Whole-cell pertussis vaccine protects against Bordetella pertussis exacerbation of allergic asthma

UNCORRECTED PROOF IMLET 4082 1–10 Immunology Letters xxx (2004) xxx–xxx Whole-cell pertussis vaccine protects against Bordetella pertussis exacerbation of allergic asthma 3 4 Darren P. Ennisaa, Joseph P. Cassidyb, Bernard P. Mahona,∗ 5 aMucosal Immunology Laboratory, Institute of Immunology, NUI Maynooth, Ireland6bDepartment of Veterinary Pathology, University College Dublin, Dublin, Ireland7 Received 23 August 2004; received in revised form 1 October 2004; accepted 4 October 2004 8 Abstract9 The prevalence of asthma and allergic disease has increased in many countries and there has been speculation that immunization promotes allergic sensitization. Bordetella pertussis infection exacerbates allergic asthmatic responses. We investigated whether whole-cell pertussis vaccine (Pw) enhanced or prevented B. pertussis induced exacerbation of allergic asthma. Groups of mice were immunized with Pw, infected with B. pertussis and/or sensitized to ovalbumin. Immunological, pathological and physiological changes were measured to assess the impact of Pw immunization on immune deviation and airway function. Pw immunization modulated ovalbumin-specific serum IgE production, and reduced local and systemic IL-13 and other cytokine responses to sensitizing allergen. Histopathological examination revealed Pw immunization reduced the severity of airway pathology and decreased bronchial hyperreactivity to methacholine exposure. Pw does not enhance airway IL-13 and consequently does not enhance but protects against the exacerbation of allergic responses. We find no evidence of Pw contributing to allergic asthma, but rather provide evidence of a mechanism whereby whole-cell pertussis vaccination has a protective role. 10 11 12 13 14 15 16 17 18 19 © 2004 Elsevier B.V. All rights reserved. 20 Keywords: Bordetella pertussis; Vaccine; Asthma; IL-13; Allergen 21 22 1. Introduction23 Asthma is a chronic disease of the respiratory tract of in-24 creasing prevalence in developed societies [1]. The current25 understandingofallergicasthmaisthatitresultsfromabreak-26 down in the normal tolerance to inhaled antigens, associated 27 with Th2 cytokine production [2,3]. The inflammatory re-28 sponse in asthma is tightly associated with airway hyperre-29 sponsiveness, increased mucus production and an infiltration30 of the bronchial mucosa with CD4+ T-cells [4]. There is ev-31 idence of an altered local T-cell response in favour of Th232 cytokinerelease(IL-4,IL-5andIL-13)resultinginB-celliso33 type switching to IgE, recruitment of eosinophils, basophils 34 andmastcellsandproductionofinflammatorymediators[5]. 35 ∗Corresponding author. Tel.: +353 1 7083835; fax: +353 1 7086337. E-mail address: [email protected] (B.P. Mahon). The murine OVA model of airway hyperresponsiveness ex36 hibits many of the features of human asthma, including air37 way hyperreactivity, inflammation and increased serum IgE 38 levels [6,7]. Th2 cells secreting IL-4, IL-5, and IL-13 play a 39 centralroleininitiatingandsustainingtheasthmaticresponse 40 in this model [8]. While Th2 cells promote airway inflamma41 tion in asthma, it has been proposed that Th1 cells protect 42 against allergic disease by antagonizing Th2 activity. Infec43 tious diseases that induce Th1 type responses, might hamper 44 the development of allergen-specific Th2 cells and prevent 45 allergy [9].46 Epidemiologicalandclinicalstudieshavesuggestedalink 47 between the relative absence of infectious diseases and the 48 increase in allergic disorders [10,11]; this is referred to as the 49 ‘hygienehypothesis’.Itpredictsthatinfectionspreventthein50 duction of allergen-specific Th2 cells through antagonism or 51 theinductionofregulatoryT-cells,particularlyduringneona52 10165-2478/$ – see front matter © 2004 Elsevier B.V. All rights reserved. 2doi:10.1016/j.imlet.2004.10.011 UNCORRECTED PROOF IMLET 4082 1–10 2D.P. Ennisa et al. / Immunology Letters xxx (2004) xxx–xxx tal and early childhood development [10,12]. However, there 53 are data that confound this interpretation, increased IFN-␥54 is seen in asthmatic patients compared with normal subjects55 [3,13]. Allergen-specific Th1 cells also fail to counteract air56 way hyperresponsiveness in murine models [14]. Further-57 more, several studies have suggested that viral/bacterial in-58 fections do not protect but exacerbate disease. Respiratory 59 syncytial virus, commonly associated with lower lung in-60 fections in infancy, is known to exacerbate asthma [15,16].61 As does Bordetella pertussis [17]. Consequently, competing 62 interpretations for the pathogenesis of asthma have been pro-63 posed [18,19].64 B. pertussis is a Gram-negative bacterium and the 65 causative agent of pertussis or “whooping cough”, a respi66 ratory disease that remains a significant cause of morbidity67 and mortality in infants worldwide. It is a highly contagious 68 disease, and can occur at any age, though severe illness is69 more common in young un-immunized children. B. pertussis 70 infection induces Th1 responses [20,21] and can be mod-71 elled by respiratory challenge of mice, which correlates well72 to responses in humans [22].73 There has been speculation about the possible promotion74 of allergy by common childhood vaccinations [23,24].A75 substantial proportion of children predisposed to allergy and 76 asthma may not be fully immunized because of public appre-77 hension surrounding immunization [25]. A number of stud78 ies have analysed the prevalence of allergic sensitization and 79 atopic disease in relation to immunization [24,26]. Gruber et 80 al. found that children with higher immunization coverage 81 seemed to acquire transient protection against development82 of atopy in the first years of life [26]. In contrast, Hurwitz83 and Morgenstern suggested that diphtheria/pertussis/tetanus84 (DTP) immunization appeared to be associated with an85 increased risk of subsequent asthma or other allergies86 [24]. 87 Two different types of pertussis vaccine have been em88 ployed in infant immunization programmes. The whole-cell 89 pertussis vaccine (Pw) consists of heat/formalin inactivated 90 virulent whole bacteria whereas the pertussis acellular vac91 cine (Pa) is composed of purified components of the bacteria92 (Pa), typically including inactivated pertussis toxin. Pw im-93 munization has a high efficacy and is associated with the94 induction of antigen-specific Th1 cells [21,27,28], but has95 been associated with reactogenicity. In contrast Pa immu96 nization induces a mixed Th1/Th2 response in children and 97 in murine models, but has reduced reactogenicity [29].Ithas 98 been suggested that promotion of allergy may occur directly, 99 by administering potentially pro-allergic vaccines, or indi-100 rectly, by hindering the Th1-promoting effect of infectious101 agents. Pertussis vaccination acts as an adjuvant for antigen-102 specificresponsesin laboratoryanimals[30];activepertussis103 toxin, is known to enhance immunoglobulin E (IgE) forma-104 tion in animal models [31] and has been linked with a shift105 toward Th2-like cytokines in humans [32,33]. 106 InfectionwithB.pertussismodulatesallergenprimingand107 the severity of airway pathology in a murine model of al108 lergic asthma [17] and we have previously shown that Pw 109 immunization induces a similar immune response to infec110 tion [34] and that although variables such as route, dose 111 and timing influence T-cell responses in animal models, Pw 112 is a consistent inducer of Th1 responses [20]. In order to 113 test whether immunization with Pw exacerbated asthma, we 114 employed a well-characterized murine model of whole-cell 115 pertussis vaccination and B. pertussis infection in combina116 tion with the murine OVA model of airway hyperresponsive117 ness. We show that although Pw induces a Th1 type im118 mune response to B. pertussis infection, it does not exac119 erbate pathology in a model of allergic asthma. Our find120 ings demonstrate that Pw immunization prevents B. pertus121 sis enhancement of OVA-induced IL-10 and IL-13, which 122 results in a subsequent decrease in airway hyperresponsive123 ness and pathology. This study finds no evidence of a mech124 anism to support speculation linking Pw immunization and 125 asthma. 126 2. Materials and methods 127 2.1. Animals and experimental approach 128 Sixto 8-week-old female BALB/c (Harlan, UK) mice 129 were used under the guidelines of the Irish Department 130 of Health and the research ethics committee of the Na131 tional University of Ireland Maynooth. The experimental 132 approach is outlined in Table 1, briefly groups of mice 133 were immunized with whole-cell pertussis vaccine (Pw), 134 infected with B. pertussis, and then sensitised to ovalbu135 min (OVA) at the peak of infection as detailed below. Con136 trol mice received similar treatment in which 0.9% (w/v) 137 (aq) NaCl (hereafter termed Saline) replaced experimental 138 treatment. 139 2.2. Immunization, sensitization and airway delivery of 140 OVA 141 Four groups of at least thirty-five 6–8-week-old female 142 BALB/c mice (Pw, PwBp, PwOVA and PwBpOVA) were 143 immunized i.p. with 0.16I.U. of whole-cell pertussis vaccine 144 (Pw) (Third International Standard, 1998, pertussis whole145 cell vaccine, NIBSC, UK), equivalent to 1/25th of the hu146 man dose according to the schedule outlined in Table 1.At 147 0 day mice were infected with B. pertussis, selected groups 148 were then sensitized with ovalbumin (OVA). Sensitization 149 involved 100␮g OVA (Grade V; Sigma, Dorset, UK) emul150 sified in Alhydrogel®adjuvant (Superfos Biosector, Swe151 den) (1mg/mouse aluminium hydroxide) administered as 152 0.2ml i.p. at 10 and 24 days. Control group (Ctrl) received 153 saline alone (i.p.). On 35, 36, and 37 days, PwOVA and 154 PwBpOVA sensitized mice received 10␮l containing 50␮g155 OVA intra-nasally (i.n.) whereas remaining groups received 156 saline only (Table 1). All experiments were repeated at least 157 twice. 158 UNCORRECTED PROOF IMLET 4082 1–10 D.P. Ennisa et al. / Immunology Letters xxx (2004) xxx–xxx 3 Table 1 Experimental design Time (days) −42 −14 0 10 24 35/36/37 Groupa Pw Pw Pw Saline aerosol Saline (i.p.) Saline (i.p. and i.n.) Saline (i.n.) Pw and B. pertussis infection (PwBp) Pw Pw B. pertussis aerosol infection Saline (i.p.) Saline (i.p. and i.n.) Saline (i.n.) Pw and OVA sensitization (PwOVA) Pw Pw Saline aerosol OVA (i.p.) OVA (i.p and i.n.) OVA (i.n.) Pw, B. pertussis infection and OVA sensitization (PwBpOVA) Pw Pw B. pertussis aerosol infection OVA (i.p.) OVA (i.p and i.n.) OVA Groupb Control (Ctrl) Saline (i.p.) Saline (i.p.) Saline aerosol Saline (i.p.) Saline (i.p. and i.n.) Saline (i.n.) B. pertussis infection (Bp) Saline (i.p.) Saline (i.p.) B. pertussis aerosol infection Saline(i.p.) Saline (i.p. and i.n.) Saline (i.n.) OVA sensitization (OVA) Saline (i.p.) Saline (i.p.) Saline aerosol OVA (i.p.) OVA (i.p and i.n.) OVA (i.n.) B.pertussis infectionandOVAsensitization (BpOVA) Saline (i.p.) Saline (i.p.) B. pertussis aerosol infection OVA (i.p.) OVA (i.p and i.n.) OVA (i.n.) aGroups of 6–8-week-old female BALB/c mice were immunized (i.p.) and boosted with whole-cell pertussis vaccine (Pw) at −42 and −14 days. On 0 day, mice were either sham infected or infected with B. pertussis (Bp) by aerosol. At 10 and 24 days selected groups were sensitised to OVA by i.p alone or i.p and i.n. routes, respectively. On 35, 36 and 37 days mice were exposed to either saline or OVA by the i.n. route (i.e. 25, 26, and 27 days post-OVA priming and after bacterial clearance). Bacterial burdens in the airways were measured between 0 and 37 days. All other readouts, including plethysmography were performed at 37 days. bFor comparison, further groups of unimmunized (sham immunized) mice were treated as follows: control (Ctrl) mice were sham infected on 0 day and sham sensitised with saline. A second group (Bp) were infected with B. pertussis at 0 day and sham sensitized. The third group (OVA) were sham infected but sensitized with OVA (100␮g, i.p.) at 10 and 24 days and again (50␮g i.n.) at 24, 35, 36, and 37 days. A separate group (BpOVA) were infected with B. pertussis on 0 day, and sensitized as above. Each experiment was repeated at least twice, on each occasion n>35 mice per group, bacterial burdens were measured between 0 and 37 days, all other readouts were at 37 days. 2.3. B. pertussis aerosol infection 159 Respiratory infection was initiated by aerosol challenge160 with B. pertussis strain W28, following growth under agita-161 tion conditions at 37◦C in Stainer-Scholte liquid medium.162 Bacteria from a log-phase culture were resuspended at a con-163 centration of 2×1010 CFU/ml in 1% (w/v) casein in 0.9%164 (w/v) saline. The challenge inoculum was administered to165 groups of mice on 0 day (Bp, PwBp and PwBpOVA groups).166 Administration was by aerosol over a period of 15min using 167 a nebulizer. Groups of four or more mice were killed at vari168 ous time points after aerosol challenge to assess the number 169 of viable B. pertussis in the lungs. Remaining mice received 170 a similar aerosol of sterile saline alone. 171 2.4. Enumeration of viable bacteria in the lungs 172 Lungs were removed aseptically into 1ml of sterile phys-173 iological saline with 1% casein. Hundred microlitres of seri-174 ally diluted homogenate from individual lungs were placed175 onto triplicate Bordet-Gengou agar plates and the number of176 CFU determined after incubation at 37◦C for 4 days. Re-177 sults are reported as the mean number of B. pertussis CFU178 (±S.E.M.)forindividuallungs,eachdeterminedintriplicate,179 from four or more mice per time point. All experiments were180 repeated twice. 181 2.5. Bronchoalveolar lavage 182 Bronchoalveolar lavage fluids (BALF) were obtained by183 cannulation of the trachea followed by repeat administration184 and aspiration of 0.5ml PBS per mouse. This was pooled 185 from five mice (total 2.5ml) per experimental group. All ex186 periments were performed at least twice. Diluted BALF was 187 assessed for the presence of cytokines. 188 2.6. Measurement of OVA and B. pertussis-specific 189 antibody 190 OVA and B. pertussis-specific IgG1, 2a, 2b, and 3 present 191 in collected sera were measured on day 37 by ELISA as pre192 viously described [35,36]. Total and OVA-specific IgE was 193 measured using a rat anti-mouse IgE monoclonal antibody 194 (BD, Pharmingen, San Diego, CA, USA). The IgE concen195 tration was expressed as ␮g/ml after comparison to murine 196 IgE standards. 197 2.7. T-cell proliferation assays 198 Spleen cells (2×106/ml) from infected, sensitized and 199 control mice (n=4 or more per group) were tested for 200 in vitro proliferation against heat-inactivated B.pertussis 201 (1×104CFU/ml), OVA (20␮g/ml), Concanavalin A (Con 202 A) (5␮g/ml, positive control), or medium alone (negative 203 control). After 72h, cell proliferation was assessed by liquid 204 scintillation counting of [3H]-thymidine incorporation and 205 resultswereexpressedasmeanCPMoftriplicatewells±S.E. 206 At the 72h time point, culture supernatants were sampled for 207 cytokine analysis, although the kinetics of cytokine produc208 tion varies this time point has previously proved acceptable 209 for detection of most cytokines [22].210 UNCORRECTED PROOF IMLET 4082 1–10 4D.P. Ennisa et al. / Immunology Letters xxx (2004) xxx–xxx 2.8. Cytokine measurement 211 Concentrations of IL-5, IL-10, IL-13 and IFN-␥from 212 spleen, and BALF were assessed by ELISA (BD, Pharmin-213 gen, San Diego, CA, USA). Cytokine concentrations were 214 calculated by comparison with known cytokine standards as215 previously described [35], all determinations were made in216 triplicate, results are presented as mean cytokine concentra-217 tion (±S.E.M.). 218 2.9. Whole body plethysmography 219 Airwayresponsivenesson37dayswasassessedbymetha220 choline (MCh) induced airflow obstruction from conscious221 mice using whole-body plethysmography (Buxco Electron-222 ics, Sharon, CT, USA) as previously described [37]. Pul223 monaryairflowobstructionwasmeasuredbyenhancedpause224 (PenH), a value determined from the ratio of expiratory time225 and relaxation time to peak expiratory flow and peak inspi-226 ratory flow and thought to correlate with airway responsive-227 ness. Measurements were obtained after exposure of mice228 for 3min to PBS (baseline) followed by incremental doses 229 (3.3mg–50mg/ml) of MCh delivered by aerosol [38]. 230 2.10. Respiratory tract histology231 Animals (n=5 per group per experiment) were sacrificed232 at 37 days. Lungs were removed, fixed in a paraformalde-233 hyde/lysine/periodate fixative, paraffin embedded, sectioned234 and stained using the haematoxylin and eosin (H&E), Dis-235 combes (identification of eosinophils), alcian blue (iden236 tification of mucus), PAS (assessment of basement mem237 brane thickness), azure-A (identification of mast cells) and 238 Van Gieson (identification of fibrosis) methods. Histopatho239 logical changes evident were graded according to a semi-240 quantitative scoring system as mild, moderate or severe by241 two researchers without prior knowledge of the treatment242 group using a previously established scoring system [17].243 All experiments were performed at least twice.244 2.11. Statistical methods 245 Results are expressed as the mean±S.E.M. of the indi-246 catednumberofanimals.AStudent’st-testwasusedtodeter247 mine significance among the groups. A value of P<0.05 was248 considered significant. Analyses were performed using the249 Graph-Pad PrismTM software (GraphPad, San Diego, CA).250 3. Results 251 3.1. Ovalbumin sensitization does not impair 252 vaccine-mediated clearance of B. pertussis253 The murine OVA model of airway hyperresponsiveness 254 induces a powerful Th2 response [7] whereas both B. per-255 Fig. 1. Course of B. pertussis infection in experimental and control mice. Groups of mice were sacrificed at intervals after challenge and the number of viable bacteria estimated by performing colony counts on individual lung homogenates. Results are representative from two experiments and are presented as mean (±S.E.M.) CFU in the lungs, determined individually from four mice at each time point, and for each experimental group. Data for Ctrl and OVA groups have been offset from zero for clarity. tussis infection and Pw immunization induce a powerful Th1 256 response [29]. In order to examine immune cross-regulation 257 and potential interference with immunization, we tested the 258 effect of OVA sensitization upon the development of a pro259 tective response to infection in Pw immunized and non260 immunized mice. Mice received combinations of OVA sen261 sitization, Pw immunization and aerosol challenge with vir262 ulent B. pertussis (Table 1). Groups of mice infected with 263 B. pertussis (Bp and BpOVA) showed similar kinetics of 264 bacterial clearance (Fig. 1), indicating that OVA sensitiza265 tion does not influence bacterial clearance rates. Likewise, 266 OVA sensitized and non-sensitized mice that had been im267 munized prior to bacterial challenge (PwBpOVA and PwBp, 268 respectively) showed identical kinetics of clearance. No bac269 teria were recovered from the OVA sensitised or control 270 (Ctrl) groups, which were uninfected but received saline by 271 aerosol (Fig. 1). The bacterial burden in the Bp and BpOVA 272 groups peaked at 10 days and declined thereafter. Pw im273 munized mice cleared subsequent infection by B. pertus274 sis by 7 days. In contrast unimmunized mice (Bp and the 275 BpOVA groups) only showed complete bacterial clearance 276 by 35 days (Fig. 1). Therefore sensitization with OVA did 277 not impair vaccine-mediated clearance of B. pertussis in this 278 model. 279 3.2. OVA-specific IgE production is modulated by Pw 280 immunization 281 The goal of this study was to examine the influence of 282 Pw immunization on responses associated with allergic sen283 sitization. Although OVA-induced sensitization does not im284 pair vaccine-mediated clearance of B. pertussis, it was pos285 sible that Pw influenced allergic sensitization. OVA-specific 286 IgG was not detected from mice infected with B. pertussis 287 UNCORRECTED PROOF IMLET 4082 1–10 D.P. Ennisa et al. / Immunology Letters xxx (2004) xxx–xxx 5 Fig. 2. Serum IgE and IgG subclasses elicited by Pw vaccination, bacterial infection and allergic sensitization. (A) B. pertussis or (B) OVA-specific serum antibody responses by IgG subclasses elicited in whole-cell B. pertussis (Pw), vaccinated plus Bp infection (PwBp), vaccinated plus sensitized (PwOVA), or in mice vaccinated then infected with B. pertussis prior to sensitization (PwBpOVA), response from mice that received OVA sensitization alone (OVA) or combined with infection (BpOVA) are shown for comparison. (C) B. pertussis-specific serum IgE and (D) OVA-specific serum IgE present from each experimental group. Results are representative of two experiments expressed as geometric mean titre or ng/ml of antibody (±S.E.M.) from four animals each determined independently in triplicate. *P<0.05 compared to the PwOVA treated group. only; similarly B. pertussis-specific IgG could not be de-288 tected in OVA sensitized animals, suggesting no significant289 cross-reaction between the two immunogens (Fig. 2A and 290 B). An analysis of the serum antibody subclasses evoked291 revealed that Pw predominantly induced pertussis-specific292 IgG2a (Fig. 2A) as previously reported [28,34]. IgG2a re-293 mained the dominant subclass of pertussis-specific anti294 body when Pw immunization was combined with OVA sen-295 sitization (PwOVA) or B. pertussis infection (PwBp) al296 though IgG1 and IgG3 was detected in the latter groups297 (Fig. 2A). Pw immunization prior to OVA sensitization298 (PwOVA) did not significantly alter the dominant OVA-299 specific IgG1 response. Infection of these mice with B. per-300 tussis (PwBpOVA) did not broaden the subclasses of OVA-301 specific antibody detected although the titer was increased302 (Fig. 2B). Taken together, these data indicate that Pw im-303 munization does not modulate serum IgG subclasses in-304 duced by allergic sensitization. In contrast to the conser-305 vative effects on serum IgG subclasses, significant differ-306 ences were observed in the induction of IgE (Fig. 2C and 307 D). Pw induces little B. pertussis-specific IgE, and this is 308 not altered by OVA sensitization or infection (Fig. 2C). 309 OVA sensitization induces high levels of OVA-specific IgE 310 but this is significantly reduced by prior Pw immunization 311 (Fig. 2D). However, a combination of immunization and in312 fection prior to OVA sensitization resulted in a significant re-313 duction of IgE (P<0.05) (Fig. 2D) compared to the PwOVA 314 group.315 3.3. Pw immunization prevents B. pertussis 316 enhancement of OVA-induced IL-10 and IL-13 317 B. pertussis infection enhances OVA-induced IL-10 and 318 IL-13 [17]. Pw immunization has hitherto been regarded as 319 inducing essentially similar immune responses to those in320 duced by infection [22]. In order to dissect the influence of 321 immunization on airway hyperresponsiveness, we examined 322 cell-mediatedimmune responses inthe variousstudy groups. 323 Pw immunization alone or in combination with B. pertussis 324 infection (Pw or PwBp) induced very little IL-5 but strong 325 IFN-␥responses (Fig. 3A and B). This was consistent with 326 the protection observed earlier (Fig. 1) and previous data 327 [34]. Pw immunization reduced levels of IL-5, IL-13 and 328 IFN-␥(Fig. 3A–C) in all immunized groups suggesting that 329 Pw immunization prevents live B. pertussis enhancement of 330 these indices; mirroring the reduction in OVA-specific IgE 331 (Fig. 2D). Interestingly, previous results have shown that B. 332 pertussis infection induced specific IL-10 as well as IL-13 333 responses [17]. Here we demonstrate that in contrast to in334 fection, Pw-immunization resulted in significantly reduced 335 levels of IL-10, and IL-13 (Fig. 3A–D). 336 To extend these findings, we examined the levels of cy337 tokines present in bronchoalveolar lavage fluid (BALF) from 338 each group of mice. Pw immunization alone induced little or 339 no detectable cytokines in BALF. As expected, OVA sensiti340 zation induced high levels of IL-5, -10 and -13 but the levels 341 of IL-10 and -13 in particular, known to rise in infected mice, 342 UNCORRECTED PROOF IMLET 4082 1–10 6D.P. Ennisa et al. / Immunology Letters xxx (2004) xxx–xxx Fig.3. Cellmediatedimmuneresponsesfromspleen,elicitedbyPwvaccination,bacterialinfectionandallergicsensitization.IL-5(A),IFN-␥(B),IL-13(C)and IL-10 (D) responses from spleen cell cultures stimulated with medium alone (−ve control, horizontal shading), heat inactivated B. pertussis at 1×104CFU/ml (hatched bar), OVA (open bar) or Con A (+ve control, black bar). Responses are representative of duplicate experiments each performed in triplicate on individual samples from four mice per group and are expressed as mean (±S.E.M.). *P<0.05 compared to the PwOVA treated group. were reduced in immunized mice that had been infected with343 B. pertussis (PwBpOVA) (Fig. 4A–D).344 3.4. Pw immunization prior to B. pertussis infection345 decreases bronchial hyperresponsiveness to sensitizing346 antigen 347 It has been proposed that prior Th1 responses to bacte348 rial infections protect against allergic disease however, Th1-349 inducing B. pertussis infection exacerbates airway hyperre-350 sponsiveness in OVA sensitized mice. It might be predicted351 that Pw, which induces a very similar immune response to B.352 pertussiswouldhaveasimilarexacerbatinginfluence.Infact, 353 this is not the case. We used whole body plethysmography as 354 a surrogate measure of airway reactivity in mice immunized 355 with Pw and infected with B. pertussis prior to OVA sensiti356 zation(Fig.5).PriorimmunizationwithPwdoesnotenhance 357 but protects against B. pertussis exacerbated airway hyperre-358 sponsiveness in comparison to controls. Statistical analysis 359 using two-way analysis of variance (ANOVA) showed that360 mice vaccinated with Pw, and sensitized to OVA following 361 B.pertussisinfection(PwBpOVA)displayedsignificantlyre-362 duced bronchial hyperreactivity compared to BpOVA sensi363 tized animals (P<0.05) (Fig. 5D). Thus demonstrating that 364 vaccination with Pw protects against B. pertussis exacerba365 tion of allergic asthma. 366 B. pertussis infection is known to modulate the quality 367 of the inflammatory influx of the respiratory tract, with a 368 marked reduction in eosinophil numbers accompanied by 369 varying degrees of epithelial hyperplasia, mucus metaplasia, 370 and airway pathology [17]. Lung tissue was assessed histo371 logically (Table 2). Minimal pathology was observed in mice 372 immunized with Pw or those immunized and infected with 373 B. pertussis (PwBp) (Fig. 6A and B). Pw and OVA sensitized 374 (PwOVA) mice illustrated moderate mural and peri-airway 375 inflammationwithaccompanyingmildmucusmetaplasiaand 376 moderate hyperplasia of the epithelium (Fig. 6C). The com377 binationofPwimmunization,B.pertussisinfectionandOVA 378 sensitizationdid notshowenhancedpathology butonlymod379 erate mucus metaplasia and moderate hyperplasia of the ep380 ithelium (Fig. 6D). Given that previous work has shown that 381 B. pertussis infection in combination with OVA sensitization 382 (BpOVA) displayed more severe airway inflammation with 383 a greater degree of both epithelial hyperplasia and mucous 384 metaplasia, it can be clearly seen here that Pw immuniza385 tion reduces the severity of airway pathology (Fig. 6D) and 386 UNCORRECTED PROOF IMLET 4082 1–10 D.P. Ennisa et al. / Immunology Letters xxx (2004) xxx–xxx 7 Fig. 4. Pw immunization modulates the local cytokine response to B. pertussis infection and OVA sensitization. Diluted BALF (0.5ml per mouse) was pooled from five mice per group and concentrations of IL-5 (A), IFN-␥(B), IL-13 (C) and IL-10 (D) were determined by EIA. Results are representative of duplicate experiments. Cytokine concentrations were assayed in triplicate with values expressed as mean cytokine concentration in diluted BALF (±S.E.M.). *P<0.05 compared to the PwOVA treated group. reduces pulmonary resistance as indicated by plethysmogra387 phy (Fig. 5D). 388 4. Discussion389 The present study demonstrates that Pw immunization 390 protects against B. pertussis exacerbation of OVA-induced391 Table 2 Histological assessment of airway pathology Treatment group Mucous metaplasia of airway epithelium Hyper-plasia of airway epithelium Smooth muscle hypertrophy of airway wall Peri-airway/vascular inflammationa Overall degree E N L M F Ctrl −−−−−−−−− Bp −+++−+++−− Ova++++++++ b+++++ c−− BpOVA +++ +++ ++ +++b++ +++ +++c−− Pw −−−−−−−−− PwBp −++−−−−−− PwOVA + ++ + ++b++ ++ ++c−− PwBpOVA ++ ++ ++ ++b++ ++ ++c−− Asemi-quantitative score(−absent, + mild, ++ moderate,+++ severe)wasassigned tofeatures of airway pathologyobserved according to previously described criteria [17]. aPeri-airway/vascular inflammation was assessed in terms of overall degree and of numbers of infiltrating eosinophils (E), neutrophils (N), lymphocytes, plasma cells and macrophages (L), mast cells (M) and in terms of circumscribing fibrosis (F). bInflammation extending into surrounding pulmonary interstitium and alveolar spaces. cMacrophage giant cells form part of inflammatory exudates within surrounding alveolar spaces. airway hyperresponsiveness in a murine model. Pw immu392 nization suppresses antibody and cell mediated responses 393 against OVA sensitization in combination with B. pertussis 394 infection at both the local and systemic level, coupled with 395 a subsequent reduction in airway reactivity and pathology. 396 It has been previously shown that B. pertussis exacerbates 397 OVA-induced airway pathology leading to the development 398 of more pronounced allergen-induced airwayinflammation UNCORRECTED PROOF IMLET 4082 1–10 8D.P. Ennisa et al. / Immunology Letters xxx (2004) xxx–xxx Fig. 5. Pw immunization prior to B. pertussis infection decreases bronchial hyperresponsiveness to sensitizing antigen. Airway hyperreactivity in response to increasing concentrations of inhaled methacholine (MCh) was measured by whole-body plethysmography. (A) Control and Bp infected mice (B), Pw and PwBp (C), OVA and PwOVA (D), BpOVA and PwBpOVA groups, respectively. Results are representative of two experiments (n=4) and values are expressed as mean enhanced pause (PenH)±S.E.M., in groups where no errors are visible, error bars are shorter than the size of the data point symbol. [17]. We demonstrate that Pw immunization protects against399 this as well as reducing airway hyperresponsiveness.400 The prevalence of asthma and allergic disease has in401 creased in many countries [39,40] and there has been spec402 ulation as to possible causes [41,42], including the possible 403 roleofimmunizationinpromotingallergicsensitization[43]. 404 For example, pertussis vaccination acts as an adjuvant for 405 antigen-specific responses in laboratory animals [30,33];a406 specific IgE response to pertussis toxin itself has been iden-407 tified in children receiving pertussis immunization [44]; and408 vaccination with some other organisms such as Haemophilus409 influenzae enhances histamine release in laboratory animals 410 [45]. Active pertussis toxin has a similar effect [46].Inad411 dition, two studies have suggested that pertussis infection 412 increased the risk of atopy [47,48]. It is therefore theoreti413 cally possible that Pw immunization might contribute to the414 development of allergic disease.415 The goal of this study was to test whether immunization416 with whole-cell pertussis (Pw) vaccine would protect against417 B. pertussis exacerbation of allergic asthma. B. pertussis in-418 fectionmodulatesallergenprimingandtheseverityofairway419 pathology in a murine model[17]. It has been proposed that 420 IL-10 plays an essential role in modulating the immune re-421 sponses by inducing towards regulatory T-cell responses[49] 422 howeverLeeetalhavedemonstratedthatIL-10inducesIL-13 423 production in vivo and that this is responsible for the mucus, 424 but not the inflammatory/fibrotic effects of IL-10 [50].In 425 the present study, we observe that Pw immunization prevents 426 induction of IL-10 and IL-13 and protects against airway hy427 perreactivity. Although IL-10 is known to act in an immune 428 regulatory manner, we and others have suggested that it has 429 broaderfunctionsthatmaynotalwaysprotectagainstinflam430 matory disease [17,51]. For example, Grunstein et al. have 431 suggested that IL-10 may play an important role in allergic 432 asthma by acting directly on the airway smooth muscle it433 self [51]. Previous studies in humans have demonstrated that 434 IL-13 mRNA and protein levels are elevated in the lungs of 435 atopic and non-atopic asthmatics [52] suggesting that over 436 expression of IL-13 may predispose toward the development 437 of both types of asthma [53]. The reduction in both IL-10 438 and IL-13 at the systemic and local levels suggests that by 439 removingthedamagingeffectsofpertussisinfectionandcon440 sequently reducing IL-10 and IL-13, Pw exerts a protective 441 effect.Interestingly,veryrecent workbyKim et al has shown 442 that components of B. pertussis can inhibit airway hyeperre443 sponsiveness [54]. That study demonstrated that unmethy444 lated CpG sequences from B. pertussis DNA inhibited Th2 445 cytokines in the airways via a TLR9 interaction [54]. Our 446 UNCORRECTED PROOF IMLET 4082 1–10 D.P. Ennisa et al. / Immunology Letters xxx (2004) xxx–xxx 9 Fig. 6. Whole-cell B. pertussis vaccine reduces the severity of airway pathology to sensitizing antigen both in the presence and absence of B. pertussis infection. Representative morphological changes at 37 days in transverse sections of bronchioles from (A) Pw immunized mice showing no changes evident; (B) Pw/Bp mice, minimal changes evident; (C) Pw/OVA treated mice, illustrating moderate mural and peri-airway inflammation with accompanying mild mucous metaplasia (blue staining goblet cells) and moderate hyperplasia of epithelium; (D) Combined Pw/Bp/OVA treated group, illustrating a moderate mural andperi-airwayinflammation, moderatemucousmetaplasiaandmoderateepithelial hyperplasia; (E)OVAsensitizedgroup illustrating moderatemural andperiairway inflammation with accompanying moderate mucus metaplasia (blue staining goblet cells) and hyperplasia of epithelium; (F) combined B. pertussis/OVA treated group illustrating severe mural and peri-airway inflammation, moderate epithelial hyperplasia and severe mucous metaplasia with accompanying mucus plugging of the lumen. All sections are representative of groups of five mice per experiment, performed at least twice. Sections stained with a combined Discombe’s/Alcian blue stain, original magnification 400×. observation of reduced inflammation and OVA-specific IgE 447 (Figs. 2 and 6) in PwBpOVA mice is consistent with that 448 finding. TLR9 mediated reduction in Th2 cytokines can be449 invoked in this situation, because although these mice will be450 exposedtobacterialDNAinacontextwherepriorimmuniza-451 tion will limit tissue damage or the other immunomodulatory452 effects of viable B. pertussis [17,54].453 The most common formulation of the hygiene hypothesis454 is based upon a lack of immunological stimuli in early in-455 fancy that results in aberrant Th2 responses [10,12]. Clearly456 infant pertussis immunization or infection could potentially457 influence this process. Studies in neonatal mice support our458 findingsthat PwisastrongTh1inducer,butitis lesseffective 459 ininducingantibodyresponses during this period [28,55,56].460 A study by Gruber et al. revealed no evidence for an allergy 461 promotingeffectofcommonchildhoodvaccinesinaprospec462 tively followed atopy risk-enhanced birth cohort [26]. More463 over, they found that children with a better vaccination cov464 erage seemed to be better protected against the development 465 of atopy in their second and third years of life. In particu466 lar, measles/mumps, pertussis, and diphtheria/tetanus immu467 nization were associated with a transient reduction of atopy, 468 whereas immunization against polio and H. influenzae had 469 no effect. Furthermore, immunization of children with Pw 470 down regulated the IgE response to co-administered diphthe471 ria and tetanus toxoids [57]. In contrast, a study involving the 472