Pertussis Toxin Improves Immune Responses to a CombinedPneumococcal Antigen and Leads to Enhanced Protection againstStreptococcus pneumoniae

Carolina Salcedo-Rivillas,a Anne-Sophie Debrie,b,c,d,e Eliane Namie Miyaji,a Jorge M. C. Ferreira, Jr.,f Isaías Raw,a Camille Locht,b,c,d,e

Paulo L. Ho,a,g Nathalie Mielcarek,b,c,d,e Maria Leonor S. Oliveiraa

Centro de Biotecnologia,a Divisão de Desenvolvimento Industrial e Produção,g and Laboratório de Imunoquímica,f Instituto Butantan, São Paulo, Brazil; Center forInfection and Immunity of Lille, Institut Pasteur de Lille, Lille, Franceb; INSERM U1019, Lille, Francec; CNRS UMR8204, Lille, Franced; Université Lille-Nord de France, Lille,Francee

Pneumococcal surface protein A (PspA) is a candidate antigen for the composition of protein-based vaccines against Streptococ-cus pneumoniae. While searching for efficient adjuvants for PspA-based vaccines, our group has described the potential of com-bining PspA with the whole-cell pertussis vaccine (wP). When given to mice through the nasal route, a formulation composed ofPspA from clade 5 (PspA5) and wP (PspA5-wP) induced high levels of antibodies and protection against challenges with differ-ent pneumococcal strains. PspA5-wP also induced the secretion of interleukin 17 (IL-17) by splenocytes and the infiltration ofleukocytes in the lungs after challenge. Here, we show that protection against a pneumococcal invasive challenge was completelyabrogated in �MT�/� mice, which are deficient in the maturation of B cells, illustrating the importance of antibodies in the sur-vival elicited by the PspA5-wP vaccine. Moreover, passive immunization showed that IgG purified from the sera of mice immu-nized with PspA5-wP conferred significant protection to naive mice, whereas the respective F(ab=)2 did not. Additionally, in vivodepletion of complement abolished protection against the pneumococcal challenge. The combination of PspA5 with wild-type ormutant Bordetella pertussis strains or with purified components showed that the pertussis toxin (PT)-containing formulationsinduced the highest levels of antibodies and protection. This suggests that the adjuvant activity of wP in the PspA5 model is me-diated at least in part by PT. The sera from mice immunized with such formulations displayed high IgG binding and induction ofcomplement deposition on the pneumococcal surface in vitro, which is consistent with the in vivo results.

Streptococcus pneumoniae is an important cause of noninvasiveinfections, such as pneumonia and otitis media, as well as of

invasive diseases, such as bacteremia and meningitis. The mostaffected population is children �5 years old, and estimates for theyear 2000 accounted for around 1 million deaths in this age grouparound the world (1). Pneumococcal conjugated vaccines havegreatly contributed to the decrease in this disease incidence inseveral countries (2, 3). However, epidemiologic studies in vacci-nated populations have shown changes in the prevalences of sero-types, which may account for the decrease in vaccine efficacy aftera period of use (4, 5).

Pneumococcal surface protein A (PspA) is a virulence factorthat mediates evasion of the immune system by inhibiting thedeposition of complement on the pneumococcal surface as well asthe bactericidal activity of apolactoferrin present on mucosal sur-faces (6, 7). Several proposals of protein-based vaccines as alter-natives to conjugated vaccines include PspA. PspA-based vaccineswere shown to be very effective against pneumococcal infectionsin animal models (8). The N-terminal part of PspA is exposed atthe bacterial surface and contains protective epitopes (9, 10).However, this region also shows sequence variability betweenstrains, and a portion at the end of the N-terminal region (theclade-defining region) is the basis for classifying PspAs in sixclades that can be grouped into three families (11). More than99% of the pneumococcal isolates around the world express PspAsfrom families 1 and 2 (12–14). Cross-reactivity between cladesfrom the same family is observed (15, 16), suggesting that usingone member from each family may be sufficient for designing abroad-coverage vaccine. In addition, some molecules, such as the

PspA from clade 5 (PspA5) used in this work, were shown toinduce antibodies with even broader cross-reactivity, as they canrecognize molecules from different families (17, 18). We haveshown that nasal immunization of mice with a formulation com-posed of PspA5 and a whole-cell pertussis vaccine (wP), used as anadjuvant, protects animals against challenges with different pneu-mococcal strains (19). Combining PspA5 with wP offers the ben-efit of the adjuvant properties of a vaccine administered to chil-dren at 2, 4, and 6 months in many countries in the world, withboosters at 15 months and 4 years of age (20).

The adjuvant properties of wP, alone or in diphtheria-teta-nus-wP (DTwP) formulations, were already reported for differentcombined antigens (both in animal models and in humans).These include influenza, hepatitis B, conjugated Haemophilus in-fluenzae B, and conjugated pneumococcal vaccines (21–26). wP isknown to modulate immune responses toward Th1- and Th17-

Received 6 March 2014 Returned for modification 7 April 2014Accepted 2 May 2014

Published ahead of print 7 May 2014

Editor: H. F. Staats

Address correspondence to Maria Leonor S. Oliveira,marialeonor.oliveira@butantan.gov.br.

Supplemental material for this article may be found at http://dx.doi.org/10.1128/CVI.00134-14.

Copyright © 2014, American Society for Microbiology. All Rights Reserved.

doi:10.1128/CVI.00134-14

972 cvi.asm.org Clinical and Vaccine Immunology p. 972–981 July 2014 Volume 21 Number 7

on July 7, 2020 by guesthttp://cvi.asm

.org/D

ownloaded from

type responses (27, 28), and several Bordetella pertussis compo-nents, such as lipopolysaccharides (LPS), pertussis toxin (PT), oradenylate cyclase toxin (ACT), contribute to this property (29–31). When nasally delivered to mice, the combination of PspA5with wP (PspA5-wP) induces high levels of mucosal and systemicanti-PspA5 antibodies, with balanced IgG1-to-IgG2a ratios, anti-gen-specific interleukin 17 (IL-17) secretion by spleen cells, andcontrolled inflammatory responses in the respiratory tract after aninvasive challenge with the S. pneumoniae ATCC 6303 strain (32).The depletion of CD4� T, CD8� T, or B lymphocytes in immu-nized mice during the pneumococcal invasive challenge did notimpair survival (32). On the other hand, passive immunization ofthe total sera from mice immunized with PspA5-wP conferredprotection to naive mice challenged with the ATCC 6303 pneu-mococcal strain (19). To further characterize the mechanisms ofprotection elicited by PspA5-wP, we address here the role of IgGand complement in this model. In addition, we evaluated thecomponents of B. pertussis that are involved in the adjuvant activ-ity to PspA5 in the wP context, and we analyzed the adjuvantactivity of purified pertussis components in combination withPspA5.

MATERIALS AND METHODSBacterial strains and growth conditions. S. pneumoniae ATCC 6303 (se-rotype 3, PspA clade 5) was grown in Todd-Hewitt broth (Difco, Detroit,MI, USA) supplemented with 0.5% yeast extract (THY) at 37°C, withoutshaking. The bacteria were plated in blood agar and grown overnight at37°C before inoculation in THY. The stocks were maintained at �80°C inTHY containing 20% glycerol. The B. pertussis strains used in this workwere BPSM (a streptomycin-resistant derivative of Tohama I) (33),BPLOW (a BPSM derivative in which the entire bvgA gene and the 5=portion of the bvgS gene, both from the Bordetella virulence control locusBvgA/S, were deleted) (34), and BPRA (a BPSM derivative in which theptx gene, which encodes the pertussis toxin, was deleted) (35). Thesestrains were grown in Bordet-Gengou medium (Difco) supplementedwith 1% glycerol, 20% sheep blood, and 100 �g/ml streptomycin, at 35°C.wP was derived from the B. pertussis NIH 137 strain and was produced atthe Instituto Butantan, São Paulo, Brazil (36).

Recombinant proteins and vaccine formulations. The N-terminalfragment of PspA from clade 5 (from strain S. pneumoniae 122/02; Insti-tuto Adolpho Lutz, São Paulo, Brazil) was expressed in Escherichia colistrain BL21-SI (Invitrogen, Carlsbad, CA, USA) and purified by chroma-tography, as previously described (17). Inactivation of the different B.pertussis strains was performed according to a protocol based on the pro-duction of wP at the Instituto Butantan (Sao Paulo, Brazil). Briefly, thebacteria were grown as described above until the exponential phase.Formaldehyde (0.2% [vol/vol]) was added, and the samples were incu-bated for 24 h at 35°C. The bacteria were centrifuged, washed in saline,suspended in 1/10 of the original volume, and maintained at 4°C until use.Purified ACT and PT oligomer B (OligB) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Pertussis toxin (PT) and filamentous hem-agglutinin (FHA) were purchased from Sigma-Aldrich and The NativeAntigen Company (Oxfordshire, United Kingdom). The purified anti-gens were suspended according to the manufacturer’s instructions andwere used immediately.

Immunization of mice and antibody responses. BALB/c andC57BL/6 specific-pathogen-free (SPF) mice were produced by the animalfacility from the Medical School of the University of São Paulo. C57BL/6�MT�/� mice were produced by the animal facility from the Institute ofBiomedical Sciences, University of São Paulo, Brazil. The animals weresupplied with food and water ad libitum, and the experimental protocolswere approved by the Ethics Committee for Animal Use from the InstitutoButantan. Immunization was performed by the nasal route in groups of 4

to 6 female mice. Each vaccine dose contained 5 �g of PspA5 alone or inthe presence of 1/8 of the human wP dose or the same amount of thedifferent inactivated B. pertussis strains. This dose corresponds to around7 � 108 inactivated bacteria and was chosen because it corresponds to thedose used for the potency assays of the wP vaccines produced at the Insti-tuto Butantan. In the experiments using purified B. pertussis components,PspA5 was combined with 1 �g of FHA, 1 �g of ACT, 1 �g of PT, or theequivalent molar mass of OligB. Before formulation, the excess LPS fromE. coli present in PspA5 preparations was removed by Triton X-114 ex-traction, as previously described (37). Before nasal immunization, themice were anesthetized through the intraperitoneal route (i.p.) with 200�l of a 0.2% xylazine and 0.5% ketamine mixture. The mice received oneto six doses of each vaccine formulation, according to the particular ex-periment, in a 10-�l volume. In the case of the six-dose schedule, the micewere immunized on days 0, 3, 14, 17, 28, and 31. The groups that receivedonly wP, inactivated B. pertussis, or PspA5, as well as the nonimmunizedmice, were used as controls. Serum samples were collected 20 days afterthe last immunization to evaluate antibody levels by enzyme-linked im-munosorbent assay (ELISA) in plates coated with PspA5, as describedearlier (38). The assay was performed using goat anti-mouse IgG andrabbit anti-goat IgG conjugated with horseradish peroxidase (HRP)(SouthernBiotech, Birmingham, AL, USA). Standard curves were gener-ated using mouse IgG (SouthernBiotech).

Passive immunization experiments. IgG was purified from the sera ofmice immunized with PspA5-wP or from the hyperimmune sera of miceimmunized with a nonrelated antigen [intimin from enteropathogenic E.coli using Al(OH)3 as an adjuvant (39)]. Purification was performed usingthe HiTrap protein G column (GE Healthcare, Waukesha, WI, USA),according to the manufacturer’s instructions. F(ab=)2 was obtained frompurified IgG using the Pierce F(ab=)2 preparation kits (Thermo Scientific,West Palm Beach, FL, USA), according to the manufacturer’s instruc-tions. The purity was verified by SDS-PAGE, and proteins were quantifiedusing the Bradford reagent (Bio-Rad, Berkeley, CA, USA). Two hoursbefore the intranasal pneumococcal challenge, naive BALB/c mice (6 pergroup) were inoculated i.p. with serum containing 20 �g to 50 �g of IgG,purified IgG (20 �g to 50 �g), or the equivalent molar mass of purifiedF(ab=)2.

In vivo depletion of complement. Cobra venom factor (CVF) fromNaja naja kaouthia (Quidel Corporation, Darmstadt, Germany) was usedfor the depletion of complement in vivo, as described previously (40, 41),and depletion was confirmed by the absence of rabbit erythrocyte hemo-lytic activity in the sera of mice treated with CVF (41). BALB/c mice (14animals) were immunized with the PspA5-wP formulation using the six-dose schedule. Additional groups (6 mice each) received wP or PspA5alone. Twenty days after the last dose, blood samples were collected toanalyze for the presence of anti-PspA5 IgG by ELISA. On the same day, 8mice that were immunized with PspA5-wP received 50 �g (approximately20 U) of CVF i.p. The additional 6 mice inoculated with PspA5-wP wereleft as controls. Intranasal pneumococcal challenge was performed, asdescribed below, 24 h after the first inoculation of CVF. The depletion ofcirculating complement was maintained by two additional i.p. injectionsof CVF at a 48-h interval. Two mice immunized with PspA5-wP andinoculated with CVF but not subjected to pneumococcal challenge wereused as controls to assess for potential deleterious effects of CVF injec-tions.

Intranasal pneumococcal challenge. The ATCC 6303 strain wasgrown in THY until mid-log phase (optical density at 600 nm [OD600], 0.4),and the aliquots were maintained at �80°C until use. Twenty-one days afterthe last immunization, the mice were anesthetized by i.p. injection of 200�l of0.2% xylazine–1.0% ketamine. The animals received 3 � 105 CFU of theATCC 6303 strain in 50 �l of saline, inoculated into one nostril, with the helpof a micropipette. Animal survival was monitored for 10 days.

Antibody binding and complement deposition assays. S. pneu-moniae ATCC 6303 was grown overnight on blood agar. The bacteria werediluted in THY, grown until an OD600 of 0.4 to 0.5 (�108 CFU/ml), and

PT Enhances Immune Responses to PspA

July 2014 Volume 21 Number 7 cvi.asm.org 973

on July 7, 2020 by guesthttp://cvi.asm

.org/D

ownloaded from

harvested by centrifugation at 3,200 � g for 10 min. The bacteria werewashed, suspended in phosphate-buffered saline (PBS), and incubatedwith 5% of the different sera for 30 min at 37°C. The samples were washedonce with PBS before incubation for 30 min on ice with fluorescein iso-thiocyanate (FITC)-conjugated goat anti-mouse IgG (MP Biomedicals,Santa Ana, CA, USA) diluted 1:100 in PBS. For the complement deposi-tion assays, the sera were previously heated at 56°C for 30 min and incu-bated with bacteria at a concentration of 5% at 37°C for 30 min. Thesamples were washed once with PBS and incubated with 10% normalmouse serum as a source of complement in gelatin veronal buffer (Sigma-Aldrich), at 37°C for 30 min. After washing, the samples were incubatedwith FITC-conjugated anti-mouse C3 IgG (MP Biomedicals) in PBS for30 min on ice. The samples were fixed with 200 �l of Cytofix (BD Biosci-ences) after two washing steps and stored at 4°C. Flow cytometry analysiswas conducted using FACSCanto II (BD Biosciences, San Jose, CA, USA),and 10,000 gated events were recorded. Fluorescence was analyzed inhistograms using the FlowJo 7.6.1 software, and the median values of thecurves were used to compare the groups.

Statistical analysis. The differences in the antibody concentrationswere analyzed by the Mann-Whitney U test. Overall survival was analyzedby Fisher’s exact test, and the survival curves were analyzed by the log ranktest. Statistical analyses were performed using the Prism 5.03 software,and a P value of �0.05 was considered significant.

RESULTSProtection elicited by PspA5-wP requires the induction of anti-bodies. Nasal immunization of mice with PspA5-wP was previ-ously shown to induce high levels of antibodies and protect miceagainst respiratory challenges with different pneumococcal strains(19). The passive transfer of sera from immunized mice to naivemice was shown to confer survival to 75% of the animals afterchallenge with the ATCC 6303 pneumococcal strain, suggestingan important role for antibodies (19). On the other hand, thedepletion of CD4� T, CD8� T, or B lymphocytes during the chal-lenge did not impair protection in this model (32). In order toevaluate the role of antibodies in protection, C57BL/6 andC57BL/6 �MT�/� mice were immunized with 6 nasal doses ofPspA5-wP. The control groups received wP only. This protocolwas chosen since it induces high levels of anti-PspA5 antibodiesand antigen-specific IL-17 secretion by splenocytes (19, 32). Highlevels of anti-PspA5 IgG were observed in the wild-type mice im-munized with PspA5-wP but not in mice immunized with wP. Asexpected, no anti-PspA5 IgG was found in C57BL/6 �MT�/�

mice immunized with PspA5-wP (see Fig. S1 in the supplementalmaterial). The mice were then challenged through the intranasalroute with a lethal dose of the ATCC 6303 pneumococcal strain.Whereas 100% of the vaccinated C57BL/6 mice survived the chal-

lenge, the vaccine did not protect C57BL/6 �MT�/� mice (Table1). Next, naive BALB/c mice were inoculated i.p. with sera fromBALB/c mice immunized with six doses of PspA5-wP (PspA5-wPserum containing 50 �g of IgG per animal), 50 �g of IgG purifiedfrom these sera (PspA5-wP IgG), or from hyperimmune sera ofmice immunized with a nonrelated antigen (NR IgG). The micewere challenged 2 h later with the ATCC 6303 pneumococcalstrain. Both the PspA5-wP sera and the respective purified IgGconferred partial but significant protection to mice (50% and41.6% survival, respectively) (Fig. 1A). No survival was observedin the group of mice immunized with NR IgG (P � 0.05 by Fish-er’s exact test, or P � 0.01 by the log rank analyses of the survivalcurves, in a comparison of mice that received PspA5-wP IgG withmice that received NR IgG). To test the capacity of the F(ab=)2

derived from the PspA5-wP IgG to confer protection, 20 �g of IgGor the same molar mass of the respective F(ab=)2 (PspA5-wPF(ab=)2 or NR F(ab=)2) was inoculated i.p. into naive mice 2 hbefore challenge with the ATCC 6303 pneumococcal strain (Fig.1B). In this experiment, reduced amounts of IgG were inoculated,since we had difficulties producing large amounts of purifiedF(ab=)2. Although only 33.3% of the mice inoculated with thePspA5-wP sera or PspA5-wP IgG survived the challenge underthese conditions, the survival curves showed a significant increasein the survival time of mice inoculated with PspA5-wP IgG com-pared to that of mice inoculated with PspA5-wP F(ab=)2 (P � 0.01,log rank survival analyses). Importantly, both purified PspA5-wP

TABLE 1 Survival of mice against challenge with the ATCC 6303pneumococcal strain: evaluation of the role of antibodies on protectioninduced by PspA5-wP vaccinea

Mousegroup

Vaccinationgiven

No. of mice alive/total no. of mice % survival Pb

�MT�/� wP 0/4 0PspA5-wP 0/4 0 1

C57BL/6 wP 0/6 0PspA5-wP 6/6 100 0.005

a For the challenge, 3 � 105 CFU of the ATCC 6303 pneumococcal strain wasinoculated through the nasal route, and survival was monitored for 10 days.b By Fisher’s exact test. Comparisons were made with the respective group immunizedwith wP.

FIG 1 Effect of passive immunization on mouse survival against the respira-tory challenge with the pneumococcal ATCC 6303 strain. Naive mice (6 pergroup) were inoculated i.p. with serum, purified IgG, or purified F(ab=)2 frommice immunized with PspA5-wP or with nonrelated antigen (NR) 2 h beforethe respiratory challenge with the ATCC 6303 pneumococcal strain. Survivalwas monitored for 10 days. The survival curves were analyzed by the log ranktest. (A) Mice received serum containing 50 �g of IgG or the same amount ofpurified IgG; **, P � 0.001 in a comparison of mice that received PspA5-wPIgG or NR IgG. The results were collected from 2 independent experiments.(B) Mice received serum containing 20 �g of IgG, the same amount of purifiedIgG, or the same molar mass of purified F(ab=)2; **, P � 0.01 in a comparisonof mice that received PspA5-wP serum or the respective purified IgG with micethat received purified F(ab=)2 derived from the same sample.

Salcedo-Rivillas et al.

974 cvi.asm.org Clinical and Vaccine Immunology

on July 7, 2020 by guesthttp://cvi.asm

.org/D

ownloaded from

IgG and PspA5-wP F(ab=)2 were able to bind to the surface of theATCC 6303 pneumococcal strain in vitro. However, the deposi-tion of complement C3 on the surface of the bacteria was inducedonly by PspA5-wP IgG (see Fig. S2 in the supplemental material).

Protection elicited by PspA5-wP depends on complement ac-tivity. Since the Fc portion of IgG may contribute to the protec-tion elicited by PspA5-wP, we next tested the role of complementin an in vivo model. BALB/c mice were vaccinated through thenasal route with 6 doses of PspA5-wP in order to induce highlevels of anti-PspA5 IgG. The control mice received PspA5 or wPalone. As expected, high concentrations of antibodies were ob-served in all animals vaccinated with PspA5-wP (see Fig. S3 in thesupplemental material). Twenty-four hours before challenge withthe ATCC 6303 pneumococcal strain, a group of mice inoculatedwith PspA5-wP received CVF (a snake toxin known to depletecomplement components in serum) (42) via i.p. injection. Thisgroup received two additional CVF injections on days one andthree after challenge. Survival monitoring showed that 100% ofthe mice vaccinated with PspA5-wP and treated with CVF suc-cumbed to the infection (Table 2). In contrast, 100% of the micevaccinated with PspA5-wP but not treated with CVF survived thechallenge, and this was the only group that was statistically differ-ent from the controls inoculated with wP (Table 2; P � 0.01,Fisher’s exact test). Two additional mice were vaccinated withPspA5-wP and treated with CVF but were not challenged. Nosigns of disease due to the three injections of CVF were observed(data not shown).

PT-deficient B. pertussis strains are less effective as adju-vants for PspA5. We have shown that the adjuvant activity of wPin PspA5-wP does not depend on the LPS present in the vaccine(19), suggesting that other components of B. pertussis exert adju-vant activity when combined with PspA5. In order to identifythese components, we first determined the minimal dose ofPspA5-wP that confers significant protection. A single nasal doseof PspA5-wP (containing 5 �g of the protein and 1/8 of the wPhuman dose) was enough to protect 80% to 100% of the BALB/cmice against challenge with the ATCC 6303 pneumococcal strain.Therefore, all the following experiments were performed underthese conditions. We first tested preparations of the B. pertussisBPLOW strain, a mutant deficient for BvgA and thereby unable toproduce different virulence factors, including PT and FHA (34).Nasal immunization of BALB/c mice with a single dose of PspA5-BPLOW induced high levels of anti-PspA5 IgG that were signifi-cantly higher than those observed in the group immunized withBPLOW or PspA5 alone (Fig. 2A). However, the levels of IgG

induced by PspA5-BPLOW were slightly but significantly lowerthan those induced by PspA5-wP (Fig. 2A). Evaluation of the IgGsubtypes showed no differences in the anti-PspA5 IgG1-to-IgG2aratios induced by immunizing mice with PspA5-BPLOW orPspA5-wP, with slightly larger amounts of IgG1 than IgG2a inboth cases (data not shown). After challenge with the ATCC 6303pneumococcal strain, only 50% of the mice immunized with asingle dose of PspA5-BPLOW were protected, and this result wasat the limit of the significance compared with the nonimmunizedgroup (P � 0.05) (Table 3), but it did not reach significance in acomparison with mice immunized with BPLOW or PspA5 (P 0.05, Fisher’s exact test). In contrast, 83% of the mice immunizedwith PspA5-wP survived the challenge (Table 3), with a significantdifference from the nonimmunized group or the groups immu-

TABLE 2 Survival of mice against challenge with the ATCC 6303pneumococcal strain: evaluation of the role of complement onprotection induced by the PspA5-wP vaccinea

Vaccination givenNo. of mice alive/total no. of mice % survival Pb

wP 1/6 16.6PspA5 4/6 66.6 0.24PspA5-wP 6/6 100 0.01PspA5-wP (CVF)c 0/6 0 1a For the challenge, 3 � 105 CFU of the ATCC 6303 pneumococcal strain wasinoculated through the nasal route, and survival was monitored for 10 days.b By Fisher’s exact test. Comparisons were made with the group immunized with wP.c Cobra venom factor (CVF) was administered 24 h before the challenge, and twoadditional injections were given at a 48-h interval.

FIG 2 Induction of anti-PspA5 IgG by immunizing mice with PspA5 com-bined with different B. pertussis preparations. Serum samples were collected 21days after the immunization. Anti-PspA5 IgG was measured by ELISA, usingIgG standard curves as references. The data for each mouse are presented, withthe median values of the groups (horizontal line). (A) Group that receivedPspA5 was significantly different from the nonimmunized group (Non); ***,P � 0.001. Significant differences (****, P � 0.0001) were also observed in acomparison of the groups that received PspA5-BPLOW or PspA5-wP with therespective adjuvant controls or with the group immunized with PspA5.PspA5-wP also showed significant differences from the group immunized withPspA5-BPLOW (**, P � 0.01). The results were collected from two indepen-dent experiments. (B) Significant differences (*, P � 0.05) were observed in acomparison of the group immunized with PspA5-BPRA with the group im-munized with BPRA or with PspA5. Immunization with PspA5-BPSM orPspA5-wP induced significantly higher levels of anti-PspA5 IgG in a compar-ison with all other groups, including PspA5-BPRA (**, P � 0.01). Statisticalanalyses were performed using the Mann-Whitney U test. The results are rep-resentative of two independent experiments.

PT Enhances Immune Responses to PspA

July 2014 Volume 21 Number 7 cvi.asm.org 975

on July 7, 2020 by guesthttp://cvi.asm

.org/D

ownloaded from

nized with wP or PspA5 (P, �0.01 and 0.05, respectively, by Fish-er’s exact test), suggesting that one or more of the B. pertussisvirulence factors may contribute to the adjuvant activity.

In order to identify this virulence factor, we tested the adjuvantproperties of the B. pertussis BPRA strain, in which the PT genehad been deleted (35). As shown in Fig. 2B, the immunization ofBALB/c mice with PspA5-BPRA induced IgG levels that werehigher than the levels induced by immunization with BPRA (P �0.05) but not statistically different from those induced by immu-nization with PspA5. The highest levels of anti-PspA5 IgG wereobserved in the mice immunized with PspA5-BPSM or PspA5-wP. BPSM was included in these experiments, as it is the parentalstrain of BPRA. No differences were observed between the PspA5-BPSM and PspA5-wP groups, and the levels of IgG induced byeither formulation were significantly higher than those of all othergroups, including the group immunized with PspA5 and thegroup immunized with PspA5-BPRA (P � 0.01). Again, no sig-nificant differences in the IgG1-to-IgG2a ratios induced by thePspA5-BPRA, PspA5-BPSM, and the PspA5-wP vaccines were ob-served. All these vaccines induced slightly larger amounts of IgG1than IgG2a (data not shown). Survival monitoring after challengewith the ATCC 6303 pneumococcal strain showed that immuniz-ing mice with one nasal dose of PspA5-BPRA did not confer pro-

tection (Table 4). Only 1 out of 6 mice from this group survivedthe challenge. In contrast, 5 out of 6 mice immunized with PspA5-BPSM and 4 out of 6 mice immunized with PspA5-wP survivedthe challenge.

Combination of PspA5 with purified PT increases the induc-tion of antibodies and protection against the pneumococcalchallenge. Since PT in the wP may be one of the molecules thatexpresses adjuvant activity in this model, we directly tested thisand other purified B. pertussis components with known immuno-modulating properties. As shown in Fig. 3A, the highest anti-PspA5 IgG levels were observed in BALB/c mice vaccinated withPspA5 combined with PT (PspA5-PT). Although the levels of an-ti-PspA5 IgG induced by the PspA5-FHA vaccine were also signif-icantly higher than those of the control groups (P � 0.01) and ofthe groups immunized with PspA5 or PspA5-ACT (P � 0.05),they were lower than those induced by PspA5-PT (P � 0.05). ACTdid not appear to display adjuvant functions, since the levels ofantibodies observed in the PspA5-ACT group were not signifi-

TABLE 4 Survival of mice against challenge with the ATCC 6303pneumococcal strain: evaluation of the adjuvant activity of a B. pertussismutant strain lacking PTa

Vaccinationgivenb

No. of mice alive/total no. of mice % survival Pc

Non 0/6 0BPRA 0/6 0 1BPSM 1/6 16.7 1wP 0/6 0 1PspA5 1/6 16.7 1PspA5-BPRA 1/6 16.7 1PspA5-BPSM 5/6 83.3 0.02PspA5-wP 4/6 66.7 0.06a For the challenge, 3 � 105 CFU of the ATCC 6303 pneumococcal strain wasinoculated through the nasal route, and survival was monitored for 10 days.b Non, nonimmunized. BPSM is a streptomycin-resistant derivative of Tohama I. BPRAis a BPSM derivative in which the ptx gene was deleted.c By Fisher’s exact test. Comparisons were made with the nonimmunized group.

FIG 3 Induction of anti-PspA5 IgG by immunization of mice with PspA5combined with different purified B. pertussis components. Serum sampleswere collected 21 days after the immunization. Anti-PspA5 IgG was measuredby ELISA, using IgG standard curves as references. The data for each mouse arepresented, with the median of the group (horizontal lines). (A) Significantdifferences (**, P � 0.01) were observed in a comparison of the groups thatreceived PspA5-ACT, PspA5-FHA, or PspA5-PT with the nonimmunizedgroup or with the respective adjuvant controls. Immunization with PspA5induced significantly higher levels of antibodies (*, P � 0.05) than those in thenonimmunized group. *, P � 0.05 and **, P � 0.01 for the comparisonsindicated. (B) Immunization with PspA5-OligB or PspA5-PT induced signif-icantly higher levels of anti-PspA5 IgG than those in all other groups **, P �0.01. Statistical analyses were performed using the Mann-Whitney U test. Theresults are representative of two independent experiments.

TABLE 3 Survival of mice against challenge with the ATCC 6303pneumococcal strain: evaluation of the adjuvant activity of a B.pertussis-attenuated straina

Vaccinationgivenb

No. of mice alive/total no. of mice % survival Pc

Non 0/6 0BPLOW 1/12 8.3 1wP 2/12 16.6 0.53PspA5 4/12 33.3 0.24PspA5-BPLOW 6/12 50.0 0.05PspA5-wP 10/12 83.3 0.002a For the challenge, 3 � 105 CFU of the ATCC 6303 pneumococcal strain wasinoculated through the nasal route, and survival was monitored for 10 days. The resultswere collected from two independent experiments.b Non, nonimmunized. BPLOW is a BPSM derivative in which the entire bvgA gene andthe 5= portion of the bvgS gene were deleted.c By Fisher’s exact test. Comparisons were made with the nonimmunized group.

Salcedo-Rivillas et al.

976 cvi.asm.org Clinical and Vaccine Immunology

on July 7, 2020 by guesthttp://cvi.asm

.org/D

ownloaded from

cantly different from those observed in mice immunized withPspA5 alone. The immunization of mice with PspA5-ACT,PspA5-FHA, or PspA5-PT induced immune responses with a drifttoward Th2, showing higher levels of IgG1 than IgG2a. However,no differences in the IgG1-to-IgG2a ratios among these groupswere observed (data not shown).

Only PspA5-PT conferred 100% survival in the mice after chal-lenge with the ATCC 6303 pneumococcal strain, although PspA5-FHA conferred partial protection, with 50% survival (Table 5).

The immunization of BALB/c mice with a combination ofPspA5 and the B oligomer of PT (PspA5-OligB) induced levels ofanti-PspA5 IgG similar to those induced by PspA5-PT (Fig. 3B),indicating that the enzymatic activity of PT is not essential for theexpression of its adjuvant activity. Both formulations conferredsignificant protection against challenge with the ATCC 6303pneumococcal strain, with around 80% of the mice surviving inboth cases (Table 6).

Sera from mice immunized with PT-containing vaccinesshowed higher capacity to bind and to induce complement de-position on pneumococcal surface in vitro. Pools of serum col-lected from the groups of mice were evaluated for IgG binding andcomplement deposition on the surface of the ATCC 6303 pneu-mococcal strain. The sera from mice immunized with formula-tions containing wild-type B. pertussis bacterial preparations(PspA5-wP and PspA5-BPSM) showed high IgG binding (medianfluorescence intensities of 1,098 and 519, respectively) and in-duced complement deposition (median fluorescence intensities of4,736 and 1,406, respectively) on the pneumococcal surface (Fig.4A and B). In contrast, the sera from mice immunized withPspA5-BPRA showed median fluorescence intensity values thatwere only slightly above the median values observed for the con-trol sera (115 for IgG binding and 349 for complement deposi-tion). Regarding the formulations composed of PspA5 and puri-fied B. pertussis components (Fig. 4C and D), only sera from miceimmunized with PspA5-PT showed significant IgG binding andcomplement deposition (median fluorescence intensity values of1,339 for binding and 806 for complement deposition). IgG bind-ing slightly above the controls was observed with the sera frommice immunized with PspA5-ACT or PspA5-FHA (median fluo-rescence intensity values of 90 and 113, respectively). No effects ofthese sera on complement deposition were observed.

DISCUSSION

Protein vaccines are being proposed as alternatives to the availablepolysaccharide conjugate vaccines, with the objective of inducingserotype-independent immunity against S. pneumoniae. A greatbody of data supports the use of PspA for these purposes (8).Although sequence variability may counteract the development ofPspA-based vaccines, some studies have shown that the choice ofPspA molecules that induce broad-reactive antibodies, or thecombination of two PspA molecules, may overcome this problem(18). The enhancement of the immune responses against the an-tigen by the addition of adjuvants may also improve cross-reac-tivity. Sera from mice immunized with PspA5 combined with wPas an adjuvant were shown to better recognize PspAs from differ-ent clades than sera from the mice immunized with PspA5 alone(19). When given to mice through the nasal route, PspA5-wPinduced protection against respiratory invasive challenges withpneumococcal strains expressing PspAs from clades 5 and 2 (S.pneumoniae ATCC 6303 and A66.1 strains, respectively) as well asagainst nasal colonization with the S. pneumoniae 0603 strain,which expresses PspA from clade 1 (19). Repeated nasal immuni-zations with PspA5-wP induce high levels of systemic and mucosalanti-PspA5 antibodies, IL-17 responses in the spleen, and a peakof proinflammatory responses in the lungs after pneumococcalchallenge (43). Here, we show that the induction of IgG by thePspA5-wP vaccine is essential for protection against invasive chal-lenge with the ATCC 6303 pneumococcal strain. PspA5-wP failedto confer protection against pneumococcal challenge in �MT�/�

mice, which showed undetectable amounts of anti-PspA5 IgG af-ter immunization. Moreover, passive immunization with IgG pu-rified from the sera of mice immunized with PspA5-wP conferredsignificant protection to naive BALB/c mice.

PspA inhibits the deposition of complement on the pneumo-coccal surface, thereby enhancing bacterial survival during sys-temic infection. The deletion of the pspA gene or the presence ofanti-PspA antibodies abolishes this inhibitory effect (6). The in-hibition of complement deposition by PspA occurs through thecholine-binding domain present at the C-terminal end of the mol-ecule. The attachment of PspA occurs via the interaction of thisdomain and the phosphocholine present on the bacterial surface.This interaction impairs the binding of the C-reactive protein, animportant activator of the complement cascade, to phosphocho-line, resulting in diminished complement deposition at the bacte-rial surface (44). However, as discussed by other authors, protec-tive antibodies against PspA are directed to its N-terminal region

TABLE 5 Survival of mice against challenge with the ATCC 6303pneumococcal strain: evaluation of the adjuvant activity of purified B.pertussis componentsa

Vaccinationgivenb

No. of mice alive/total no. of mice % survival Pc

Non 0/6 0ACT 0/6 0 1FHA 0/6 0 1PT 0/6 0 1PspA5 0/6 0 1PspA5-ACT 1/6 16.6 1PspA5-FHA 3/6 50 0.18PspA5-PT 6/6 100 0.002a For the challenge, 3 � 105 CFU of the ATCC 6303 pneumococcal strain wasinoculated through the nasal route, and survival was monitored for 10 days.b Non, nonimmunized; ACT, adenylate cyclase toxin; FHA, filamentous hemagglutinin;PT, pertussis toxin.c By Fisher’s exact test. Comparisons were made with the nonimmunized group.

TABLE 6 Survival of mice against challenge with the ATCC 6303pneumococcal strain: evaluation of the adjuvant activity of PTa

Vaccinationgivenb

No. of mice alive/total no. of mice % survival Pc

Non 0/6 0OligB 1/6 16.7 1PT 0/6 0 1PspA5 1/6 16.7 1PspA5-OligB 5/6 83.3 0.02PspA5-PT 4/5 80.0 0.02a For the challenge, 3 � 105 CFU of the ATCC 6303 pneumococcal strain wasinoculated through the nasal route, and survival was monitored for 10 days.b Non, nonimmunized; OligB, pertussis toxin oligomer B; PT, pertussis toxin.c By Fisher’s exact test. Comparisons were made with the nonimmunized group.

PT Enhances Immune Responses to PspA

July 2014 Volume 21 Number 7 cvi.asm.org 977

on July 7, 2020 by guesthttp://cvi.asm

.org/D

ownloaded from

(9, 10), suggesting that the protective effects of anti-PspA antibod-ies are probably related to the direct binding of their Fc portion tothe complement C1q, activating the classical complement path-way (44). The PspA5 antigen used here comprises the N-terminalregion up to the proline-rich region and lacks the choline-bindingregion. Passive immunization with IgG purified from PspA5-wPsera conferred partial but increased survival against the pneumo-coccal invasive challenge, whereas the respective F(ab=)2 did not,supporting this hypothesis.

Repeated nasal immunizations with PspA5-wP induce highlevels of anti-PspA5 IgG and IgA in the serum and in the respira-tory mucosa of mice (19), but the depletion of complement inmice immunized with up to 6 doses of PspA5-wP completely ab-rogated protection against respiratory invasive challenge with theATCC 6303 strain. This result supports the importance of thecomplement system in the protection elicited by PspA5-wP. Sim-ilarly, a nasal vaccine composed of a hybrid protein that targetsPspA to the human Fc receptor type I (anti-hFcRI-PspA) in-

duced high levels of anti-PspA antibodies and complement-de-pendent protection against a challenge with a serotype 3 pneumo-coccal strain in transgenic mice expressing hFcRI (40).

We previously showed that the LPS present in wP is not essen-tial to its adjuvant activity. A modified wP vaccine containing lowlevels of LPS, produced at the Instituto Butantan (São Paulo, Bra-zil), showed adjuvant properties similar to those of wP. In addi-tion, PspA5-wP induced high levels of antibodies and protectionagainst the pneumococcal invasive challenge in C3H/HeJ mice,which lack signaling through Toll-like receptor 4 (TLR4) (19).Therefore, we decided to evaluate the role of other B. pertussiscomponents in the adjuvant activity in this model. Here, we foundthat even a single nasal dose of PspA5-wP consistently inducedhigh levels of anti-PspA5 IgG and conferred survival to around80% of the BALB/c mice after invasive challenge with the ATCC6303 pneumococcal strain. The inactivated virulent B. pertussisBPSM strain, derived from Tohama I, displayed adjuvant activityequivalent to that of our previous wP preparations. In contrast,

FIG 4 In vitro antibody binding and complement deposition on pneumococcal surface. Sera were collected 21 days after the last immunization, and pools (from6 mice) were incubated with the ATCC 6303 strain. The samples were incubated with FITC-conjugated anti-mouse IgG (A and C). For complement deposition,normal mouse serum was added to the samples, followed by incubation with FITC-conjugated anti-mouse C3 (B and D). Results were analyzed by flow cytometrywith 10,000 events gated. Sera from nonimmunized animals or mice immunized with the adjuvants or PspA5 alone are represented as gray areas, since all of themshowed similar results of antibody binding and complement deposition. (A and B) Sera from mice immunized with PspA5-BPRA are represented by solid blacklines, sera from mice immunized with PspA5-BPSM are represented by dashed black lines, and sera from mice immunized with PspA5-wP are represented bydotted black lines. (C and D) Sera from mice immunized with PspA5-ACT are represented by dotted black lines, sera from mice immunized with PspA5-FHAare represented by solid black lines, and sera from mice immunized with PspA5-PT are represented by dashed black lines. The median fluorescence intensity(numbers in the graph) is indicated for each sample.

Salcedo-Rivillas et al.

978 cvi.asm.org Clinical and Vaccine Immunology

on July 7, 2020 by guesthttp://cvi.asm

.org/D

ownloaded from

BPLOW, the attenuated mutant BPSM derivative, showed a de-crease in the induction of anti-PspA5 IgG compared with that ofwP. Moreover, immunizing mice with a single dose of PspA5-BPLOW induced only partial protection (50% survival) in miceafter challenge with the ATCC 6303 pneumococcal strain. InBPLOW, the two-component regulatory system BvgA/S, whichregulates the expression of most B. pertussis virulence genes, in-cluding those encoding PT, FHA, and ACT, was disrupted (34,45). Since these proteins are known to possess immunomodulat-ing properties (46–48), we investigated their adjuvant propertiesin the PspA5 model. Decreased adjuvant activity was observed forthe BPRA strain, in which the ptx gene was deleted. The levels ofanti-PspA5 IgG were significantly lower in the mice immunizedwith PspA5-BPRA than in the mice immunized with PspA5-BPSM or PspA5-wP, and no protection was observed in thisgroup, suggesting that the PT contributes to the adjuvant activity.The combination of PspA5 with purified B. pertussis componentsconfirmed the adjuvant activity of PT in this model. Although asingle nasal immunization of PspA5 combined with FHA inducedhigher levels of anti-PspA5 IgG than the immunization withPspA5 alone, the highest levels of antibodies were observed whenPT was used as the adjuvant. The levels of protection correlatedwith the levels of antibodies, with the highest survival percentagesobserved in the mice immunized with PspA5-PT. Immunizationwith PspA5-FHA protected 50% of the mice compared to no sig-nificant protection after immunization with PspA5 alone. How-ever, vaccination with PspA5-PT showed significantly enhancedprotection compared to vaccination with PspA5-FHA. ACT waspreviously shown to express adjuvant activity in mice when ad-ministered in combination with ovalbumin or pertactin throughthe nasal route, inducing antibodies and T-cell responses againstthe antigens (49). However, in our experiments, we did not ob-serve such adjuvant activity. The levels of anti-PspA5 IgG inducedby PspA5-ACT were not different from those induced by PspA5.The possible reasons for these differences are the administrationof a single dose using 1 �g of ACT in our experiments, whereas inprevious work, three doses with 10 �g of ACT were used.

In agreement with the in vivo results that indicate the impor-tance of IgG and complement in the protection mediated byPspA5-wP, the highest levels of complement deposition on thepneumococcal surface in vitro were induced by the sera from miceimmunized with PspA5 combined with preparations from viru-lent B. pertussis strains (wP and BPSM) or with PT. No differencesin the IgG1-to-IgG2a ratios were observed when PspA5 was com-bined with the different B. pertussis preparations. All the combi-nations induced balanced Th1/Th2 responses. On the other hand,immunization with PspA5 combined with the purified B. pertussiscomponents induced responses with a Th2 character with higherlevels of IgG1 than IgG2a. PspA-based vaccines able to induce Th1responses were shown to be more efficient at eliciting protectionagainst pneumococcal infections in mice due to the induction ofrelatively high levels of IgG2a, a subtype with a higher capacity tomediate complement deposition on the pneumococcal surface(50, 51). However, in our experiments, this may have been com-pensated by the induction of high levels of total IgG (includingIgG1 and IgG2a) observed in mice immunized with PspA5-wP,PspA5-BPSM, or PspA5-PT.

PT is composed of two moieties called the A protomer and theB oligomer. While the A subunit (also known as S1) expressesenzymatic activity, catalyzing the ADP-ribosylation of signal-

transducing G proteins in the host cells, the B oligomer (com-posed of five subunits named S2 to S5) promotes binding to thePT receptors at the host cell surface. Enzymatically inactive PTwith mutations in the S1 subunit but that is able to bind to hostcells was shown to enhance Th1 and Th2 responses to coinjectedantigens. This activity depends on the receptor-binding activity ofthe B moiety (46). In agreement with these observations, OligBdisplayed adjuvant activity similar to that of PT in the PspA5model. High levels of anti-PspA5 IgG and protection againstpneumococcal challenge were observed in mice immunized with asingle dose of PspA5-OligB.

Several studies have shown the effect of formaldehyde treat-ment on PT structure and activity (52, 53). Increasing concentra-tions of formaldehyde can differently affect the enzymatic andcarbohydrate-binding activities, as well as the presence of nativeepitopes (52). In addition, reversion of the formaldehyde effects,mainly with respect to the carbohydrate-binding activity, has beenobserved (53). These studies highlighted the importance ofachieving a balance between low toxicity and immunogenicitywhen developing PT-containing pertussis vaccines. We did notassess the properties of PT in the formaldehyde-inactivatedwhole-cell pertussis preparations, but the results presented hereshow that the PT-deficient strain displays reduced adjuvant activ-ity compared with those of the PT-producing strains, suggestingthat formaldehyde treatment did not affect the adjuvant activity ofPT in a major manner. In summary, we show here that the pro-tective effect of PspA5-wP against pneumococcal invasive chal-lenge is mainly conferred by the induction of anti-PspA5 IgG anddepends on complement. PT-containing formulations can im-prove the immune responses against PspA5, leading to enhancedprotection.

ACKNOWLEDGMENTS

This work was supported by CNPq/INSERM grant 490391/2010-9,FAPESP grant 2011/24019-7, and Fundação Butantan. C.S.-R. is a Ph.D.student from the Interunidades em Biotecnologia (IPT/USP/Butantan) pro-gram and has a Ph.D. fellowship from FAPESP (process 2012/10132-9).

REFERENCES1. O’Brien KL, Wolfson LJ, Watt JP, Henkle E, Deloria-Knoll M, McCall

N, Lee E, Mulholland K, Levine OS, Cherian T, Hib and PneumococcalGlobal Burden of Disease Study Team. 2009. Burden of disease caused byStreptococcus pneumoniae in children younger than 5 years: global esti-mates. Lancet 374:893–902. http://dx.doi.org/10.1016/S0140-6736(09)61204-6.

2. Dagan R. 2009. New insights on pneumococcal disease: what we havelearned over the past decade. Vaccine 27(Suppl 3):C3–C5. http://dx.doi.org/10.1016/j.vaccine.2009.06.002.

3. Pilishvili T, Lexau C, Farley MM, Hadler J, Harrison LH, Bennett NM,Reingold A, Thomas A, Schaffner W, Craig AS, Smith PJ, Beall BW,Whitney CG, Moore MR, Active Bacterial Core Surveillance/EmergingInfections Program Network. 2010. Sustained reductions in invasivepneumococcal disease in the era of conjugate vaccine. J. Infect. Dis. 201:32– 41. http://dx.doi.org/10.1086/648593.

4. Richter SS, Heilmann KP, Dohrn CL, Riahi F, Diekema DJ, Doern GV.2013. Pneumococcal serotypes before and after introduction of conjugatevaccines, United States, 1999 –2011. Emerg. Infect. Dis. 19:1074 –1083.http://dx.doi.org/10.3201/eid1907.121830.

5. Weinberger DM, Malley R, Lipsitch M. 2011. Serotype replacement indisease after pneumococcal vaccination. Lancet 378:1962–1973. http://dx.doi.org/10.1016/S0140-6736(10)62225-8.

6. Ren B, Szalai AJ, Hollingshead SK, Briles DE. 2004. Effects of PspA andantibodies to PspA on activation and deposition of complement on thepneumococcal surface. Infect. Immun. 72:114 –122. http://dx.doi.org/10.1128/IAI.72.1.114-122.2004.

PT Enhances Immune Responses to PspA

July 2014 Volume 21 Number 7 cvi.asm.org 979

on July 7, 2020 by guesthttp://cvi.asm

.org/D

ownloaded from

7. Shaper M, Hollingshead SK, Benjamin WH, Jr, Briles DE. 2004. PspAprotects Streptococcus pneumoniae from killing by apolactoferrin, andantibody to PspA enhances killing of pneumococci by apolactoferrin.Infect. Immun. 72:5031–5040. http://dx.doi.org/10.1128/IAI.72.9.5031-5040.2004.

8. Miyaji EN, Oliveira ML, Carvalho E, Ho PL. 2013. Serotype-independent pneumococcal vaccines. Cell. Mol. Life Sci. 70:3303–3326.http://dx.doi.org/10.1007/s00018-012-1234-8.

9. McDaniel LS, Ralph BA, McDaniel DO, Briles DE. 1994. Localization ofprotection-eliciting epitopes on PspA of Streptococcus pneumoniae be-tween amino acid residues 192 and 260. Microb. Pathog. 17:323–337. http://dx.doi.org/10.1006/mpat.1994.1078.

10. Roche H, Hakansson A, Hollingshead SK, Briles DE. 2003. Regions ofPspA/EF3296 best able to elicit protection against Streptococcus pneu-moniae in a murine infection model. Infect. Immun. 71:1033–1041. http://dx.doi.org/10.1128/IAI.71.3.1033-1041.2003.

11. Hollingshead SK, Becker R, Briles DE. 2000. Diversity of PspA: mosaicgenes and evidence for past recombination in Streptococcus pneumoniae.Infect. Immun. 68:5889 –5900. http://dx.doi.org/10.1128/IAI.68.10.5889-5900.2000.

12. Hollingshead SK, Baril L, Ferro S, King J, Coan P, Briles DE, Pneumo-coccal Proteins Epi Study Group. 2006. Pneumococcal surface protein A(PspA) family distribution among clinical isolates from adults over 50years of age collected in seven countries. J. Med. Microbiol. 55:215–221.http://dx.doi.org/10.1099/jmm.0.46268-0.

13. Pimenta FC, Ribeiro-Dias F, Brandileone MC, Miyaji EN, Leite LC,Sgambatti de Andrade AL. 2006. Genetic diversity of PspA types amongnasopharyngeal isolates collected during an ongoing surveillance study ofchildren in Brazil. J. Clin. Microbiol. 44:2838 –2843. http://dx.doi.org/10.1128/JCM.00156-06.

14. Vela Coral MC, Fonseca N, Castañeda E, Di Fabio JL, Hollingshead SK,Briles DE. 2001. Pneumococcal surface protein A of invasive Streptococcuspneumoniae isolates from Colombian children. Emerg. Infect. Dis. 7:832–836. http://dx.doi.org/10.3201/eid0705.017510.

15. Tart RC, McDaniel LS, Ralph BA, Briles DE. 1996. Truncated Strepto-coccus pneumoniae PspA molecules elicit cross-protective immunityagainst pneumococcal challenge in mice. J. Infect. Dis. 173:380 –386. http://dx.doi.org/10.1093/infdis/173.2.380.

16. Nabors GS, Braun PA, Herrmann DJ, Heise ML, Pyle DJ, GravensteinS, Schilling M, Ferguson LM, Hollingshead SK, Briles DE, Becker RS.2000. Immunization of healthy adults with a single recombinant pneumo-coccal surface protein A (PspA) variant stimulates broadly cross-reactiveantibodies to heterologous PspA molecules. Vaccine 18:1743–1754. http://dx.doi.org/10.1016/S0264-410X(99)00530-7.

17. Darrieux M, Moreno AT, Ferreira DM, Pimenta FC, de Andrade AL,Lopes AP, Leite LC, Miyaji EN. 2008. Recognition of pneumococcalisolates by antisera raised against PspA fragments from different clades. J.Med. Microbiol. 57:273–278. http://dx.doi.org/10.1099/jmm.0.47661-0.

18. Moreno AT, Oliveira ML, Ferreira DM, Ho PL, Darrieux M, Leite LC,Ferreira JM, Jr, Pimenta FC, Andrade AL, Miyaji EN. 2010. Immuni-zation of mice with single PspA fragments induces antibodies capable ofmediating complement deposition on different pneumococcal strains andcross-protection. Clin. Vaccine Immunol. 17:439 – 446. http://dx.doi.org/10.1128/CVI.00430-09.

19. Oliveira ML, Miyaji EN, Ferreira DM, Moreno AT, Ferreira PC, LimaFA, Santos FL, Sakauchi MA, Takata CS, Higashi HG, Raw I, KubruslyFS, Ho PL. 2010. Combination of pneumococcal surface protein A(PspA) with whole cell pertussis vaccine increases protection againstpneumococcal challenge in mice. PLoS One 5:e10863. http://dx.doi.org/10.1371/journal.pone.0010863.

20. Ho PL, Miyaji EN, Oliveira ML, Dias WDO, Kubrusly FS, TanizakiMM, Martins EAL, Raw I. 2011. Economical value of vaccines for thedeveloping countries–the case of Instituto Butantan, a public institutionin Brazil. PLoS Negl. Trop. Dis. 5:e1300. http://dx.doi.org/10.1371/journal.pntd.0001300.

21. Berstad AK, Andersen SR, Dalseg R, Dromtorp S, Holst J, Namork E,Wedege E, Haneberg B. 2000. Inactivated meningococci and pertussisbacteria are immunogenic and act as mucosal adjuvants for a nasal inac-tivated influenza virus vaccine. Vaccine 18:1910 –1919. http://dx.doi.org/10.1016/S0264-410X(99)00442-9.

22. Dagan R, Goldblatt D, Maleckar JR, Yaich M, Eskola J. 2004. Reduction ofantibody response to an 11-valent pneumococcal vaccine coadministered

with a vaccine containing acellular pertussis components. Infect. Immun.72:5383–5391. http://dx.doi.org/10.1128/IAI.72.9.5383-5391.2004.

23. Nicol M, Huebner R, Mothupi R, Käyhty H, Mbelle N, Khomo E,Klugman KP. 2002. Haemophilus influenzae type b conjugate vaccinediluted tenfold in diphtheria-tetanus-whole cell pertussis vaccine: a ran-domized trial. Pediatr. Infect. Dis. J. 21:138 –141. http://dx.doi.org/10.1097/00006454-200202000-00010.

24. Prikazsky V, Bock HL. 2001. Higher anti-hepatitis B response with com-bined DTPw-HBV vaccine compared with separate administration inhealthy infants at 3, 4 and 5 months of age in Slovakia. Int. J. Clin. Pract.55:156 –161.

25. Tamizifar H, Jennings R, Ali SA, Potter CW. 1997. Induction of IL-2 andIFN-gamma in BALB/c mice immunised with subunit influenza A vaccinein combination with whole cell or acellular DTP vaccine. J. Med. Micro-biol. 46:61– 66. http://dx.doi.org/10.1099/00222615-46-1-61.

26. Tamizifar H, Robinson A, Jennings R, Potter CW. 1995. Immuneresponse and protection against influenza A infection in mice immunisedwith subunit influenza A vaccine in combination with whole cell or acel-lular DTP vaccine. Vaccine 13:1539 –1546. http://dx.doi.org/10.1016/0264-410X(95)00092-F.

27. Berstad AK, Oftung F, Korsvold GE, Haugen IL, Froholm LO, Holst J,Haneberg B. 2000. Induction of antigen-specific T cell responses in hu-man volunteers after intranasal immunization with a whole-cell pertussisvaccine. Vaccine 18:2323–2330. http://dx.doi.org/10.1016/S0264-410X(00)00024-4.

28. Mills KH, Barnard A, Watkins J, Redhead K. 1993. Cell-mediatedimmunity to Bordetella pertussis: role of Th1 cells in bacterial clearance ina murine respiratory infection model. Infect. Immun. 61:399 – 410.

29. Fedele G, Nasso M, Spensieri F, Palazzo R, Frasca L, Watanabe M,Ausiello CM. 2008. Lipopolysaccharides from Bordetella pertussis andBordetella parapertussis differently modulate human dendritic cell func-tions resulting in divergent prevalence of Th17-polarized responses. J.Immunol. 181:208 –216. http://dx.doi.org/10.4049/jimmunol.181.1.208.

30. Fedele G, Spensieri F, Palazzo R, Nasso M, Cheung GY, Coote JG,Ausiello CM. 2010. Bordetella pertussis commits human dendritic cells topromote a Th1/Th17 response through the activity of adenylate cyclasetoxin and MAPK-pathways. PLoS One 5:e8734. http://dx.doi.org/10.1371/journal.pone.0008734.

31. Nasso M, Fedele G, Spensieri F, Palazzo R, Costantino P, Rappuoli R,Ausiello CM. 2009. Genetically detoxified pertussis toxin induces Th1/Th17 immune response through MAPKs and IL-10-dependent mecha-nisms. J. Immunol. 183:1892–1899. http://dx.doi.org/10.4049/jimmunol.0901071.

32. Lima FA, Ferreira DM, Moreno AT, Ferreira PC, Palma GM, FerreiraJM, Jr, Raw I, Miyaji EN, Ho PL, Oliveira ML. 2012. Controlledinflammatory responses in lungs is associated with protection elicited by apneumococcal surface protein A-based vaccine against a lethal respiratorychallenge with Streptococcus pneumoniae in mice. Clin. Vaccine Immunol.19:1382–1392. http://dx.doi.org/10.1128/CVI.00171-12.

33. Menozzi FD, Mutombo R, Renauld G, Gantiez C, Hannah JH, Lein-inger E, Brennan MJ, Locht C. 1994. Heparin-inhibitable lectin activityof the filamentous hemagglutinin adhesin of Bordetella pertussis. Infect.Immun. 62:769 –778.

34. Antoine R, Alonso S, Raze D, Coutte L, Lesjean S, Willery E, Locht C,Jacob-Dubuisson F. 2000. New virulence-activated and virulence-repressed genes identified by systematic gene inactivation and generationof transcriptional fusions in Bordetella pertussis. J. Bacteriol. 182:5902–5905. http://dx.doi.org/10.1128/JB.182.20.5902-5905.2000.

35. Antoine R, Locht C. 1990. Roles of the disulfide bond and the carboxy-terminal region of the S1 subunit in the assembly and biosynthesis ofpertussis toxin. Infect. Immun. 58:1518 –1526.

36. Higashi HG, Luna E, Precioso AR, Vilela M, Kubrusly FS, Dias WO,Raw I. 2009. Acellular and “low” pertussis vaccines: adverse events and therole of mutations. Rev. Inst. Med. Trop. Sao Paulo. 51:131–134. http://dx.doi.org/10.1590/S0036-46652009000300002.

37. Aida Y, Pabst MJ. 1990. Removal of endotoxin from protein solutions byphase separation using Triton X-114. J. Immunol. Methods 132:191–195.http://dx.doi.org/10.1016/0022-1759(90)90029-U.

38. Ferreira DM, Darrieux M, Silva DA, Leite LC, Ferreira JM, Jr, Ho PL,Miyaji EN, Oliveira ML. 2009. Characterization of protective mucosaland systemic immune responses elicited by pneumococcal surface proteinPspA and PspC nasal vaccines against a respiratory pneumococcal chal-

Salcedo-Rivillas et al.

980 cvi.asm.org Clinical and Vaccine Immunology

on July 7, 2020 by guesthttp://cvi.asm

.org/D

ownloaded from

lenge in mice. Clin. Vaccine Immunol. 16:636 – 645. http://dx.doi.org/10.1128/CVI.00395-08.

39. Ferreira PC, Campos IB, Abe CM, Trabulsi LR, Elias WP, Ho PL,Oliveira ML. 2008. Immunization of mice with Lactobacillus casei express-ing intimin fragments produces antibodies able to inhibit the adhesion ofenteropathogenic Escherichia coli to cultivated epithelial cells. FEMS Im-munol. Med. Microbiol. 54:245–254. http://dx.doi.org/10.1111/j.1574-695X.2008.00471.x.

40. Bitsaktsis C, Iglesias BV, Li Y, Colino J, Snapper CM, Hollingshead SK,Pham G, Gosselin DR, Gosselin EJ. 2011. Mucosal immunization with anunadjuvanted vaccine that targets S. pneumoniae PspA to human Fc re-ceptor type I protects against pneumococcal infection through comple-ment- and lactoferrin-mediated bactericidal activity. Infect. Immun. 80:1166 –1180. http://dx.doi.org/10.1128/IAI.05511-11.

41. Cunnion KM, Lee JC, Frank MM. 2001. Capsule production and growthphase influence binding of complement to Staphylococcus aureus. Infect.Immun. 69:6796 – 6803. http://dx.doi.org/10.1128/IAI.69.11.6796-6803.2001.

42. Van den Berg CW, Aerts PC, Van Dijk H. 1991. In vivo anti-complementary activities of the cobra venom factors from Naja naja andNaja haje. J. Immunol. Methods 136:287–294. http://dx.doi.org/10.1016/0022-1759(91)90015-8.

43. Lima FA, Miyaji EN, Quintilio W, Raw I, Ho PL, Oliveira ML. 2013.Pneumococcal surface protein A does not affect the immune responses toa combined diphtheria tetanus and pertussis vaccine in mice. Vaccine31:2465–2470. http://dx.doi.org/10.1016/j.vaccine.2013.03.026.

44. Mukerji R, Mirza S, Roche AM, Widener RW, Croney CM, Rhee DK,Weiser JN, Szalai AJ, Briles DE. 2012. Pneumococcal surface protein Ainhibits complement deposition on the pneumococcal surface by competingwith the binding of C-reactive protein to cell-surface phosphocholine. J. Im-munol. 189:5327–5335. http://dx.doi.org/10.4049/jimmunol.1201967.

45. Hot D, Antoine R, Renauld-Mongénie G, Caro V, Hennuy B, LevillainE, Huot L, Wittmann G, Poncet D, Jacob-Dubuisson F, Guyard C,Rimlinger F, Aujame L, Godfroid E, Guiso N, Quentin-Millet MJ,Lemoine Y, Locht C. 2003. Differential modulation of Bordetella pertussisvirulence genes as evidenced by DNA microarray analysis. Mol. Genet.Genomics 269:475– 486. http://dx.doi.org/10.1007/s00438-003-0851-1.

46. Ryan M, McCarthy L, Rappuoli R, Mahon BP, Mills KH. 1998. Pertussistoxin potentiates Th1 and Th2 responses to co-injected antigen: adjuvantaction is associated with enhanced regulatory cytokine production andexpression of the co-stimulatory molecules B7-1, B7-2 and CD28. Int.Immunol. 10:651– 662. http://dx.doi.org/10.1093/intimm/10.5.651.

47. Poulain-Godefroy O, Menozzi FD, Alonso S, Vendeville C, Capron A,Locht C, Riveau G. 2003. Adjuvant activity of free Bordetella pertussisfilamentous haemagglutinin delivered by mucosal routes. Scand. J. Immu-nol. 58:503–510. http://dx.doi.org/10.1046/j.1365-3083.2003.01336.x.

48. Poulain-Godefroy O, Mielcarek N, Ivanoff N, Remoué F, Schacht AM,Phillips N, Locht C, Capron A, Riveau G. 1998. Bordetella pertussisfilamentous hemagglutinin enhances the immunogenicity of liposome-delivered antigen administered intranasally. Infect. Immun. 66:1764 –1767.

49. Orr B, Douce G, Baillie S, Parton R, Coote J. 2007. Adjuvant effects ofadenylate cyclase toxin of Bordetella pertussis after intranasal immunisa-tion of mice. Vaccine 25:64 –71. http://dx.doi.org/10.1016/j.vaccine.2006.07.019.

50. Arulanandam BP, Lynch JM, Briles DE, Hollingshead S, Metzger DW.2001. Intranasal vaccination with pneumococcal surface protein A andinterleukin-12 augments antibody-mediated opsonization and protectiveimmunity against Streptococcus pneumoniae infection. Infect. Immun. 69:6718 – 6724. http://dx.doi.org/10.1128/IAI.69.11.6718-6724.2001.

51. Ferreira DM, Darrieux M, Oliveira ML, Leite LC, Miyaji EN. 2008. Opti-mized immune response elicited by a DNA vaccine expressing pneumococcalsurface protein A is characterized by a balanced immunoglobulin G1 (IgG1)/IgG2a ratio and proinflammatory cytokine production. Clin. Vaccine Immu-nol. 15:499–505. http://dx.doi.org/10.1128/CVI.00400-07.

52. Fowler S, Xing DK, Bolgiano B, Yuen CT, Corbel MJ. 2003. Modifica-tions of the catalytic and binding subunits of pertussis toxin by formalde-hyde: effects on toxicity and immunogenicity. Vaccine 21:2329 –2337.http://dx.doi.org/10.1016/S0264-410X(03)00104-X.

53. Oh H, Kim BG, Nam KT, Hong SH, Ahn DH, Choi GS, Kim H, HongJT, Ahn BY. 2013. Characterization of the carbohydrate binding andADP-ribosyltransferase activities of chemically detoxified pertussis toxins.Vaccine 31:2988 –2993. http://dx.doi.org/10.1016/j.vaccine.2013.04.060.

PT Enhances Immune Responses to PspA

July 2014 Volume 21 Number 7 cvi.asm.org 981

on July 7, 2020 by guesthttp://cvi.asm

.org/D

ownloaded from