Ia

LEa

b

c

a

ARR2AA

KMPC

1

cpheitavnfNaw

9

0d

Vaccine 29 (2011) 1431–1437

Contents lists available at ScienceDirect

Vaccine

journa l homepage: www.e lsev ier .com/ locate /vacc ine

dentifying long-term memory B-cells in vaccinated children despite waningntibody levels specific for Bordetella pertussis proteins

otte H. Hendrikxa,b,∗, Kemal Öztürka, Lia G.H. de Ronda, Reinier H. Veenhovenb,lisabeth A.M. Sandersc, Guy A.M. Berbersa, Anne-Marie Buismana

Centre for Infectious Disease and Control (Cib), National Institute for Public Health and the Environment, Bilthoven, The NetherlandsDepartment of Pediatrics, Spaarne Hospital Hoofddorp, The NetherlandsDepartment of Pediatric Immunology and Infectious Diseases, University Medical Centre, Utrecht, The Netherlands

r t i c l e i n f o

rticle history:eceived 25 August 2010eceived in revised form3 November 2010ccepted 13 December 2010vailable online 25 December 2010

eywords:emory B-cells

ertussis

a b s t r a c t

Whooping cough is a respiratory disease caused by Bordetella pertussis. Since the 1950s in developedcountries pertussis vaccinations are included in the national immunization program. However, anti-body levels rapidly wane after both whole cell and acellular pertussis vaccination. Therefore protectionagainst pertussis may depend largely on long-term B- and T-cell immunities. We investigated long-termpertussis-specific memory B-cell responses in children who were primed at infant age with the DutchwP-vaccine (ISRCTN65428640). Purified B-cells were characterized by FACS-analysis and after polyclonalstimulation memory B-cells were detected by ELISPOT-assays specific for pertussis toxin, filamentoushaemagglutinin, pertactin and tetanus. In addition, plasma IgG levels directed to the same antigens weremeasured by a fluorescent bead-based multiplex immunoassay. Two and 3 years after wP priming as

hildren well as 2 and 5 years after the aP booster at the age of 4, low plasma IgG levels to the pertussis proteinswere found. At the same time, however pertussis protein-specific memory B-cells could be detected andtheir number increased with age. The number of tetanus-specific memory B-cells was similar in all agegroups, whereas IgG-tetanus levels were high 2 years after tetanus booster compared to pre- and 5 yearspost-booster levels. This study shows the presence of long-term pertussis protein-specific memory B-cells in children despite waning antibody levels after vaccination, which suggests that memory B-cells

may c

in addition to antibodies

. Introduction

Despite a high pertussis vaccination coverage in developedountries, Bordetella pertussis is reemerging in the last decade. B.ertussis causes whooping cough, a respiratory disease which isighly contagious. Since antibacterial treatment of pertussis is gen-rally given in a later phase of the disease and therefore mostly isneffective, vaccination against pertussis is of great importance. Inhe Netherlands infants are immunized at 2, 3, 4 and 11 months ofge and from 2001 onwards boosted with an acellular pertussis (aP)accine at 4 years of age. The introduction of the aP booster vacci-ation resulted in shifting the peak incidence of whooping cough

rom children 4–5 years to children 10–11 years of age in 2008 [1].owadays, a higher incidence of whooping cough is observed indolescents and adults who have been vaccinated with the Dutchhole cell pertussis (wP) vaccine. From 2005 onwards the wP

∗ Corresponding author at: RIVM, LIS, Postbak 22, Antonie van Leeuwenhoeklaan, 3720 BA Bilthoven, The Netherlands. Tel.: +31 30 2743944; fax: +31 30 274418.

E-mail address: Lotte.Hendrikx@rivm.nl (L.H. Hendrikx).

264-410X/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.oi:10.1016/j.vaccine.2010.12.033

ontribute to protection against pertussis.© 2010 Elsevier Ltd. All rights reserved.

vaccine at infant age was replaced by the aP vaccine. Although ado-lescents and adults mostly suffer from mild symptoms, they are themain source for infection of the vulnerable young infants, in partic-ular those not yet fully vaccinated, in whom whooping cough cancause serious complications and mortality [2,3].

Vaccination is presumed to introduce protective antibodies andimmunological memory that is characterized by the ability of theimmune system to respond rapidly and specifically upon a subse-quent encounter with a pathogen or antigen. Ideally, vaccinationagainst disease should provide protection for life by a continu-ous production of protective antibodies. The detection of specificserum antibodies is the most widely applied method to investi-gate immunity against vaccine-preventable diseases like pertussis,although a specific correlate of protection against pertussis hasnot been established yet. Besides antibodies, long-term memory B-and T-cell immunities might play an important role in protection

against pertussis as has been demonstrated in mouse models [4,5].Furthermore, different vaccines may achieve protection againstpertussis in different ways, but the exact mechanism remainsunclear [4,6]. In contrast to aP, the wP vaccine contains compo-nents like lipopolysaccharide that serves as adjuvants stimulating

1 accine

tltwmaIwyy

2

2

copyifbital(b4Aobs

(Iw6

2

c(tuu

2

Fuoac

2

dpcp

432 L.H. Hendrikx et al. / V

he innate immune system. Despite this, most wP vaccines elicitower antibody responses against the major virulence factors thanhe aP vaccines. However, antibody levels wane rapidly after bothP and aP vaccinations. Since little is known about long-term B-cellemory following childhood vaccinations, we recently developedmethod to identify pertussis-protein specific memory B-cells [7].

n the present study, we investigated memory B-cell responses asell as antibody levels in blood of wP primed children in the first

ear of life at age 3 and 4 years before booster and at age 6 and 9ears after aP booster vaccinations.

. Subjects and methods

.1. Study population

The children presented in this study form a subset of a largerross-sectional study (ISRCTN65428640) performed from 2007nwards in the Netherlands that investigates the immunity to B.ertussis following wP and aP vaccinations in groups of children 3–9ears of age (total number = 414). After both parents had signed annformed consent, a single blood sample (8–15 ml) was collectedrom each child. All children have received DTwP-IPV/Hib (com-ined vaccine containing diphtheria, tetanus, whole cell pertussis,

nactivated polio, heamophilus influenza type b) (NVI, Bilthoven,he Netherlands) at 2, 3, 4 and 11 months of age and those 6nd 9 years also a 3-component aP vaccine (aP, GlaxoSmithK-ine, containing pertussis toxin (PT), filamentous heamagglutininFHA) pertactin (Prn)) simultaneously administered with a DT-IPVooster vaccine (NVI) at 4 years of age (blood samples from childrenyears of age were taken just before they received their booster).ll children were healthy and had no prolonged coughing peri-ds as symptom of infection with B. pertussis within the two yearsefore sampling as measured by a parental questionnaire (data nothown).

In this study, 4 groups of children of 3, 4, 6 and 9 years of agen = 61–65 per group) were used for measuring antibody responses.n a randomly selected subset of these children memory B-cells

ere measured (n = 19, 19, 16 and 16 for respectively children 3, 4,and 9 years of age).

.2. Isolation of lymphocytes

PBMCs were isolated from 4 ml vacutainer cell preparation tubesontaining sodium citrate (Becton Dickinson Biosciences, CA, USABD)) according to the procedures recommended by the manufac-urer. Cells were isolated as described [7] and stored at −135 ◦Cntil further testing. Plasmas were isolated and frozen at −20 ◦Cntil further use.

.3. Serology

IgG levels directed against the B. pertussis vaccine antigens (PT,HA and Prn) and tetanus toxoid were measured in plasma bysing the fluorescent bead-based multiplex immunoassay as previ-usly described [8]. Data were expressed as EU/ml for the pertussisntigens and IU/ml for tetanus and were given in geometric meanoncentrations (GMC).

.4. B-cell purification and stimulation

B-cells were purified and polyclonal stimulated as recentlyescribed [7]. In short, B-cells were purified by an anti-CD19ositive selection cocktail (EasySep, Stemcell, France). Purified B-ells (90%) were resuspended at a concentration of 5 × 104 cellser 200 �l and cultured in 96-well round bottom tissue culture

29 (2011) 1431–1437

plates (Greiner, Invitrogen, Breda, Netherlands). B-cells were stim-ulated polyclonally with 3 �g/ml CpG ODN2006 in the presence of10 ng/ml IL-2 (Milteny Biotec, Auburn, USA), IL-10 (Pharmingen,BD) and IL-15 (Biosource, Camarillo, USA) for 5 days at 37 ◦C and5% CO2, before testing them in antigen-specific ELISPOT assays.

2.5. ELISPOT assay

Antigen-specific ELISPOT assays were performed as described[7]. In short, multiscreen-IP plates (MISPS4510, Millipore) werecoated with 50 �l PBS containing either 10 �g/ml anti-humanIgG-Fc-specific, 30 �g/ml PT, 20 �g/ml FHA, 80 �g/ml recombinantPrn and 7LF/ml tetanus toxoid. Stimulated B-cell suspensionswere added to the plates (1 × 105 per well) at least in triplicateand were incubated ON at 37 ◦C and 5% CO2. After incubationwith alkaline phosphatase (AP)-conjugated goat anti-human IgGand substrate solution, plaques appearing as blue spots wereenumerated by using an ELISPOT reader (CTL Europe, Bonn,Germany).

Mean spot values of the non-coated wells were used as neg-ative control. From all mean spot values of the antigen-coatedwells, the negative controls were subtracted. Numbers of antigen-specific memory B-cells were given per 105 B-cells. In figures with alogarithmic scale, a value of 0.1 was given to a value of zero antigen-specific spots. Percentages of antigen-specific memory B-cells weregiven per total IgG-producing cells. All individual values are shownor data are expressed as geometric mean (GM) with or without1-standard deviation (±) and average values were indicated.

2.6. Flow cytometry

Subpopulations of B-cells were determined by flow cytometry(FACS Calibur, BD). Cells were incubated for 30 min with a mixtureof phycoerythrin (PE) labeled anti-human CD19 (BD) and fluores-cein isothiocyanate (FITC) labeled anti-human CD27 (BD).

2.7. Statistics

Comparison of the results per birth cohort was done byMann–Whitney test. Spearman correlations (rs) were done to ana-lyze the correlation between variables. p < 0.05 was consideredsignificantly different.

3. Results

3.1. Antibody levels

Antibody levels to all pertussis antigens were low at 3 and 4years of age and significantly higher in children 6 and 9 years ofage following the booster at 4 years of age (Table 1). In the majorityof the children of all age groups low IgG-PT levels were found (94%of all 250 children had IgG-PT levels below 50 EU/ml). An IgG-PTvalue above 50 EU/ml is considered indicative for a recent infectionwith B. pertussis [9–11]. All children had IgG-tetanus levels abovethe protective concentration of 0.01 IU/ml [12]. Children 6 years ofage had significantly higher IgG-tetanus levels than children 3, 4and 9 years of age. In 9 years old children higher antibody levels totetanus were found than in children 3 and 4 years old (Table 1).

3.2. Characterization of B-cells before and after polyclonalstimulation

In children an average of 1.2 × 106 ± 0.5 × 106 PBMCs per mlblood was isolated. After purification 12.9 ± 8.2% of the PBMCs werefound to be B-cells. The percentage of B-cells in children 9 yearsof age (9.5%) was significantly lower than that in children 4 years

L.H. Hendrikx et al. / Vaccine 29 (2011) 1431–1437 1433

Table 1Geometric mean concentrations (GMC) and 95% confidence interval (CI) of the IgG levels to the pertussis antigens (given in EU/ml) and to tetanus (given in IU/ml) in groupsof 3–9 years old children.

Age group (years), number

3, n = 61 4, n = 61 6, n = 63 9, n = 65

IgG-PT (EU/ml)GMC 5.7 4.5 10.6*,# 8.9*,#

95% CI 4.1–8.0 3.3–6.0 7.8–14.4 6.4–12.3IgG-FHA (EU/ml)

GMC 6.8 8.8 44.5*,# 31.9*,#,†

95% CI 4.4–10.4 5.8–13.3 30.9–64.0 25.1–40.6IgG-Prn (EU/ml)

GMC 4.2 3.1 19.4*,# 16.6*,#

95% CI 3.0–5.8 2.3–4.3 13.3–28.4 12.4–22.1IgG-tetanus (IU/ml)

GMC 0.33 0.23 1.64*,# 0.46*,†

0.1

f age.f age.

oyasBap

bFc

3

imaitcfnfcBdawlwspgpti0

bfpp

B5

In the subset of children in which B-cell responses were mea-sured, comparable IgG levels were observed as those for thecomplete groups of children (data not shown). Since the smallsubsets of children in this study did not show a Gaussian distribu-tion, the Spearman correlation was used to analyze the correlation

95% CI 0.23–0.46

# Significant higher IgG-level in children 6 and/or 9 years than children 3 years o* Significant higher IgG-level in children 6 and/or 9 years than children 4 years o† Significant lower IgG level in children 9 years than children 6 years of age.

f age (15.1%) and non-significantly lower than children 3 and 6ears of age (respectively 13.6% and 13.8%). The B-cells prolifer-ted on average 2.6 ± 1.4- fold within a period of 5 days. After B-celltimulation a similar percentage of 14.2 ± 7.0 total IgG-producing-cells was observed in all age groups. After purification a percent-ge of 5.7 death cells was found and after 5 days of stimulation thisercentage was 14.7.

The memory B-cells (CD19+/CD27+) increased from 22.9%efore to 36.9% after stimulation of the total B-cell pool.ig. 1 illustrates the FACS analysis for one representativehild.

.3. Numbers and frequencies of antigen-specific memory B-cells

Pertussis- and tetanus-specific memory B-cells were detectablen all samples, except for one child 3 years of age in whom no

emory B-cells against the pertussis antigens were detected atll. Fig. 2 shows that children 6 and 9 years of age had signif-cantly higher numbers of PT- and Prn-specific memory B-cellshan children 3 and 4 years of age. The only exception was theomparison between children 6 years versus 3 years of age thatailed to reach significance for Prn-specific memory B-cells. Sig-ificantly higher numbers of FHA-specific memory B-cells were

ound in children of 6 and 9 years compared with 3 year-oldhildren. No significant differences in tetanus-specific memory-cells were measured between all age groups. From all 67 chil-ren tested, in 36% no memory B-cells specific for one or twontigens were detected; especially PT-specific memory B-cellsere absent before the booster at 3 and 4 years. Neverthe-

ess, in these subjects, memory B-cells to the other antigensere detectable in various numbers. Frequencies of pertussis-

pecific memory B-cells per total IgG-producing memory B-cellool (Table 2) were significantly lower in the pre booster ageroups (3 and 4 years old children) when compared with theost booster age groups (6 and 9 years old children), whereashe frequency of tetanus-specific memory B cells did not signif-cantly differ between all age groups (average for all age groups.13%).

High Spearman correlations were found between the num-ers of pertussis-specific memory B-cells per 105 B-cells and therequencies of pertussis-specific memory B-cells per total IgG-

roducing B-cells (for all pertussis proteins on average rs = 0.87,< 0.0001).

To prevent any influence of a possible recent infection with. pertussis we excluded 3 children with an IgG-PT level above0 EU/ml for further analysis of the B-cell responses.

8–0.30 1.28–2.11 0.37–0.57

3.4. Correlation between the number of antigen-specific memoryB-cells and IgG plasma levels

Fig. 1. A representative FACS figure: surface staining of CD19 and CD27 on purifiedB-cells of a 4 years old child before (a) and after (b) 5 days of polyclonal stimulation.

1434 L.H. Hendrikx et al. / Vaccine 29 (2011) 1431–1437

F tactin-e metrit

bpmmt

TP(

ig. 2. Numbers (log) of pertussis toxin- (a), filamentous heamagglutinin- (b), perach child in the different age groups (x-axis). The horizontal line indicates the geohe x-axis. *Significant difference between the two age groups (p < 0.05).

etween the numbers of antigen-specific memory B-cells and the

lasma IgG levels. In all children tested significant, very low Spear-an correlations were found between numbers of antigen-specificemory B-cells and circulating plasma IgG levels for FHA, Prn and

etanus, but not for PT (Fig. 3). Importantly, in the majority of the

able 2ercentages of pertussis- and tetanus-specific memory B-cells of the total IgG-producingCIs) in the different age groups.

% Antigen-specific memory B-cells PT FHA

Age group (years) n GM 95% CI n GM

3 14 0.01 0.00–0.05 15 0.034 14 0.01 0.00–0.02 15 0.036 14 0.03*,# 0.01–0.05 14 0.06#

9 15 0.03*,# 0.01–0.10 15 0.07#

# Significant higher percentage in children 6 and/or 9 years than in children 3 years of* Significant higher percentage in children 6 and/or 9 years than in children 4 years of

(c) and tetanus- (d) specific memory B-cells per 105 B-cells are given (y-axis) forc mean value of the specific age group that is also given for each age group below

children low IgG-PT or -Prn levels were measured, whereas most

of these children had detectable PT- or Prn-specific memory B-cells(as indicated with the red circle in Fig. 3a and c). In 21 predomi-nantly 3 and 4 years old children PT-specific memory B-cells couldnot be measured and IgG-PT levels were low (GMC 5.8 EU/ml). How-

B-cell population are given in geometric mean (GM) and 95% confidence intervals

Prn Tetanus

95% CI n GM 95% CI n GM 95% CI

0.02–0.06 17 0.02 0.01–0.05 12 0.09 0.05–0.180.01–0.07 15 0.01 0.01–0.04 14 0.05 0.02–0.160.02–0.19 14 0.04 0.02–0.06 12 0.11 0.07–0.200.03–0.13 15 0.04* 0.02–0.08 15 0.06 0.02–0.17

age.age.

L.H. Hendrikx et al. / Vaccine 29 (2011) 1431–1437 1435

F cells (a ory B

eBfldcIao

4

Bsyfiasceor

tiw

ig. 3. Spearman correlations (rs) between numbers of antigen-specific memory B-ll children tested (n = 67). The circle indicates detectable PT- and Prn-specific mem

ver, in these 21 children, FHA-, Prn- and tetanus-specific memory-cells and IgG levels were detectable in various frequencies, except

or one child 3 years of age in whom no memory B-cells and low IgGevels against the pertussis antigens were detected. Just a few chil-ren (n = 10) showed the absence of FHA- or Prn-specific memory Bells together with low IgG levels for these antigens. In all childrengG-tetanus levels were above the level of protection (0.01 IU/ml)nd in only 2 children (4 and 9 years old) no tetanus-specific mem-ry B cells could be detected.

. Discussion

This is the first study that demonstrated long-term memory-cell responses in vaccinated children of different ages. Pertussis-pecific memory B-cells were found in blood of most children 3ears after infant vaccinations with the Dutch wP vaccine in therst year of life and higher levels at 6 and 9 years following theP booster vaccination at 4 years of age. Importantly, pertussis-pecific memory B-cells were detected in the majority of thehildren despite low plasma antibody levels. This indicates thatven though antibody levels wane, circulating long lasting mem-ry B-cells remain that may show a rapid booster response upon

enewed encounter with pertussis antigens.

Previously, only few studies performed in adults have shownhe presence of memory B-cells despite low antibody levels for var-ous antigens [12–19]. In children, antigen-specific memory B-cells

ere demonstrated despite rapidly waning antibody levels already

y-axis) and plasma IgG levels (x-axis) for PT (a), FHA (b), Prn (c) and tetanus (d) for-cells despite low IgG-values.

shortly after meningococcal C- and pneumococcal-conjugated vac-cinations [20–22]. For pertussis waning antibody levels have beenshown after vaccination in only several previous studies [23–27]and the prevalence of lasting immunity has only been demon-strated for cellular-mediated proliferative responses, which aredominated by T-lymphocyte responses [24,28–30]. Overall, no dataare available on long-term memory B-cell responses in vaccinatedchildren.

We demonstrated an increase in pertussis protein-specificmemory B-cells with age. This is partly due to the acellular boosteradministered at 4 years of age. The number of PT, FHA- and Prn-specific memory B-cells seemed to reach a plateau in children 6and 9 years of age. This is in agreement with the findings of Sasakiet al. after influenza vaccination [18] who demonstrated a ratherconstant frequency over time of approximately 0.5% influenza-specific memory B-cells in children 6 years and older after an initialincrease with age. This was supposedly due to a continued annualcirculation of influenza viruses with a resulting high percentage ofinfluenza-specific memory B-cells. Indeed, we found a more than10-fold lower frequency of pertussis protein-specific memory B-cells in children of the same age (on average 0.06%) comparedto that of influenza. Other studies showed various frequencies

of antigen-specific memory B-cells shortly after vaccination inchildren depending on age, time after vaccination and type of vac-cination [21,31].

In our study, a relatively large proportion of children haveundetectable PT-specific memory B-cells, predominantly before

1 accine

talwmlRcdlordoNcf

ioTwiwoisNaivtmtTbatTt

ow[mpSced

aiFcopadaehrttt

[

436 L.H. Hendrikx et al. / V

he booster at 4 years of age. This might be explained by the very lowmounts of inactivated PT in the Dutch wP vaccine [32], which is inine with the weak antibody response to PT. After acellular booster

ith a relatively high PT concentration at 4 years of age, PT-specificemory B-cells are detected in most children, however IgG-PT

evels were low both before and years after the acellular booster.ecently, a significantly higher number of PT-specific memory B-ells in adults compared with children was reported [7] possiblyue to the increased circulation of B. pertussis since the 1990s simi-

ar to recurrent boosting with influenza [1]. We used a cut-off valuef 50 EU/ml for plasma IgG-PT levels to exclude children with aecent infection [9–11]. In addition, a questionnaire was used toiagnose possible clinical symptoms of whooping cough in a periodf 12 months before blood sampling. No cases could be registered.evertheless, booster responses after subclinical natural infectionannot be excluded from these measurements as illustrated by theew children with very high IgG-PT concentrations.

By comparing pertussis with tetanus antigens that are includedn the same vaccines given previously, the number and frequencyf tetanus-specific memory B-cells are similar in all age groups.etanus comprises on average 0.13% per total IgG-producing cells,hich is higher than that for pertussis but lower compared to

nfluenza, however consistent with the study of Crotty et al. [15]ho showed the long term persistence of tetanus-specific mem-

ry B-cells in adults for decades after vaccination. Remarkably, noncrease in tetanus-specific memory B-cells was measured in ourtudy 2–5 years after the booster vaccination. In another study,anan et al. [33] even showed a decreased proliferative responsefter antigenic re-challenge in frequently tetanus-re-immunizedndividuals. In our study, children had received multiple tetanusaccinations at the same time as the pertussis vaccinations nexto a MenC-tetanus conjugated vaccine at 14 months of age that

ight explain the stable tetanus-specific memory B-cell popula-ion. Furthermore, no natural boosting of Clostridium tetani occurs.he variant behavior between vaccine antigens is also underpinnedy the fact that despite similar tetanus-specific memory B-cells atll ages, IgG-tetanus levels are significantly higher 2 years after theetanus-booster at 4 years of age compared to the pre booster levels.his confirms the presence of a limited memory B-cell pool causinghis antigen-specific immunological booster response [34].

We measured 12.7% B-cells in the total PBMC pool in blood ofur children, which is consistent with previous literature [35,36],hereas in adults a much lower percentage of 4.5% is found

7]. However, in our children on average 14.7% of the stimulatedemory B-cell population produced IgG which is lower than the

ercentage in adults (22.4%), in accordance with the findings ofiegrist and Aspinall [37]. Children might have more circulating B-ells and may need more B-cells to be able to react to the multiplencountered antigens, since their memory immunity has not beeneveloped as much as in adults.

We found very low correlations between the numbers ofntigen-specific memory B-cells and the corresponding circulat-ng antibody levels for the pertussis vaccine proteins FHA and Prn.or PT however, no correlation was observed. Moreover, in somehildren high antibody levels with low or no numbers of mem-ry B-cells were found. This strengthens the idea that long-livedlasma cells maintain antibody levels and that memory B-cellsre a distinct population of cells [12,38,39]. In steady state con-itions, memory B-cells do not secrete antibodies. Upon renewedntigen stimulation, memory B-cells may rapidly respond, prolif-rate and differentiate in antibody secreting cells. At that time, a

igher memory B-cell dependent antibody induction is expected,esulting into a correlation between memory B-cell and short-erm antibody levels. The ongoing circulation of B. pertussis amonghe Dutch population causing antigenic re-challenge might explainhe low but significant correlations between circulating pertussis-

[

29 (2011) 1431–1437

specific memory B-cells and antibody levels measured in our study.Importantly, in some children no pertussis-specific memory B-cellsas well as low pertussis-specific antibody levels were detectablewhich might result in a higher susceptibility for infection with B.pertussis.

A low correlation between tetanus-specific memory B-cells andantibody levels was measured that was not reported from studies inadults [7,12]. In our case this may be due to the relative recent andalready 5th or 6th tetanus vaccination these children had receiveda few years before.

To summarize, we showed despite waning antibody levels,pertussis-specific memory B-cells in blood of Dutch wP primed chil-dren that increase with age during childhood due to the booster at 4years of age next to potential natural boosting with B. pertussis. Thisis an important finding to understand the protective mechanism ofpertussis vaccination, since pertussis-specific memory B-cells canintroduce a rapid booster response. Further studies on the effectof the different aP vaccines on long term B- and T-cell immunitieswill be performed as this may differ from wP primed individuals.Greater insight in memory immunity will be important for futurepertussis vaccine design as well as for future decisions concern-ing pertussis vaccination policies to sustain protective levels ofimmunity.

Acknowledgements

We would like to thank all children who participated in theMEMORY study. In addition we would like to thank all researchstaff from the Spaarne hospital Hoofddorp, in special Greetje vanAsselt and Jacqueline Zonneveld. At the National Institute for PublicHealth and the Environment in Bilthoven we would like to thankRose-minke Schure for technical assistance.

Funding: This study is funded by the Dutch government.Conflict of interest statement: There is no conflict of interest.

References

[1] de Greeff SC, Mooi FR, Schellekens JF, de Melker HE. Impact of acellular per-tussis preschool booster vaccination on disease burden of pertussis in TheNetherlands. Pediatr Infect Dis J 2008;27(3):218–23.

[2] Schellekens J, von Konig CH, Gardner P. Pertussis sources of infection androutes of transmission in the vaccination era. Pediatr Infect Dis J 2005;24(Suppl.5):S19–24.

[3] Wirsing von Konig CH, Postels-Multani S, Bock HL, Schmitt HJ. Pertus-sis in adults: frequency of transmission after household exposure. Lancet1995;346(8986):1326–9.

[4] Canthaboo C, Williams L, Xing DK, Corbel MJ. Investigation of cellular andhumoral immune responses to whole cell and acellular pertussis vaccines.Vaccine 2000;19(6):637–43.

[5] Mahon BP, Brady MT, Mills KH. Protection against Bordetella pertussis in micein the absence of detectable circulating antibody: implications for long-termimmunity in children. J Infect Dis 2000;181(6):2087–91.

[6] Olin P, Rasmussen F, Gustafsson L, Hallander HO, Heijbel H. Randomised con-trolled trial of two-component, three-component, and five-component acel-lular pertussis vaccines compared with whole-cell pertussis vaccine. Ad HocGroup for the Study of Pertussis Vaccines. Lancet 1997;350(9091):1569–77.

[7] Buisman AM, de Rond CG, Ozturk K, Ten Hulscher HI, van Binnendijk RS. Long-term presence of memory B-cells specific for different vaccine components.Vaccine 2009;28(1):179–86.

[8] van Gageldonk PG, van Schaijk FG, van der Klis FR, Berbers GA. Developmentand validation of a multiplex immunoassay for the simultaneous determinationof serum antibodies to Bordetella pertussis, diphtheria and tetanus. J ImmunolMethods 2008;335(1–2):79–89.

[9] Baughman AL, Bisgard KM, Edwards KM, Guris D, Decker MD, Holland K, et al.Establishment of diagnostic cutoff points for levels of serum antibodies to per-tussis toxin, filamentous hemagglutinin, and fimbriae in adolescents and adultsin the United States. Clin Diagn Lab Immunol 2004;11(6):1045–53.

10] de Melker HE, Versteegh FG, Conyn-Van Spaendonck MA, Elvers LH, Berbers GA,

van Der Zee A, et al. Specificity and sensitivity of high levels of immunoglobulinG antibodies against pertussis toxin in a single serum sample for diagnosis ofinfection with Bordetella pertussis. J Clin Microbiol 2000;38(2):800–6.

11] Wirsing von Konig CH, Gounis D, Laukamp S, Bogaerts H, Schmitt HJ. Eval-uation of a single-sample serological technique for diagnosing pertussis inunvaccinated children. Eur J Clin Microbiol Infect Dis 1999;18(5):341–5.

accine

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

L.H. Hendrikx et al. / V

12] Amanna IJ, Carlson NE, Slifka MK. Duration of humoral immunity to commonviral and vaccine antigens. N Engl J Med 2007;357(19):1903–15.

13] Pinna D, Corti D, Jarrossay D, Sallusto F, Lanzavecchia A. Clonal dissection ofthe human memory B-cell repertoire following infection and vaccination. EurJ Immunol 2009;39(5):1260–70.

14] Crotty S, Aubert RD, Glidewell J, Ahmed R. Tracking human antigen-specificmemory B cells: a sensitive and generalized ELISPOT system. J Immunol Meth-ods 2004;286(1–2):111–22.

15] Crotty S, Felgner P, Davies H, Glidewell J, Villarreal L, Ahmed R. Cutting edge:long-term B cell memory in humans after smallpox vaccination. J Immunol2003;171(10):4969–73.

16] Tuaillon E, Tabaa YA, Petitjean G, Huguet MF, Pajeaux G, Fondere JM, et al.Detection of memory B lymphocytes specific to hepatitis B virus (HBV) sur-face antigen (HBsAg) from HBsAg-vaccinated or HBV-immunized subjects byELISPOT assay. J Immunol Methods 2006;315(1–2):144–52.

17] Rojas OL, Narvaez CF, Greenberg HB, Angel J, Franco MA. Characterization ofrotavirus specific B cells and their relation with serological memory. Virology2008;380(2):234–42.

18] Sasaki S, Jaimes MC, Holmes TH, Dekker CL, Mahmood K, Kemble GW, et al. Com-parison of the influenza virus-specific effector and memory B-cell responsesto immunization of children and adults with live attenuated or inactivatedinfluenza virus vaccines. J Virol 2007;81(1):215–28.

19] Ward SM, Phalora P, Bradshaw D, Leyendeckers H, Klenerman P. Direct ex vivoevaluation of long-lived protective antiviral memory B cell responses againsthepatitis B virus. J Infect Dis 2008;198(6):813–7.

20] Kelly DF, Snape MD, Cutterbuck EA, Green S, Snowden C, Diggle L, et al. CRM197-conjugated serogroup C meningococcal capsular polysaccharide, but not thenative polysaccharide, induces persistent antigen-specific memory B cells.Blood 2006;108(8):2642–7.

21] Clutterbuck EA, Oh S, Hamaluba M, Westcar S, Beverly PCL, Pollard AJ.Serotype specific and age dependent generation of pneumococcal polysac-charide specific memory B cells and antibody response to immunizationwith a pneumococcal conjugate vaccine. Clin Vaccine Immunol 2007;15(2):182–93.

22] Blanchard Rohner G, Snape MD, Kelly DF, John T, Morant A, Yu LM, et al. Themagnitude of the antibody and memory B cell responses during priming with aprotein–polysaccharide conjugate vaccine in human infants is associated withthe persistence of antibody and the intensity of booster response. J Immunol2008;180(4):2165–73.

23] Hallander HO, Gustafsson L, Ljungman M, Storsaeter J. Pertussis antitoxin decay

after vaccination with DTPa. Response to a first booster dose 3 1/2–6 1/2 yearsafter the third vaccine dose. Vaccine 2005;23(46–47):5359–64.

24] Guiso N, Njamkepo E, Vie le Sage F, Zepp F, Meyer CU, Abitbol V, et al. Long-term humoral and cell-mediated immunity after acellular pertussis vaccinationcompares favourably with whole-cell vaccines 6 years after booster vaccinationin the second year of life. Vaccine 2007;25(8):1390–7.

[

[

29 (2011) 1431–1437 1437

25] Hendrikx LH, Berbers GA, Veenhoven RH, Sanders EA, Buisman AM. IgGresponses after booster vaccination with different pertussis vaccines inDutch children 4 years of age: effect of vaccine antigen content. Vaccine2009;27(47):6530–6.

26] Edelman K, He Q, Makinen J, Sahlberg A, Haanpera M, Schuerman L, et al. Immu-nity to pertussis 5 years after booster immunization during adolescence. ClinInfect Dis 2007;44(10):1271–7.

27] Giuliano M, Mastrantonio P, Giammanco A, Piscitelli A, Salmaso S, Wassilak SG.Antibody responses and persistence in the two years after immunization withtwo acellular vaccines and one whole-cell vaccine against pertussis. J Pediatr1998;132(6):983–8.

28] Zepp F, Knuf M, Habermehl P, Schmitt HJ, Meyer C, Clemens R, et al. Cell-mediated immunity after pertussis vaccination and after natural infection. DevBiol Stand 1997;89:307–14.

29] Tran Minh NN, He Q, Edelman K, Olander RM, Viljanen MK, Arvilommi H, et al.Cell-mediated immune responses to antigens of Bordetella pertussis and protec-tion against pertussis in school children. Pediatr Infect Dis J 1999;18(4):366–70.

30] Rieber N, Graf A, Belohradsky BH, Hartl D, Urschel S, Riffelmann M, et al. Dif-ferences of humoral and cellular immune response to an acellular pertussisbooster in adolescents with a whole cell or acellular primary vaccination. Vac-cine 2008;16:6929–35.

31] Kelly DF, Pollard AJ, Moxon ER. Immunological memory: the role of Bcells in long-term protection against invasive bacterial pathogens. JAMA2005;294(23):3019–23.

32] Thalen M, van der Ark A, van den Ijssel J, van Straaten I, Jansen D, Beuvery C, et al.Improving the cellular pertussis vaccine: increased potency and consistency.Vaccine 2008;26(5):653–63.

33] Nanan R, Heinrich D, Frosch M, Kreth HW. Acute and long-term effectsof booster immunisation on frequencies of antigen-specific memory B-lymphocytes. Vaccine 2001;20(3–4):498–504.

34] Kline GH, Hayden TA, Klinman NR. B cell maintenance in aged mice reflectsboth increased B cell longevity and decreased B cell generation. J Immunol1999;162(6):3342–9.

35] Huck K, Feyen O, Ghosh S, Beltz K, Bellert S, Niehues T. Memory B-cells in healthyand antibody-deficient children. Clin Immunol 2009;131(1):50–9.

36] van Gent R, van Tilburg CM, Nibbelke EE, Otto SA, Gaiser JF, Janssens-KorpelaPL, et al. Refined characterization and reference values of the pediatric T- andB-cell compartments. Clin Immunol 2009;133(1):95–107.

37] Siegrist CA, Aspinall R. B-cell responses to vaccination at the extremes of age.Nat Rev Immunol 2009;9(3):185–94.

38] Traggiai E, Puzone R, Lanzavecchia A. Antigen dependent and independentmechanisms that sustain serum antibody levels. Vaccine 2003;21(Suppl.2):S35–7.

39] Bernasconi NL, Traggiai E, Lanzavecchia A. Maintenance of serologicalmemory by polyclonal activation of human memory B cells. Science2002;298(5601):2199–202.