reduced 50-kDa subunit yielded the amino acid sequence SVTLINNENNERYYFET. Both the 50- and 52-kDa subunits have identical sequences up to the 10th residue. This protein was identi¢ed by BLAST analysis, using the SwissProt database, as the H. pylori citrate synthase homologue. There was no detectable urease activity in the puri¢ed protein preparation. 3.2. Recovery of live H. pylori SS1 from mice
The quest for an e¡ective H. pylori vaccine has resulted in the puri¢cation and testing of many H. pylori proteins for e¤cacy as vaccine antigens. The 50/52-kDa protein identi¢ed as the citrate synthase homologue is a likely vaccine antigen as it is present in signi¢cant quantities in the H. pylori strain NCTC 11637 and is easily puri¢ed. The protein was demonstrated to be a potent vaccine antigen when used to pre-immunise mice by IPP injection. This suggests that this protein will be an appropriate antigen for inclusion in a human oral vaccine using an appropriate delivery system. The IPP route of immunisation is useful for screening antigens as potential oral vaccine candidates as it is independent of the e¤cacy of various oral delivery systems. E¡ective antigens identi¢ed in this way can then be incorporated into various delivery systems. Soluble proteins are not usually immunogenic when delivered directly to the gut lumen as, unlike particulate antigens, they are not readily taken up by Peyer's patch M cells [10,11]. Appropriate delivery systems are necessary to deliver the antigen to the Peyer's patches which are the intestinal immune induction site. The 50/52kDa protein is currently being tested in various oral delivery systems.
Acknowledgements Ms. Sarah Cooper is thanked for assistance with H. pylori preparations.
References
The pooled data from two experiments in which mice were pre-immunised IPP with the 50/52-kDa protein are shown in Table 1. In experiment 1, sig-
[1] Marshall, B.J., Armstrong, J.A., McGechie, D.B. and Glancy, R.J. (1985) Attempts to ful¢ll Koch's postulates for pyloric campylobacter. Med. J. Aust. 142, 436^439.
FEMSIM 1021 6-5-99
M.L. Dunkley et al. / FEMS Immunology and Medical Microbiology 24 (1999) 221^225 [2] Hentschel, E., Brandstatter, G., Dragostics, B., Hirschl, A.M., Neme, H. and Schutzek, K. et al. (1993) E¡ect of ranitidine and amoxycillin plus metronidazole on the eradication of Helicobacter pylori and the recurrence of duodenal ulcer. New Engl. J. Med. 328, 308^312. [3] Chen, M., Lee, A. and Hazell, S. (1992) Immunization against gastric Helicobacter infection in a mouse Helicobacter felis model. Lancet 339, 1120^1121. [4] Czinn, S.J. and Nedrud, J.G. (1991) Oral immunisation against Helicobacter pylori. Infect. Immun. 59, 2359^2363. [5] Blanchard, T.G., Nedrud, J.G., Zagorski, S.J. and Czinn, S.J. (1997) Antimicrobial therapy as an adjunct to oral immunization signi¢cantly enhances the resolution of gastritis. Gut 41, (Suppl. 1) A63. [6] Cripps, A.W., McShane, L., Beagley, L.G., Clancy, R.L., Beagley, K.W., Dunkley, M.L., Smith, C.J., Cresswell, M. and Tyreman, D.R. (1997) Identi¢cation and characterization of an antigen preparation of Helicobacter pylori, containing novel proteins and its use in an ELISA assay for the detection of Helicobater pylori infection. In: Mucosal Solutions ^ Advances in Mucosal Immunology (Husband, A., Beagley, K.,
[7]
[8]
[9]
[10]
[11]
225
Clancy, R., Collins, A., Cripps, A. and Emery, D., Eds.), Vol. 2. pp. 127^143. University of Sydney Press, Sydney. Dunkley, M.L. and Husband, A.J. (1986) The induction and migration of antigen-speci¢c helper cells for IgA responses in the intestine. Immunology 57, 379^385. Cripps, A.W., Dunkley, M.L. and Clancy, R.L. (1994) Mucosal and systemic immunizations with killed Pseudomonas aeruginosa protect against acute respiratory infection in rats. Infect. Immun. 62, 1427^1436. Lee, A., O'Rourke, J., Ungria, M.C.D., Robertson, B., Daskalopoulos, G. and Dixon, M.F. (1997) A standardized mouse model of Helicobacter pylori infection : Introducing the Sydney strain. Gastroenterology 112, 1386^1397. O' Hagen, D.T., Palin, K.J., Davis, S.S. and Cho, M.J. (1987) Intestinal absorption of proteins and macromolecules and the immunological response. CRC Crit. Rev. Ther. Drug Carrier Syst. 4, 197^220. Childers, N.K., Denys, F.R., McGee, N.F. and Michalek, S.M. (1990) Ultrastructural study of liposome uptake by M cells of rat Peyer's patch : An oral vaccine system for delivery of puri¢ed antigen. Reg. Immunol. 3, 8^16.
FEMSIM 1021 6-5-99