Microbiology (2014), 160, 962–969

DOI 10.1099/mic.0.074690-0

Antibody-mediated inhibition of Bordetella pertussis adenylate cyclase–haemolysin-induced macrophage cytotoxicity is influenced by variations in the bacterial population N. Hegerle1,2 and N. Guiso1,2 Correspondence N. Hegerle [email protected]

Received 4 November 2013 Accepted 17 February 2014

1

Prevention and Molecular Therapies of Human Diseases Unit, Institut Pasteur, Paris, France

2

Institut Pasteur, CNRS-URA3012, Paris, France

Whooping cough is a vaccine-preventable disease presenting with epidemic cycles linked to natural and/or vaccine-driven evolution of the aetiological agent of the disease, Bordetella pertussis. Adenylate cyclase–haemolysin (AC-Hly) is a major toxin produced by this pathogen, which mediates macrophage apoptosis in vitro and in vivo. While current acellular pertussis vaccine (APV) formulations do not include AC-Hly, they all contain pertussis toxin and can comprise filamentous haemagglutinin (FHA), which interacts with AC-Hly, and pertactin (PRN), which has been hypothesized also to interact with AC-Hly. We aimed to study the capacity of specific antibodies to inhibit the in vitro B. pertussis AC-Hly-mediated cytotoxicity of J774A.1 murine macrophages in a background of a changing bacterial population. We demonstrate that: (i) clinical isolates of different types or PRN phenotype are all cytotoxic and lethal in the mouse model of respiratory infection; (ii) lack of PRN production does not impact AC-Hly-related phenotypes; (iii) anti-AC-Hly antibodies inhibit cell lysis whatever the phenotype of the isolate, while anti-PRN antibodies significantly inhibit cell lysis provided the isolate produces this antigen, which might be relevant in vivo for APV-induced immunity; and (iv) anti-FHA antibodies only inhibit lysis induced by isolates collected in 2012, maybe indicating specific characteristics of epidemic lineages of B. pertussis.

INTRODUCTION Bordetella pertussis is the Gram-negative bacterium responsible for whooping cough, a severe acute respiratory disease life-threatening for partially or unvaccinated newborns. The colonization of the human respiratory tract by B. pertussis is achieved through expression of an array of virulence factors that promote adhesion to the respiratory epithelium and subversion of the immune system to avoid rapid clearance of the pathogen (Mattoo & Cherry, 2005). While adenylate cyclase–haemolysin (AC-Hly) and pertussis toxin (PT) are well-known toxins involved in immune modulation (Carbonetti, 2010; Fedele et al., 2013), filamentous haemagglutinin (FHA), pertactin (PRN), fimbriae 2 (FIM2) and FIM3 are adhesins involved in the mechanical adherence of B. pertussis to the respiratory epithelium (Jacob-Dubuisson & Locht, 2007). Colonization of the respiratory tract of partially or unvaccinated newborns is accompanied by characteristic clinical manifestations and can be fatal (Cherry et al., 2012; Mattoo & Cherry, 2005). To protect Abbreviations: AC, adenylate cyclase; APV, acellular pertussis vaccine; FHA, filamentous haemagglutinin; Hly, haemolysin; PRN, pertactin; PT, pertussis toxin; WPV, whole-cell pertussis vaccine.

962

this population from the disease, whole-cell pertussis vaccine (WPV) was introduced in 1959 in France for primary and booster immunization of children. This strategy managed to control circulating isolates similar to WPV strains, and dramatically decreased the morbidity and mortality of the disease in this age group (Baron et al., 1998; Bonmarin et al., 2007; Bouchez et al., 2008; Caro et al., 2005, 2006; Hegerle & Guiso, 2013; Weber et al., 2001). In 1998, acellular pertussis vaccines (APVs) were introduced for adolescent booster vaccination (Institut de Veille Sanitaire, 1998) and afterwards replaced WPV for child immunization as well. Adult APV booster vaccination was later introduced (Institut de Veille Sanitaire, 2004, 2008), with the aim of protecting newborn children from pertussis by increasing the herd immunity. But despite vaccination, whooping cough still remains a major issue and resurgence of this disease has occurred in several countries (He & Mertsola, 2008; Mooi et al., 2014; Plotkin, 2014). WPV is composed of killed bacteria and induces a broadrange immune response against B. pertussis, while all licensed APVs include PT in addition to FHA, or FHA and PRN, or FHA, PRN, FIM2 and FIM3. APV-induced immunity is thus narrower, but targets every circulating isolate producing the 074690 G 2014 The Authors

Printed in Great Britain

Inhibition of B. pertussis-induced macrophage lysis

antigens included in the vaccine. Although demonstrated to be a major toxin in the murine model of respiratory infection (Bouchez et al., 2009; Fedele et al., 2013; Khelef et al., 1994), AC-Hly has not been included in any APV formulation so far. This toxin drives CD11b/CD18-expressing eukaryotic cells (Osickova et al., 2010) to an apoptotic state in vivo (Gueirard et al., 1998) and in vitro (Khelef et al., 1993; Osickova et al., 2010), and can thus directly target airway macrophages in its human host. Although capable of inducing apoptosis on its own (Khelef & Guiso, 1995), AC-Hly has been demonstrated to interact with FHA in vitro in a non-covalent manner (Zaretzky et al., 2002), the lack of this adhesin causing an increased secretion of AC-Hly in the culture supernatant (Zaretzky et al., 2002). Interactions between PRN and AC-Hly have only been hypothesized so far (Khelef & Guiso, 1995), but could draw attention in a background of an increasing prevalence of natural PRN-deficient isolates in countries vaccinating with APVs (Barkoff et al., 2012; Hegerle & Guiso, 2013; Miyaji et al., 2013; Otsuka et al., 2012; Pawloski et al., 2014; Queenan et al., 2013; Quinlan et al., 2013). The loss of this adhesin does not impact the virulence or transmission of the bacterium (Bodilis & Guiso, 2013) and, despite the effectiveness of current APVs in preventing severe pertussis in children less than 6 months of age (Bodilis & Guiso, 2013), great concerns have arisen regarding the duration of APV-induced immunity if the prevalence of PRN-deficient isolates increases. We thus sought to analyse the possible interaction existing between PRN and AC-Hly, and the changes that can occur in the B. pertussis AC-Hly-related phenotype by means of enzymic activity determination and AC-Hly-mediated macrophage cytotoxicity. To study possible variations in antibody recognition of B. pertussis isolates from different vaccine era or pertussis cycles, we performed inhibition assays with specific antisera targeting AC-Hly, PT, PRN and FHA. The murine model of respiratory infection was also used to assess differences in the in vivo virulence of the isolates.

METHODS Isolates and growth conditions. B. pertussis reference strain Tohama and WPV strains Bp1414 and Bp1416 were part of the collection of the Institut Pasteur (Njamkepo et al., 2002). PRNdeficient and PRN-producing clinical isolates belonged to the collection of the National Reference Centre held in our laboratory since 1993. Strain and isolate characteristics are summarized in Table 1.

Bacteria were grown on Bordet–Gengou agar (BGA; Difco) supplemented with 15 % defibrinated sheep blood at 36 uC for 72 h and plated again for 24 h before each cytotoxicity assay. When necessary, isolates and reference strains were cultured in modified Stainer–Scholte medium until OD650 1, starting from OD650 0.2. Characterization of clinical isolates was performed by PFGE, genotyping and serotyping as described elsewhere (Mooi et al., 2000, 2009). Determination of AC activity. For AC activity measurement,

isolates were grown as previously described (Hegerle et al., 2013) in http://mic.sgmjournals.org

enriched Stainer–Scholte medium and harvested at mid-exponential phase at OD650 1±0.2. For secreted AC activity, the bacterial suspension was centrifuged at 6000 g for 10 min in a bench centrifuge and the resulting supernatant tested for AC activity by radioactive assay as previously described (Ladant et al., 1986). The proportion of membrane-bound AC activity was estimated as the ratio between the total AC activity (total ACact) minus the supernatant AC activity (supernatant ACact) and the total AC activity. Membrane-bound activity (%) =

Total ACact − Supernatant ACact ×100 Total ACact

Production of murine polyclonal antisera. Specific polyclonal

mouse antisera directed against AC-Hly, PT, FHA and PRN were obtained as described by Weber et al. (2001). Briefly, 4-week-old BALB/c mice were immunized subcutaneously four times at 4 week intervals with 10 mg of each purified antigen adsorbed onto aluminium hydroxide and were bled 7 days after the last injection. PT (PtxA2), FHA and PRN (PRN1) were purified from B. pertussis in their native form, while AC-Hly was produced as a recombinant protein (Betsou et al., 1993). PtxA2- and PRN1 type antigens were used since they are the vaccine type antigens of currently licensed APVs. The specificity of each polyclonal antiserum was verified by Western blotting against the purified antigens and whole B. pertussis bacterial extracts. All murine immune and naive sera were heat inactivated for 30 min at 56 uC prior to use in the inhibition assays. Western blot analysis of virulence factor production. Clinical

isolates were tested for virulence factor production by Western blot using the mouse polyclonal antisera obtained after immunizing mice with purified antigens as described above and by Weber et al. (2001). J774A.1 cell culture and cytotoxic/inhibition assays. J774A.1

(ATCC TIB-67) murine monocyte/macrophage cells were cultured and cytotoxic assays were performed as previously described (Bouchez et al., 2009). Briefly, the J774A.1 murine macrophages were cultured in TPP T75 filter-capped flasks at 37 uC with 5 % CO2 in RPMI 1640+Glutamax 1 (Invitrogen) complemented with 1 % sodium pyruvate 100 mM solution, 1 % HEPES 1 M solution, 1 % antibiotic– antimycotic solution (Invitrogen) and 10 % FBS (BioWest). Cells were transferred into 92-well plates 24 h before lysis experiments were conducted at a final concentration of 56104 cells per well. On the day of infection, adherent cells were washed three times with the same culture medium lacking antibiotic–antimycotic solution, and the number of cells was determined by counting after Trypan blue staining. B. pertussis isolates were suspended in cell culture medium and were added to the cells at a final m.o.i. of 100 : 1 in a final volume of 100 ml per well of the 96-well culture plate. Bacteria were deposited by light centrifugation at 600 g for 5 min at room temperature prior to incubation. Bacteria and cells were left in contact for 8 h and cell lysis was assessed with the CytoTox 96 non-radioactive assay kit (Promega). Cytotoxicity was determined as a percentage of cell lysis determined by OD495 measurement as presented in the formula below, where ‘OD495 isolate’ is the measured optical density of the well containing the B. pertussis isolate, ‘OD495 control’ is the optical density of the control well containing only cells, and ‘OD495 total lysis’ is the optical density measured in the well where all cells are lysed by adding the Promega lysis buffer. Cell lysis (%) =

OD495 isolate − OD495 control OD495 total lysis − OD4955 control

×100

For inhibition assays, the bacterial suspensions were incubated for 30 min at 37 uC with 5 % CO2 and with the appropriate polyclonal immune serum at a final dilution of 1 : 500. This serum dilution was chosen as it causes 50 % lysis inhibition of the strain Tohama 963

N. Hegerle and N. Guiso

Table 1. Characteristics of selected B. pertussis strains and isolates Isolates shown in bold were collected during the last pertussis cycle in France (2012). Isolate

Year

Tohama Bp1414 Bp1416 FR3713 FR3749 FR4929 FR5133 FR5392 FR3693 FR3793 FR4596 FR4684 FR4953 FR5187 FR5388

,1954 ,1954 ,1954 2007 2007 2011 2012 2012 2007 2007 2009 2010 2011 2012 2012

WB* + + + + PT2 + + + PRN2 (IS) PRN2 (D) PRN2 (IS) PRN2 (D84) PRN2 (stop) PRN” (D) PRN” (IS)

ND,

Not determined;

NL,

Not lethal.

FIMD

Type (PFGE/ptxP/ptxA/PRN) d

ACact/ACbound (U ml”1/% total)§

LD50 (c.f.u. per mouse)

Reference

2 2/3 3 3 3 3 3 3 3 3 3 3 3 3 3

II/1/1/1 II/1/4/1 III/1/2/1 IVc/3/1/2 IVc/D/D/2 IVc/3/1/ 2 IVb/3/1/2 IVb/3/1/2 IVa/3/1/2 IVa/3/1/2 IVa/3/1/2 III/1/2/1 IVb/3/1/2 IVa/3/1/2 IVa/3/1/2

43.1±3.2/89.1±1.3 44.2±1.5/88.8±3.8 41.0±4.5/88.2±5.8 54.0±8.2/98.8±1.0 53.6±5.6/96.9±0.4 ND/ND 49.1±7.4/100±0.0 58.8±9.9/94.0±5.1 46.8±7.8/97.3±0.3 55.9±1.8/98.6±0.4 56.7±9.4/93.9±0.8 51.7±7.4/94.8±3.6 ND/ND 62.7±6.2/100±0.0 50.3±9.9/89.5±8.1

86107 86106 36107 26107

Bouchez et al. (2009) Hegerle et al, (2012) Hegerle et al. (2012) Hegerle et al. (2012) Bouchez et al. (2009) This study This study This study Bouchez et al. (2009 Bouchez et al. (2009) Hegerle et al. (2012) This study This study This study This study

NL

56107 6¾107 3¾107 66107 26108 36107 16107 86107 2¾107 4¾107

*Western blot (WB) analyses were conducted using specific anti-FHA, anti-PRN, anti-PT and anti-AC-Hly antisera. If not specified, the isolate expressed all tested virulence factors and is designated by +. D, Deletion of gene; D84, deletion of 84 bp at the beginning of the gene; IS, insertion sequence inside the gene of interest; stop, SNP causing a stop codon at position 1479. DSerotype of the isolate is given as the type of fimbriae (FIM) produced. dIsolate type is given by its PFGE group, PT operon promoter (ptxP) allele type, PT subunit A (ptxA) allele type and prn allele type as follows. §Total AC activity (ACact) of the bacterial suspension is given as the amount (nmol) of cAMP formed min21 ml21 (U ml21), while the ‘bound’ activity corresponds to the proportion of ACact that remains bound to the cell surface. Values are given as the mean±SD for at least three independent measurements.

after 8 h of incubation. The specific polyclonal immune sera used for inhibition were those used for Western blotting and thus directed against AC-Hly, PT, FHA and PRN. As FHA and PT were unable to inhibit Tohama-induced cell lysis, they were used at an arbitrary 1 : 500 dilution. Naive serum at a final dilution of 1 : 500, obtained from un-immunized mice, was used as the reference for the determination of the statistical significance of cell lysis inhibition. As a CO2-responsive regulon exists in B. pertussis (Hester et al., 2012), the control condition consisting of the same isolate at the same m.o.i. without serum was incubated under the same conditions to avoid any bias. Inhibition was calculated as the ratio between lysis without serum minus the lysis in the presence of serum and the lysis without serum, as shown below. Lysis inhibition (%) =

Isolate lysis − Isolate lysis with serum ×100 Isolatte lysis

RESULTS AC activity and secretion All isolates produced comparable amounts of AC activity in vitro regardless of their PRN phenotype (Table 1) and were haemolytic on BGA plates. AC-Hly toxin thus conserved both its pore-forming haemolytic activity and its AC activity whether it was produced in a PRN-deficient or in a PRN-producing isolate. The absence of PRN production does not impact the distribution of AC activity as over 90 % remained surface bound whatever the phenotype of the isolate (Table 1), which is consistent with published observations (Zaretzky et al., 2002). B. pertussis-induced macrophage lysis

Murine intra-nasal model of infection. All procedures involving

animals were conducted in accordance with the Institut Pasteur Animal Care and Use Committee guidelines. LD50 determinations were conducted as described elsewhere (Bouchez et al., 2009). Statistical analysis. All cytotoxicity results presented are the mean±SE for a minimum of three independent assays. For each isolate, a one-way ANOVA followed by a Dunnett comparison posttest against the inhibition achieved by naive serum was performed using GraphPad Prism version 5.0b for Mac (GraphPad Software). A normality test was performed for Tohama and confirmed the Gaussian distribution of the results.

964

All isolates were cytotoxic for J774A.1 murine macrophages in vitro whatever their phenotype (Fig. 1). Although variable, no significant difference or trend was found in cell cytotoxicity. The loss of PRN production thus does not impair the toxicity of B. pertussis isolates to macrophages. Inhibition of B. pertussis-induced macrophage lysis Serum obtained from unvaccinated naive mice was heat inactivated and used as a control and reference for the Microbiology 160

Inhibition of B. pertussis-induced macrophage lysis

calculation of cell lysis inhibition. Co-incubating bacteria with the naive serum did not impact their cytotoxicity (Fig. 2a); neither did the serum impact J774A.1 cell viability throughout the incubation time (not shown).

produce PRN and dropped to a mean of 5.0±1.0 % inhibition for PRN-deficient isolates (Fig. 2c). Murine model of intra-nasal infection

As expected, immune serum containing polyclonal antibodies directed against AC-Hly significantly inhibited cell lysis in every case (Fig. 2a), with a mean inhibition of 67.0±3.2 % for the tested isolates. In contrast, anti-PT antibodies never inhibited cell lysis (not shown), which was in accordance with our expectations, as it does not play a role in macrophage lysis.

As previously reported (Bouchez et al., 2009; Hegerle et al., 2012), apart from the PT-deficient isolate FR3749, all isolates remained capable of establishing a lethal colonization in vivo in the mouse model of respiratory infection. Despite inter-isolate differences such as observed between Bp1414 and FR3793 (PRN-deficient), intra-species variability was low and LD50 values were quite homogeneous whatever the phenotype of the tested B. pertussis isolate (Table 1). The PRN deficiency of FR3793 cannot alone explain the increased LD50, which is probably linked to other unknown genetic and/or proteomic differences.

Anti-FHA antibodies presented variable results depending more on the isolate’s era of collection than its PRN phenotype (Fig. 2b). It thus appeared that the anti-FHA antiserum significantly inhibited cell lysis induced by FR5133 (51.5±8.2 %), FR5187 (PRN2; 59.2±4.1 %), FR5388 (PRN2; 56.3±8.9 %) and FR5392 (63.4±10.1 %), collected during the last French epidemic of pertussis in 2012, while the other isolates remained unaffected by the addition of anti-FHA antibodies.

DISCUSSION AC-Hly is the secreted B. pertussis toxin responsible for macrophage apoptosis (Gueirard et al., 1998; Khelef et al., 1993; Osickova et al., 2010) and is mandatory for establishment of successful colonization in the mouse model of respiratory infection (Carbonetti et al., 2005; Khelef et al., 1992). Its importance for the pathogenesis of pertussis is further underlined by the rarity of isolates not producing

The results obtained with anti-PRN antiserum are more surprising as this serum was able to significantly inhibit B. pertussis-induced cell lysis with a mean inhibition of 72.3± 4.1 % for isolates producing PRN. The inhibitory capacity of anti-PRN antiserum was abolished if the isolates did not

100

J774A.1 lysis (%)

80

60

40

20

+

+

+ APV

+

+







– APV







To

ha

m a Bp 14 14 Bp 14 16 FR 37 13 FR 37 49 FR 49 29 FR 51 33 FR 53 92 FR 36 93 FR 37 93 FR 45 96 FR 46 84 FR 49 53 FR 51 87 FR 53 88

0 + + PRN + Era Pre-vaccine

Isolate

Fig. 1. Lysis of J774A.1 macrophages (%) after 8 h of incubation with the selected strains or isolates of B. pertussis. Results presented are the mean±SE for at least three independent experiments. Statistical analysis did not reveal any significant difference between isolates, with a confidence interval of 95 %. http://mic.sgmjournals.org

965

N. Hegerle and N. Guiso ***

100 ***

***

80 ***

***

60

***

***

***

***

***

***

***

***

***

***

40 20 0 + + PRN + Era Pre-vaccine

+

+ + ApV

+







– – ApV





14 14 14 1 FR 6 37 1 FR 3 37 4 FR 9 49 2 FR 9 51 3 FR 3 53 9 FR 2 36 9 FR 3 37 9 FR 3 45 9 FR 6 46 8 FR 4 49 5 FR 3 51 8 FR 7 53 88

+

100 80

*** ***

60



– 88

53

87

FR

53

51

FR

84



– ApV





49

FR

96

46

93

93

37

92



36



***

*** ***

***

***



FR

***

– ApV

45

80



FR

***



FR

100



FR

53

33

+

FR

51

29

+

FR

49

FR

37

FR

37

+ ApV

49

+

13

+

FR

Bp

Bp

14

14

m ha

16

+ + Pre-vaccine

a

+

To

***

60 40 20 0 PRN Era

+

+

+ ApV

+

+



Bp 14

To

ha

m

a

+ + + Pre-vaccine

1 Bp 4 14 1 FR 6 37 1 FR 3 37 4 FR 9 49 2 FR 9 51 3 FR 3 53 9 FR 2 36 9 FR 3 37 9 FR 3 45 9 FR 6 46 8 FR 4 49 5 FR 3 51 8 FR 7 53 88

Anti-PRN inhibition of cell lysis (%)

***

20 0 PRN Era

(c)

***

40

14

Anti-FHA inhibition of cell lysis (%)

(b)

Bp

Bp

To ha

m a

Anti-AC-Hly inhibition of cell lysis (%)

(a)

Fig. 2. Inhibition of J774A.1 cell lysis by (a) naive serum [black circles], and polyclonal immune sera (grey bars) targeting (a) ACHly, (b) FHA and (c) PRN. Results are the mean±SE for at least three independent experiments. Inhibition was calculated against the control condition without serum as presented in Methods. Statistical significance was determined between the inhibition of the naive serum and the inhibition of the polyclonal immune serum being considered; ***P,0.05. 966

Microbiology 160

Inhibition of B. pertussis-induced macrophage lysis

this toxin (C.-H. Wirsing von Koenig, Direktor des Instituts fu¨r Hygiene und Labormedizin, HELIOS-Klinikum Krefeld, Krefeld, Germany, personal communication), as well as the genetic invariability of cyaA, the gene encoding this virulence factor (Chenal-Francisque et al., 2009; unpublished data). The determination of AC enzymic activity of clinical PRNdeficient and PRN-producing isolates demonstrates that this particular adhesin was not involved in the distribution or amount of AC activity produced (Table 1). All isolates were also cytotoxic for J774A.1 murine macrophages in vitro and no statistically significant difference was found whatever the phenotype of the isolate (Fig. 1). Thus, there is no evidence of any PRN–AC-Hly interaction being involved in the process of macrophage cytotoxicity in vitro. These results are consistent with the in vivo murine model of respiratory infection since all tested isolates are evenly lethal, apart from FR3793, for which there is no obvious reason explaining this phenotype (Table 1). Nonetheless, while the expected inhibition of cell lysis with specific anti-AC-Hly antiserum occurred whatever the isolate’s phenotype (Fig. 2a), specific anti-PRN antiserum was capable of inhibiting macrophage lysis induced by B. pertussis isolates that still produced PRN, the lack of which impaired this inhibitory capacity (Fig. 2c). As these two virulence factors did not seem to directly interact according to previous observations, this could be linked to anti-PRN antibodies preventing the correct secretion and presentation of AC-Hly to the target cell, but it remains that these antibodies are thus capable of inhibiting macrophage cytotoxicity in vitro and therefore might also act in vivo. Despite the known interaction between FHA and AC-Hly, anti-FHA antiserum had no effect on cell lysis induced by the majority of tested isolates (Fig. 2b). FR5133, FR5187 (PRN-deficient), FR5388 (PRN-deficient) and FR5392, collected during the last French cycle of pertussis in 2012, were the only isolates sensitive to anti-FHA antiserum. No mutation was found in fhaB, the gene encoding FHA, in FR5133 or in FR5187 as compared with the reference strain Tohama, whereas sphB1, the subtilisin-like protease involved in FHA processing and maturation (Coutte et al., 2001), presented a non-synonymous SNP. This mutation was also found in the isolate FR4929 (unpublished data), which was not inhibited by anti-FHA antiserum, thus ruling out possible involvement of this SNP in the observed anti-FHAmediated inhibition of cell lysis. Besides vaccine-driven evolution of B. pertussis populations (Octavia et al., 2011), epidemic lineages circulate (Advani et al., 2009; Weber et al., 2001) and need to be considered as well since they are possibly responsible for the characteristic cycles of pertussis, as they might present a particular phenotype and/or fitness. The result obtained with anti-FHA antiserum might reflect such a population of B. pertussis, as FR5133 and FR5187 were collected during the same pertussis cycle and are closely related, as evidenced by their segregation in a neighbour-joining tree built from the concatenated sequences of their SNPs obtained against B. http://mic.sgmjournals.org

pertussis reference strain Tohama (unpublished data). These two isolates, as well as FR5388 and FR5392, may thus share specific characteristics, yet to be determined, that impact the anti-FHA-mediated inhibition of cell cytotoxicity. While PT, FHA and PRN can be targeted by APV-induced immunity, AC-Hly cannot, and any related phenotype should, therefore, remain unaffected during infection in an APV-immunized host. In vitro studies, however, demonstrated that AC-Hly interacts with FHA (Zaretzky et al., 2002) and it has been hypothesized that PRN might interact with this toxin as well (Khelef & Guiso, 1995). These interactions may participate in pertussis pathogenesis since a higher concentration of AC-Hly is localized at the site of adhesion through non-covalent interaction with FHA (Zaretzky et al., 2002), while AC-Hly seems mandatory for optimal FHA binding activity to epithelial cells (Perez Vidakovics et al., 2006). We here demonstrate that loss of PRN production does not impact AC-Hly-related phenotypes, whereas the presence of specific anti-FHA or antiPRN antibodies does. Although these inhibitory capacities only occur if the isolates were collected during the last cycle of pertussis in 2012 or produce PRN, active immunization with APVs may impact AC-Hly-related phenotypes in vivo despite the lack of this toxin in all currently licensed APVs. Loss of PRN production might thus be an advantage for this pathogen in an APV-immunized population, while the sudden inhibition obtained with anti-FHA antiserum for isolates collected in 2012 remains puzzling. Further in vivo and field studies are needed to explore the impact of PRN deficiency on the effectiveness of APVinduced immunity, but all licensed APVs include PT and monovalent PT APV is capable of inducing a protective immunity in the population (Petersen et al., 2012). Most APVs also induce anti-FHA antibodies, which seem to impair in vitro macrophage cytotoxicity induced by currently circulating isolates, which may also be relevant in vivo. Although development of new vaccines and evolution of vaccination strategies will probably occur in the next decade, high vaccine coverage with effective APVs will increase herd immunity and should enable control of the disease and protect children in the meanwhile.

ACKNOWLEDGEMENTS We thank GlaxoSmithKline laboratories for the gift of purified antigens. This work was supported by the Institut Pasteur Foundation (URA CNRS3012) and GlaxoSmithKline Biologicals, Rixensart, Belgium.

REFERENCES Advani, A., Van der Heide, H. G., Hallander, H. O. & Mooi, F. R. (2009). Analysis of Swedish Bordetella pertussis isolates with three

typing methods: characterization of an epidemic lineage. J Microbiol Methods 78, 297–301. Barkoff, A. M., Mertsola, J., Guillot, S., Guiso, N., Berbers, G. & He, Q. (2012). Appearance of Bordetella pertussis strains not expressing the

967

N. Hegerle and N. Guiso vaccine antigen pertactin in Finland. Clin Vaccine Immunol 19, 1703– 1704.

Hegerle, N., Paris, A. S., Brun, D., Dore, G., Njamkepo, E., Guillot, S. & Guiso, N. (2012). Evolution of French Bordetella pertussis and

Baron, S., Njamkepo, E., Grimprel, E., Begue, P., Desenclos, J. C., Drucker, J. & Guiso, N. (1998). Epidemiology of pertussis in French

Bordetella parapertussis isolates: increase of bordetellae not expressing pertactin. Clin Microbiol Infect 18, E340–E346.

hospitals in 1993 and 1994: thirty years after a routine use of vaccination. Pediatr Infect Dis J 17, 412–418.

Hegerle, N., Rayat, L., Dore, G., Zidane, N., Bedouelle, H. & Guiso, N. (2013). In-vitro and in-vivo analysis of the production of the

Betsou, F., Sebo, P. & Guiso, N. (1993). CyaC-mediated activation

Bordetella type three secretion system effector A in Bordetella pertussis, Bordetella parapertussis and Bordetella bronchiseptica. Microbes Infect 15, 399–408.

is important not only for toxic but also for protective activities of Bordetella pertussis adenylate cyclase–hemolysin. Infect Immun 61, 3583–3589. Bodilis, H. & Guiso, N. (2013). Virulence of pertactin-negative

Bordetella pertussis isolates from infants, France. Emerg Infect Dis 19, 471–474. Bonmarin, I., Levy-Bruhl, D., Baron, S., Guiso, N., Njamkepo, E., Caro, V. & Renacoq (2007). Pertussis surveillance in French hospitals:

Hester, S. E., Lui, M., Nicholson, T., Nowacki, D. & Harvill, E. T. (2012). Identification of a CO2 responsive regulon in Bordetella. PLoS

ONE 7, e47635. Institut de Veille Sanitaire (1998). Calendrier vaccinal 1998. Bull

Epide´miol Hebd 1998, 15. Institut de Veille Sanitaire (2004). Calendrier vaccinal 2004. Bull

results from a 10 year period. Euro Surveill 12, 678.

Epide´miol Hebd 2004, 28–29.

Bouchez, V., Caro, V., Levillain, E., Guigon, G. & Guiso, N. (2008).

Institut de Veille Sanitaire (2008). Calendrier vaccinal 2008. Bull

Genomic content of Bordetella pertussis clinical isolates circulating in areas of intensive children vaccination. PLoS ONE 3, e2437.

Jacob-Dubuisson, F. & Locht, C. (2007). The Bordetella adhesins.

Bouchez, V., Brun, D., Cantinelli, T., Dore, G., Njamkepo, E. & Guiso, N. (2009). First report and detailed characterization of B. pertussis isolates

In Bordetella: Molecular Microbiology. pp. 69–97. Edited by F. C. Locht. Wymondham: Horizon Bioscience.

not expressing pertussis toxin or pertactin. Vaccine 27, 6034–6041.

Khelef, N. & Guiso, N. (1995). Induction of macrophage apoptosis by

Carbonetti, N. H. (2010). Pertussis toxin and adenylate cyclase toxin:

Bordetella pertussis adenylate cyclase–hemolysin. FEMS Microbiol Lett 134, 27–32.

key virulence factors of Bordetella pertussis and cell biology tools. Future Microbiol 5, 455–469.

Epide´miol Hebd 2008, 16–17.

Khelef, N., Sakamoto, H. & Guiso, N. (1992). Both adenylate cyclase

Carbonetti, N. H., Artamonova, G. V., Andreasen, C. & Bushar, N. (2005). Pertussis toxin and adenylate cyclase toxin provide a one–

and hemolytic activities are required by Bordetella pertussis to initiate infection. Microb Pathog 12, 227–235.

two punch for establishment of Bordetella pertussis infection of the respiratory tract. Infect Immun 73, 2698–2703.

Khelef, N., Zychlinsky, A. & Guiso, N. (1993). Bordetella pertussis

Caro, V., Njamkepo, E., Van Amersfoorth, S. C., Mooi, F. R., Advani, A., Hallander, H. O., He, Q., Mertsola, J., Riffelmann, M. & other authors (2005). Pulsed-field gel electrophoresis analysis of Bordetella

Khelef, N., Bachelet, C. M., Vargaftig, B. B. & Guiso, N. (1994).

pertussis populations in various European countries with different vaccine policies. Microbes Infect 7, 976–982. Caro, V., Hot, D., Guigon, G., Hubans, C., Arrive´, M., Soubigou, G., Renauld-Monge´nie, G., Antoine, R., Locht, C. & other authors (2006). Temporal analysis of French Bordetella pertussis isolates by

comparative whole-genome hybridization. Microbes Infect 8, 2228– 2235. Chenal-Francisque, V., Caro, V., Boursaux-Eude, C. & Guiso, N. (2009). Genomic analysis of the adenylate cyclase–hemolysin C-

terminal region of Bordetella pertussis, Bordetella parapertussis and Bordetella bronchiseptica. Res Microbiol 160, 330–336. Cherry, J. D., Tan, T., Wirsing von Ko¨nig, C. H., Forsyth, K. D., Thisyakorn, U., Greenberg, D., Johnson, D., Marchant, C. & Plotkin, S. (2012). Clinical definitions of pertussis: summary of a Global

Pertussis Initiative Roundtable Meeting, February 2011. Clin Infect Dis 54, 1756–1764. Coutte, L., Antoine, R., Drobecq, H., Locht, C. & Jacob-Dubuisson, F. (2001). Subtilisin-like autotransporter serves as maturation protease

in a bacterial secretion pathway. EMBO J 20, 5040–5048. Fedele, G., Bianco, M. & Ausiello, C. M. (2013). The virulence factors

of Bordetella pertussis: talented modulators of host immune response. Arch Immunol Ther Exp (Warsz) 61, 445–457. Gueirard, P., Druilhe, A., Pretolani, M. & Guiso, N. (1998). Role of

adenylate cyclase–hemolysin in alveolar macrophage apoptosis during Bordetella pertussis infection in vivo. Infect Immun 66, 1718–1725. He, Q. & Mertsola, J. (2008). Factors contributing to pertussis

resurgence. Future Microbiol 3, 329–339. Hegerle, N. & Guiso, N. (2013). Epidemiology of whooping cough &

typing of Bordetella pertussis. Future Microbiol 8, 1391–1403. 968

induces apoptosis in macrophages: role of adenylate cyclase– hemolysin. Infect Immun 61, 4064–4071. Characterization of murine lung inflammation after infection with parental Bordetella pertussis and mutants deficient in adhesins or toxins. Infect Immun 62, 2893–2900. Ladant, D., Brezin, C., Alonso, J. M., Crenon, I. & Guiso, N. (1986).

Bordetella pertussis adenylate cyclase: purification, characterization, and radioimmunoassay. J Biol Chem 261, 16264–16269. Mattoo, S. & Cherry, J. D. (2005). Molecular pathogenesis,

epidemiology, and clinical manifestations of respiratory infections due to Bordetella pertussis and other Bordetella subspecies. Clin Microbiol Rev 18, 326–382. Miyaji, Y., Otsuka, N., Toyoizumi-Ajisaka, H., Shibayama, K. & Kamachi, K. (2013). Genetic analysis of isolates from the 2008–2010

pertussis epidemic in Japan. PLoS ONE 8, e77165. Mooi, F. R., Hallander, H., Wirsing von Ko¨nig, C. H., Hoet, B. & Guiso, N. (2000). Epidemiological typing of Bordetella pertussis isolates:

recommendations for a standard methodology. Eur J Clin Microbiol Infect Dis 19, 174–181. Mooi, F. R., van Loo, I. H., van Gent, M., He, Q., Bart, M. J., Heuvelman, K. J., de Greeff, S. C., Diavatopoulos, D., Teunis, P. & other authors (2009). Bordetella pertussis strains with increased toxin

production associated with pertussis resurgence. Emerg Infect Dis 15, 1206–1213. Mooi, F. R., VAN DER Maas, N. A. T. & De Melker, H. E. (2014). Pertussis resurgence: waning immunity and pathogen

adaptation - two sides of the same coin. Epidemiol Infect 142, 685– 694. Njamkepo, E., Rimlinger, F., Thiberge, S. & Guiso, N. (2002). Thirty-

five years’ experience with the whole-cell pertussis vaccine in France: vaccine strains analysis and immunogenicity. Vaccine 20, 1290– 1294. Microbiology 160

Inhibition of B. pertussis-induced macrophage lysis

Octavia, S., Maharjan, R. P., Sintchenko, V., Stevenson, G., Reeves, P. R., Gilbert, G. L. & Lan, R. (2011). Insight into evolution of

Petersen, R. F., Dalby, T., Dragsted, D. M., Mooi, F. & Lambertsen, L. (2012). Temporal trends in Bordetella pertussis populations, Denmark,

Bordetella pertussis from comparative genomic analysis: evidence of vaccine-driven selection. Mol Biol Evol 28, 707–715.

1949–2010. Emerg Infect Dis 18, 767–774.

Osickova, A., Masin, J., Fayolle, C., Krusek, J., Basler, M., Pospisilova, E., Leclerc, C., Osicka, R. & Sebo, P. (2010). Adenylate cyclase toxin

translocates across target cell membrane without forming a pore. Mol Microbiol 75, 1550–1562. Otsuka, N., Han, H. J., Toyoizumi-Ajisaka, H., Nakamura, Y., Arakawa, Y., Shibayama, K. & Kamachi, K. (2012). Prevalence and genetic

characterization of pertactin-deficient Bordetella pertussis in Japan. PLoS ONE 7, e31985. Pawloski, L. C., Queenan, A. M., Cassiday, P. K., Lynch, A. S., Harrison, M. J., Shang, W., Williams, M. M., Bowden, K. E., Burgos-Rivera, B. & other authors (2014). Prevalence and molecular characterization of

pertactin-deficient Bordetella pertussis in the United States. Clin Vaccine Immunol 21, 119–125. Perez Vidakovics, M. L., Lamberti, Y., van der Pol, W. L., Yantorno, O. & Rodriguez, M. E. (2006). Adenylate cyclase influences filamentous

haemagglutinin-mediated attachment of Bordetella pertussis to epithelial alveolar cells. FEMS Immunol Med Microbiol 48, 140– 147.

http://mic.sgmjournals.org

Plotkin, S. A. (2014). The pertussis problem. Clin Infect Dis 58, 830–

833. Queenan, A. M., Cassiday, P. K. & Evangelista, A. (2013). Pertactin-

negative variants of Bordetella pertussis in the United States. N Engl J Med 368, 583–584. Quinlan, T., Musser, K. A., Currenti, S. A., Zansky, S. M. & Halse, T. A. (2013). Pertactin-negative variants of Bordetella pertussis in New York

State: a retrospective analysis, 2004–2013. Mol Cell Probes. Weber, C., Boursaux-Eude, C., Coralie, G., Caro, V. & Guiso, N. (2001). Polymorphism of Bordetella pertussis isolates circulating

for the last 10 years in France, where a single effective whole-cell vaccine has been used for more than 30 years. J Clin Microbiol 39, 4396–4403. Zaretzky, F. R., Gray, M. C. & Hewlett, E. L. (2002). Mechanism of

association of adenylate cyclase toxin with the surface of Bordetella pertussis: a role for toxin-filamentous haemagglutinin interaction. Mol Microbiol 45, 1589–1598. Edited by: P. Langford

969

Antibody-mediated inhibition of Bordetella pertussis adenylate cyclase-haemolysin-induced macrophage cytotoxicity is influenced by variations in the bacterial population.

Whooping cough is a vaccine-preventable disease presenting with epidemic cycles linked to natural and/or vaccine-driven evolution of the aetiological ...
220KB Sizes 0 Downloads 3 Views