Protection and humoral immune responses against Bordetella pertussis infection in mice immunized with acellular or cellular pertussis immunogens

Vaccine 19 (2001) 1018 – 1128 www.elsevier.com/locate/vaccine

Protection and humoral immune responses against Bordetella pertussis infection in mice immunized with acellular or cellular pertussis immunogens Bernard M. van den Berg a,*, Silke David b, Henry Beekhuizen a, Frits R. Mooi b, Ralph van Furth a a

Department of Infectious Diseases, Leiden Uni6ersity Medical Center, Building 1, C5 -P, PO Box 9600, 2300 RC, Leiden, The Netherlands b Research Laboratory for Infectious Diseases, The Netherlands, National Institute of Public Health and the En6ironment, Biltho6en, The Netherlands Received 14 January 2000; received in revised form 24 January 2000; accepted 17 August 2000

Abstract In the present study, protection against Bordetella pertussis infection and humoral immunological responses in mice has been assessed upon immunization with custom-made acellular pertussis vaccines (ACVs) and whole-cell pertussis vaccine (WCV). Mice were immunized, next intranasally infected with B. pertussis and during 14 days the number of bacteria in the trachea and lungs and the level of serum antibodies were determined. ACV contained five immunogens, filamentous hemagglutinin, pertactin, fimbriae serotypes 2 and 3, and chemically detoxified pertussis toxin (PMC-5), or three immunogens, filamentous hemagglutinin, pertactin, and genetically detoxified (BC-3) or chemically detoxified pertussis toxin (SKB-3). Immunization with a high or low dose of ACV or WCV resulted in significant protection against B. pertussis, with differences in the degree of protection between the vaccines. The lowest protection was found with a low dose of SKB-3 and WCV. The pattern of cytokine production by spleen cells of immunized, non-infected, mice indicated that T-helper 1 cells are activated by vaccination with WCV, and T-helper 1 and T-helper 2 cells are involved in the immune response upon vaccination with ACVs. Each vaccine stimulated the production of IgG, but not IgA, antibodies. In mice immunized with ACV, elimination of B. pertussis from trachea and lungs correlated significantly with the titre of IgG1, but not IgG2a, antibodies. © 2000 Elsevier Science Ltd. All rights reserved. Keywords: Acellular pertussis immunogens; Bordetella pertussis; Bordetella pertussis vaccines; Pertussis immunogens; Respiratory infection in mice

1. Introduction At present, in most countries protection against whooping cough is based on the use of a whole-cell pertussis vaccine (WCV) composed of one or more Bordetella pertussis strains. However, acellular pertussis vaccines (ACVs) consisting of different combinations of B. pertussis virulence factors, such as detoxified pertussis toxin, filamentous hemagglutinin, pertactin, and fimbriae, have been developed to replace WCVs. In Abbre6iations: ACV, acellular pertussis vaccine; AUC, area under the curve; WCV, whole-cell pertussis vaccine. * Corresponding author. Present address: Department of Medical Physics, University of Amsterdam, The Netherlands. Tel.: +31-205665211; fax: +31-20-6917233. E-mail address: [email protected] (B.M. van den Berg).

various countries, an ACV is used to limit the adverse reactions associated with WCV. Recent clinical trials have demonstrated that ACVs can confer protection against whooping cough [1–6] in varying degrees. Several studies in children vaccinated with WCV, or recovered from whooping cough, have indicated a correlation between the presence of serum antibodies against filamentous hemagglutinin, pertactin or fimbriae and protection against whooping cough [7–10]. In recently completed efficacy trials with ACV or WCV, in Sweden and in Germany, high antibody titres against pertactin, and to a lesser extent against fimbriae and pertussis toxoid, in serum of children correlated with protection [11,12]. In addition, peripheral blood mononuclear cells, i.e. lymphocytes and monocytes, obtained from children

0264-410X/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 2 6 4 - 4 1 0 X ( 0 0 ) 0 0 3 2 9 - 7

B.M. 6an den Berg et al. / Vaccine 19 (2001) 1118–1128

during infection with B. pertussis or children convalescent from whooping cough, produced cytokines such as IL-2 and IFN-g, but little or no IL-4 and IL-5 [13–15], which suggests that the immune response induced by infection involves Th1 CD4-positive T-helper cells [16– 18]. To elucidate further the immune response after vaccination with WCV or ACV, or infection with B. pertussis, studies have been performed using murine aerosol-infection models. In these studies, complementary roles for Th1 and Th2 CD4-positive T-helper cells have been suggested in protection against B. pertussis [19 – 22]. Together, these studies argue for the existence of differences in the immune response after infection and vaccination. It has been observed that the course of respiratory infection with B. pertussis in mice has many characteristics in common with the course of whooping cough in children, such as colonization of respiratory epithelium causing local alteration of target cells and induction of systemic effects, such as leucocytosis, hyperinsulinaemia, and sensitization to histamine [21,23 – 26]. Furthermore, bacterial clearance from lungs of mice immunized with WCV or ACV and challenged by an aerosol of B. pertussis correlated with the estimated vaccine efficacy in children [21]. In the present study mice were immunized with vaccines composed of various combinations of B. pertussis immunogens or with WCV, but all devoid of immunogens from other microorganisms, and subsequently infected intranasally with B. pertussis. Next, the number of bacteria in trachea and lungs and the antibody responses in serum were determined after different intervals. The aim of the study was to measure the degree of protection against a B. pertussis infection after immunization with various ACV or WCV and to compare these results with the amount of antibodies during infection.

2. Materials and methods

2.1. Bacterial strain and culture condition B. pertussis strain BP536 is a streptomycin-resistant derivative of Tohama I, which was obtained by selecting bacteria from plates containing increasing concentrations of streptomycin (up to 30 mg/ml) [27]. Bacteria were cultured at 35°C for 3 days on Bordet – Gengou agar plates (Difco Laboratories, Detroit, MI) supplemented with 15% sheep blood and 30 mg/ml of streptomycin. Before use, bacteria were harvested and suspended in Verwey medium [28]. The number of bacteria was determined with a spectrophotometer at 600 nm and then adjusted to 5× 108 CFU/ml in Verwey medium. The number of bacteria was confirmed by colony counts after plating on Bordet – Gengou agar. It

1119

has been found that the streptomycin-resistant strain BP536 is as virulent as the parental strain (data not shown).

2.2. Mice Specific-pathogen-free mice (BALB/c/RIVM) were used and kept in protective isolators. The mice were routinely checked for contamination with Gram-negative bacteria such as Bordetella bronchiseptica, Klebsiella pneumoniae, members of the family Pasteurellaceae, Pseudomonas aeroginosa, Salmonella sp., and Yersinia enterocolitica, according to standard protocols.

2.3. Vaccines and immunization All vaccines contained only pertussis components. In order to obtain pertussis vaccines without immunogens from other microorganisms, such as tetanus and diphtheria toxoids, ACVs were custom-made with purified immunogens donated by various pharmaceutical manufacturers. The composition of each stock of ACV was identical to the composition of the pertussis immunogens of the commercial vaccines manufactured at the time the experiments were performed. One three-component ACV contained 25 mg formaldehyde- and glutaraldehyde-detoxified pertussis toxin, 25 mg filamentous hemagglutinin and 8 mg pertactin (obtained from SmithKline Beecham; Rixensart, Belgium) in 0.5 ml saline, designated SKB-3. Another three-component ACV contained 7.5 mg genetically detoxified pertussis toxin [29,30], 10 mg filamentous hemagglutinin and 10 mg pertactin (obtained from Biocine Chiron; Siena, Italy) in 0.5 ml saline, designated BC-3. The five-component ACV contained 20 mg glutaraldehydedetoxified pertussis toxin, 20 mg filamentous hemagglutinin, 3 mg pertactin and 5 mg fimbriae serotype 2 and 3 (obtained from Pasteur Me´rieux Connaught; North York, Canada) in 0.5 ml saline, designated PMC-5. Each ACV was adsorbed to aluminium hydroxide (25% Alu-Gel-S; Serva, Heidelberg, Germany) in phosphatebuffered saline (PBS; 10 mM sodium phosphate, 155 mM sodium chloride, pH 7.4). WCV obtained from the National Institute of Public Health and the Environment (Bilthoven, The Netherlands), contained 1.6× 1010 of two B. pertussis strains (16 opacity units) in 0.5 ml saline containing the same amount aluminium hydroxide. The vaccine dose of ACV and WCV used in our study is either a 1:5 dilution of the stock vaccine, called high vaccine dose, or a 1:50 dilution, called low vaccine dose. Mice were immunized subcutaneously on days 0 and 14 with a high or low dose of the indicated vaccines, or with aluminium hydroxide in PBS (25% Alu-Gel-S), i.e. control mice.

1120

B.M. 6an den Berg et al. / Vaccine 19 (2001) 1118–1128

2.4. Intranasal infection of mice Intranasal infection of mice was performed as described [31]. At 2 weeks after the second immunization mice were lightly anaesthetized with ether, and a drop of 20 ml of the inoculum, containing 1× 107 B. pertussis, was carefully placed in each nostril, and allowed to be inhaled by the animal. For the determination of bacterial colonization, mice were killed at 2 h and 3, 7, and 14 days after infection by an intraperitoneal injection of an overdose of a barbiturate (Nembutal, Sanofi/Algin, Maassluis, The Netherlands). Blood samples were taken for the assessment of antibodies. Trachea and lungs were excised aseptically, after which the trachea suspended in 200 ml Verwey medium were treated with glass pearls on a vortex and both lungs suspended in 1 ml Verwey medium were homogenized using a homogenizer (Pro200, ProScientific, Monroe, CT) for 10 s at 20 000 rpm. Viable bacteria in trachea and lung homogenates were determined by plating serial dilutions of the tissue suspension on Bordet – Gengou agar plates containing 30 mg streptomycin/ml. The number of bacteria per millilitre suspension was adjusted for the amount of tissue added to the Verwey medium and the results are expressed as the number of B. pertussis per trachea or total lung. The limit of detection was determined to be 10 bacteria/ml (log10 0.95) for lung homogenates and 0.4 bacteria/ml (log10 −0.46) per trachea.

2.5. Detection of antibodies in mouse serum Blood was obtained by cardiac puncture on day 0 (2 h after infection) and on days 3, 7, and 14. Serum antibody titres were determined by enzyme-linked immunosorbent assay (ELISA), as described [32]. Polysorp 96-well plates (Nalgene Nunc International, Rochester, NY) were coated with 1×108 heat-killed B. pertussis per millilitre (30 min at 56°C) in 50 mM sodium carbonate buffer (pH 9.6) and incubated overnight at 37°C. After washing with PBS containing 0.05% Tween-20, the plates were incubated with PBS containing 0.5% BSA and 0.05% Tween-20 for 1 h at 37°C to reduce non-specific binding. Next, serial dilutions of mouse immune sera were added and the plates were incubated for 90 min at 37°C and thereafter washed with PBS containing 0.05% Tween-20. Immunoglobulin IgG1 and IgG2a subclass antibodies were determined by incubation with peroxidase-conjugated goat anti-mouse IgG1 or IgG2a [Southern Biotechnology Associates Inc. (SBA), Birmingham, AL] for 1 h at 37°C. Finally, 100 ml of the peroxidase substrate 3,3,5,5%-tetramethyl benzidine (0.4 mM; SigmaAldrich Chemie bv, Zwijndrecht, The Netherlands) and H2O2 (0.003%) in 110 mM sodium acetate buffer (pH 5.5) (TMB-reaction medium) was added and al-

lowed to develop for 10 min at room temperature. The reaction was stopped with 50 ml of 2 M H2SO4 and absorbance was determined at an optical density of 450 nm. The highest dilution corresponding with three times the blank value was taken as the antibody titre, which is expressed as − log10.

2.6. Cytokine assays The levels of cytokines secreted by spleen cells of non-infected mice was determined 2 weeks after the second immunization. The mice were killed by an intraperitoneal injection of an overdose of Nembutal (Sanofi/Algin), spleens were aseptically excised, and single-cell suspensions were prepared using an openfilter chamber (NPBI; Emmer-Compascum, The Netherlands). 1×106 cells per well were cultured in RPMI-1640 medium (Gibco, Grand Island, NY), containing 10% heat-inactivated foetal calf serum (Gibco), 1000 U/ml penicillin-G and 50 mg/ml streptomycin, in 96-wells tissue culture plates (Greiner Labortechniek, Frickenhausen, Germany). Spleen cells were cultured in the presence of 10 mg/ml homologues ACV, 1.0 opacity units/ml WCV, 2 mg/ml concanavalin A (Sigma– Aldrich Chemie), which served as positive control, or RPMI medium alone. In pilot experiments, these concentrations were shown to be the lowest to induce a cytokine response (data not shown). Supernatants were removed after 48 h to determine cytokines by ELISA, according to the protocol of the manufacturer (Pharmingen, San Diego, CA). In short, Maxisorp 96wells plates (Nalgene Nunc) were coated with 0.5 mg/ml or 1.0 mg/ml of the respective rat anti-mouse mAb against IFN-g, IL-2, IL-4, IL-5, or IL-10 (Pharmingen) in sodium carbonate buffer (50 mM, pH 9.6) and incubated overnight at 4°C. After washing with PBS containing 0.5% Tween-20, the plates were incubated with PBS containing 1.0% BSA and 0.5% Tween-20 for 2 h at 37°C, to reduce non-specific binding. Next, serial dilutions of culture supernatants were added and the plates were incubated for 2 h at 37°C and thereafter washed with PBS containing 0.5% Tween-20. The cytokine content was determined by incubation for 1 h at room temperature with corresponding biotinylated rat anti-mouse cytokine mAb (Pharmingen). Plates were washed with PBS containing 0.5% Tween-20, then 0.1 mg/ml multi-labelled peroxidase streptavidine/ avidine (CLB, Amsterdam, The Netherlands) was added and incubated for 1 h at room temperature. Finally, 100 ml of TMB-reaction medium was added and allowed to develop for 10 min at room temperature. The reaction was stopped with 50 ml of 2 M H2SO4 and absorbance was determined at an optical density of 450 nm. The concentration of the various cytokines was determined using murine recombinant cytokines (Pharmingen) for generation of the standard curves.

B.M. 6an den Berg et al. / Vaccine 19 (2001) 1118–1128

2.7. Statistical calculations and analysis The number of viable B. pertussis in trachea and lungs of mice and the antibody titres are expressed as mean 9standard deviation (SD). The weighted means by formulation and by day of the number of bacteria in trachea or lungs of mice during the 14 day period of infection, together with the number of mice in each group, were used to calculate the area under the curve (AUC: mean log10 number of B. pertussis/ml× days)9 standard error (SE). Similarly the AUC9 SE of the antibody titres (mean log10 antibody titre× days) are calculated. First, the variance between the various immunized groups was evaluated using a multivariate analysis of variance using Pillai’s trace on the mean log10 number of bacteria/ml and log10 antibody titre (fixed factors: group and time). When statistically significant, differences between AUC of immunized and control mice were further evaluated by means of a two-tailed t-test. Correlation between bacterial elimination from trachea or lungs of vaccinated, infected, mice and the antibody response was assessed by means of the two-tailed Spearman’s non-parametric correlation test r. The cytokine concentration produced by spleen cells is expressed as the mean9SD and differences were

1121

evaluated by means of analysis of variance (ANOVA) and Bonferroni multiple-comparison test. Differences are not significant when P\ 0.05.

3. Results

3.1. Number of B. pertussis in trachea and lungs of mice immunized with a high dose of 6accine Since in the murine respiratory-infection model the number of B. pertussis in trachea reflects the association of the bacteria with respiratory epithelium, and the number of B. pertussis in lungs represents the sum of bacteria associated with epithelium of the bronchial tree and the alveolar spaces, the numbers of bacteria in trachea and lungs were determined separately. On day 0 (2 h after infection) the trachea of control mice contained log10 5.24 B. pertussis, which after a delay declined to 2.01 at day 14 (Fig. 1). After day 0 the total numbers of B. pertussis in the trachea of mice immunized with a high dose of WCV, PMC-5, BC-3, or SKB-3, were significantly (PB 0.001) less in comparison with the numbers of control mice during the 14 days of infection (Table 1). The reduction in the

Fig. 1. B. pertussis colonization of trachea or lungs of mice with a high dose (High) or low dose (Low) of WCV (closed squares), PMC-5 (closed circles), BC-3 (closed triangles), SKB-3 (closed diamonds), or aluminium hydroxide in PBS (control; open squares). Values are the mean log10 CFU9 SD.

B.M. 6an den Berg et al. / Vaccine 19 (2001) 1118–1128

1122

Table 1 AUC of B. pertussis of trachea and lungs of mice immunized with a high or low dose of vaccine during 14 days of infectiona Vaccine

Control WCV PMC-5 BC-3 SKB-3

No. of mice with H/Lb

44/43 43/43 43/20 43/43 44/22

AUC of B. pertussis in trachea ( 9SE)c

AUC of B. pertussis in lungs ( 9 SE)

High-vaccine dose

Low-vaccine dose

High-vaccine dose

Low-vacine dose

57.09 1.93g,i,k,m 10.09 1.29e,k,m 15.893.33e,f,m 17.5 93.94e,g,l 25.2 91.65e,g,i,j

57.9 9 2.14g,i,k,m 34.2 9 3.74e,h,k,l 27.2 9 3.75e,f,m 23.2 9 3.16e,g,m 42.2 9 9.18e,f,i,k

80.3 9 0.78g,i,k,m 40.2 9 2.43e,k 33.4 92.03e,f,m 33.0 9 4.99e,g,m 38.4 9 8.53e,f,i,k

81.5 9 1.18g,i,k,l 58.0 9 4.48e,j,l 52.8 917.9e,l 46.8 9 10.7e,f,m 69.5 9 25.8d,f,h,k

a

Mice were immunized twice with a high or low dose of WCV, PMC-5, BC-3 or SKB-3, or aluminium hydroxide in PBS (control). Number of mice immunized with a high (H) or low (L) dose of vaccine. c Calculated with data of mean log10 CFU9 SD at days 0, 3, 7, 14 and the number of mice in each group and expressed as AUC 9 SE (mean log10 bacteria/ml×days). For details, see Ref. [33]. Differences between AUC of immunized and control mice were evaluated by means of two-tailed t-test. d PB0.05 versus control. e PB0.001 versus control. f PB0.05 versus WCV. g PB0.001 versus WCV. h PB0.05 versus PMC-5. i PB0.001 versus PMC-5. j PB0.05 versus BC-3. k PB0.001 versus BC-3. l PB0.05 versus SKB-3. m PB0.001 versus SKB-3. b

numbers of B. pertussis was the highest in trachea of WCV-immunized mice and was significantly larger than in PMC-5 (P B 0.05), BC-3 (P B0.001), and SKB-3 (P B0.001) immunized mice. Reduction in the number of B. pertussis in the trachea of PMC-5 immunized mice was similar to BC-3 immunized mice, and significantly (P B0.001 and PB 0.05 respectively) greater than in SKB-3 immunized mice. The number of viable B. pertussis in lungs of control mice decreased gradually during the 14 days of infection (Fig. 1). In lungs of all immunized mice the reduction of B. pertussis was significantly (P B 0.001) larger than the decline of bacteria in control mice, but not significantly different than from each other (Table 1).

3.2. Number of B. pertussis in trachea and lungs of mice immunized with a low dose of 6accine The above experiments showed that immunization of mice with either of the pertussis vaccines resulted in a high overall reduction in the number of B. pertussis from both trachea and lungs during 14 days of infection. To evaluate further the protective activity of these vaccines, mice were immunized with a ten times lower dose. The results showed that the rate of elimination of B. pertussis from trachea of these mice during the 14 days of infection (Fig. 1) was lower than in mice immunized with a high dosage of the respective vaccines, but are all significantly (P B 0.001) larger than the reduction of B. pertussis in control mice (Table 1).

In trachea of BC-3 immunized mice the reduction in the number of B. pertussis was the highest and significantly (PB 0.001) greater in comparison with WCV and SKB3 immunized mice; the reduction in B. pertussis in the trachea of PMC-5 immunized mice was similar to BC-3 immunized mice, but significantly higher than from WCV (PB 0.05) and SKB-3 (P B 0.001) immunized mice. The reduction in the number of bacteria in lungs during 14 days of infection was significantly larger in mice immunized with WCV (P B 0.001), PMC-5 (P B 0.001), BC-3 (PB 0.001), or SKB-3 (PB 0.05) than control mice (Table 1). In lungs of BC-3 immunized mice the reduction of B. pertussis was the highest and significantly greater than in WCV (P B 0.05) and SKB3-immunized (PB0.001) mice. The reduction in the number of bacteria in the lungs of PMC-5 immunized mice was similar to BC-3 immunized mice, but significantly (PB 0.05) greater than from SKB-3 immunized mice.

3.3. B. pertussis-specific antibody responses Serum of mice immunized with a high dose of each vaccine, but not infected, contained considerably higher amounts of IgG and IgM anti-B. pertussis, but no IgA, antibodies compared with normal mouse serum (data not shown). In immunized and infected mice, anti-B. pertussis antibodies were determined at days 0 (2 h after infection), 3, 7, and 14 of infection. At day 0 the control

B.M. 6an den Berg et al. / Vaccine 19 (2001) 1118–1128

mice contained virtually no antibodies against B. pertussis (data not shown). At day 7 of infection, these control mice contained IgG2a antibodies that increased further at day 14; IgG1 antibodies were found only at day 14. In mice immunized with a high or low dose of WCV, PMC-5, or BC-3 the titre of IgG1 antibodies was significantly (PB 0.001) higher compared with control mice; mice immunized with a high dose, but not a low dose, of SKB-3 contained a significant higher amount of IgG1 than control mice (Table 2). In WCV-immunized mice the amount of IgG1 antibodies was significantly higher than in mice immunized with each of the ACVs. Mice immunized with a high or low dose of PMC-5 contained significantly (P B 0.001) higher amounts of IgG1 antibodies than SKB-3 immunized mice, and immunization with a high dose of PMC-5 resulted in a significantly (P B 0.05) higher amount of IgG1 antibodies compared with BC-3 immunized mice. The titre of IgG1 antibodies in BC-3 immunized mice was significantly (PB0.001) higher than in SKB-3 immunized mice. In mice immunized with WCV, PMC-5, or BC-3, but not with SKB-3, the titre of IgG2a antibodies was significantly higher than in control mice (Table 2). In WCV-immunized mice, the amount of IgG2a antibodies was significantly (PB 0.001) higher than in mice immunized with each of the ACVs. The titre of IgG2a antibodies in PMC-5 immunized mice was significantly higher than of BC-3 (P B0.05) immunized and of SKB3 (P B 0.001) immunized mice. The titre of IgG2a antibodies in BC-3 immunized mice was not significantly higher than in SKB-3 immunized mice.

1123

3.4. Effect of immunization on the number of bacteria in trachea and lungs in relation to the antibody response The protective activity of the four immunogens applied in a high or low dose (Table 1) was evaluated further. To reduce experiment-to-experiment variation, the decrease in the number of B. pertussis in trachea and lungs of immunized versus control mice during the 14 days of infection was defined and calculated as the difference in AUC between these groups (DAUC: Table 3). To distinguish between the antibody response upon a B. pertussis infection and to immunization with the various immunogens (Table 2), the difference between immunized and control mice was calculated and expressed as DAUC (Table 3). Comparison of the DAUC of antibody titres with the DAUC of bacteria in the trachea and lungs revealed a significant positive correlation between the IgG1-antibody titre in serum and the reduction in number of B. pertussis in trachea (r 0.886; PB0.05) and in lungs (r 0.986; PB0.01) of mice immunized with a high or low dose of ACV (Fig. 2). There was no significant correlation between the IgG2a-antibody levels in serum and bacterial elimination in trachea (r 0.314; P=0.544) and lungs (r 0.174; P=0.742) of these mice. In mice immunized with a high or low dose of WCV, the DAUCs of IgG1 and IgG2a-antibody levels were much higher than in ACVimmunized mice (Table 3). When the DAUCs of antibody titre and the number of B. pertussis in trachea and lungs of mice immunized with WCV (Table 3) were also

Table 2 AUC of antibody titre in sera of mice immunized with a high or low dose of vaccine during 14 days of infectiona Vaccine

Control WCV PMC-5 BC-3 SKB-3 a

No. of mice with H/L

44/43 43/43 43/20 43/43 44/22

AUC of B. pertussis specific IgG1 ( 9 SE)b

AUC of B. pertussis specific IgG2a ( 9SE)b

High-vaccine dose

Low-vaccine dose

High-vaccine dose

Low-vaccine dose

8.23 9 1.00f,h,j,l 86.090.82d,h,j,l 66.1 9 3.83d,f,i,l 72.9 9 1.74d,f,g,l 50.7 9 8.14d,f,h,j

5.22 9 1.60e,h,j 69.1 9 0.69d,h,j,k 45.9 9 24.5d,f,l 46.8 9 8.16d,f,l 9.14 9 3.72e,h,j

17.5 9 4.82f,h,j 70.1 9 1.76d,h,j,l 39.3 9 3.94d,f,i,l 31.4 9 6.95d,f,g,l 25.6 9 13.6d,h,j

21.9 9 5.67f,h,i,k 57.4 9 5.48d,h,j,l 45.4 9 2.36d,f,j,l 31.1 9 5.83c,f,h 30.6 9 7.43c,f,h

Mice were immunized twice with a high or low dose of WCV, PMC-5, BC-3 or SKB-3, or aluminium hydroxide in PBS (control). AUC9SE (mean log10 antibody titre×days). See further legends of Table 1. c PB0.05 versus control d PB0.001 versus control. e PB0.05 versus WCV. f PB0.001 versus WCV. g PB0.05 versus PMC-5. h PB0.001 versus PMC-5. i PB0.05 versus BC-3. j PB0.001 versus BC-3. k PB0.05 versus SKB-3. l PB0.001 versus SKB-3. b

B.M. 6an den Berg et al. / Vaccine 19 (2001) 1118–1128

1124

Table 3 The log10 number of B. pertussis in trachea and lungs and the log10 serum antibody titres in immunized, infected, mice versus control, infected, micea Mice immunized with

Vaccine dose

DAUC B. pertussis inb

WCV PMC-5 BC-3 SKB-3

High Low High Low High Low High Low

Antibodiesc

Trachea

Lungs

IgG1

IgG2a

47 9 1.64 24 93.05 41 92.71 31 92.11 40 93.09 35 92.70 32 91.79 16 95.62

40 91.80 24 93.26 47 91.53 29 910.1 47 93.55 35 97.61 42 96.06 12 915.0

789 0.92 64 91.23 58 9 2.78 419 13.9 65 91.41 42 9 5.88 439 5.80 4 9 2.53

53 93.63 36 9 5.58 22 94.56 24 94.87 14 9 5.98 9 9 5.75 8 98.83 9 96.32

a Expressed as difference between DAUC of the mean log10 number of B. pertussis in trachea or lungs, and the mean log10 antibody titre of IgG1 or IgG2a antibodies of mice immunized with a high-vaccine dose or a low-vaccine dose of WCV, PMC-5, BC-3 or SKB-3 during the 14 day period of infection (data from Tables 1 and 2) b Calculated as AUCcontrol minus AUCvaccine. c Calculated as AUCvaccine minus AUCcontrol.

included in these calculations, the correlations between DAUC of IgG1 antibodies and DAUC of bacteria in trachea (r 0.690; P =0.058) and lungs (r 0.551; P= 0.157) were not significant.

produced high amounts of IFN-g, IL-2, IL-4, IL-5, and IL-10 (data not shown).

4. Discussion

3.5. Cytokine production after immunization No detectable levels of cytokines were produced by spleen cells of the various immunization groups when cultured in the presence of RPMI medium (data not shown). Spleen cells of control mice and mice immunized with WCV, PMC-5, or SKB-3 produced high amounts of IFN-g, IL-2, IL-4, IL-5, and IL-10 upon stimulation with concanavalin A, which served as positive control (data not shown and Fig. 3). Spleen cells of control mice produced IFN-g, IL-2, and IL-10 upon incubation with WCV, IL-2 in the presence of PMC-5, and IFN-g and IL-2 when stimulated with SKB-3 (Fig. 3). Cells of WCV-immunized mice stimulated with its homologous antigens (WCV) resulted in a significantly higher amount of IFN-g, IL-2, and IL-10 in comparison with cells incubated with medium alone. No IL-4 and IL-5 production by these cells was detected. Cells of PMC-5 immunized mice stimulated with PMC-5 produced a significantly higher amount of IFN-g, IL-2, IL-5, and IL-10 than cells incubated with medium alone, and a small amount of IL-4. Cells of SKB-3 immunized mice produced upon stimulation with SKB3 equal amounts of IFN-g, IL-2 and IL-4 in comparison with cells of PMC-5 immunized mice, and no IL-5 and IL-10. In cultures of spleen cells from mice immunized with BC-3 stimulated with BC-3, all cells died due to preservatives in the BC-3 stock; concanavalin A stimulation of spleen cells from BC-3 immunized mice

In the present study we compared in mice the protection against B. pertussis in both trachea and lungs after immunization with three different ACVs and a WCV. We concluded that immunization with each of the pertussis vaccines resulted in a high level of protection.

Fig. 2. Correlation between the DAUC of the number of bacteria in trachea (closed symbols) and lungs (open symbols) and the DAUC IgG1 antibody titres in serum of mice immunized with a high or low dose of PMC-5 (circles), BC-3 (triangles), or SKB-3 (diamonds). Data from Table 3. Correlation was assessed by means of two-tailed Spearman’s non-parametric test and expressed as r.

B.M. 6an den Berg et al. / Vaccine 19 (2001) 1118–1128

1125

Fig. 3. Cytokine production by spleen cells of mice immunized twice with a high dose of WCV, PMC-5, SKB-3, or aluminium hydroxide in PBS (-). Spleen cells were cultured with 10 mg/ml of ACV, 1.0 opacity units/ml of WCV or 2 mg/ml concanavalin A (conA) and supernatants were removed after 48 h to determine IFN-g, IL-2, IL-4, IL-5 and IL-10. Concentrations are the mean 9standard deviation from four mice per group. Significance and P values were obtained by ANOVA and Bonferroni multiple comparison test, *P B0.05; **PB0.001, with respect to cells cultured with medium alone. \LD: above level of detection; −: no cytokine detected.

However, significant differences were observed in the elimination of bacteria from trachea and lungs of mice immunized with the various vaccines (Table 1). Furthermore, bacterial elimination from trachea and lungs and the titre of serum IgG1 antibodies correlated significantly in ACV-immunized mice. Mice immunized with a high dose of WCV cleared B. pertussis better from their trachea than from the lungs,

since B. pertussis was not detectable in trachea of the majority of these mice at day 7 of infection, but their lungs still contained B. pertussis at day 14. The longer persistence of B. pertussis in lungs compared with the trachea of WCV-immunized mice was also observed in mice challenged by an aerosol of B. pertussis [19]. In mice immunized with a high dose of PMC-5 or BC-3 almost all B. pertussis were eliminated from the trachea

1126

B.M. 6an den Berg et al. / Vaccine 19 (2001) 1118–1128

and lungs at day 14 of infection, but the trachea and lungs of mice immunized with a high dose of SKB-3 still contained B. pertussis. However, mice immunized with a low dose of PMC-5 or BC-3 cleared B. pertussis well from both trachea and lungs, but mice immunized with either a low dose of SKB-3 or WCV were significantly less able to clear these bacteria. From our results it is difficult to decide whether elimination of B. pertussis from the trachea or lungs of immunized mice provides a better prediction of the quality of a given vaccine. Can differences in protection be explained by differences in detoxification methods, vaccine composition and/or the amount of immunogens of the vaccines? Our results revealed that PMC-5, which contains glutaraldehyde-detoxified pertussis toxin, and BC-3, containing genetically detoxified pertussis toxin, eliminated B. pertussis significantly better from trachea and lungs than SKB-3, which contains formaldehyde- and glutaraldehyde-detoxified pertussis toxin. Apparently the method of detoxification does affect the protective power of the vaccines, as exemplified by the high level of protection after immunization with BC-3, especially when used in a low dose of vaccine. Since PMC-5 differs from BC-3 and SKB-3 because it contains fimbriae serotypes 2 and 3, the enhanced protection by PMC-5 compared with the other two ACVs may be due to the presence of fimbriae in the former. Another study came to a similar conclusion, i.e. that the number of components and the methods of purification and detoxification of the immunogens affect the potency of the ACV [21]. Furthermore, difference in the efficacy of a bicomponent ACV (custom-made with PT and FHA) and a commercial tricomponent ACV (Infanrix™) has been reported as well [34,35]. Antibodies produced in mice upon immunization with each of the pertussis vaccines were mainly of the IgG isotype and consisted of more IgG1 than IgG2a antibodies, whereas serum of control, infected, mice contained more IgG2a than IgG1 antibodies against B. pertussis. Elimination of B. pertussis from trachea and lungs of mice immunized with the various ACVs correlated significantly with IgG1 antibody levels. Immunogen recipients already have antibodies against B. pertussis at the time of challenge, and infection may induce a secondary antibody response for some antigens, but also a primary response for other antigens. However, the similarity in the increasing amount of ACV-induced IgG2a titres during the 14 days of infection and of IgG2a antibodies in control, infected, mice suggested that these antibodies were raised in response to the bacterial infection rather than being induced by the various ACVs. A correlation between antibody response and protection against an intranasal or aerosol challenge with B. pertussis has been also reported for mice immunized with either filamentous hemagglu-

tinin [36–38], pertussis toxin [38–40], fimbriae [31,41,42] or pertactin [37,39,43]. Immunization with WCV resulted in an excellent protection and very high titres of IgG1 and IgG2a antibodies, but no correlation between protection against the B. pertussis infection and the AUC of IgG1 or IgG2a antibodies was found. Conceivably, vigorous stimulation of antibody production by the various antigens of WCV, which includes (intracellular) antigens other than present in ACVs and lipopolysaccharide, elicit also the production of antibodies that are measured with the ELISA using heatkilled B. pertussis as antigen. In children vaccinated with WCV, a low correlation was found between protection against whooping cough and the IgG antibody titre against filamentous hemagglutinin, fimbriae or pertactin [7–10]. In an earlier study we demonstrated that surface molecules other than filamentous hemagglutinin, fimbriae and pertactin reduced the adherence of B. pertussis to human respiratory epithelial cells [44]. In children vaccinated with ACV or WCV, serum-antibody concentrations against pertactin, and to a lesser extent against fimbriae and pertussis toxoid, correlated well with protection against whooping cough [11,12]. In accordance with earlier studies in mice vaccinated with WCV [19–22], we observed a Th1-type of immune response, i.e. the high amounts of IgG2a antibodies and production of IFN-g and IL-2 by in-vitro-stimulated spleen cells. However, immunization with ACVs resulted in activation of both Th1 and Th2 cells. These results agree with the observation that in ACV-vaccinated children Th1 and Th2 cells were activated upon in vitro stimulation with killed B. pertussis or purified B. pertussis antigens [13,45,46]. Taken together, a murine intranasal infection model is useful to compare various combinations of B. pertussis immunogens or commercial B. pertussis vaccines in their ability to protect against infection with B. pertussis.

Acknowledgements The authors wish to express their gratitude to Henk Gielen for technical assistance. This work was financially supported by the Praeventie Fonds grant 2825450.

References [1] Ad Hoc Group for the study of Pertussis Vaccines. A randomized controlled trial of a two-component, a three-component and a five-component acellular pertussis vaccine and a British wholecell pertussis vaccine. Lancet 1997;350:1569 – 77.

B.M. 6an den Berg et al. / Vaccine 19 (2001) 1118–1128 [2] Centers for Disease Control and Prevention. Pertussis vaccination: use of acellular pertussis vaccines among infants and young children-recommendations of the advisory committee on immunization practices (ACIP). Morbidity & Mortality Weekly Report Surveillance Summaries 1997;46:(RR)1–(RR)25. [3] Greco D, Salmaso S, Mastrantonio P, Guiliano M, et al. A controlled trial of two acellular vaccines and one whole-cell vaccine against pertussis. New England Journal of Medicine 1996;334:341 – 55. [4] Gustafsson L, Hallander HO, Olin P, Reizenstein E, Storsaeter J. A controlled trial of a two-component acellular, a five-component acellular, and a whole cell pertussis vaccine. New England Journal of Medicine 1996;334:349–55. [5] Simondon F, Preziosi M-P, Yam A, Kane CT, et al. A randomized double blind trial comparing a two-component acellular to a whole-cell pertussis vaccine in Senegal. Vaccine 1997;15:1606 – 12. [6] Trollfors B, Taranger J, Lagergard T, Lind L, et al. A placebocontrolled trial of a pertussis-toxoid vaccine. New England Journal of Medicine 1995;333:1045–50. [7] Ashworth LAE, Robinson A, Irons LI, Morgan CP, Isaacs D. Antigens in whooping cough vaccine and antibody levels induced by vaccination of children. Lancet 1983;1:878–81. [8] Blumberg DA, Pineda E, Cherry JD, Caruso A, Scott JV. The agglutinin response to whole-cell and acellular pertussis vaccines is Bordetella pertussis-strain dependent. American Journal Diseases of Children 1992;146:1148–50. [9] He Q, Viljanen MK, Olander R, Bogaerts H, et al. Antibodies to filamentous hemagglutinin of Bordetella pertussis and protection against whooping cough in schoolchildren. Journal of Infectious Diseases 1994;170:705–8. [10] Mink CM, O’Brien CH, Wassilak S, Deforest A, Meade BD. Isotype and antigen specificity of pertussis agglutinins following whole-cell pertussis vaccination and infection with Bordetella pertussis. Infection and Immunity 1994;62:1118–20. [11] Cherry JD, Gornbein J, Heininger U, Stehr K. A search for serologic correlates to Bordetella pertussis cough illnesses. Vaccine 1998;16:1901 – 6. [12] Storsaeter J, Hallander HO, Gustafsson L, Olin P. Levels of anti-pertussis antibodies related to protection after household exposure to Bordetella pertussis. Vaccine 1998;16:1907–16. [13] He Q, Tran Minh NN, Edelman K, Viljanen MK, Arvilommi H, Mertsola J. Cytokine mRNA expression and proliferative responses induced by pertussis toxin, filamentous hemagglutinin, and pertactin of Bordetella pertussis in the peripheral blood mononuclear cells of infected and immunized schoolchildren and adults. Infection and Immunity 1998;66:3796–801. [14] Ryan M, Murphy G, Gothefors L, Nilsson L, Storsaeter J, Mills KHG. Bordetella pertussis respiratory infection in children is associated with preferential activation of type 1 T helper cells. Journal of Infectious Diseases 1997;175:1246–50. [15] Ryan M, Murphy G, Ryan E, Nilsson L, et al. Distinct T-cell subtypes induced with whole cell and acellular pertussis vaccines in children. Immunology 1998;93:1–10. [16] Mosman TR, Coffman RL. Heterogeneity of cytokine secretion patterns and functions of helper T cells. Advances in Immunology 1989;46:111 – 47. [17] Mosman TR, Sad S. The expanding universe of T-cell subsets: Th1, Th2 and more. Immunology Today 1996;17:138–46. [18] Romagnani S. Lymphokine production by human T cells in disease states. Annual Review of Immunology 1994;12:227 – 57. [19] Barnard A, Mahon BP, Watkins J, Redhead K, Mills KHG. Th1/Th2 cell dichotomy in acquired immunity to Bordetella pertussis: variables in the in vivo priming and the in vitro cytokine detection techniques affect the classification of T-cell subsets as Th1, Th2 or Th0. Immunology 1996;87:372–80. [20] Mills KHG, Barnard A, Watkins J, Redhead K. Cell-mediated

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

1127

immunity to Bordetella pertussis: role of Th1 cells in bacterial clearance in a murine respiratory infection model. Infection and Immunity 1993;61:399 – 410. Mills KHG, Ryan M, Ryan E, Mahon BP. 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. Infection and Immunity 1998;66:594 – 602. Redhead K, Watkins J, Barnard A, Mills KH. Effective immunization against Bordetella pertussis respiratory infection in mice is dependent on induction of cell-mediated immunity. Infection and Immunity 1993;61:3190 – 8. Alonso JM, Ravisse P, Brezin C. Bordetella pertussis respiratory infection in C57Bl/6 and Balb/c mice: pathophysiology and immune responses. FEMS Microbiology Letters 1987;40:273–7. Cherry JD, Brunnel PA, Goldon GS, Karzon DT. Report of the task force on pertussis and pertussis immunization-. Pediatrics ;81 1988;1988:939– 84. Pittman M, Furman BL, Wardlaw AC. Bordetella pertussis respiratory tract infection in the mouse: pathophysiological responses. Journal of Infectious Diseases 1980;142:56 – 66. Sato Y, Izumiya K, Sato H, Cowell JL, Manclark CR. Aerosol infection of mice with Bordetella pertussis. Infection and Immunity 1980;29:261 – 6. Relman DA, Domenighini M, Tuomanen E, Rappuoli R, Falkow S. Filamentous hemagglutinin of Bordetella pertussis: nucleotide sequence and crucial role in adherence. Proceedings of the National Academy of Sciences United States of America 1989;86:2637 – 41. Verwey WF, Thiele EH, Sage DN, Schuckhardt LT. A simplified liquid culture medium for the growth of Haemophilus pertussis. Journal of Bacteriology 1949;50:127 – 34. Peppoloni S, Nencioni L, Di Tommaso A, Tagliabue A, et al. Lymphokine secretion and cytotoxic activity of human CD4+ T-cell clones against Bordetella pertussis. Infection and Immunity 1991;59:3768 – 73. Podda A, Nencioni L, Marsili I, Peppoloni S, et al. Phase I clinical trial of an acellular pertussis vaccine composed of genetically detoxified pertussis toxin combined with FHA and 69 kDa. Vaccine 1991;9:741 – 5. Willems RJL, Kamerbeek J, Geuijen CAW, Top J, et al. The efficacy of a whole cell pertussis vaccine and fimbriae against Bordetella pertussis and Bordetella parapertussis infections in a respiratory mouse model. Vaccine 1998;16:410 – 6. Willems R, Paul A, van der Heide HGJ, ter Avest AR, Mooi FR. Fimbrial phase variation in Bordetella pertussis: a novel mechanism for transcriptional regulation. The EMBO Journal 1990;9:2803 – 9. van den Berg BM. Protection and humoral immune response against Bordetella pertussis infection in mice immunized with acellular pertussis vaccines and a whole-cell pertussis vaccine. In: Role of antibodies against virulence factors of Bordetella pertussis in the protection against infection: a study in vitro and in mice. Thesis ISBN 90-9012438-1: 1999, pp. 102 – 124. Guiso N, Capiau C, Carletti G, Poolman J, Hauser P. Intranasal murine model of Bordetella pertussis infection. I. Prediction of protection in human infants by acellular vaccines. Vaccine 1999;17:2366 – 76. Boursaux-Eude C, Thiberge S, Carletti G, Guiso N. Intranasal murine model of Bordetella pertussis infection: II. Sequence variation and protection induced by a tricomponent acellular vaccine. Vaccine 1999;17:2651 – 60. Cahill ES, O’Hagan DT, Illum L, Barnard A, Mills KHG, Redhead K. Immune response and protection against Bordetella pertussis infection after intranasal immunization of mice with filamentous haemagglutinin in solution or incorporated in biodegradable microparticles. Vaccine 1995;13:455 – 62.

1128

B.M. 6an den Berg et al. / Vaccine 19 (2001) 1118–1128

[37] Roberts M, Cropley I, Chatfield S, Dougan G. Protection of mice against respiratory Bordetella pertussis infection by intranasal immunization with P.69 and FHA. Vaccine 1993;11:866 – 72. [38] Sato H, Sato Y. Bordetella pertussis infection in mice: correlation of specific antibodies against two antigens, pertussis toxin, and filamentous hemagglutinin with mouse protectivity in an intracerebral or aerosol challenge system. Infection and Immunity 1984;46:415 – 21. [39] Capiau C, Carr SA, Hemling ME, Plainchamp D, et al. Purification, characterization, and immunological evaluation of the 69kDa outer membrane protein of Bordetella pertussis. In: Manclark CR, editor. Proceedings of the Sixth International Symposium on Pertussis. Bethesda, MD: Department of Health and Human Services, 1990:75 – 86 DHHS publication no (FDA) 90-1164. [40] Shahin RD, Witvliet MH, Manclark CR. Mechanism of pertussis toxin B oligomer-mediated protection against Bordetella pertussis respiratory infection. Infection and Immunity 1990;58:4063 – 8. [41] Jones DH, McBride BW, Thornton C, O’Hagan DT, Robinson A, Farrar GH. Orally administered microencapsulated Bordetella pertussis fimbriae protect mice from B. pertussis infection. Infection and Immunity 1996;64:489–94.

.

[42] Robinson A, Gorringe AR, Funnel SGP, Fernandez M. Serospecific protection of mice against intranasal infection with Bordetella pertussis. Vaccine 1989;7:321 – 4. [43] Shahin RD, Brennan MJ, Li ZM, Meade BD, Manclark CR. Characterization of the protective capacity and immunogenicity of the 69-kD outer membrane protein of Bordetella pertussis. Journal of Experimental Medicine 1990;171:63– 73. [44] van den Berg BM, Beekhuizen H, Mooi FR, van Furth R. Role of antibodies against Bordetella pertussis virulence factors in adherence of Bordetella pertussis and Bordetella parapertussis to human bronchial epithelial cells. Infection and Immunity 1999;67:1050 – 5. [45] Ausiello CM, Urbani F, La Sala A, Lande R, Cassone A. Vaccine- and antigen-dependent type 1 and type 2 cytokine induction after primary vaccination of infants with whole-cell or acellular pertussis vaccines. Infection and Immunity 1997;65:2168 – 74. [46] Zepp F, Knuf M, Habermehl P, Schmitt HJ, et al. Pertussis-specific cell-mediated immunity in infants after vaccination with a tricomponent acellular pertussis vaccine. Infection and Immunity 1996;64:4078 – 84.