FEMS Immunology and Medical Microbiology 31 (2001) 153^162
www.fems-microbiology.org
Bordetella pertussis ¢mbriae are e¡ective carrier proteins in
Neisseria meningitidis serogroup C conjugate vaccines
Karen M. Reddin *, Annette Crowley-Luke, Simon O. Clark, Philip J. Vincent,
Andrew R. Gorringe, Michael J. Hudson, Andrew Robinson
Centre for Applied Microbiology and Research, Salisbury SP4 0JG, Wiltshire, UK
Received 19 March 2001 ; received in revised form 18 June 2001; accepted 21 June 2001
First published online 20 July 2001
Abstract
Serogroup C meningococcal conjugate vaccines generally use diphtheria or tetanus toxoids as the protein carriers. The use of alternative
carrier proteins may allow multivalent conjugate vaccines to be formulated into a single injection and circumvent potential problems of
immune suppression in primed individuals. Bordetella pertussis fimbriae were assessed as carrier proteins for Neisseria meningitidis
serogroup C polysaccharide. Fimbriae were conjugated to the polysaccharide using modifications of published methods and characterised
by size exclusion chromatography; co-elution of protein and polysaccharide moieties confirmed conjugation. The conjugates elicited
boostable IgG responses to fimbriae and serogroup C polysaccharide in mice, and IgG:IgM ratios indicated that the responses were
thymus-dependent. High bactericidal antibody titres against a serogroup C strain of N. meningitidis were also observed. In a mouse infection
model, the conjugate vaccine protected against lethal infection with N. meningitidis. Therefore, B. pertussis fimbriae are effective carrier
proteins for meningococcal serogroup C polysaccharide and could produce a vaccine to protect against meningococcal disease and to
augment protection against pertussis. ß 2001 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All
rights reserved.
Keywords : Carrier protein ; Conjugate vaccine; Bordetella pertussis ¢mbria; Neisseria meningitidis
1. Introduction
The capsular polysaccharide constitutes a major virulence factor of many pathogenic bacteria that cause invasive disease, including Neisseria meningitidis. Antibodies to
the capsular polysaccharide of serogroup C N. meningitidis
are protective and bactericidal [1^3]. Polysaccharides are
thymus-independent (TI) antigens because they do not require mature T-cells to elicit a humoral response in vivo.
However, whilst polysaccharide antigens are immunogenic
in adults, they are only poorly or non-immunogenic in
infants and young children, who are highly susceptible
to infection caused by encapsulated bacteria [4]. In addition, TI antigens generally fail to elicit a memory response
or a¤nity maturation. In contrast, the ability to respond
to thymus-dependent antigens (TD), which is present at
* Corresponding author. Tel. : +44 (1980) 61 28 77;
Fax: +44 (1980) 61 13 10.
E-mail address : [email protected] (K.M. Reddin).
birth, results in the formation of memory cells; the antibody response subsequently undergoes a¤nity maturation
following boosting [5]. For TI antigens, IgG3 and IgM are
the major antibody isotypes expressed in mice, even after
secondary immunisation. In contrast, for TD antigens, the
ratio of IgG to IgM increases after boosting, with IgG1
being the major subclass in mice [6].
Polysaccharide antigens, or hydrolysed fragments of
these polysaccharides, can be converted to TD antigens
by conjugation to a protein carrier. The Haemophilus in£uenzae type b capsular polysaccharide conjugate vaccine
has been successfully introduced in many national childhood vaccination programmes [7]. N. meningitidis serogroup C polysaccharide conjugate vaccines have now
been developed. Clinical trials for these vaccines proved
successful [8] and, as a consequence, a number of these
vaccines were introduced into the UK vaccination schedules in 1999. These conjugate vaccines have already had a
signi¢cant impact on the incidence of disease in immunised groups [9]. However, these vaccines use either diphtheria or tetanus toxoids (DT and TT respectively) as the
0928-8244 / 01 / $20.00 ß 2001 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved.
PII: S 0 9 2 8 - 8 2 4 4 ( 0 1 ) 0 0 2 5 4 - 1
FEMSIM 1336 30-8-01
154
K.M. Reddin et al. / FEMS Immunology and Medical Microbiology 31 (2001) 153^162
protein carrier [10^13] which does not broaden protection
as these proteins are components of other vaccines, and
infants may have maternal antibody to these proteins.
While the combination of puri¢ed capsular polysaccharide vaccines does not generally alter the immunogenicity
of the individual polysaccharides, a reduced response to
the polysaccharide moiety of individual conjugates has
been observed when administered in multivalent formulations. In addition, where mice have been pre-primed with
the carrier protein, a reduction in antibody responses has
been observed [14]. This indicated that epitopic load rather
than the antigenic load at the site of injection resulted in
reduced immunity to conjugates. In addition, individual
components of multivalent capsular polysaccharide vaccines conjugated to the same carrier proteins could compete for a limited number of speci¢c carrier proteinprimed T cells. Therefore, the development and use of
multiple carrier proteins should be considered as an approach to reduce interference when more than one conjugate vaccine is used, especially if they are to be formulated
into a single injection.
In this study, Bordetella pertussis ¢mbriae were assessed
as carrier proteins for the capsular polysaccharide of
N. meningitidis serogroup C. Although these proteins are
not components of some acellular pertussis vaccines, recent clinical trials have shown that ¢mbriae are important
protective antigens against whooping cough [15]. Use of
B. pertussis ¢mbriae as protein carriers could, therefore,
produce a vaccine that would protect against meningococcal disease and also augment protection against whooping
cough where acellular vaccines de¢cient in ¢mbriae are
used. It would also introduce an alternative carrier to
DT and TT in conjugate vaccines and be potentially useful
as a carrier in other conjugate vaccines, such as Hib [16]
and pneumococcal vaccines, and would be an advantage if
more than one vaccine were to be used or given as multivalent formulations.
2. Materials and methods
2.1. Puri¢cation of antigens
Fimbriae (a mixture of ¢m2 and ¢m3) were puri¢ed
from B. pertussis strain Wellcome 28 using the method
of Robinson et al. [17], and meningococcal serogroup C
polysaccharide (menCPs) was puri¢ed in-house from
N. meningitidis strain L91 543 (C:2a:P1.2R) using standard methodology [18]. The molecular masses of puri¢ed
¢mbriae and menCPs were estimated by size exclusion
chromatography (SEC) using a Hiload Superdex 200 HR
(16/60) column (Amersham Pharmacia). The void volume
(V0 ) was determined using Blue dextran and the column
calibrated for proteins using a gel ¢ltration molecular
weight calibration kit (Amersham Pharmacia). The column was calibrated for polysaccharides using Polysaccha-
ride SAC-10 dextran standards (Polymer Laboratories,
Shropshire, UK). The apparent molecular masses of native
¢mbriae and menCPs were s 400 000 Da as both eluted in
the void volume (V0 ) of the column.
2.2. Dissociation of ¢mbriae
Native ¢mbriae in 0.5 M phosphate bu¡er, pH 7.2, were
precipitated by the addition of 2 vols. of ice-cold acetone.
The precipitate was recovered by centrifugation in a bench
centrifuge at 4000 rpm for 20 min. The acetone was allowed to evaporate from the pellet, which was then resuspended in 6 M guanidine HCl (Gdn HCl), 0.125 M
ethanolamine, pH 10.5, and allowed to equilibrate overnight at 4³C to dissociate the ¢mbriae [19]. The dissociated
¢mbriae (dFim) were dialysed against at least three
changes of 0.1 M ammonium bicarbonate, pH 7.2, and
concentrated by freeze-drying prior to conjugation [16].
SEC (Section 2.1) of dFim gave a molecular mass of
approx. 25 000 Da, indicating dissociation into ¢mbral
subunits which have estimated molecular masses of
22 500 Da (¢m2) and 22 000 (¢m3) [19].
2.3. Preparation of conjugates
Dissociated ¢mbriae (dFim) were conjugated to menCPs
using carbodiimide-mediated coupling [20] or reductive
amination [21,22], producing the conjugates dFim-menCPs(EDC) and dFim-menCPs(RA), respectively.
2.3.1. Carbodiimide coupling
Carbodiimide coupling involved derivatising dFim using
adipic acid dihydrazide (ADH) as a C6 spacer and 1-ethyl3-(3-dimethylaminopropyl)carbodiimide-HCl (EDAC), facilitating coupling of the amino group of ADH to cyanogen bromide-activated menCPs. Brie£y, ADH derivatives
of proteins were prepared by reacting protein solutions
(1 mg ml31 ) and ADH (3.45 mg per mg protein) with
1-ethyl-3-(3-dimethylaminopropyl)carbodiimide-HCl
(EDAC) (0.3 mg per mg protein). The reaction mixture
was maintained at pH 4.7 þ 0.2 by careful addition of
0.1 M HCl on ice. The reaction was allowed to proceed
at room temperature for 3 h and terminated by dialysing
(MW cuto¡ 12 000; Sigma) at 3^8³C against at least three
changes of 5 l of 20 mM ammonium bicarbonate, pH 7.2.
The protein derivatives were then freeze-dried.
To activate menCPs with cyanogen bromide (CNBr), a
solution of menCPs (45 mg ml31 ) was equilibrated on ice,
in 250 mM Na2 CO3 , pH 10.5. CNBr (500 mg ml31 in
acetonitrile) was added to a ¢nal concentration of 0.4
mg per mg polysaccharide and the pH maintained at
10.5 for 6 min. The reaction mixture pH was then lowered
to pH 8.5 by careful addition of 1 M HCl. The derivatised
ADH-protein was added to the CNBr-activated menCPs
at a protein:polysaccharide molar ratio of 1:10. The reaction mixture was mixed gently overnight at 3^8³C, after
FEMSIM 1336 30-8-01
K.M. Reddin et al. / FEMS Immunology and Medical Microbiology 31 (2001) 153^162
which time the mixture was dialysed (MW cuto¡ 12 000;
Sigma) against at least three changes of 5 l of 50 mM
ammonium bicarbonate, 150 mM NaCl, pH 7.2, to remove hydrolysed, unconjugated polysaccharide before
freeze-drying and puri¢cation by size exclusion chromatography. The conjugate was puri¢ed using a Hiload
Superdex 200 HR (16/60) column (Amersham Pharmacia);
the conjugate eluted in the void volume (V0 ). Protein was
detected by absorbance at 280 nm and polysaccharide was
detected by changes in refractive index (RI) using an RI
detector (Viscotek) [23]. Co-elution of protein and polysaccharide in V0 con¢rmed conjugation. The concentrations of protein and polysaccharide were determined using
the appropriate assays (Sections 2.4 and 2.5).
2.3.2. Reductive amination
To couple spaced, surface lysines to dFim [21], protein
(10 mg dFim) was dissolved in 9 ml of 50 mM HEPES,
pH 8.5 (1.1 mg ml31 ), in a glass bottle, to which 225 Wl of
lysine solution (2 mg ml31 ) was added, followed by 1 ml
dimethyl suberimidate-2 HCl (DMS, 25 mg ml31 ). The
solution was mixed for 1 h at room temperature before
adding a further 2 ml of the lysine solution to stop the
reaction. The material was dialysed (MW cuto¡ 12 000;
Sigma) overnight against at least three changes of 5 l of
0.1 M PBS, pH 7.4. The dialysed derivatised protein was
then concentrated at 25³C using polyethylene glycol
(PEG, 8000 Da molecular mass) until the volume was
reduced to 1^2 ml. The concentrate was then dialysed
(MW cuto¡ 12 000, Sigma) against at least three changes
of 5 l of 100 mM sodium bicarbonate, pH 9.0, overnight
at 4³C.
Polysaccharide was used at a molar ratio (protein :polysaccharide) of 1:10, based on the number of moles of
starting material. Oxidation of the polysaccharide [22]
was achieved by dissolving 150 mg menCPs in 1 ml of
1 mM sodium periodate solution and mixed in the dark
for 30 min at 4³C, 1 ml of ethylene glycol was then added
to stop the reaction ; oxidisation with sodium periodate
hydrolysed the majority of the polysaccharide (data not
shown). To couple the polysaccharide to dFim, this mixture was immediately added to the derivatised protein solution and incubated at 4³C for 2 h. 200 Wl of a freshly
prepared sodium borohydride solution (50 mg ml31 ) was
added to the reaction mixture and incubated at 4³C overnight. The product was dialysed (MW cuto¡ 12 000; Sigma) against at least three changes of 5 l of PBS pH 7.4 to
remove unconjugated, hydrolysed polysaccharide. The
conjugate was puri¢ed and conjugation con¢rmed by
size exclusion chromatography (Section 2.3.1).
2.4. Protein determination
Protein concentrations were determined using a BCA
assay (Pierce).
155
2.5. Polysaccharide determination
Sialic acid concentrations were determined using the
Warren assay [24] and total polysaccharide determined
by the resorcinol assay [25].
2.6. Immunisations
To determine the immune response to conjugates,
groups of ¢ve Balb/c mice (female, 6^9 weeks old) were
immunised with two doses of conjugate vaccine containing
10 Wg polysaccharide or unconjugated control antigens;
protective e¤cacy was determined after groups of mice
received three doses of the above preparations. For assessment of dose response, Balb/c mice (¢ve per group) were
immunised with dFim-menCPs(RA) conjugate containing
either 0.1 Wg, 1.0 Wg or 10 Wg of polysaccharide per dose.
All vaccines were adjuvanted with aluminium phosphate
(Adjuphos, Superfos-Biosector, Frederikssund, Denmark)
to give a ¢nal concentration of 4 mg ml31 aluminium
phosphate. Subcutaneous injections were performed on
days 1 and 28 for immune response and on days 1, 21
and 28 for assessment of protective e¤cacy. Sera were
collected on days 0, 14, 21 and 35 and assayed for antibodies against native ¢mbriae, dFim and menCPs by
ELISA for whole IgG and antibody isotype.
2.7. Immunoassays
2.7.1. ELISA
96-Well microtitration plates (Immulon) were coated
with 1 Wg ml31 native ¢mbriae, dFim or de-O-acetylated
menCPs-methylated human serum albumin (menCPsmHSA) in PBS overnight at 4³C and incubated with serum dilutions. Speci¢c antibodies were detected using goat
anti-mouse IgG-HRP and anti-mouse IgM and IgG isotype-HRP conjugates (Jackson) and TMBlue (Universal
Biologicals). The antibody titre was de¢ned as the reciprocal of the dilution of serum corresponding to the midpoint of the dose^response curve. This was calculated
using interpolation software (Genesis ; Labsystems) on
dose^response curves generated from eight dilutions of
each serum. Interplate variation was corrected for by using
a pool of day 14 sera, and sera showing a titre less than
the detection limit were assigned an arbitrary titre of 50
for calculation of geometric means. An avidity ELISA
using thiocyanate as the chaotropic agent [26,27] for
anti-menCPs antibodies was carried out on day 35 sera.
2.7.2. Serum bactericidal assay
Serum bactericidal antibodies (SBA) were determined
according to standardised methodology [28] using a complement source of 25% baby rabbit complement (Pelfreeze)
and N. meningitidis serogroup C strain C11 as the target
strain. Titres are expressed as the reciprocal of the ¢nal
FEMSIM 1336 30-8-01
156
K.M. Reddin et al. / FEMS Immunology and Medical Microbiology 31 (2001) 153^162
dilution producing v 50% bactericidal killing after 60 min
[29].
2.7.3. Inhibition assay
Inhibition assays were performed for the detection of
polysaccharide-speci¢c antibodies in murine sera 35 days
after immunisation with dFim-menCPs(RA) conjugate.
Brie£y, dilutions of puri¢ed menCPs (Section 2.1) ranging
from 3.9 Wg ml31 to 500 Wg ml31 were prepared in a
microtitre plate (Sterilin) with PBS, 0.01% (v/v) Tween
20, 10% (v/v) newborn calf serum blocking bu¡er as diluent. Murine sera were diluted 1/100 in blocking bu¡er
and an equal volume was added to each well of the microtitre plate containing the polysaccharide. The plate was
incubated overnight at 4³C with gentle shaking. Untreated
sera were used as a control. Polysaccharide-treated and
untreated sera were then assayed by ELISA against
menCPs-mHSA. Binding of speci¢c antibody was detected
using anti-mouse IgG-HRP (Jackson) and TMBlue (Universal Biologicals). Using the absorbance at 450 nm of the
untreated sera as 100% binding, results were expressed as
percentage inhibition of antibody binding against concentration of menCPs (Wg ml31 ) used in the preincubation
step with the murine sera.
2.8. Protection studies
2.8.1. Murine intranasal infection model for B. pertussis
2.8.1.1. Immunisations. Harlan NIH mice (female, 6^8
weeks old) were immunised subcutaneously with native or
dissociated puri¢ed B. pertussis ¢mbriae. Each mouse received 200 Wl antigen (50 Wg ml31 ) mixed 1:1 with
Freund's complete adjuvant. Mice were boosted 2 weeks
later with the same amount of ¢mbriae but mixed 1:1 with
Freund's incomplete adjuvant.
(w/v) casamino acids. The organs were homogenised using
a Silverson homogeniser and serial dilutions (1031 , 1032 ,
1033 ) of the homogenised tissue were prepared. The numbers of viable bacteria in organs were determined by inoculating 4U25 Wl from each dilution onto selective CA
plates containing 40 Wg ml31 of cephalexin and plates incubated for 6 days at 35³C in an humidi¢ed atmosphere.
2.8.2. Murine intraperitoneal infection model for
N. meningitidis
To determine the protective e¤cacy of dFim-menCPs(RA) conjugate, Balb/c mice (female 6^8 weeks old, Harlan) were immunised on days 1, 21 and 28 with conjugate
or an unconjugated mixture of polysaccharide and protein
containing 10 Wg ml31 polysaccharide or dFim. The mice
were infected on day 35 using the intraperitoneal (i.p.)
infection model [30]. Bacteria were grown in MHB broth
for 4 h, adjusted to the required density with the same
medium, and mixed with an equal volume of sterile human
transferrin (40 mg ml31 , Sigma) in PBS. Mice received the
appropriate challenge dose i.p. in a 0.5 ml suspension and
24 h later a second i.p. injection of 0.2 ml saline containing
human transferrin (50 mg ml31 ) was administered. Susceptibility to infection was monitored for 4 days after infection and the endpoint was reached when mice exhibited
symptoms of closed eyes, ru¥ed fur and immobility, at
which time mice were euthanatised.
All procedures involving animals were conducted according to the requirements of the UK Home O¤ce Ani-
2.8.1.2. Preparation of inocula. Bacterial inocula were
prepared for the infection study by growing B. pertussis on
charcoal agar (CA) plates for 48 h at 35³C. Bacteria were
then subcultured onto fresh CA plates and incubated for a
further 24 h at 35³C. Growth from these plates was dispersed into 10 ml of 1% (w/v) casamino acids to give an
OD550 of approx. 1.0. The suspension was then adjusted in
casamino acids to an OD550 of 0.1, which was estimated to
contain approx. 108 colony forming units (cfu) ml31 .
2.8.1.3. Determination of protective e¤cacy. Two
weeks after the second dose of antigen, mice were challenged intranasally with B. pertussis. Under light anaesthesia (0.1 ml of Hypnorm/Hypnovel), 50 Wl of bacterial inoculum containing approx. 108 cfu ml31 was administered
into the nares and allowed to be inhaled. Seven days post
challenge, mice were sacri¢ced by cervical dislocation and
lungs and trachea were removed. Each organ was placed
into separate glass bottles containing 10 ml of sterile 1%
Fig. 1. FPLC pro¢les of dFim (A) and dFim-menCPs(RA) conjugate
(B). A: SEC of dFim analysed using Superdex 200 Hi-load 16/60 column in 0.2 M ammonium acetate, pH 7.2. P1a is re-associated ¢mbriae
and P2a is dFim. B: SEC of dFim-menCPs(RA) conjugate analysed using a Superdex 200 Hi-load 16/60 column in 0.2 M ammonium acetate,
pH 7.2. P1b is dFim-menCPs(RA) conjugate and P2b is unconjugated
dFim and unconjugated oligosaccharide.
FEMSIM 1336 30-8-01
K.M. Reddin et al. / FEMS Immunology and Medical Microbiology 31 (2001) 153^162
157
Fig. 2. Serum IgG response to menCPs in Balb/c mice. Titres are expressed as the geometric mean of pooled murine sera (n = 5). dFim-menCPs(RA)
elicits a boostable response and high avidity antibodies in day 35 (D35) sera.
mals (Scienti¢c Procedures) Acts, 1986 and are subject to
ethical review.
3. Results
The dFim-menCPs(EDC) conjugate had a polysaccharide:protein ratio of 1:1.75. The menCPs used to produce
the conjugate was also analysed by NMR spectroscopy
and con¢rmed to be de-O-acetylated (data not shown).
3.2. Immune response to conjugates
3.1. Puri¢cation of conjugates
Conjugates were puri¢ed by SEC on a Superdex 200
column. Fig. 1A shows the SEC pro¢le for dFim (P2a);
some dFim has reassociated and eluted close to the void
volume for the column (P1a). The conjugate eluted in the
void volume (P1b, Fig. 1B), while unconjugated dFim and
any remaining oligosaccharide eluted in fractions 20^23
(P2b, Fig. 1B), the pro¢le indicates that the majority of
the dFim was conjugated to the menCPs. Immunoblot
analysis of SEC fractions using antibody speci¢c for
dFim or menCPs was used to con¢rm co-elution of protein and polysaccharide (data not shown). There was some
variation in the ratio of polysaccharide:protein in successive preparations of conjugates; dFim-menCPs(RA) conjugate with a polysaccharide :protein ratio of 3.2:1 was
used for assessment of the immune response, whereas a
conjugate with a polysaccharide:protein ratio of 5.2:1
was used for the determination of protective e¤cacy.
3.2.1. ELISA
Sera from immunised Balb/c mice were assayed by
ELISA to determine the antibody response to native ¢mbriae and menCPs. The antibody titres in Fig. 2 demonstrate a speci¢c, boostable response to menCPs. In mice
immunised with d¢m-menCPs(RA), high antibody titres
were observed after a boost dose of conjugate and the
antibodies were also of high avidity. High serum IgG antibody levels against native ¢mbriae were elicited by all
vaccines (Fig. 3), indicating that ¢mbrial subunit structure
is su¤ciently conserved throughout the dissociation and
conjugation process to maintain important epitopes. Speci¢c antibodies to dissociated ¢mbriae were observed with
all vaccines except menCPs (data not shown).
3.2.2. Isotype ELISA
Immunoglobulin isotype was determined in murine sera
after boosting and determined that the response to con-
Table 1
Ratio of IgG to IgM in pre-boost and post-boost murine sera tested against d¢m-menCPs conjugate
Sera
D21 (pre-boost)
D35 (post-boost)
Increase in ratioa (D35 versus D21)
Immunogen
dFim-menCPs(EDC)
dFim-menCPs(RA)
Unconjugated mixture
menCPs
dFim
182:1
2686:1
15-fold
2:1
119:1
60-fold
25:1
19:1
0
ND
2:1
0
1:1
147:1
147-fold
a
The increase in ratio of IgG:IgM was calculated from the geometric mean ELISA titres of IgG and IgM in pooled murine sera (n = 5). The ratio of
IgG:IgM titre in day 35 (post-boost) sera versus the ratio in day 21 (pre-boost) sera was used.
FEMSIM 1336 30-8-01
158
K.M. Reddin et al. / FEMS Immunology and Medical Microbiology 31 (2001) 153^162
Fig. 3. Serum IgG response to native ¢mbriae in Balb/c mice. Titres are expressed as the geometric mean of pooled sera (n = 5).
jugates was primarily IgG1, indicating a predominantly
Th2 response. In addition, the ratio of IgG to IgM was
increased following boosting with the conjugate but not
with polysaccharide or an unconjugated mixture, indicating that the response to the conjugate is TD, in contrast
with the TI response elicited by the polysaccharide and the
unconjugated mixture. This observation was noted when
sera were assayed against the conjugate or the polysaccharide (Table 1 and Table 2, respectively). The greatest increase in ratio was observed in the dFim-menCPs(RA)
conjugate sera.
Ps, with the greatest response in mice given 10 Wg menCPs
in the conjugate dose. However, a boostable anti-¢mbriae
response was observed only at the higher dose (Fig. 4).
3.2.3. Dose response
Sera from Balb/c mice immunised with dFim-menCPs(RA) conjugate containing either 0.1 Wg, 1.0 Wg or 10 Wg
Ps were assayed by ELISA to determine the IgG response
to menCPs and native ¢mbriae. A dose^response e¡ect
was observed for both antigens. A boostable response
was observed against menCPs at both 1 Wg and 10 Wg
3.2.5. Inhibition ELISA
In the inhibition ELISA, 200 Wg ml31 menCPs was suf¢cient to produce s 90% inhibition of binding of the conjugate antibody to conjugate bound to the assay plate
(Fig. 6). These results demonstrate that immunisation
with dFim-menCPs(RA) conjugate elicits polysaccharidespeci¢c antibody.
3.2.4. Serum bactericidal assay
Bactericidal antibody titres were determined in the sera
of immunised mice after boosting. The highest serum bactericidal response was found in mice immunised with d¢mmenCPs(RA) conjugate. After three doses of this conjugate, the serum bactericidal titre increased to 1/2048 (Fig.
5).
Table 2
Ratio of IgG to IgM in pre-boost and post-boost murine sera tested against menCPs
Sera
D21 (pre-boost)
D35 (post-boost)
Increase in ratioa (D35 versus D21)
a
Immunogen
dFim-menCPs(EDC)
dFim-menCPs(RA)
Unconjugated mixture
menCPs
1:1
2:1
2-fold
1:1
15:1
15-fold
2:1
1:2
0
2:1
2:1
0
Refer to Table 1.
FEMSIM 1336 30-8-01
K.M. Reddin et al. / FEMS Immunology and Medical Microbiology 31 (2001) 153^162
159
Fig. 6. Inhibition ELISA of D35 murine sera after three doses of dFimmenCPs(RA) conjugate. The sera from ¢ve female Balb/c mice were
pooled prior to the inhibition assay.
with the homologous serogroup C strain using two challenge doses. The infection was monitored over a period of
4 days (Fig. 7). The dFim-menCPs conjugate a¡orded
100% protection at both challenge doses compared with
menCPs alone which gave 100% protection at the low dose
Fig. 4. Dose response in Balb/c mice to dFim-menCPs(RA) conjugate.
Titres are expressed as the geometric mean of pooled murine sera
(n = 5). Panel A demonstrates the dose response to menCPs and panel B
the dose response to native ¢mbriae.
3.3. Protective e¤cacy of dFim-menC conjugate
The dFim-menCPs(RA) gave the highest avidity and
serum bactericidal antibodies and was therefore used to
immunise mice in a murine i.p. challenge model of meningococcal disease (Section 2.6). Mice were immunised with
either conjugate, menCPs alone or dFim and challenged
Fig. 5. Serum bactericidal antibody titre in murine sera comparing conjugates prepared using di¡erent chemistries. SBA titre was determined
in pooled murine sera (n = 5). The SBA titre is expressed as the geometric mean of duplicate serum bactericidal assays.
Fig. 7. Protective e¤cacy of dFim-menCPs(RA) conjugate. A: Murine
challenge with 2.4U106 cfu N. meningitidis serogroup C L91 543
(C:2a:P1.2R) (n = 5, except in dFim group where n = 4). B: Murine
challenge with 7.5U108 cfu (n = 5). At both challenge doses, the di¡erence in protection between mice immunised with conjugate and unvaccinated controls was signi¢cant (P90.01).
FEMSIM 1336 30-8-01
160
K.M. Reddin et al. / FEMS Immunology and Medical Microbiology 31 (2001) 153^162
Fig. 8. Recovery of B. pertussis from lungs and trachea of NIH mice
after intranasal infection.
but no protection at the higher dose of 7.5U108 cfu. For
mice immunised with conjugate, the di¡erence in protection when compared with the unvaccinated controls was
signi¢cant at both challenge doses (P 9 0.01, Finney contingency tables).
3.4. Protective e¤cacy of dFim
The ability of dissociated ¢mbriae to protect mice
against intranasal colonisation with B. pertussis was determined. Unimmunised animals were more susceptible to
infection with B. pertussis, whereas all animals immunised
with either native or dissociated ¢mbriae were protected
(Fig. 8). Although there was some variability in the levels
of colonisation of lungs and trachea, particularly in the
unvaccinated group, protection was observed and dissociated ¢mbriae induced similar protection against respiratory colonisation to that induced by immunisation with
native ¢mbriae.
4. Discussion
Enhancement of the immunogenicity of polysaccharides
by coupling polysaccharides or oligosaccharides to proteins is well established [31^33]. Typically, polysaccharides
have been coupled to large immunogenic proteins such as
TT or DT. These proteins were chosen in most cases because both have been used for human vaccination for
many years without widely reported adverse side e¡ects.
However, since most individuals are immunised with TT
and DT, the response against an antigen coupled to these
carriers may be inhibited by anti-carrier immunity. This
immunomodulatory e¡ect is termed carrier-induced epitopic suppression [34] and has been demonstrated with a
number of synthetic peptide-protein conjugates using TT
as the carrier protein [35^37]. For example, Peeters et al.
[38] observed that when mice were primed with a high
dose of carrier protein (25 Wg) before immunisation with
polysaccharide-protein conjugates, the anti-polysaccharide
antibody was suppressed. They also found that the consequences of carrier priming persisted after repeated immunisations with the conjugate vaccine and occurred with
both high molecular mass polysaccharides and oligosaccharides. This indicates that high dose carrier priming
may cause signi¢cant suppression of the anti-polysaccharide IgG antibody response induced by protein-polysaccharide conjugates. Although the mechanism of this suppression is unclear, involvement of T suppressor cells and
carrier-speci¢c B memory cells has been reported [38].
Therefore, if more than one conjugate vaccine is used, or
if they are given as multivalent formulations with the same
carrier protein, the likelihood of carrier-induced epitopic
suppression is increased. The immunological history of the
vaccinee would also be important [39]. Therefore, there is
a need for alternative carrier proteins for use in conjugate
vaccines.
B. pertussis components in a vaccine may modulate immune suppression and administration of pertussis vaccine
or puri¢ed components has been found to block the establishment of epitopic suppression. Administration of
B. pertussis components enhanced antibody responses to
both carrier and synthetic peptides [36]. In this study, immunisation of mice with dissociated ¢mbriae demonstrated a protective e¡ect in an intranasal colonisation
model for B. pertussis infection. Dissociated ¢mbriae
were used as carrier proteins for menCPs because native
¢mbriae conjugates were di¤cult to characterise, principally due to their large molecular mass and tendency to
aggregate [16]. By using dFim, the conjugation process
proved to be more e¤cient and it was possible to detect
the increase in molecular mass when conjugated to
menCPs. Native, unhydrolysed menCPS was used as the
starting material as previous studies have demonstrated a
direct relationship between the molecular size of a polysaccharide and its ability to induce an immune response in
man [2].
Rubenstein and Stein [40] reported that the immune
response to menCPs in Balb/c mice provides a model system which closely parallels the response in humans. Conjugating dFim to menCPs by reductive amination resulted
in a conjugate that produced high, boostable IgG responses in our Balb/c mouse model. This was not observed
when mice were immunised with menCPs alone or an unconjugated mixture of dFim and menCPs. Responses to a
conjugate produced by carbodiimide coupling were less
pronounced; the reasons for this are unclear, but may
be due to the lower polysaccharide :protein ratio in this
conjugate or contaminating unconjugated unhydrolysed
menCPs which was not removed by dialysis. The responses
were TD, as demonstrated by an increase in the IgG:IgM
FEMSIM 1336 30-8-01
K.M. Reddin et al. / FEMS Immunology and Medical Microbiology 31 (2001) 153^162
ratio after boosting. The antibodies were of high avidity
and speci¢c for the polysaccharide moiety; in addition,
antibodies were found to be bactericidal against a heterologous serogroup C strain. The conjugate provided 100%
protection in mice against challenge with 7.5U108 cfu of
the homologous meningococcal strain.
Conjugates of menCPs and TT using long chain saccharides or short chain oligosaccharides conjugated by reductive amination [6,33,41] have been used in previous studies
to immunise Balb/c mice. These studies showed that the
conjugate, compared with unconjugated menCPs, resulted
in an isotype shift from predominantly IgM and IgG3 to
predominantly IgG1, accompanied by a large (10-fold)
increase in bactericidal activity and an increase in antibody a¤nity. Fukasawa et al. [42] conjugated menCPs
to N. meningitidis serogroup B outer membrane vesicles
by carbodiimide coupling using ADH as a C6 spacer.
They studied the immune response in C3H/Hepas mice
and found that this conjugate elicited boostable IgG titres
and bactericidal antibodies. The bactericidal antibody response was increased approx. 2-fold when comparing sera
raised against menCPs alone and the conjugate vaccine.
These responses are similar to those observed in this study
where a v 10-fold increase in bactericidal titre was seen
with the reductive amination conjugate, but a much smaller increase in titre with the conjugate produced by carbodiimide coupling. It is, however, di¤cult to make comparisons with the study of Fukasawa et al. [42] because a
di¡erent mouse strain was used.
This study indicates that immunisation with a dFimmenCPs conjugate vaccine is protective in animal models
and elicits speci¢c and functional TD antibody responses.
Therefore, this vaccine should protect against serogroup C
meningococcal disease and augment protection against
pertussis when acellular vaccines de¢cient in ¢mbriae are
used. Furthermore, it has been recommended that acellular pertussis vaccine should be given to adults in order to
maintain life-long protection [43]. Therefore, a conjugate
vaccine based upon B. pertussis ¢mbriae and menCPs,
which could be given to young adults as well as infants
to protect against meningococcal disease and pertussis,
would have clear advantages.
Acknowledgements
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
This work was funded by the UK Department of
Health. The authors would like to thank Dr Ray Borrow,
Manchester PHL for his advice on serological assays and
Dr Chris Jones, NIBSC for NMR spectroscopy of
menCPs.
[17]
References
[19]
[1] Goldschneider, I., Gotschlich, E.C. and Artenstein, M.S. (1969) Hu-
[18]
161
man immunity to the meningococcus I. The role of humoral antibodies. J. Exp. Med. 129, 1307^1326.
Gotschlich, E.C., Goldschneider, I. and Artenstein, M.S. (1969) Human immunity to the meningococcus IV. Immunogenicity of group A
and group C polysaccharides in human volunteers. J. Exp. Med. 129,
1367^1384.
Reller, L.B., MacGregor, R.R. and Beaty, H.N. (1973) Bactericidal
antibody after colonisation with Neisseria meningitidis. J. Infect. Dis.
127, 56^62.
Goldschneider, I., Gotschlich, E.C. and Artenstein, M.S. (1969) Human immunity to the meningococcus II. Development of natural
immunity. J. Exp. Med. 129, 1327^1347.
Stein, K.E. (1992) Thymus-independent and thymus-dependent responses to polysaccharide antigens. J. Infect. Dis. 165, S49^S52.
Garc|¨a-O'Jeda, P.A., Monser, M.E., Rubinstein, L.J., Jennings, H.J.
and Stein, K.E. (2000) Murine immune response to Neisseria meningitidis group C capsular polysaccharide: analysis of monoclonal antibodies generated in response to a thymus dependent toxoid conjugate
vaccine. Infect. Immun. 68, 239^246.
Wenger, J.D., Heath, P.T., Moxon, R. and Booy, R. (1997) Epidemiological impact of conjugate vaccines against invasive disease
caused by H. in£uenzae type b. In: New Generation Vaccines, 2nd
edn. (Levine, M.M., Woodrow, G.C., Kaper, J.B. and Cobon, G.S.,
Eds.), pp. 489^502. Marcel Dekker, New York.
Maclennan, J.M., Shackley, F., Heath, P.T., Deeks, J.J., Flamank,
C., Herbert, M., Gri¤ths, H., Hatzmann, E., Goilav, C. and Moxon,
E.R. (2000) Safety, immunogenicity, and induction of immunological
memory by a serogroup C meningococcal conjugate vaccine in infants: a randomised controlled trial. J. Am. Med. Assoc. 283, 2795^
2801.
Ramsey, M.E., Andrews, N., Kaczmarski, E.B. and Miller, E. (2001)
E¤cacy of meningococcal serogroup C conjugate vaccine in teenagers
and toddlers in England. Lancet 357, 195^196.
Beuvery, E.C., Miedema, F., Van Delft, R. and Haverkamp, K.
(1983) Preparation and immunochemical characterisation of meningococcal serogroup C polysaccharide-tetanus toxoid conjugates as a
new generation of vaccines. Infect. Immun. 40, 39^45.
Maclennan, J.M., Deeks, J.J., Obaro, S., Williams, D., Carlone,
G.M., Moxon, E.R. and Greenwood, B.M. (2001) Immunologic
memory 5 years after meningococcal A/C conjugate vaccination in
infancy. J. Inf. Dis. 39, 97^104.
Anderson, P. (1983) Antibody response to Hib and dp toxin induced
by conjugate of oligosaccharide of the type b capsule with the nontoxic protein CRM197. Infect. Immun. 39, 233^238.
Goldblatt, D. (1998) Recent developments in bacterial conjugate vaccines. J. Med. Microbiol. 47, 563^567.
Fattom, A., Hee Cho, Y., Chu, C., Fuller, S., Fries, L. and Naso, R.
(1999) Epitopic overload at the site of injection may result in suppression of the immune response to combined capsular polysaccharide conjugate vaccines. Vaccine 17, 126^133.
Olin, P., Rasmussen, F., Gustafsson, L., Hallander, H. and Helijebel,
H. (1997) Randomised control trial of two-component, three-component, and ¢ve component acellular pertussis vaccine. Lancet 350,
1569^1577.
Crowley-Luke, A., Reddin, K.M., Gorringe, A.R., Hudson, M.J. and
Robinson, A. (2001) Formulation and characterisation of Bordetella
pertussis ¢mbriae as novel carrier proteins for Hib conjugate vaccines. Vaccine 19, 3399^3407.
Robinson, A., Gorringe, A.R., Funnell, S.G. and Fernandez, M.
(1989) Serospeci¢c protection of mice against intranasal infection
with Bordetella pertussis. Vaccine 7, 321^324.
Gotschlich, E.C. (1975) Development of polysaccharide vaccines for
the prevention of meningococcal diseases. Monogr. Allergy 9, 245^
258.
Pearce, A., Irons, L., Robinson, A. and Seabrook, R. (1992) E¡ects
of guanidinium hydrochloride on the structure and immunological
properties of Bordetella pertussis ¢mbriae. Biochem. J. 283, 823^828.
FEMSIM 1336 30-8-01
162
K.M. Reddin et al. / FEMS Immunology and Medical Microbiology 31 (2001) 153^162
[20] Schneerson, R., Barrera, O., Sutton, A. and Robbins, J.B. (1980)
Preparation, characterisation, and immunogenicity of Haemophilus
in£uenzae type b polysaccharide protein conjugates. J. Exp. Med.
152, 361^376.
[21] Hand, E.S. and Jencks, W.P. (1962) Mechanism of the reaction of
imidoesters with amines. J. Am. Chem. Soc. 84, 3505^3514.
[22] Wilson, M.B. and Nakane, P.K. (1978). In: Immuno£uorescence and
Related Techniques (Knapp, W., Holubar, H. and Wick, G., Eds.),
p. 215. Elsevier/North-Holland, Amsterdam.
[23] Plumb, J. and Yost, S. (1996) Molecular size characterisation of
Haemophilus in£uenzae type b polysaccharide-protein conjugate vaccines. Vaccine 14, 399^404.
[24] Chaplin, M.F. (1994) Monosaccharides ^ the Warren assay for sialic
acid. In: Carbohydrate Analysis, a Practical Approach Chaplin, M.F.
and Kennedy, J.F., Eds.) IRL Press, 7^8.
[25] Monsigny, M., Petit, C. and Roche, A.C. (1988) Colorimetric determination of neutral sugars by a resorcinol sulfuric acid micromethod.
Anal. Biochem. 175, 525^530.
[26] Grano¡, D.M., Maslanka, S.E., Carlone, G., Plikaytis, B.D., Santos,
G.F., Mokatrin, A. and Ra¡, H.V. (1998) A modi¢ed enzyme-linked
immunosorbent assay of measurement of antibody responses to meningococcal C polysaccharide that correlate with bactericidal responses. Clin. Diagn. Lab. Immunol. 5, 479^485.
[27] Gheesling, L.L., Carlone, G.M., Pais, L.B., Holder, P.F., Maslanka,
S.E., Plikaytis, B.D., Achtman, M., Densen, P., Frasch, C.E.,
Kayhty, H., Mays, J.P., Nencioni, L., Peeters, C., Phipps, D.C., Poolman, J.T., Rosenqvist, E., Siber, G.R., Thiesen, B., Tai, J., Thompson, C.M., Vella, P.P. and Wenger, J.D. (1994) Multicenter comparison of Neisseria meningitidis serogroup C anti-capsular
polysaccharide antibody levels measured by a standardised enzymelinked immunosorbent assay. J. Clin. Microbiol. 32, 1475^1482.
[28] Centers for Disease Control (1992) Report of the 2nd International
Workshop on Meningococcal Immunology and Serology, 17^18 June
(Decatur, GA), CDC, Atlanta, GA.
[29] Maslanska, S.E., Gheesling, L.L., Libutti, D.E., Donaldson, K.B.,
Harakeh, H.S., Dykes, J.K., Arhin, F.F., Devi, S.J., Frasch, C.E.,
Huang, J.C., Kris-Kuzemenska, P., Lemmon, R.D., Lorange, M.,
Peeters, C.C., Quataert, S., Tai, J.Y. and Carlone, G.M. (1997)
Standardisation and a multilaboratory comparison of Neisseria meningitidis serogroup A and C serum bactericidal assays. The Multilaboratory Study Group. Clin. Diagn. Lab. Immunol. 4, 156^167.
[30] Wilks, K.E., Dunn, J.L., Farrant, J.L., Reddin, K.M., Gorringe,
A.R., Langford, P.R. and Kroll, J.S. (1998) Periplasmic superoxide
dismutase in meningococcal pathogenicity. Infect. Immun. 66, 213^
217.
[31] Beuvery, E.C., van Rossum, F. and Nagel, J. (1982) Comparison of
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
the induction of IgM and IgG antibodies in mice with puri¢ed pneumococcal type 3 and meningococcal group C polysaccharides and
their protein conjugates. Infect. Immun. 37, 15^22.
Einhorn, M.S., Weinberg, G.A., Anderson, E.L., Grano¡, P.D. and
Grano¡, D.M. (1986) Immunogenicity in infants of Hib polysaccharide in a conjugate vaccine with Neisseria meningitidis outer membrane protein. Lancet ii, 1415^1418.
Jennings, H.J. and Lugowski, C. (1981) Immunochemistry of groups
A, B and C meningococcal polysaccharide-tetanus toxoid conjugates.
J. Immunol. 127, 1011^1018.
Muller, E. and Apicella, M.A. (1988) T cell modulation of the murine
antibody response to Neisseria meningitidis group A capsular polysaccharide. Infect. Immun. 56, 259^266.
Pollard, A.J. and Levine, M. (2000) Vaccines for the prevention of
meningococcal disease. Pediatr. Infect. Dis. J. 19, 333^345.
Vogel, F.R., LeClerc, C., Schutze, M.P., Jolivet, M., Audibet, F.,
Klein, T.W. and Chedid, L. (1987) Modulation of carrier-induced
epitopic suppression by Bordetella pertussis components and muramyl peptide. Cell. Immunol. 107, 40^51.
Schutze, M.P., LeClerc, C., Jolivet, M., Audibert, F. and Chedid, L.
(1985) Carrier-induced epitopic suppression, a major issue for synthetic vaccines. J. Immunol. 135, 2319^2322.
Peeters, C.C.A.M., Tenbergen-Meekes, A.M., Poolman, J.T., Beurret, M., Zegers, B.J.M. and Rijkers, G.T. (1991) E¡ect of carrier
priming on immunogenicity of saccharide-protein conjugate. Vaccine
59, 3504^3510.
Herzenberg, L.A. and Tokuhisa, T. (1982) Epitopic speci¢c regulation I. Carrier-speci¢c induction of suppression for IgG anti-hapten
antibody responses. J. Exp. Med. 155, 1730^1740.
Rubenstein, L.J. and Stein, K.E. (1988) Murine immune response to
the Neisseria meningitidis group C capsular polysaccharide. I. Ontogeny. J. Immunol. 141, 4352^4356.
Rubenstein, L.J., Garcia-Ojeda, P.A., Michon, F., Jennings, H.J. and
Stein, K.E. (1998) Murine immune responses to Neisseria meningitidis
group C capsular polysaccharide and a thymus-dependent toxoid
conjugate vaccine. Infect. Immun. 66, 5450^5456.
Fukasawa, L.O., Gorla, M.C.O., Schenkman, R.P.F., Garcia, L.R.,
Carneiro, S.M., Raw, I. and Tanizaki, M.M. (1999) Neisseria meningitidis serogroup C polysaccharide and serogroup B outer membrane
vesicle conjugate as a bivalent meningococcus vaccine candidate.
Vaccine 17, 2951^2958.
Granstrom, M. (1996) The history of pertussis vaccination : from
whole-cell to subunit vaccines. Vaccinia, vaccination, vaccinology.
Jenner, Pasteur and their successors. International Meeting on the
History of Vaccinology, Paris (Plotkin, S.A. and Fantini, B., Eds.
Published by Elsevier). Pages 107^114.
FEMSIM 1336 30-8-01