INFECTION AND IMMUNITY, Nov. 2001, p. 6981–6986 0019-9567/01/$04.00⫹0 DOI: 10.1128/IAI.69.11.6981–6986.2001 Copyright © 2001, American Society for Microbiology. All Rights Reserved.
Vol. 69, No. 11
Reciprocal Protective Immunity against Bordetella pertussis and Bordetella parapertussis in a Murine Model of Respiratory Infection MINEO WATANABE1*
AND
MASAAKI NAGAI2
Department of Microbiology and Biologics, Daiichi College of Pharmaceutical Sciences, Fukuoka 815-8511,1 and Research Center for Biologicals, The Kitasato Institute, Kitamoto 364-0026,2 Japan Received 23 April 2001/Returned for modification 25 June 2001/Accepted 20 August 2001
The protective immunity induced by infection with Bordetella pertussis and with Bordetella parapertussis was examined in a murine model of respiratory infection. Convalescent mice that had been infected by aerosol with B. pertussis or with B. parapertussis exhibited a protective immune response against B. pertussis and also against B. parapertussis. Anti-filamentous hemagglutinin (anti-FHA) serum immunoglobulin G (IgG) and anti-FHA lung IgA antibodies were detected in both mice infected with B. pertussis and those infected with B. parapertussis. Antibodies against pertussis toxin (anti-PT) and against killed B. pertussis cells were detected in mice infected with B. pertussis. Pertactin-specific antibodies and antibodies against killed B. parapertussis cells were detected in mice infected with B. parapertussis. Spleen cells from mice infected with B. pertussis secreted interferon-␥ (IFN-␥) in response to stimulation by FHA or PT. Spleen cells from mice infected with B. parapertussis also secreted IFN-␥ in response to FHA. Interleukin-4 was not produced in response to any of the antigens tested. The profiles of cytokine secretion in vitro revealed induction of a Th1-biased immune response during convalescence from infection by B. pertussis and by B. parapertussis. It is possible that Th1 and Th2 responses against FHA might be related to the reciprocal protection achieved in our murine model. during convalescence after infection with B. pertussis (20). We postulated that immunization by natural infection of the two species might clarify the relationship between protection against B. pertussis and protection against B. parapertussis. To test our hypothesis, we infected mice by exposing them to an aerosol of B. pertussis or B. parapertussis. After mice had recovered, convalescent mice were investigated for protective responses against the two species of Bordetella, for levels of antigen-specific antibodies, and for splenocyte proliferation and cytokine secretion responses in vitro after stimulation by antigens.
Whooping cough caused by Bordetella pertussis is a serious disease in children. Commercial pertussis vaccines, which consist of killed B. pertussis cells or derived antigens, are very effective and have reduced the incidence of whooping cough very considerably. However, in addition to B. pertussis, Bordetella parapertussis also causes symptoms typical of whooping cough (22). The illness caused by B. parapertussis is sometimes as severe as that caused by B. pertussis (10). Outbreaks of infection by B. parapertussis have been reported in several countries (8, 11, 18). B. parapertussis is closely related to B. pertussis in terms of virulence and attachment factors, such as filamentous hemagglutinin (FHA), adenylate cyclase toxin, heat-labile toxin, and pertactin (PRN) (29). However, several reports suggest that pertussis vaccine has no or limited ability to protect against B. parapertussis (9, 13, 15, 27, 32). Stehr et al. reported that the efficacy of the acellular pertussis component diphtheria-tetanus-pertussis (DTP) vaccine and the whole-cell pertussis component DTP vaccine in children was 31% and ⫺6%, respectively (27). Khelef et al. suggested that immunization with antigens derived from B. pertussis induce no protection against B. parapertussis in mice (13). These reports suggested that reciprocal protective immunity between the two species might not be induced. However, in these studies, subcutaneous or peritoneal injections were commonly used as methods of immunization. Mills et al. suggested that there might be a difference, in terms of the profiles of the protective immune response against B. pertussis, between the response after immunization by injection with vaccines and the response
MATERIALS AND METHODS Mice. Specific-pathogen-free female dd-Y mice were obtained from Japan SLC (Hamamatsu, Japan). All mice were 3.5 weeks old at the start of experiments. Bacterial strains and culture conditions. The phase I strain of B. pertussis strain 18-323 and B. parapertussis strain 23054 were used in this study. Cells were grown on Bordet-Gengou (BG) agar supplemented with 20% (vol/vol) defibrinated horse blood at 37°C. Bacterial antigens. Killed whole-cell B. pertussis or B. parapertussis antigens were prepared as described below. B. pertussis or B. parapertussis was cultured on BG plates for 30 h at 37°C. Cells were harvested in phosphate-buffered saline (PBS) on ice, and suspensions of cells were adjusted to 1010 cells/ml after measurement of the optical density at 660 nm (OD660) of the suspension. The bacterial suspension was supplemented with formalin to a final concentration of 0.2 M. After incubation for 1 h at 37°C, the suspension of formalin-killed whole cells was supplemented with 0.2 M lysine and then it was dialyzed against PBS for 2 days at 4°C. FHA and pertussis toxin (PT) were purified from the culture supernatant of B. pertussis by modified versions of the methods of Menozzi et al., Chong and Klein, and Sekura et al. (5, 17, 26, 30). PRN was purified from a heated extract of B. pertussis cells by a modified version of the method of Gould-Kostka et al. (7). Purified FHA, PT, and PRN were examined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis using a modified version of Laemmli’s method (14). No contaminants were detected in each purified preparation (data not shown). Detoxified pertussis toxin (PTd) was prepared as described previously (31).
* Corresponding author. Mailing address: Department of Microbiology and Biologics, Daiichi College of Pharmaceutical Sciences, 22-1 Tamagawa-cho, Minami-ku, Fukuoka 815-8511, Japan. Phone: 81-92541-0161. Fax: 81-92-553-5698. E-mail:
[email protected]. 6981
6982
WATANABE AND NAGAI
Aerosol infection. Infection by aerosols of B. pertussis or B. parapertussis was achieved by a modified version of the method of Oda et al. (21, 30, 31). B. pertussis or B. parapertussis was cultured on BG plates for 30 h at 37°C. The bacteria were then harvested in PBS on ice and then each suspension of bacteria was adjusted to 1010 cells/ml after measurement of the OD660. Mice were allowed to inhale the suspension for 45 min in a sealed aerosol chamber within a biosafety cabinet (MHE-130B1; Sanyo Electric, Moriguchi, Japan). The number of viable Bordetella cells in each mouse lung after such treatment was approximately 105 CFU. Quantitation of bacteria in lungs. After sacrifice, the lungs of mice were dissected and homogenized in 10 ml of PBS per lung in a Teflon homogenizer on ice. After dilution of each lung homogenate, it was spread on BG plates and incubated for 4 days at 37°C. The number of CFU was used to estimate the number of viable bacteria. The limit of detection was 102 CFU/lung by this method (31). Assay of protective immunity. Protective immunity was determined as described previously (30, 31). Convalescent mice, which were maintained in individual cages for 6 weeks after primary infection with an aerosol of B. pertussis or B. parapertussis, were infected via the respiratory tract by an aerosol of B. pertussis or B. parapertussis. Two weeks after the second infection, the lungs of each mouse were surgically removed and homogenized in PBS in a Teflon homogenizer on ice. The number of CFU was measured as described above. The significance of the difference between the nonimmunized control group and each immunized (convalescent) group was examined by Student’s t test. Probability values of ⬍0.05 were considered evidence of statistical significance (30, 31). Quantitation of antibodies by enzyme-linked immunosorbent assays (ELISAs). Sera and lungs of convalescent mice which had been maintained for 6 weeks after the first infection with B. pertussis or B. parapertussis were obtained for determinations of levels of antigen-specific serum immunoglobulin G (IgG) and lung IgA antibodies. Lungs were homogenized in 10 ml of PBS per lung that contained 0.1 mM phenylmethylsulfonyl fluoride and 8% (vol/vol) fetal calf serum in a Teflon homogenizer on ice. Homogenates were centrifuged (25,000 ⫻ g, 30 min, 4°C), and supernatants were used for determinations of levels of antigen-specific IgA antibodies in lungs (24). FHA- and PT-specific antibodies were quantitated by ELISAs, as described previously (30, 31). Levels of antibodies were expressed in terms of mean absorbance at 492 nm (OD492). Levels of PRN-specific antibodies were determined by a modified version of the ELISA method of Manghi et al. (16) and were expressed as mean absorbance values at 405 nm (OD405). Assay of FHA-neutralizing antibodies. Samples of serum from mice infected with B. pertussis and mice infected with B. parapertussis were obtained and diluted serially with PBS. Aliquots of 50 l of each dilution were mixed with a solution of purified FHA that contained 16 hemagglutinating units of purified FHA. After incubation for 1 h at 37°C, each sample was supplemented with 50 l of a suspension of chicken erythrocytes (0.5% [vol/vol]). After mixing and incubation for 2 h at room temperature, the maximum dilution at which hemagglutination was inhibited was taken as the titer of FHA-neutralizing antibodies. Agglutination test. Levels of serum antibodies against B. pertussis cells or B. parapertussis cells were determined by an agglutination test (25). Samples of serum from convalescent mice were serially diluted in 96-well round-bottom plates (50 l/well). Then 50 l of the suspension of killed whole cells (1010 cells/ml) was added to each well. After mixing and incubation for 2 days at 4°C, the maximum dilution that induced agglutination was taken as the titer of specific antibodies. Proliferation of spleen cells. Proliferation of spleen cells upon stimulation by B. pertussis or B. parapertussis antigens was examined by a modified version of the method of Ahmed et al. (1). Spleens were surgically and aseptically removed from convalescent mice. Single-cell suspensions (106 cells/ml) of spleen cells were prepared in RPMI 1640 medium supplemented with 10% (vol/vol) fetal calf serum, 50 g of gentamicin (Life Technologies, Rockville, Md.)/ml, and 10 g of polymyxin B (Sigma, St. Louis, Mo.)/ml. In order to neutralize the cytokineinducing activity of the endotoxin in the preparation of whole-cell antigens, the medium was supplemented with polymyxin B (3, 6). The suspensions of spleen cells were supplemented with FHA (5 g/ml), PTd (5 g/ml), PRN (5 g/ml), and killed whole B. pertussis cells (106 cells/ml) or killed whole B. parapertussis cells (106 cells/ml). Then, 200-l aliquots of the mixture that contained spleen cells and antigens were placed in wells of a flat-bottom 96-well plate. Control wells contained the spleen cells but no antigens for stimulation. Blank wells contained medium only (no spleen cells and no antigens for stimulation). After incubation of the plate in a CO2 incubator (5% CO2 in air) for 24 h at 37°C, 20 l of Alamar Blue (AccuMed International, Chicago, Ill.) was added to each well
INFECT. IMMUN.
FIG. 1. Time course of numbers of CFU in lungs of mice after infection with B. pertussis or B. parapertussis. Mice were infected by exposure to an aerosol of B. pertussis (BP) or of B. parapertussis (BPP) as described in Materials and Methods. At 0, 1, 2, 4, 5, or 6 weeks after infection, viable bacteria in mouse lungs were counted. Results are presented in terms of CFU. The results shown are mean values per lung, as estimated from individual lungs of five mice for each group at each time point. Each symbol with a vertical line represents a mean ⫾ standard error.
and the plate was incubated for 48 h under the same conditions. The OD570 and OD600 of the contents of each well were measured with a microplate reader (MTP-120; Corona Electric, Hitachinaka, Japan). A proliferation index (PI) was calculated from the following formula: PI ⫽ [(OD570 ⫺ OD600)sample ⫺ (OD570 ⫺ OD600)blank]/[(OD570 ⫺ OD600)control ⫺ (OD570 ⫺ OD600)blank]. Secretion of cytokines by spleen cells in vitro. Spleen cells of convalescent mice were tested in vitro for secretion of cytokines in response to B. pertussis or B. parapertussis antigens by a modified version of the method of Redhead et al. (23). Single-cell suspensions of spleen cells (2.0 ⫻ 106 cells/ml) were incubated with FHA (5 g/ml), PTd (5 g/ml), PRN (5 g/ml), and killed whole B. pertussis cells (106 cells/ml) or killed whole B. parapertussis cells (106 cells/ml) at 37°C for 72 h in an atmosphere of 5% CO2 in air. Culture supernatants were obtained by centrifugation (450 ⫻ g, 5 min, 4°C) and stored at ⫺80°C prior to assays. Mouse gamma interferon (IFN-␥) and mouse interleukin-4 (IL-4) were quantitated by ELISAs. A commercial Cytoscreen Immunoassay Kit (BioSource International Inc., Camarillo, Calif.) was used according to the manufacturer’s recommendations. Quantitation of protein. Proteins were quantitated by Bradford’s method (4) with egg albumin (Sigma) as the standard protein. Statistical analysis. The statistical significance of differences between results from different groups was examined by Student’s t test.
RESULTS Time course of numbers of bacterial cells in mouse lungs after primary infection with B. pertussis or with B. parapertussis. Mice were infected with B. pertussis or with B. parapertussis as described in Materials and Methods. Initial counts of viable bacteria in lungs of mice infected with B. pertussis and with B. parapertussis were 104.8 and 105.3 CFU/lung, respectively (Fig. 1). There was no significant difference between these values (P ⬎ 0.05). Numbers of bacteria increased for 1 week after infection and then declined slowly. The mice recovered within 6 weeks of infection, at which time no bacteria were detected in their lungs (Fig. 1). These mice were taken as the convalescent groups for this study. There were no significant differences in terms of CFU in lungs at each time point between mice infected with B. pertussis and those infected with B. parapertussis. Protective effects of previous infection against B. pertussis. Mice infected with B. pertussis or mice infected with B. para-
VOL. 69, 2001
RECIPROCAL IMMUNITY AGAINST TWO SPECIES OF BORDETELLA
FIG. 2. Protection of convalescent mice against B. pertussis or B. parapertussis. Control mice, mice infected with B. pertussis (BP), and mice infected with B. parapertussis (BPP) were challenged by exposure to an aerosol of B. pertussis (A) or B. parapertussis (B). Two weeks after the challenge, numbers of B. pertussis cells or B. parapertussis cells in each mouse lung were counted. The results shown are mean values per lung, as estimated from individual lungs of five mice for each group. Each column and vertical line represent a mean and standard error. ⴱ, P ⬍ 0.05 versus the control group.
pertussis were challenged by an aerosol of B. pertussis or B. parapertussis. Two weeks after the challenge, the number of CFU in the lungs of each mouse was measured as described in Materials and Methods. The number of CFU in the lungs of control mice was approximately 106.5 2 weeks after aerosol infection with B. pertussis (Fig. 2A). In the case of mice infected with B. pertussis, the number of CFU in the lungs was approximately 102.1, which was much more than 10,000-fold lower than that in lungs of control mice. The number of CFU in lungs of mice infected with B. parapertussis was approximately 102.6. There were significant differences between the results for the control group and each convalescent group (P ⬍ 0.05). The results demonstrated that protection against B. pertussis was established during convalescence from infection by B. pertussis or by B. parapertussis. Protective effects of previous infection against B. parapertussis. Since we had found that mice that had recovered from infection not only with B. pertussis but also with B. parapertussis exhibited protective immunity against B. pertussis, we next examined protection against B. parapertussis. We found that protective immunity against B. parapertussis was induced in mice infected with B. pertussis and in those infected with B. parapertussis (Fig. 2B). The numbers of CFU in lungs of control mice, mice infected with B. pertussis, and mice infected with B. parapertussis were 104.7, 102.1, and 102.2/lung, respectively. There were significant differences between the results for the control group and those for each convalescent group (P ⬍ 0.05). Levels of antibodies in convalescent mice. We examined the antibody responses against B. pertussis and B. parapertussis
6983
FIG. 3. Antigen-specific serum IgG responses in mice infected with B. pertussis and mice infected with B. parapertussis. Sera of control mice, mice infected with B. pertussis (BP), and mice infected with B. parapertussis (BPP) were examined by ELISAs. Levels of antibodies were expressed as mean values of absorbance at the indicated wavelength. (A) FHA-specific serum IgG. (B) PT-specific serum IgG. (C) PRN-specific serum IgG. Each column with a vertical line represents a mean and standard error of results from individual samples of serum from five mice for each group. ⴱ, P ⬍ 0.05 versus the control group.
antigens and found FHA-specific serum IgG antibodies not only in mice infected with B. pertussis but also in mice infected with B. parapertussis 6 weeks after primary infection (Fig. 3A). These antibodies were not detected in control mice. There was no significant difference between the levels of FHA-specific antibodies in mice infected with B. pertussis and in mice infected with B. parapertussis (P ⬎ 0.05). The calculated titer of FHA-neutralizing antibodies in the serum of both convalescent groups was 1,024. As shown in Fig. 3B, a significant level of PT-specific serum IgG was produced only by mice infected with B. pertussis (P ⬍ 0.05). PRN-specific IgG was detected in the sera of mice infected with B. parapertussis but not in those of mice infected with B. pertussis (Fig. 3C). The titer in the agglutination test against killed B. pertussis cells was high for the sera of mice infected with B. pertussis but not for sera of mice infected with B. parapertussis (Fig. 4A). A high level of
FIG. 4. Whole-cell agglutination titers in mice infected with B. pertussis and mice infected with B. parapertussis. Sera of control mice, mice infected with B. pertussis (BP), and mice infected with B. parapertussis (BPP) were examined as described in Materials and Methods. The titers of antibodies are expressed as mean values. (A) Antibodies against B. pertussis. (B) Antibodies against B. parapertussis. See legend to Fig. 3 for details.
6984
INFECT. IMMUN.
WATANABE AND NAGAI
FIG. 5. Antigen-specific lung IgA responses in mice infected with B. pertussis and mice infected with B. parapertussis. Levels of antigenspecific IgA in lungs of control mice, mice infected with B. pertussis (BP), and mice infected with B. parapertussis (BPP) were determined by ELISAs. The levels of antibodies are expressed in terms of mean absorbance. (A) FHA-specific lung IgA. (B) PT-specific lung IgA. (C) PRN-specific lung IgA. See legend to Fig. 3 for details.
agglutinating antibodies against B. parapertussis was found only in the sera of mice infected with B. parapertussis (Fig. 4B). We detected a response in terms of FHA-specific lung IgA both in mice infected with B. pertussis and in mice infected with B. parapertussis (Fig. 5A). Mice infected with B. parapertussis produced lower levels of the antibodies than mice infected with B. pertussis but the difference from control mice was significant (P ⬍ 0.05). Furthermore, PT-specific lung IgA was produced only by mice infected with B. pertussis (Fig. 5B). No PRNspecific lung IgA was detected in any samples examined (Fig. 5C). Responses of spleen cells to B. pertussis and to B. parapertussis antigens. We examined the proliferative responses in vitro of antigen-stimulated spleen cells from mice infected with B. pertussis and mice infected with B. parapertussis. The proliferation of spleen cells of mice infected with B. pertussis was induced to a greater or lesser extent upon stimulation with FHA, PTd, and killed B. pertussis cells (Fig. 6). In the case of
FIG. 7. Secretion of IFN-␥ by spleen cells in vitro in response to antigens of B. pertussis or of B. parapertussis. Spleen cells were prepared from control mice, mice infected with B. pertussis (BP), and mice infected with B. parapertussis (BPP). The results represent mean concentrations of IFN-␥, as estimated for individual cultures of spleen cells from five mice in each group. Each column with a vertical line represents a mean and standard error. ⴱ, P ⬍ 0.05 versus the control group.
spleen cells from mice infected with B. parapertussis, proliferation was induced by FHA and by killed B. parapertussis cells, but not by PTd. No PRN-specific proliferative response was detected in either convalescent group. As shown in Fig. 7, spleen cells derived from mice infected with B. pertussis secreted significant levels of IFN-␥ in response to stimulation in vitro by FHA or by PTd (P ⬍ 0.05). Stimulation by PRN, killed whole B. pertussis cells, or killed whole B. parapertussis cells did not induce any significant secretion of IFN-␥ from the spleen cells of mice infected with B. pertussis (P ⬎ 0.05). Spleen cells from mice infected with B. parapertussis secreted IFN-␥ upon stimulation by FHA (P ⬍ 0.05). Although some secretion of IFN-␥ was detected upon stimulation by killed B. parapertussis cells, the level was not significant (P ⬎ 0.05). There was no significant difference in terms of secretion of IFN-␥ between the spleen cells of mice infected with B. parapertussis and those of control mice after stimulation of spleen cells with PTd, PRN, or killed B. pertussis cells. The spleen cells of mice infected with B. pertussis and of mice infected with B. parapertussis did not release IL-4 upon stimulation by any of the antigens that we tested (data not shown). DISCUSSION
FIG. 6. Proliferative responses of spleen cells in vitro against antigens of B. pertussis and of B. parapertussis. Spleen cells were prepared from control mice, mice infected with B. pertussis (BP), and mice infected with B. parapertussis (BPP). The results represent a mean value of the increase in percentage in the proliferation index, calculated as indicated in Materials and Methods. Values were estimated for individual cultures of spleen cells from five mice for each group. Each column with a vertical line represents a mean and standard error.
Our data provide a demonstration of the relationship between the protective immunity induced by natural infection with B. pertussis and that induced by infection with B. parapertussis. We demonstrated that reciprocal protection is induced in mice infected with B. pertussis and those infected with B. parapertussis in a murine model of respiratory infection. Stehr et al. reported that the commercial pertussis vaccine prepared from antigens of B. pertussis had no or little efficacy against B. parapertussis in children (27). Khelef et al. suggested that reciprocal immunity might not be established between the two species of Bordetella in a murine model of respiratory infection (13). In their experiment, children or mice were immunized by subcutaneous or peritoneal injection. Immunization by injection of antigens effectively induces a serum antibody response. However, it does not effectively activate mucosal and Th1
VOL. 69, 2001
RECIPROCAL IMMUNITY AGAINST TWO SPECIES OF BORDETELLA
responses, which are important for protection against B. pertussis (12, 19, 20, 23). Furthermore, Mills et al. suggested a difference in terms of protective immune responses between mice that have recovered from infection by B. pertussis and mice that have been vaccinated by injection (20). Thus, the method of immunization might be important for detection of the reciprocal immune responses induced by infection with B. pertussis and B. parapertussis. We detected increases in levels of FHA-specific serum IgG and FHA-specific lung IgA in mice infected with B. pertussis and mice infected with B. parapertussis. Furthermore, we detected FHA-neutralizing antibodies in the sera not only of mice infected with B. pertussis but also of mice infected with B. parapertussis. It is known that both these species of Bordetella produce FHA, and the antigenicities of the FHAs are similar (13). Moreover, FHA is important for the attachment to host cells in the case of both B. pertussis and B. parapertussis (28). He et al. suggested that antibodies against FHA of B. pertussis in serum IgG might be responsible for protection against B. parapertussis in outbreaks of infection in West Finland (8). Thus, previous reports and our data suggest that FHA-specific antibodies in sera and lungs might be important in the establishment of reciprocal protection. We detected PT-specific serum IgG and lung IgA antibodies in mice infected with B. pertussis but not in mice infected with B. parapertussis. The genome of B. parapertussis includes a ptx gene but this gene is not transcribed because of mutations in the promoter region (2). PT is a major protective antigen against B. pertussis infection and an important component of pertussis vaccines. Although PT should be included in pertussis vaccines, PT cannot function for protection against B. parapertussis. We found no PRN-specific serum IgG and no PRN-specific lung IgA in the mice infected with B. pertussis in this study, confirming the reports of Mills et al. (19) and Redhead et al. (23). It is thought that PRN-specific serum antibodies played a minor role in the protection against B. pertussis in our convalescent mice. However, we detected an increase in the level of PRN-specific serum IgG antibodies in our mice infected with B. parapertussis, which might indicate the importance of PRN in the protective immune response in mice infected with B. parapertussis. No PRN-specific IgA antibodies were produced in the lungs of mice infected with B. pertussis or of mice infected with B. parapertussis. In both groups of convalescent mice, the mucosal immune response to PRN might have played only a minor protective role. In our presentation of the results of agglutination tests, we did not detect cross-reactions between the sera from mice infected with B. pertussis and those from mice infected with B. parapertussis. There are considerable immunological differences in terms of surface structures, which include fimbriae, between B. pertussis and B. parapertussis. Willems et al. reported that type-2 and type-3 fimbriae from B. pertussis partially protected mice against infection by B. parapertussis (32). They postulated that the partial protection against B. parapertussis infection might have been based on structural differences between the cell surfaces, including fimbriae. It has been suggested that the antibodies against fimbriae might be responsible for reciprocal protection (32). Nevertheless, cross-reaction in the agglutination test was not detected in our convalescent mice. Although agglutinating antibodies against B. pertussis
6985
might have a role in protection against B. pertussis infection, the role of the agglutinating antibodies in reciprocal protection might have been minor. It has been reported that Th1 responses play a major role in protection against B. pertussis in convalescent mice (19, 20, 23). We confirmed the induction of Th1 responses in mice infected with B. pertussis. In the case of mice infected with B. pertussis, proliferation of spleen cells in vitro was detected after stimulation with FHA and PT. The stimulation of the cells by FHA and PT induced the secretion of IFN-␥, but not of IL-4. In the case of spleen cells from mice infected with B. parapertussis, both the proliferation of spleen cells and the secretion of IFN-␥ were detected after stimulation with FHA. However, the amount of IFN-␥ secreted after stimulation by FHA from spleen cells of mice infected with B. parapertussis was lower than that from spleen cells of mice infected with B. pertussis. The results might be explained by differences in epitopes, recognized by the system for Th1, between FHA of B. pertussis and that of B. parapertussis. No secretion of IL-4 from spleen cells of mice infected with B. pertussis or of mice infected with B. parapertussis was detected. Our results reveal the induction of a Th1-biased immune response in mice infected with B. parapertussis, as well as in mice infected with B. pertussis. In this study, both mice infected with B. pertussis and mice infected with B. parapertussis exhibited a Th1 response against FHA, a result that might suggest the participation of Th1 responses against FHA in reciprocal protection. No proliferation and no secretion of IFN-␥ and IL-4 were detected upon stimulation by PRN of spleen cells from mice infected with B. pertussis and from mice infected with B. parapertussis. PRN might have played only a minor role on Th1 responses in our convalescent mice. We observed proliferative responses against killed B. pertussis cells and killed B. parapertussis cells by spleen cells of mice infected with B. pertussis and of mice infected with B. parapertussis, respectively. The results suggest that cell surface antigens of B. pertussis and B. parapertussis might play a role in Th1 responses against B. pertussis and B. parapertussis, respectively. However, it is unlikely that Th1 responses against cell surface antigens play a major role in reciprocal protection because of differences in the antigenicity of the cell surface between B. pertussis and B. parapertussis. For this study, we used antigens produced exclusively by B. pertussis, with the exception only of killed B. parapertussis cells. Further experiments are necessary to clarify virulence and immunological relationships between B. pertussis and B. parapertussis using the corresponding antigens produced by B. parapertussis. The virulence factors of B. parapertussis, including FHA, PRN, and fimbriae, have not been fully characterized in terms of biological activity and antigenicity. We are now characterizing the virulence factors of B. parapertussis and comparing them with those of B. pertussis. It is also necessary to confirm that our data in a murine model of respiratory infection correlate with the data in humans. B. parapertussis, as well as B. pertussis, causes whooping cough. Although pertussis vaccines decrease the incidence of whooping cough, several studies reported that the pertussis vaccines have no or limited efficacy against B. parapertussis (9, 13, 15, 27, 32). In order to prevent whooping cough caused by the two species, pertussis vaccines should have sufficient effi-
6986
WATANABE AND NAGAI
cacy against not only B. pertussis but also B. parapertussis. In this study, we have shown that reciprocal immunity can be induced between B. pertussis and B. parapertussis in a murine model of respiratory infection. It appears that Th1 and Th2 responses against FHA may be related to this reciprocal protection. It is thought that the data are important for the study of a vaccine which is effective against not only B. pertussis but also B. parapertussis. Studies of immunization which induce reciprocal protection against B. pertussis and B. parapertussis are proceeding in our laboratories. ACKNOWLEDGMENT We thank M. Endoh for helpful discussions. REFERENCES 1. Ahmed, S. A., R. M. Gogal, Jr., and J. E. Walsh. 1994. A new rapid and simple non-radioactive assay to monitor and determine the proliferation of lymphocytes: an alternative to [3H]thymidine incorporation assay. J. Immunol. Methods 170:211–224. 2. Arico, B., and R. Rappuoli. 1987. Bordetella parapertussis and Bordetella bronchiseptica contain transcriptionally silent pertussis toxin genes. J. Bacteriol. 169:2847–2853. 3. Blanchard, D. K., J. Y. Djeu, T. W. Klein, H. Friedman, and W. E. Stewart, Jr. 1986. Interferon-gamma induction by lipopolysaccharide: dependence on interleukin-2 and macrophages. J. Immunol. 136:963–970. 4. Bradford, M. M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248–254. 5. Chong, P., and M. Klein. 1989. Single-step purification of pertussis toxin and its subunits by heat-treated fetuin-Sepharose affinity chromatography. Biochem. Cell Biol. 67:387–391. 6. Giacomini, E., F. Urbani, C. M. Ausiello, and A. L. Luzzati. 1999. Induction of a specific antibody response to Bordetella pertussis antigens in cultures of human peripheral blood mononuclear cells. J. Med. Microbiol. 48:1081– 1086. 7. Gould-Kostka, J. L., D. L. Burns, M. J. Brennan, and C. R. Manclark. 1990. Purification and analysis of the antigenicity of a 69,000 Da protein from Bordetella pertussis. FEMS Microbiol. Lett. 55:285–289. 8. He, Q., K. Edelman, H. Arvilommi, and J. Mertsola. 1996. Protective role of immunoglobulin G antibodies to filamentous hemagglutinin and pertactin of Bordetella pertussis in Bordetella parapertussis infection. Eur J. Clin. Microbiol. Infect. Dis. 15:793–798. 9. Heininger, U., K. Stehr, P. Christenson, and J. D. Cherry. 1999. Evidence of efficacy of the Lederle/Takeda acellular pertussis component diphtheria and tetanus toxoids and pertussis vaccine but not the Lederle whole-cell component diphtheria and tetanus toxoids and pertussis vaccine against Bordetella parapertussis infection. Clin. Infect. Dis. 28:602–604. 10. Heininger, U., K. Stehr, S. Schmitt-Grohe, C. Lorenz, R. Rost, P. D. Christenson, M. Uberall, and J. D. Cherry. 1994. Clinical characteristics of illness caused by Bordetella parapertussis compared with illness caused by Bordetella pertussis. Pediatr. Infect. Dis. J. 13:306–309. 11. Iwata, S., T. Aoyama, A. Goto, H. Iwai, Y. Sato, H. Akita, Y. Murase, T. Oikawa, T. Iwata, S. Kusano, C. Kawashima, and K. Sunakawa. 1991. Mixed outbreak of Bordetella pertussis and Bordetella parapertussis in an apartment house. Dev. Biol. Stand. 73:333–341. 12. Jabbal-Gill, I., A. N. Fisher, R. Rappuoli, S. S. Davis, and L. Illum. 1998. Stimulation of mucosal and systemic antibody responses against Bordetella pertussis filamentous haemagglutinin and recombinant pertussis toxin after nasal administration with chitosan in mice. Vaccine 16:2039–2046. 13. Khelef, N., B. Danve, M. J. Quentin-Millet, and N. Guiso. 1993. Bordetella pertussis and Bordetella parapertussis: two immunologically distinct species. Infect. Immun. 61:486–490.
Editor: J. D. Clements
INFECT. IMMUN. 14. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680–685. 15. Lautrop, H. 1971. Epidemics of parapertussis. 20 years’ observations in Denmark. Lancet i:1195–1198. 16. Manghi, M. A., N. I. Dellepiane, P. V. Eriksson, G. Di Paola, and F. H. De Dios. 1990. Antibody response to pertussis toxin and filamentous haemagglutinin in NIH mice immunized with the International Standard for Pertussis Vaccine. Biologicals 18:165–172. 17. Menozzi, F. D., C. Gantiez, and C. Locht. 1991. Interaction of the Bordetella pertussis filamentous hemagglutinin with heparin. FEMS Microbiol. Lett. 62:59–64. 18. Mertsola, J. 1985. Mixed outbreak of Bordetella pertussis and Bordetella parapertussis infection in Finland. Eur. J. Clin. Microbiol. 4:123–128. 19. Mills, K. H., A. Barnard, J. Watkins, and K. Redhead. 1993. Cell-mediated immunity to Bordetella pertussis: role of Th1 cells in bacterial clearance in a murine respiratory infection model. Infect. Immun. 61:399–410. 20. Mills, K. H. G., M. Ryan, E. Ryan, and B. P. Mahon. 1998. A murine model in which protection correlates with pertussis vaccine efficacy in children reveals complementary roles for humoral and cell-mediated immunity in protection against Bordetella pertussis. Infect. Immun. 66:594–602. 21. Oda, M., J. L. Cowell, D. G. Burstyn, and C. R. Manclark. 1984. Protective activities of the filamentous hemagglutinin and the lymphocytosis-promoting factor of Bordetella pertussis in mice. J. Infect. Dis. 150:823–833. 22. Preston, N. W. 1988. Pertussis today, p. 1–18. In A. C. Wardlaw and R. Parton (ed.), Pathogenesis and immunity in pertussis. John Wiley & Sons, Ltd., Chichester, United Kingdom. 23. Redhead, K., J. Watkins, A. Barnard, and K. H. Mills. 1993. Effective immunization against Bordetella pertussis respiratory infection in mice is dependent on induction of cell-mediated immunity. Infect. Immun. 61:3190– 3198. 24. Ryan, E. J., E. McNeela, G. A. Murphy, H. Stewart, D. O’Hagan, M. Pizza, R. Rappuoli, and K. H. Mills. 1999. Mutants of Escherichia coli heat-labile toxin act as effective mucosal adjuvants for nasal delivery of an acellular pertussis vaccine: differential effects of the nontoxic AB complex and enzyme activity on Th1 and Th2 cells. Infect. Immun. 67:6270–6280. 25. Sekiya, K., M. Kawahira, and Y. Nakase. 1983. Protection against experimental Bordetella bronchiseptica infection in mice by active immunization with killed vaccine. Infect. Immun. 41:598–603. 26. Sekura, R. D., F. Fish, C. R. Manclark, B. Meade, and Y. L. Zhang. 1983. Pertussis toxin. Affinity purification of a new ADP-ribosyltransferase. J. Biol. Chem. 258:14647–14651. 27. Stehr, K., J. D. Cherry, U. Heininger, S. Schmitt-Grohe, M. Uberall, S. Laussucq, T. Eckhardt, M. Meyer, R. Engelhardt, and P. Christenson. 1998. A comparative efficacy trial in Germany in infants who received either the Lederle/Takeda acellular pertussis component DTP (DTaP) vaccine, the Lederle whole-cell component DTP vaccine, or DT vaccine. Pediatrics 101: 1–11. 28. van den Akker, W. M. 1998. The filamentous hemagglutinin of Bordetella parapertussis is the major adhesin in the phase-dependent interaction with NCI-H292 human lung epithelial cells. Biochem. Biophys. Res. Commun. 252:128–133. 29. Wardlaw, A. C., and R. Parton. 1988. The host-parasite relationship in pertussis, p. 327–352. In A. C. Wardlaw and R. Parton (ed.), Pathogenesis and immunity in pertussis. John Wiley & Sons, Ltd., Chichester, United Kingdom. 30. Watanabe, M., K. Funaishi, T. Takeo, and M. Endoh. 2000. Efficacy of pertussis vaccines consisted of antigens detoxified with tea-leaf catechins. Vaccine 19:1204–1210. 31. Watanabe, M., M. Nagai, K. Funaishi, and M. Endoh. 2000. Efficacy of chemically cross-linked antigens for acellular pertussis vaccine. Vaccine 19: 1199–1203. 32. Willems, R. J., J. Kamerbeek, C. A. Geuijen, J. Top, H. Gielen, W. Gaastra, and F. R. Mooi. 1998. The efficacy of a whole cell pertussis vaccine and fimbriae against Bordetella pertussis and Bordetella parapertussis infections in a respiratory mouse model. Vaccine 16:410–416.
