Antigenicity and immunogenicity of a synthetic oligosaccharide-protein conjugate vaccine against Haemophilus influenzae type b.

Abstract

Polysaccharide-protein conjugates as vaccines have proven to be very effective in preventing Haemophilus influenzae type b infections in industrialized countries. However, cost-effective technologies need to be developed for increasing the availability of anti-H. influenzae type b vaccines in countries from the developing world. Consequently, vaccine production with partially synthetic antigens is a desirable goal for many reasons. They may be rigidly controlled for purity and effectiveness while at the same time being cheap enough that they may be made universally available. We describe here the antigenicity and immunogenicity of several H. influenzae type b synthetic oligosaccharide-protein conjugates in laboratory animals. The serum of H. influenzae type b-immunized animals recognized our synthetic H. influenzae type b antigens to the same extent as the native bacterial capsular polysaccharide. Compared to the anti-H. influenzae type b vaccine employed, these synthetic versions induced similar antibody response patterns in terms of titer, specificity, and functional capacity. The further development of synthetic vaccines will meet urgent needs in the less prosperous parts of the world and remains our major goal.

Full text

Page 1

INFECTION AND IMMUNITY, Dec. 2004, p. 7115–7123
0019-9567/04/$08.00?0DOI: 10.1128/IAI.72.12.7115–7123.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.
Vol. 72, No. 12
Antigenicity and Immunogenicity of a Synthetic
Oligosaccharide-Protein Conjugate Vaccine against
Haemophilus influenzae Type b
V. Ferna ´ndez-Santana,1* Fe ´lix Cardoso,1Arlene Rodriguez,2Tania Carmenate,1Luis Pen ˜a,1
Yuri Valde ´s,1Eugenio Hardy,2Fatme Mawas,3Lazaro Heynngnezz,2Maria C. Rodríguez,1
Ignacio Figueroa,1Janoi Chang,1Maria E. Toledo,4Alexis Musacchio,2Ibis Herna ´ndez,4
Mabel Izquierdo,2Karelia Cosme,2Rene Roy,5and V. Verez-Bencomo1
Center for the Study of Synthetic Antigens, Facultad de Química, Universidad de la Habana,1Center for Genetic
Engineering and Biotechnology, Cubanacan, Playa,2and Institute for Tropical Medicine “Pedro Kouri,”
Autopista Novia del Mediodía, La Lisa,4Havana, Cuba; National Institute for Biological
Standard and Control, Blanche Lane, South Mimms, United Kingdom3; and
Department of Chemistry, Universite ´ du Que ´bec a ` Montre ´al,
Montreal, Que ´bec, Canada5
Received 6 April 2004/Returned for modification 1 June 2004/Accepted 6 July 2004
Polysaccharide-protein conjugates as vaccines have proven to be very effective in preventing Haemophilus
influenzae type b infections in industrialized countries. However, cost-effective technologies need to be devel-
oped for increasing the availability of anti-H. influenzae type b vaccines in countries from the developing world.
Consequently, vaccine production with partially synthetic antigens is a desirable goal for many reasons. They
may be rigidly controlled for purity and effectiveness while at the same time being cheap enough that they may
be made universally available. We describe here the antigenicity and immunogenicity of several H. influenzae
type b synthetic oligosaccharide-protein conjugates in laboratory animals. The serum of H. influenzae type
b-immunized animals recognized our synthetic H. influenzae type b antigens to the same extent as the native
bacterial capsular polysaccharide. Compared to the anti-H. influenzae type b vaccine employed, these synthetic
versions induced similar antibody response patterns in terms of titer, specificity, and functional capacity. The
further development of synthetic vaccines will meet urgent needs in the less prosperous parts of the world and
remains our major goal.
The advent of conjugate vaccines against Haemophilus influ-
enzae type b-associated diseases has opened a new era in vac-
cinology (15). These vaccines, which demonstrate excellent
immunogenicity, safety, and efficacy in infants, have been ob-
tained by conjugating the capsular polysaccharide polyribosyl-
ribitolphosphate from H. influenzae type b to carrier proteins
(1, 3, 8, 12). They have been particularly useful in preventing
infection with H. influenzae type b in high-risk infant popula-
tions. In fact, the almost complete disappearance of H. influ-
enzae type b disease in the developed world together with the
corresponding reduction in H. influenzae type b pharyngeal
carriage testify to the usefulness of these conjugate vaccines.
However, it is extremely difficult to accomplish the progress
brought by these commercial vaccines in many poor countries
because their high cost reduces both their acquisition and their
availability. More than 118 million children are without pro-
tection, and only ?2% of cases of H. influenzae type b disease
are actually prevented worldwide (15).
Given this, H. influenzae type b vaccination in developing
countries is urgent but limited by cost and the availability of
vaccines. The availability of vaccines depends on producing
them with improved technologies and making them affordable
to even the poorest societies. In 1989, we embarked on a
project to produce a new conjugate anti-H. influenzae type b
vaccine from a fully synthetic fragment of the capsular poly-
saccharide. We have now successfully completed the produc-
tion, preclinical, and clinical development stages for this new
vaccine (17). Here we describe the main preclinical studies that
have shown the potential of this new vaccine. Several synthetic
H. influenzae type b oligosaccharide-protein conjugates were
prepared and their immunological properties (e.g., antigenicity
and inmunogenicity) were evaluated with laboratory animals.
MATERIALS AND METHODS
HbO-HA antigen. Plates coated with H. influenzae type b conjugated to human
albumin (HA) (HbO-HA antigen) (lot 17) was supplied by the National Institute
for Biological Standards and Control, London, United Kingdom. The H. influ-
enzae type b conjugates were obtained by controlled periodic oxidation to degree
of polymerization of 30 followed by reductive amination to human serum albu-
min.
Licensed vaccines. Polyribosylribitolphosphate conjugated to tetanus toxoid
(PRP-TT; Hiberix, lot Hib 900A11) was from Smith Kline Beecham. The vaccine
is composed of the capsular polysaccharide activated by cyanogen bromide and
coupled to tetanus toxoid through an adipic acid hydrazide spacer. PRP-
CRM197 (cross-reacting mutant 197; Vaxem-Hib Lot 3204) was from Chiron.
The vaccine is composed of oligosaccharide fragments obtained from the cap-
sular polysaccharide by acid hydrolysis (DP 10) and coupled to CRM197 through
adipic acid hydrazide.
Synthetic polyribosylribitolphosphate. The synthetic oligosaccharides that we
used were produced in our facilities on a multigram scale under good manufac-
turing practice conditions. The chemical synthesis process starts from D-ribose
(18). As shown in Fig. 1, the synthetic oligosaccharides (sPRP) contained be-
* Corresponding author. Mailing address: Center for the Study of
Synthetic Antigens, Facultad de Química, Universidad de La Habana,
Havana, Cuba 10400. Phone: 5378702331. Fax: 5378792331. E-mail:
violeta@fq.uh.cu.
7115

Page 2

tween six and nine repeating units of the Haemophilus influenzae type b capsular
polysaccharide. The synthetic compound was made with a spacer arm with a
terminal maleimido function ready for conjugation to protein carriers.
Carrier proteins. Tetanus toxoid (TT; lot P-007?00) and Neisseria meningitidis
outer membrane protein complex (OMP; lot 0006B) were produced as compo-
nents of licensed vaccines by the Finlay Institute for Serum and Vaccines,
Havana, Cuba. Human serum albumin (HSA) and bovine serum albumin (BSA)
were from Sigma.
Native polyribosylribitolphosphate. The capsular polysaccharide polyribosyl-
ribitolphosphate (lot 00-Hib350-01) was supplied by RIVM, Biltjoven, The Neth-
erlands.
Protein conjugates. Several conjugates containing the synthetic antigens de-
scribed below were obtained with a two-step procedure (7).
(i) Step 1. To a solution of human serum albumin (24.5 mg, 0.36 ?mol) in
phosphate-buffered saline (PBS; pH 8, with EDTA at 1.86 g/liter, 5.0 ml), a
solution of N-hydroxysuccinimide dithiopropionate (2.9 mg, 7.2 ?mol) in di-
methyl sulfoxide (50 ?l) was added under N2atmosphere. After 2 h, dithiothre-
itol (19.3 mg, 325 mmol/liter) was added (under N2gas) and the mixture was
stirred at 4°C for 1 h. The resulting solution was diafiltered with N2as the
pressure source (pH 7.2, regenerated cellulose membrane, 30-kDa cutoff). Pro-
tein and SH content were analyzed by the methods of Lowry (11) and Ellman (6),
respectively. A 20 to 25% molar substitution with the oligosaccharide was usually
attained on human serum albumin.
When the Neisseria meningitidis outer membrane protein complex was used,
the procedure was modified at the step of diafiltration. The reaction mixture was
precipitated twice with 80% ethanol (in purified water) followed by centrifuga-
tion (1,500 rpm, 4°C, 10 min).
(ii) Step 2. To a solution of thiolated human serum albumin (20.8 mg) in PBS
(pH 7.2, EDTA 1.86 g/liter, 5 ml) a solution of synthetic polyribosylribitolphos-
phate (16.8 mg) previously dissolved in PBS (pH 7.2, 0.4 ml) was added under an
N2atmosphere.The conditions used with the other proteins may be seen in
Table 2. The resulting solution was gently stirred for several hours at 4 to 8°C.
The reaction was quenched with ethylmaleimide (1 mg, 8 ?mol) and then
diafiltered against PBS (pH 7.2, cellulose acetate membrane, 30-kDa cutoff).
The protein and carbohydrate contents in the final conjugate were determined
by the methods of Lowry et al. (11) and orcinol (4), respectively. The oligosac-
charide-protein conjugates were analyzed by sodium dodecyl sulfate-polyacryl-
amide gel electrophoresis (10).
Immunization methods. Four-week-old female Sprague Dawley rats were
maintained at the animal facilities at the National Institute for Biological Stan-
dards and Control, London, United Kingdom. Groups of five rats were immu-
nized subcutaneously at weeks 0 and 4, with 2 ?g of immunogen (based on
polyribosylribitolphosphate). Serum samples were collected at week 6.
Six- to eight-week-old female BALB/c mice were supplied by the Center for
the Production of Laboratory Animals (Bejucal, Havana, Cuba) and maintained
FIG. 1. Structure of synthetic Haemophilus influenzae type b oligosaccharides (synthetic polyribosylribitolphosphate).
TABLE 1. Conditions and results obtained during the thiolation of
proteins (step 1 in our conjugation process)
Protein
Wt
(mg)
Reaction conditionsa
SH/protein ratio
(?mol/?mol)
Mass
(Da)
Reaction
vol (ml)
DSP
(mg)
DTT
(mg)
HSA
BSA
TT
OMPc
24.5
13.3
18.5
30.0
68,000
67,000
150,000
5.0
2.8
4.0
6.0
2.9
2.4
1.5
5.4
19.3
10.8
15.4
23.1
25
20
32
0.24
aDSP, N-hydroxysuccinide dithiopropionate. DTT, dithiothreitol.
bSH, thiol.
cOMP, for which the SH/protein ratio is expressed in micromoles per milli-
gram.
TABLE 2. Experiments under typical conditions for conjugation of
sPRP to thiolated proteins
Conjugate
Reaction conditions
Thiolated protein/
protein ratio (mg)
Yield based
on sPRP
(%)
Thiolated
protein
(mg)
sPRP
(mg)
sPRP-HSA
sPRP-BSA
sPRP-TT
sPRP-OMP
20.8
13.0
14.0
17.8
16.8
7.0
13.3
5.0
1.0:3.0
1.0:2.9
1.0:2.2
1.0:9.2
42.7
63.0
40.9
31.2
7116FERNA ´NDEZ-SANTANA ET AL.INFECT. IMMUN.

Page 3

gn our animal facilities. Groups of 10 mice were immunized intraperitoneally
with 2.5 ?g of immunogen (based on polyribosylribitolphosphate). Immuniza-
tions were performed at weeks 0, 2, and 4, and serum samples were collected at
week 6.
Four-week-old female New Zealand rabbits were supplied by the Center for
the Production of Laboratory Animals (Bejucal, Havana, Cuba) and maintained
in our animal facilities. Groups of three rabbits were immunized subcutaneously
with 5 ?g of immunogen (based on polyribosylribitolphosphate). The immuni-
zation was performed with either two doses at 0 and 2 weeks or three doses at 0,
2, and 4 weeks. Serum was collected at 0 and 3 weeks or 0, 1, 3, and 5 weeks,
respectively.
ELISA. Antibody titers against polyribosylribitolphosphate were determined
by enzyme-linked immunosorbent assay (ELISA). The ELISA method used is
one based on that described by Phipps et al. (16) adapted for the evaluation of
animal serum. Plates were coated with one of the following conjugates for
antigenicity studies: HbO-HA, sPRP-HSA, or sPRP-BSA. For immunogenicity
studies only HbO-HA was used. The albumin conjugates were dissolved at 1
?g/ml (based on polyribosylribitolphosphate) in PBS and incubated overnight at
37°C. After that, plates were washed four times with 0.05% Tween 20 aqueous
solution. Plates were blocked with 1% bovine serum albumin in PBS and incu-
bated for 30 min at 37°C. The wells were washed four times and incubated with
a serial twofold dilution of serum sample diluted in PBS solution containing 0.3%
Tween 20, 10 mM EDTA, and 1% bovine serum albumin for 90 min at room
temperature. The wells were then rinsed four times, and the corresponding
anti-immunoglobulin G whole molecule horseradish peroxidase conjugate was
added to each well. Plates were washed again after 90 min of incubation at room
temperature, and the substrate solution was added: O-phenylendiamine, H2O2in
citrate buffer, pH 5. After 20 min in darkness, the reaction was stopped with 3 M
HCl and read at 492 nm with an ELISA Sunrise reader.
Antibody titers were defined as the log of the highest dilution giving twice the
absorbance value calculated against that of sera from control animals (immu-
nized with buffer), with a minimum value of 0.2.
Inhibition studies were performed with the same procedure with the following
modification. Serum dilution was selected in order to obtain an optical density of
1 and was incubated separately with a serial dilution of inhibitor. The procedure
was continued as above.
FIG. 2. Coomassie brilliant blue-stained sodium dodecyl sulfate-polyacrylamide gel of conjugates. A, B, and C, gels at 10% polyacrylamide; D,
gel at 7.5% polyacrylamide. Lanes: 1, BSA; 2, sPRP-BSA; 3, human serum albumin; 4, sPRP-HSA; 5, OMP; 6, sPRP-OMP; 7, TT; and 8, sPRP-TT.
TABLE 3. Comparative recognition of synthetic and natural PRP oligosaccharide conjugatesa
Rabbit no.VaccinePolysaccharidesPRP-HSA HbO-HA sPRP-BSA
1
2
3
4
5
6
7
8
PRP-CRM197
PRP-CRM197
PRP-CRM197
PRP-TT
PRP-TT
PRP-TT
PRP-TT
PRP-TT
Oligosaccharide
Oligosaccharide
Oligosacchide
Polysaccharide
Polysaccharide
Polysaccharide
Polysaccharide
Polysaccharide
1.98
3.51
2.93
2.63
1.54
3.25
2.08
2.72
2.14
3.66
3.00
2.73
1.13
3.10
1.73
2.60
11
12
13
14
15
16
17
18
PRP-CRM197
PRP-CRM197
PRP-CRM197
PRP-TT
PRP-TT
PRP-TT
PRP-TT
PRP-TT
Oligosacchide
Oligosacchide
Oligosacchide
Polysaccharide
Polysaccharide
Polysaccharide
Polysaccharide
Polysaccharide
3.76
3.23
2.91
2.62
2.27
2.72
3.29
2.76
3.73
3.23
2.94
2.46
2.42
2.83
3.37
2.84
aRabbits were immunized with PRP-TT or PRP-CRM197. For rabbits 1 to 8, sPRP- and PRP-HSA conjugates were used as coating reagent; for rabbits 11 to 18,
the coating antigen used was sPRP with two different protein carriers, HSA and BSA. Values are for individual sera.
VOL. 72, 2004OLIGOSACCHARIDE-PROTEIN CONJUGATES AGAINST H. INFLUENZAE7117

Page 4

Measurement of relative avidity by ELISA. The measurement of relative
avidity by ELISA was followed as described above, with slight modifications. The
serum dilution was chosen to obtain an optical density of 1. After the plates were
washed following the incubation of the serial serum dilution, ammonium thio-
cyanate in PBS was added to the appropriate wells in concentrations ranging
from 0.1 to 1 M. The plates were then allowed to stand for 15 min at room
temperature before washing and continuing the assay as described above. The
avidity index was calculated by the formula avidity index ? C ? f, where C is the
concentration of ammonium thiocyanate destabilizing 50% of antigen-antibody
interaction and f is the dilution factor.
Serum bactericidal assay. The functional activity of the antibodies obtained
was measured by a serum bactericidal assay (14). Serum samples, heat inacti-
vated at 56°C for 30 min, were tested in twofold dilutions in cold Hanks’ balanced
salt solution with 0.15 mM CaCl2and 0.5 mM MgCl2. The reaction mixtures
contained 12.5 ?l of serum sample, 12.5 ?l of human serum lacking bactericidal
activity as a complement source, and 12.5 ?l of an 8 ? 103/ml log-phase H.
influenzae type b cell suspension (strain Eagan). The final volume was completed
to 60 ?l with Hanks’ balanced salt solution. All reactions were performed in
duplicate, and the mixtures were incubated for 30 min at 37°C. After that, the
reactions were plated onto brain heart infusion plates supplemented with NAD
and hemin. Colonies were counted the next day, and bactericidal titer was
determined as the reciprocal of the highest dilution of serum capable of killing
more than 50% of the initial bacterial inoculum. For measurement of the specific
inhibition of bactericidal activity, 100 ?l of serum was incubated with 100 ?l of
native polyribosylribitolphosphate (0.05 mg of polyribosylribitolphosphate/ml,
final concentration) at 37°C for 1 h before the serum bactericidal assay was
performed.
RESULTS
Conjugation of synthetic oligosaccharides to protein. Syn-
thetic oligosaccharides are small molecules differing from the
capsular polysaccharide by their low absorption on ELISA
plates and by the presence (usually) of a single epitope. To
better understand their antigenic and immunogenic properties,
they were conjugated to several proteins. The first step in this
conjugation process is the thiolation of the carrier protein,
which was performed initially with N-hydroxysuccinimide di-
thiopropionate followed by reduction with dithiothreitol. The
procedure is a general one and can be applied to different
proteins. The results of derivatization are shown in Table 1.
The maleimido group from the spacer in the synthetic
polyribosylribitolphosphate has a strong thiophilic character
that facilitated conjugation to proteins. The conjugation reac-
tion proceeds smoothly, giving yields ranging from 30 to 60%
based on synthetic polyribosylribitolphosphate. The results are
summarized in Table 2. All conjugates were characterized by
polyacrylamide gel electrophoresis (Fig. 2). These show an
increase in the molecular mass as well as a broadening in the
band typical for this random introduction of several carbohy-
drate chains.
Antigenicity of synthetic polyribosylribitolphosphate conju-
gates. To determine if synthetic oligosaccharides mimic the
capsular polysaccharide epitopes and are recognized to the
same extent, synthetic polyribosylribitolphosphate-albumin
conjugates were prepared and used as the coating reagent for
ELISA plates. In the first experiment, the ability of sPRP-
HSA, prepared from synthetic antigen, and HbO-HA (Na-
tional Institute for Biological Standard and Control, United
Kingdom) to bind anti-H. influenzae type b antibodies was
compared. The serum of rabbits immunized with two types of
vaccines were used: PRP-TT, having a native capsular polysac-
charide (full size) conjugated to tetanus toxoid, and PRP-
CRM197, having oligosaccharides obtained by partial hydroly-
sis of the capsular polysaccharide conjugated to CRM197
protein.
Table 3 shows comparative titers for the two coating anti-
gens calculated for the same serum. A general correlation of
0.972 was obtained between synthetic and native oligosaccha-
rides conjugated to the same carrier (Table 3). A similar value,
0.978, was obtained for the synthetic antigen conjugated with
two different carriers (Table 3). Furthermore, the reaction
between the anti-H. influenzae type b antibodies (pools from
rabbit serum PRP-CRM197 and PRP-TT) and sPRP-HSA or
sPRP-BSA was inhibited by the native capsular polysaccharide,
thereby demonstrating the reaction is specific for polyribosyl-
ribitolphosphate (Fig. 3).
Immunogenicity of synthetic oligosaccharide-protein conju-
gates. To study the ability of synthetic oligosaccharide to in-
duce an immune response useful in vaccine development, we
selected two different bacterial proteins successfully used as
FIG. 3. Inhibition of the ELISA shown in Table 3 with the capsular H. influenzae type b polysaccharide. The serum samples used are a pool
of sera from rabbits immunized with PRP-CRM197 and PRP-TT.
7118FERNA ´NDEZ-SANTANA ET AL.INFECT. IMMUN.

Page 5

carriers in conjugate vaccines, Neisseria meningitidis outer
membrane protein and tetanus toxoid. Our conjugates were
studied initially in BALB/c mice and Sprague Dawley rats. Two
sPRP-OMP conjugates with different carbohydrate/protein ra-
tios (1:4 and 1:12) were used in this preliminary study. The
response to both conjugates was strong (Fig. 4) and compara-
ble to the commercially available PRP-CRM197 vaccine. The
response to sPRP-TT in a wide carbohydrate/protein ratio (1:2
to 1:12) was weak and inconsistent (not shown). sPRP-OMP
with a ratio of 1:12 was selected for further studies in rats. This
serum was completely inhibited by the native capsular polysac-
charide (Fig. 5), showing that the synthetic polyribosylribitol-
phosphate conjugate with a suitable carrier was also immuno-
genic.
In other experiments we sought to find a suitable animal
model for developing both types of conjugates as vaccine
candidates and for comparing their immunogenic proper-
ties. New Zealand rabbits consistently responded to both
sPRP-TT (1:2) and sPRP-OMP (1:9), in the presence or
even in the absence of aluminum hydroxide (Fig. 6). Despite
the strong differences reported for such conjugates in the
literature (5), the immune response was very similar. In
another experiment, a group of rabbits were immunized
with both types of conjugates containing the synthetic
polyribosylribitolphosphate. No major differences were
found in the kinetics or in the responses compared to con-
jugate containing the capsular antigen PRP-TT (Fig. 7).
The rabbit sera were further selected for a more detailed
study including relative avidity for the second and third doses
(Table 4). The behavior of sPRP-TT resembled that of
FIG. 4. Comparative immunogenicity in Sprague Dawley rats and BALB/c mice of synthetic polyribosylribitolphosphate conjugated to Neisseria
meningitidis outer membrane protein and a commercial vaccine, PRP-CRM197.
FIG. 5. Inhibition of anti-polyribosylribitolphosphate serum elicited in rats by sPRP-OMP (1:12). HbO-HA was used as the coating antigen.
Sera 1, 2, and 3 are from individual rats. The capsular polysaccharide was used at increasing concentrations.
VOL. 72, 2004 OLIGOSACCHARIDE-PROTEIN CONJUGATES AGAINST H. INFLUENZAE7119

Page 6

PRP-TT with a small increase in avidity, sPRP-OMP inducing
a stronger increase in avidity after the third dose.
The specificity of the antibodies in a serum pool was studied
by determining their inhibition with four different compounds:
capsular polyribosylribitolphosphate, sPRP-BSA, synthetic
polyribosylribitolphosphate, and unrelated capsular polysac-
charide from Neisseria meningitidis group A (MenA). For sim-
plicity, results are shown in two different panels (Fig 8A and
B). Inhibition was independent of the source of polyribosyl-
ribitolphosphate in the vaccine or in the inhibitor. All three
sera display similar results. The best inhibitors were a capsular
polysaccharide and sPRP-BSA. Synthetic polyribosylribitol-
phosphate required a higher concentration for the same inhi-
bition, as expected because of its monovalent nature. The low
inhibition even at a high concentration of MenA capsular poly-
saccharide demonstrated once again the specificity of the in-
teraction.
The bactericidal power of rabbit serum in the presence of
complement was also studied. As can be seen from Fig. 9, the
sera of rabbits immunized with sPRP-OMP, sPRP-TT, or con-
trol PRP-CRM197 display strong bactericidal activity inhibited
by the capsular polysaccharide. Rabbit serum displayed a non-
specific residual bactericidal activity against H. influenzae type
b that was not inhibited by the capsular polysaccharide. The
results displayed in Fig. 9 showed the overall and the nonspe-
cific bactericidal activity after inhibition with the capsular poly-
saccharide. The difference corresponds to specific bactericidal
activity.
FIG. 6. Titers of anti-H. influenzae type b serum elicited in New Zealand rabbits (pooled serum) by sPRP-TT and sPRP-OMP alone or with
aluminum hydroxide. PRP-CRM197 was used as a control.
FIG. 7. Kinetics of anti-H. influenzae type b antibody formation in New Zealand rabbits (pooled sera) by sPRP-OMP and sPRP-TT. PRP-TT
was used as a positive control.
7120 FERNA ´NDEZ-SANTANA ET AL.INFECT. IMMUN.

Page 7

DISCUSSION
Capsular polysaccharides and their oligosaccharide fragments
have been used for the preparation of conjugate vaccines against
H. influenzae type b (8, 12). In parallel, several attempts to repro-
duce the fragment of the capsular polysaccharide by chemical
synthesis have been made successfully (2, 9, 13). The synthetic
fragments are then conjugated to protein carriers and tested in
laboratory animals for their potential as vaccine candidates (2,
14). None of these synthetic antigens were previously tested in
humans. This was probably because the efficiency of the synthe-
tic process is too low for competing with conventional, well-
established antigen production processes from native sources.
We therefore developed a new improved procedure for the
synthesis of H. influenzae type b oligosaccharide fragment (17).
This synthetic antigen could be produced and conjugated to
carrier proteins as efficiently as happens with the use of the
capsular polysaccharide (Tables 1 and 2).
FIG. 8. (a) Comparative inhibition of anti-H. influenzae type b serum elicited in rabbits against sPRP-OMP, sPRP-TT, and control vaccine
PRP-TT. HbO-HA was used in all cases as the coating reagent. The inhibitors used were capsular polysaccharide (polyribosylribitolphosphate) and
synthetic polyribosylribitolphosphate (sPRP). (b) Comparative inhibition of anti-H. influenzae type b serum elicited in rabbits against sPRP-OMP,
sPRP-TT, and control vaccine PRP-TT. HbO-HA was used in all cases as the coating reagent. The inhibitors used were sPRP-BSA conjugate and
nonrelated MenA capsular polysaccharide.
TABLE 4. Average avidity index of antibodies after ammonium
thiocyanate dissociation
Inmunogen
Avidity index (log)
Dose 2Dose 3
sPRP-OMP
sPRP-TT
PRP-TT
1.53 ? 0.09
1.76 ? 0.45
1.65 ? 0.05
2.48 ? 0.40
2.01 ? 0.36
1.82 ? 0.14
VOL. 72, 2004 OLIGOSACCHARIDE-PROTEIN CONJUGATES AGAINST H. INFLUENZAE7121

Page 8

The first experiments were designed to determine if the
synthetic polyribosylribitolphosphate displayed the same epi-
topes as the fragment of capsular polysaccharides. The results
obtained in those experiments showed an excellent correlation
between the synthetic and capsular oligosaccharides, demon-
strating that the epitopes exposed in synthetic polyribosylribi-
tolphosphate are very similar to those of fragments from native
polysaccharide. Furthermore, the reaction was inhibited by a
very low polysaccharide concentration, 3 ? 10?3to 1.6 ? 10?4
?g/ml for 50% inhibition (Fig. 3). This confirmed that the
synthetic polyribosylribitolphosphate reproduces the epitopes
recognized by anti-H. influenzae type b antibodies.
Among the proteins that are currently in use as carriers
for anti-H. influenzae type b vaccine are Neisseria meningi-
tidis outer membrane protein, tetanus toxoid, diphtheria
toxoid, and CRM197. We selected the first two for their
ready availability and also for the sharp differences found in
clinical trials with these two conjugates. The conjugation
proceeded in both cases with similar good yields, and al-
though tetanus toxoid is more readily available, we were
interested in obtaining a general immunological profile for
both conjugates.
The first and the most important response differences were
found in rodents. Usually the response to sPRP-OMP was
stronger in titer and also in number of animals responding.
Our experience in using sPRP-OMP as an immunogen in mice
indicates that animals rarely failed to respond. This may be
associated with the ability of outer membrane proteins to in-
duce a Th1-type response, especially in BALB/c mice. The
same was not true for sPRP-TT; many animals failed to re-
spond, and the titers were usually very weak. This initial ex-
periment with rodents demonstrated that the synthetic antigen
conjugated to a suitable carrier was able to induce anti-H.
influenzae type b antibodies.
In a search for a more suitable animal model independent of
the protein carrier, we found that New Zealand rabbits re-
spond equally well to both types of conjugates.
The anti-H. influenzae type b immune response in rabbits
was very good with or without aluminum hydroxide, regardless
of the protein carrier used (Fig. 6). The antibody titers always
increased after the second dose, and an additional increment
occurred after the third dose. Both conjugates were able to
induce a strong anti-H. influenzae type b response with a small
increase in the avidity index for both vaccines and control. This
is associated with the vaccine’s ability to induce maturation in
the anti-H. influenzae type b antibody response. The low cap-
sular polysaccharide concentration needed for 50% inhibition
indicated that a specific reaction was induced by synthetic
polyribosylribitolphosphate conjugates. On the other hand the
very low sPRP-BSA concentration, similar to capsular polysac-
charide needed for 50% inhibition of the reaction between
anti-PRP-TT serum and HbO-HA, indicated that synthetic
antigen as a BSA conjugate with multiple copies interacts in a
similar fashion to the capsular polysaccharide. The nonrelated
MenA capsular polysaccharide showed only a low, nonspecific
inhibition even at high concentrations. Finally, in both cases,
sPRP-TT and sPRP-OMP, the antibodies displayed bacteri-
cidal activity similar to that induced by the commercially avail-
able anti-H. influenzae type b vaccine, demonstrating that de-
spite the synthetic origin, the antigen possess all the relevant
properties of their native counterparts.
In conclusion, our data show that these synthetic polyribosyl-
ribitolphosphate fragments display both antigenicity and immu-
nogenicity patterns in laboratory animals similar to those of their
native analogues currently used in other commercial vaccine
preparations. This finding supports further pharmaceutical devel-
opment and clinical evaluations of our conjugates as alternatives
to conventional vaccines against H. influenzae type b.
ACKNOWLEDGMENTS
This work was supported by grants from the Cuban State Council
and the Ministry of Science, Technology and Environment and also
from the Pan-American Health Organization (PAHO).
We gratefully recognize J. L. DiFabio (PAHO) for many helpful
discussions and support, M. Beurret (RIVM, Biltjoven, The Nether-
lands) for providing the native polyribosylribitolphosphate, and J. B.
Robbins (NIH) for providing the H. influenzae type b Eagan strain.
C. H. Fox is gratefully acknowledged for critical reading of the manu-
script.
FIG. 9. Overall and residual nonspecific bactericidal activity of serum against the Eagan strain of Haemophilus influenzae type b expressed as
the last serum dilution killing ?50% of bacteria. The differences correspond to specific anti-H. influenzae type b bactericidal activity.
7122FERNA ´NDEZ-SANTANA ET AL.INFECT. IMMUN.

Page 9

REFERENCES
1. Anderson, P. 1983. Antibody responses to Haemophilus influenzae type b and
diphtheria toxin induced by conjugates of oligosaccharides of the type b
capsule with the nontoxic protein CRM197. Infect. Immun. 39:233–238.
2. Chong, P., N. Chan, A. Kandil, B. Triplet, O. James, Y. P. Yang, S. P. Shi,
and M. Klein. 1997. A strategy for rational design of fully synthetic glyco-
peptide conjugate vaccines. Infect. Immun. 65:4918–4925.
3. Chu, C., R. Schneerson, J. B. Robins, and S. C. Rastorgi. 1983. Further
studies on the immunogenicity of Haemophilus influenzae type b and pneu-
mococcal type 6A polysaccharide-protein conjugates. Infect. Immun. 40:
245–246.
4. Dische, Z. 1962. Color reactions of pentoses, p. 484–488. In R. L. Whistler,
M. L. Wolfrom, J. N. BeMiller, and F. Shafizadeh (ed.), Methods in carbo-
hydrate chemistry, vol. 1. Academic Press, New York, N.Y.
5. Donelly, J. J., R. Randall Deck, and M. A. Liu. 1990. Immunogenicity of a
Haemophilus influenzae polysaccharide-Neisseria meningitidis outer mem-
brane protein complex conjugate vaccine. Immunology 145:3071.
6. Ellman G. L. 1958. A colorimetric method for determining low concentra-
tions of mercaptans. Arch. Biochem. Biophys. 74:443.
7. Fernandez-Santana, V., R. Gonzalez-Lio, J. Sarracent-Perez, and V. Verez
Bencomo. 1998. Synthesis of inmunogens containing lewisaand lewisbhap-
tens by the use of glycosides of 5-azido-3-oxapentanol. Glycoconj. J. 306:
163–170.
8. Gordon, L. K. 1984. Characterization of a hapten-carrier conjugate vaccine:
H. influenzae diphtheria conjugate vaccine, p. 393–396. In R. M. Chanock
and R. A. Lerner (ed.), Modern approaches to vaccines. Cold Spring Harbor
Laboratory Press, Cold Spring Harbor, N.Y.
9. Hoogerhout P., D. Evenberg, C. A. A. van Boeckel, J. T. Poolman, E. C.
Beuvery, C. A. van der Marel, and J. H. van Boom. 1987. Synthesis of
fragments of the capsular polysaccharide of Haemophilus influenzae type b,
comprising two or three repeating units. Tetrahedron Lett. 28:1553.
10. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of
the head of bacteriophage T4. Nature 227:680-685.
11. Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein
measurement with the Folin phenol reagent. J. Biol. Chem. 193:265–275.
12. Marburg, S. D. Jorn, R. L. Tolman, B. Arison, J. McCauley, P. J. Kniskern,
A. Hagopian, and P. P. Vella. 1986. Bimolecular chemistry of macromole-
cules-synthesis of bacterial polysaccharide conjugates with Neisseria menin-
gitidis membrane protein. J. Am. Chem. Soc. 108:5282–5297.
13. Nilsson, S., M. T. Bengtsson, and O. Norberg. 1992. Solid-phase synthesis of
a fragment of the capsular polysaccharide of Haemophilus influenzae using
H-phosphonate intermediates. J. Carbohydr. Chem. 11:265–285.
14. Peeters, C. C. A. M., D. Evenberg, P. Hoogerhout, H. Kayhty, L. Saarinen,
C. A. A. van Boeckel, G. A. van der Marel, J. H. van Boom, and J. T.
Poolman. 1992. Synthetic trimer and tetramer of 3-?-D-ribose-(1–1)-D-ribi-
tol-5-phosphate conjugated to protein induce antibody responses to Hae-
mophilus influenzae type b capsular polysaccharide in mice and monkeys.
Infect. Immun. 60:1826–1833.
15. Peltola, H. 2000. Worldwide Haemophilus influenzae type b disease at the
beginning of the 21st century: global analysis of the disease burden 25 years
after the use of the polysaccharide vaccine and a decade after the advent of
conjugates. Clin. Microbiol. Rev. 13:302–317.
16. Phipps, D. C., J. West, R. Eby, M. Koster, D. V. Madore, and S. A. Quataert.
1990. An ELISA employing a Haemophilus influenzae type b oligosaccharide-
human serum albumin conjugate correlates with the radioantigen binding
assay. J. Immunol. Methods 135:121–128.
17. Verez-Bencomo, V., V. Ferna ´ndez-Santana, E. Hardy, M. E. Toledo, M. C.
Rodríguez, L. Heynngnezz, A. Rodriguez, A. Baly, L. Herrera, M. Izquierdo,
A. Villar, Y. Valde ´s, K. Cosme, M. L. Deler, M. Montane, E. Garcia, A.
Ramos, A. Aguilar, E. Medina, G. Toran ˜o, I. Sosa, I. Hernandez, R. Mar-
tínez, A. Muzachio, A. Carmenates, L. Costa, F. Cardoso, C. Campa, M.
Diaz, and R. Roy. 2004. A synthetic capsular polysaccharide vaccine against
Haemophilus influenzae type b polysaccharide. Science 305:522–524.
18. Verez-Bencomo, V., and R. Roy. August 1999. Oligosaccharides derived from
ribose-ribitol-phosphate, and vaccines containing them. Patent Community
Treaty WO 01/16146. U.S. patent 6,765,091.
Editor: D. L. Burns
VOL. 72, 2004OLIGOSACCHARIDE-PROTEIN CONJUGATES AGAINST H. INFLUENZAE7123