VI Polysaccharide

NIH Public Access
Author Manuscript
Vaccine. Author manuscript; available in PMC 2014 September 15.
Published in final edited form as:
NIH-PA Author Manuscript
Vaccine. 2011 January 17; 29(4): 712–720. doi:10.1016/j.vaccine.2010.11.022.
Vi-CRM197 as a new conjugate vaccine against Salmonella Typhi

F. Micolia,*, S. Rondinia, I. Pisonia, D. Proiettib, F. Bertib, P. Costantinob, R. Rappuolib, S.
Szuc, A. Saula, and L.B. Martina
aNovartis Vaccines Institute for Global Health, Via Fiorentina 1, 53100 Siena, Italy
bNovartis Vaccines and Diagnostics Research Center, via Fiorentina 1, 53100 Siena, Italy
cNational
Institute of Child Health and Human Development, National Institutes of Health,
Bethesda, MD 20892, USA
Abstract
An efficacious, low cost vaccine against typhoid fever, especially for young children, would make
a major impact on disease burden in developing countries. The virulence capsular polysaccharide
of Salmonella Typhi (Vi) coupled to recombinant mutant Pseudomonas aeruginosa exoprotein A
(Vi-rEPA) has been shown to be highly efficacious. We investigated the use of carrier proteins
included in infant vaccines, standardized the conjugation process and developed key assays
required for routine lot release at production scale. Vi from a BSL1 organism, Citrobacter
freundii, strain WR7011, was used as an alternative to Vi from S. Typhi. We showed that Vi
conjugated to CRM197, a non-toxic mutant of diphtheria toxin, widely used in commercial
vaccines, was produced at high yield. Vi-CRM197 proved immunogenic in animal studies, even
without adjuvant. Thus, Vi-CRM197 appears to be a suitable candidate for the development of a
commercially viable, effective typhoid vaccine for developing countries.
Keywords
Vi; Conjugate vaccine; Salmonella Typhi
1. Introduction
Typhoid fever is a common serious disease in many parts of the world and remains a major
health problem in developing countries with an estimated 21 million cases and 200,000
deaths in 2000 [1,2]. The highest incidence is in South Asia (India, Pakistan and
Bangladesh) but cases occur throughout Asia, Africa and the Americas. Travellers visiting
these areas are at risk of contracting typhoid fever and approximately 400 cases are recorded
annually in the USA and in the UK, mostly with a travel history to India [3,4]. In highly
endemic areas, children are at particular risk with the peak age inversely proportional to the
incidence in the community [5,6]. Although commonly quoted as a disease of school age
children [7,8], one study from Bangladesh showed that the most common age of infection in
© 2010 Elsevier Ltd. All rights reserved.

*
Corresponding author. Tel.: +39 0577 539087; fax: +39 0577 243540. [email protected] (F. Micoli).
Micoli et al. Page 2
hospitalised children was 1–2 years [5]. As Salmonella enterica serovar Typhi only infects
humans, vaccines targeting young children would give protection and also reduce
transmission of typhoid fever in nonvaccinated members of the community, as was seen in a
recent vaccine trial in Kolkuta, India [9].
The capsular polysaccharide of Salmonella Typhi (Vi) is a linear homopolymer of α1,4-N-

acetylgalactosaminouronic acid, 60–90% O-acetylated at the C-3 position [10].
Unconjugated Vi polysaccharide is one of the two widely available licensed vaccines
together with an oral live attenuated vaccine (Ty21a). The Ty21a vaccine is distributed as
enteric coated capsules, licensed only for people 6 years and older [11]. Several
manufacturers produce unconjugated Vi vaccine, licensed for adults and children 2 years
and older [12]. There is no typhoid vaccine that is licensed for use in infants.
A recent meta-analysis of both Ty21a (oral) and Vi polysaccharide (parenteral) vaccines

estimated that the cumulative efficacy is 51% (95% CI 36–62%) for Ty21a and 55% (95%
CI 30–70%) for Vi [13,14]. The duration of protection is not well determined, with estimates
of five to seven years for the Ty21a vaccine and three years for Vi vaccination [13,14].
Despite these limitations, several studies have illustrated the importance of vaccination
against typhoid fever for populations at risk [11]. The World Health Organization and GAVI
have recommended, but not yet funded, introduction of the existing Vi vaccine, and support
the development of more effective vaccines [15]. A vaccine that could be administered to
infants would be highly beneficial.
As observed with other polysaccharides [16,17], conjugation of Vi to a carrier protein

substantially increases the antibody response. A conjugated vaccine of Vi coupled to
recombinant mutant of Pseudomonas aeruginosa exoprotein A (Vi-rEPA) was shown to be
safe in all ages including infants [18]. Vaccination provided excellent and long lasting
immunity with 92% protection over two years post vaccination in a randomized, two-dose
placebo controlled trial in 2–5-year-old children in Vietnam, and with 89% protection over
46 months [19,20].
Diphtheria toxoid (DT), tetanus toxoid (TT), and CRM197 (a non-toxic variant of diphtheria
toxin ), a outer-membrane protein complex (OMPC) from Neisseria meningitides and
Haemophilus influenzae outer membrane protein D (OMPD) are used as protein carriers in
licensed glycoconjugate vaccines [17,21]. This study reports use of CRM197 for the
preparation of a Vi conjugate, its characterization and immunogenicity in mice as part of a
program to develop a consistent and affordable conjugate vaccine for use in all ages in
developing countries. Unlike DT or TT, CRM197 does not require detoxification with
formaldehyde and homogeneous preparations of purified antigen can be readily obtained.
CRM197 is a precisely defined protein, consistent from batch to batch. Unlike rEPA,
CRM197 is licensed for human use in several efficacious conjugate vaccines already used in
hundreds of millions of children [22–24]. Its use as carrier protein should facilitate the
manufacturability of a Vi conjugate vaccine and simplify its pathway to licensure.

In order to study the influence of a different carrier protein on the efficacy of a Vi conjugate
vaccine, the same chemistry has been used for the synthesis of Vi-CRM197 and Vi-TT
vaccines and their immunogenicity compared in mice.
Citrobacter freundii WR7011 has been chosen as source of Vi instead of Salmonella

enterica serotype Typhi (Ty-2). Vi from Citrobacter is structurally similar and
immunologically indistinguishable to Vi from S. Typhi [25,26]. Vi from Citrobacter
freundii WR7011 has been successfully used as the Vi source in studies assessing
immunogenicity of Vi-vaccine conjugates [10,27,28]. As a low risk organism and a high Vi
yield strain, C. freundii constitutes a safer and more economic source for Vi production than
BSL3 S. Typhi.
2. Materials and methods

Polysaccharide
Vi polysaccharide from C. freundii WR7011 was from the Program in Developmental and
Molecular Immunity, the National Institute of Child Health and Human Development,
National Institutes of Health. Characterization of the polysaccharide was done at Novartis
Vaccines Institute for Global Health (NVGH) by A260 for nucleic acid content, micro BCA
for protein estimation, 1H NMR for Vi identity and O-acetylation level. O-acetyl groups
were also estimated by the Hestrin method [29]. Thermogravimetric analysis and Karl Fisher
were used for dried weight and residual moisture measure, respectively. Na+ content was
evaluated by atomic absorption spectroscopy [30].
Proteins
CRM197 and tetanus toxoid were from Novartis Vaccines and Diagnostics (NV&D). Tetanus
toxoid was further purified by gel filtration through Sephacryl S-300 (GE Healthcare)
equilibrated in 0.15 M NaCl, 10 mM NaH2PO4, pH 7.2. The fractions, corresponding to the
monomeric molecular weight of tetanus toxoid, as verified by MALDI-TOF (average
molecular mass of 155.3 kDa), were pooled.
Reagents and materials

The following materials were used in this study: adipic acid dihydrazide (ADH),N-(3-
dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride (EDAC), formic acid, 4-

morpholine ethanesulfonic acid (MES) and trifluoroacetic acid (TFA) [Sigma]; sodium
chloride (NaCl) [Merck]; sodium hydroxide (NaOH, 50% solution) [J. T. Baker]; sodium
phosphate monobasic monohydrate (NaH2PO4-H2O) [Carlo Erba]; acetonitrile [LC–MS
Chromasolv]; sodium nitrate (NaNO3) [Fluka]; Spectra/Por dialysis membrane MWCO
6000–8000 Da, 10 mm diameter [Invitrogen]; and Sephacryl S-1000 [GE Healthcare].
2.1. Analytical methods

Protein analysis—Protein concentration was measured by micro BCA, using bovine
serum albumin (BSA) as a reference following the manufacturer’s instructions [Thermo
Scientific].

Derivatized proteins and conjugates were examined by sodium dodecyl sulfate-

polyacrylamide gel electrophoresis (SDS-PAGE) using 7% Tris-acetate gels (NuPAGE,
from Invitrogen). The samples (5–20 µL with a protein content of 5 µg) were added to 0.5 M
dithiothreitol (1/5, v/v) and NuPAGE LDS sample buffer (1/5, v/v). The mixtures were
heated at 100 °C for 1 min and the samples loaded into the wells of the gel. The gel was
subjected to electrophoresis at 45 mA in NuPAGE Tris-acetate SDS running buffer (20×,
Invitrogen) and stained with Simply Blue Safe Stain (Invitrogen).
Nucleic acid analysis—Nucleic acid content was determined by UV spectroscopy at a

wavelength of 260 nm. Nucleic acid concentration of 50 µg/mL produces an OD260 = 1 .
Thermogravimetric analysis—Vi sample was analysed in duplicate for volatile content

estimation using a Perkin-Elmer TGA-6. Keeping the sample in inert atmosphere,
temperature was increased from 25 °C to 150 °C at 5 °C/min and then left at 150 °C for 15
min.
Karl Fisher analysis—Residual moisture was determined by the Karl Fisher coulometric
method with Metrohm 756.
Mass spectrometry—Protein samples were analysed by MS Q-TOF and MALDI-TOF.

For Q-TOF analysis, sample of protein at 2 mg/mL was ultrafiltered using a Microcon 3
(Millipore) for 15 min at 10 °C and 10,000 × g. The treatment was repeated three times,
using 5 mM ammonium acetate buffer pH 6.8 to reconstitute the retentate. The sample was
then diluted 1:4 with 50% acetonitrile/H2O containing 0.1% formic acid and injected by
direct infusion in a Q-TOF micro (Micromass) at a flow rate of 10 µL/min. The mass
spectrometer was tuned as follows: ESI positive mode, capillary 3000 V, sample cone 30 V,
desolvation temperature 250 °C, source temperature 100 °C, scan range from 500 to 2500
Da.
For MALDI-TOF analysis, the protein was diafiltrated, using a Centricon-10 (Millipore),
against 10% acetonitrile containing 0.1% TFA. Two microliters of protein (about 80
pmoles) were mixed with 2 µL of a saturated solution of sinapinic acid in 30% acetonitrile
solution containing 0.1% TFA. One microliter of the mix was spotted on a MTP 384
stainless steel target (Bruker Daltonics GmbH, Bremen, Germany) and allowed to air-dry.
Measurements were recorded on an Ultraflex (Bruker Daltonics) MALDI-TOF/TOF MS in

linear mode. External calibration was performed by spotting 0.5 µL of protein calibration
standard II (Bruker Daltonics) containing the following proteins: trypsinogen (23,982 Da),
protein A (44,613 Da) and bovine serum albumin (66,431 Da). Data were collected using an
Accelerating Voltage of 25 kV, Ion Source1 20 kV, Ion Source2 17 kV and Lens 9.60 kV.
The laser power was adjusted between 30 and 50% intensity, using a 337 nm nitrogen laser
with a frequency of 50 Hz. All mass spectra were recorded by summing up to 400 laser
shots. The Flex Analysis version 2.4 software packages provided by the manufacturer were
used for data processing.
1H NMR spectroscopy—Nuclear Magnetic Resonance (NMR) experiments were

recorded at 25 °C on Bruker DRX 500 MHz spectrometer, equipped with a Bruker 5 mm

dual 1H/13C Z gradient probe. For data acquisition and processing XWINNMR 2.6 software
package (Bruker Daltonics) was used. 1D proton NMR spectra were collected using a
standard one-pulse experiment. Chemical shifts were referenced to HDO at 4.79 ppm. 32k
points were collected on 4000 Hz spectral window. 1H NMR spectra were obtained in
quantitative matter using a total recycle time to ensure a full recovery of each signal (5 ×
Longitudinal Relaxation Time T1). FID (free induction decay) were Fourier transformed
applying 0.1 Hz of line-broadening.
Dried Vi sample was solubilized in deuterated water (Aldrich) and transferred to 5 mm

NMR tube (Wilmad, 535-PP-7). Two spectra were collected: the first one in deuterated
water and the second one after addition of sodium deuteroxide to a final concentration of
200 mM. The first spectrum was recorded in order to assure that other impurities did not fall
at the same chemical shift of the acetate anion (released after de-O-acetylation of the
sample), with implications for the quantification of O-acetyl content.
Addition of sodium deuteroxide to the sample caused rapid de-O-acetylation (the sample
was heated at 37 °C for 20 min to have complete de-O-acetylation) and consequently
sharper peaks appeared in the proton spectrum. The spectrum of the de-O-acetylated
polysaccharide in alkaline medium was better suited for use as an identification test than that
of the untreated sample, based on the presence of five typical signals (from about 5 to about
4 ppm) corresponding to the protons of the carbohydrate ring. Well resolved N-acetyl and
acetate resonances were observed so that the degree of O-acetylation of Vi was determined
by comparison of the integrals of these two peaks [31].
HPLC-SEC—Analytical size exclusion high pressure liquid chromatography (HPLC-SEC)

analysis was used for proteins and conjugates characterization. The HPLC-SEC system
consisted of an Ultimate 3000 Dionex equipped with a G6000PW (30 cm × 7.5 mm) column
(particle size 17 µm; Sigma 8-05765) with a TSK gel PWH guard column (7.5 mm ID × 7.5
cm L; particle size 13 µm; Sigma 8–06732) (TosohBioscience). The effluent was monitored
with a Photodiode Array Detector and a RF2000 Fluorescence Detector (Dionex). CRM197
and its derivative CRMADH were also analysed using a G3000PW (30 cm × 7.5 mm)
column (particle size 10 µm; Sigma 8–05762). Separation was performed with a flow rate of
0.5 mL/min using isocratic elution of 0.1 M NaCl, 0.1 M NaH2PO4, 5% CH3CN at pH 7.2.
Column void and bed volumes were measured with λ.-DNA ( λ.-DNA Molecular Weight
Marker III, 0.12–21.2 kbp, Roche) and sodium azide (NaN3, Merck), respectively. Eluted
polysaccharides and proteins were detected setting UV detector at 214nm (polysaccharide
and protein) and at 280 nm (protein). Protein peaks were also detected using tryptophan
fluorescence (emission spectrum at 336 nm, with excitation wavelength at 280 nm).
HPAEC-PAD—After evaluating different conditions of hydrolysis, Vi samples were

treated with NaOH at a final concentration of 2 M. Samples were heated at 110 °C for 4 h in
a closed screw-cap test tube, then chilled at 2–8 °C for about 30 min, filtered with 0.45 µm
Acrodisc (PALL) filter and analysed. Vi polysaccharide was used for building the
calibration curve, set up with standards in the range of 15–200 µg/mL (Vi was considered as
acid form fully acetylated in C-3 position). High Performance Anion Exchange
Chromatography with Pulsed Amperometric Detection (HPAEC–PAD) was performed with

a Dionex ICS3000 equipped with a CarboPac PA1 column (4 mm × 250 mm) coupled with
PA1 guard column (4 mm × 50 mm). Samples were run with a flow rate of 1 mL/min, using
a gradient from 40 mM to 150 mM NaNO3 in 100 mM NaOH. The effluent was monitored
using an electrochemical detector in the pulse amperometric mode with a gold working
electrode and an Ag/AgCl reference electrode. A quadruple-potential waveform for
carbohydrates was applied. The resulting chromatographic data were processed using
Chromeleon software 6.8.
Acridine orange method—Vi content in free and conjugated samples was measured also
by acridine orange binding according to the procedure reported in the literature [32]. Vi
samples were analysed in water, while conjugates were dialysed against 2 mM phosphate
buffer before the analysis.
2.2. Immunogenicity in mice

Immunization—Fourteen groups of 8 BALB/c female mice 5–6 weeks of age (Table 1)
were immunized with Vi, Vi-conjugates (Vi-CRM197, Vi-TT) or a physical mixture of Vi
and carrier protein (CRM197 or TT). Three subcutaneous injections of 200 µL each
containing 2.5 or 10 µg of Vi were given at two weeks intervals and blood was drawn prior
to and two weeks after each immunization. Mice received Vi conjugates formulated either
without adjuvant, or adjuvanted with 400 µg aluminum hydroxide per injection (Alum,
manufactured by Novartis Vaccines and Diagnostics) or Freund’s adjuvant (complete
Freund’s adjuvant (CFA), on the first injection; and incomplete Freund’s adjuvant (IFA), on
the second and third injections).
Analysis by ELISA—The wells of 96-well ELISA plates (Maxisorp, Nunc) were coated
overnight with 100 µL of 1 µg/mL Vi or 2 µg/mL CRM197 in 0.05 M carbonate buffer, pH
9.6 at 4 °C. The plates were blocked with 200 µL/well of 5% fat-free milk in phosphate-
buffered saline containing 0.05% Tween 20 (PBST) for 1 h at room temperature (RT). After
washing with PBST, 100 µL/well of mouse sera (diluted 1:200 in PBST containing 0.1%
BSA) were incubated for 2h at RT. After three more washes, 100 µL/well of alkaline
phosphatase-conjugated goat anti-mouse IgG secondary antibody (Sigma A3438, diluted
1:10,000 in PBST containing 0.1% BSA) was incubated for 1 h at RT. After another wash,
alkaline phosphatase substrate (p-nitrophenol phosphate tablets, Sigma N2765), dissolved in
1 M diethanolamine buffer, pH 9.8, was added to the plates and incubated for 1 h at RT.
Absorbance at 490 nm and 405 nm were obtained using an ELISA reader (ELx800, BioTek)
and OD405 minus OD490 values were used for subsequent analysis. Antibody units were
expressed relative to a mouse anti-Vi or anti-CRM197 standard serum curve, with best 4
parameter fit determined by modified Hill Plot. One ELISA unit was defined as the
reciprocal of the dilution of the standard serum that gave an absorbance value equal to 1 in
this assay. Each mouse serum was run in triplicate. Data are presented as scatter plots of
individual mice ELISA units, and bars represent the geometric mean of each group.
Statistical methods—Kruskal–Wallis ANOVA was used to compare the anti-Vi ELISA

units elicited by the various Vi-conjugates.

2.3. Protein derivatization with ADH

CRM197 and TT were derivatized by treatment with ADH and EDAC [27]. Derivatization
was performed at pH 6.0–6.2 rather than at the usual pH of 4.9–5.1 used for this kind of
reaction in order to avoid protein precipitation. MES buffer was used in order to maintain
constant pH during the entire reaction time. In the standard reaction, ADH (ADH/protein =
3.5, w/w) was added to either CRM197 or TT (10–12 mg/mL in 60–80 mM MES buffer pH
6.0–6.2) and mixed. When the solution became clear, 1.5–1.8 mg/mL EDAC (EDAC/protein
= 0.15, w/w) was added. The reaction was carried out for 1h at RT, then dialyzed at 2–8 °C
first against 0.2 M NaCl, 10 mM MES buffer pH 7.2 and then against 5 mM MES buffer pH
7.2. Protein recovery was estimated using micro BCA while the level of derivatization was
measured by mass spectrometry.
2.4. Vi conjugates
Conjugates were synthesized as described by Kossaczka et al. [27], with slight
modifications.
Vi conjugation to derivatized CRM197—Multiple Vi-CRM197 lots were synthesized.

For all lots, EDAC was added to Vi (molar ratio EDAC/Vi of 0.9–1.4, referred to Vi
repeating unit) in 100 mM MES buffer pH 6.0 and mixed for approximately 5 min at RT.
The individual lots also varied in the ratio of Vi to derivatized CRM197 used. The details of
the various synthesis are given in Table 2.
Derivatized CRM197 in 5 mM MES buffer pH 7.2 was added to the solution under mixing
and allowed to react with activated Vi for 3h at RT. Final conjugation mixture contained
>40 mM MES buffer pH 6, in order to maintain a constant pH during the reaction. The
reaction mixture was then dialyzed against 10 mM NaH2PO4, 0.2 M NaCl, pH 7.0 at 4 °C
and purified by size exclusion chromatography on a 1.6 cm × 90 cm Sephacryl S-1000
column eluted at 0.2 mL/min in either 10 mM NaH2PO4, 0.2 M NaCl, pH 7.0 (conjugation
lots 1–6) or 10 mM NaH2PO4, 5 mM NaCl pH 7.0 (conjugate lots 7–11). Fractions were
analysed by SDS-PAGE and those showing presence of conjugate without free protein were
collected. Two different pools were collected according to polysaccharide profile on the
same column in the same eluting conditions: the first of higher molecular weight did not
containing free Vi, while the second of lower molecular weight overlapped with
unconjugated Vi.
Vi conjugation to derivatized tetanus toxoid—A similar procedure was used to

synthesize Vi-TT. EDAC was added to Vi in 100 mM MES buffer pH 6.0 and allowed to
react for approximately 5 min at RT. Derivatized TT (Vi/TT, w/w, ratio of 0.5:1) in 5 mM
MES buffer pH 7.2 was added over few seconds to activated Vi solution and mixed slowly
at RT for 3h. The conjugation reaction mixture contained 3.15 mg/mL TTADH, 1.53 mg/mL
Vi and 1.47 mg/mL EDAC. The reaction was dialyzed against 0.2 M NaCl, 10 mM
NaH2PO4 pH 7.0 at 4 °C and purified by Sephacryl S-1000 (1.6 cm × 90 cm column) in 10
mM NaH2PO4, 0.2 M NaCl pH 7.0 eluted at 0.2 mL/min. Fractions were analysed by SDS-
PAGE and those not containing free protein were collected. As was done for Vi-CRM197,
two different pools were collected according to polysaccharide profile on the same column

in the same eluting conditions: the first of higher molecular weight did not contain free Vi,
while the second of lower molecular weight overlapped with unconjugated Vi.
3. Results
3.1. Characterization of Vi polysaccharide
Characterization of Vi showed that this preparation contained 1% of nucleic acids, 0.3% of
proteins, an O-acetylation level of 68% (according to 1H NMR) and of 63.4% (according to
Hestrin method). Spectral assignments of 1H NMR of de-O-acetylated Vi saccharide
(5.10ppm H1, 4.70ppm H5, 4.44ppm H4, 4.2ppm H2, 4.13 ppm H3, 2.06 ppm N-acetyl in
non-O-acetylated residues (3H), 1.91 ppm acetate anion arising from de-O-acetylation (3H))
were in good agreement with published results [31]. Na+ content was <0.08%.
Thermogravimetric and Karl Fisher analysis gave 76% of dried weight and 21.6% of
residual moisture, respectively. This Vi with a purity of 76% by weight was used as standard
for High Performance Anion Exchange Chromatography with Pulsed Amperometric
Detection (HPAEC-PAD) analysis.
3.2. Assay for unconjugated and conjugated Vi

To measure the concentration of Vi in solution or in glycoconjugate samples, we developed

an assay based on the measurement of the product released following hydrolysis of the Vi
polysaccharide based on HPAEC-PAD. During the development the following aspects were
investigated:
Hydrolysis conditions—Hydrolysis of Vi in 4 M TFA at 120 °C for periods up to 2 h

resulted in a mixture of oligosaccharides that were separated on HPAEC and detected by
PAD (Fig. 1A). Hydrolysis of Vi in 4 M TFA at 120 °C after de-O-acetylation or reduction
of COOH groups [33,10] also resulted in a mixture of oligosaccharides. Hydrolysis with
TFA after treatment for complete de-acetylation in 2 M NaOH at 110 °C for 6 h [33]
resulted in a single peak on HPAEC-PAD. Only the step of hydrolysis with 2 M NaOH at
110 °C resulted in this single peak (Fig. 1B). Kinetic of hydrolysis was studied (data not
shown) and, as a result of this investigation, 4 h at 110 °C in 2 M NaOH were selected as
conditions of hydrolysis.
Range of detection—Calibration curves generated from hydrolysis of purified Vi and

HPAEC-PAD analysis showed excellent linearity over the range of 1–200 µg/mL when 25
µL samples were analysed (0.025–5 µg per sample) (R2 > 0.999).
Under the conditions described above, the HPAEC-PAD method gave a close correlation
with the acridine orange method [32,34] for unconjugated Vi samples and for some Vi-
CRM197 conjugates tested (Table 3). In contrast to other reported procedures for Vi
quantification, our new method was more sensitive. For example, the intensities of the Vi
absorbance peaks observed by FT-IR spectroscopy were proportional to Vi in the range of
0.25–2.0 mg [32], and using acridine orange method Vi in the range of 20–700 µg/mL can
be assayed. Furthermore, the HPAEC-PAD method can be applied to samples in different
and complex matrixes unlike FT-IR that requires salt-free solids or the acridine orange assay

that is sensitive to ionic strengths greater than 3 mM. In the latter assay and in the presence
of higher salt concentrations or impurities, the binding of the cation dye acridine orange to
Vi is inhibited. Attempts to quantify Vi in culture supernatants or in intermediate steps of
purification using acridine orange method failed, as did the quantification of Vi in

conjugates without including a step of dialysis. We also had difficulty obtaining quantitative
results with some Vi conjugates when using spectrophotometric titration with acridine
orange, resulting in underestimations of Vi concentration. The reason for this has not been
determined.
3.3. Protein derivatization with ADH

CRM197 derivatizated with ADH (CRMADH) resulted in a similar mobility on SDS-PAGE
and similar retention time on HPLC-SEC as the underivatized CRM197 (Fig. 2). Yields were
>85%. There was no significant formation of protein–protein dimers or higher oligomers, a
theoretical side reaction with the chemistry chosen (Figs. 2 and 3). The level of CRM197
derivatization with ADH was measured by mass spectrometry. Q-TOF analysis showed the
formation of several products characterized by the presence of a different number of ADH
linkers ranging from 3 to 10 with the major peak containing 6 bound linkers (Fig. 3A).
Analysis by MALDI-TOF showed a broad peak shifted by a mass of 830 Da compared with
un-derivatized CRM197. As shown by Kossaczka et al. [27], the level of derivatization

depended on the concentration of EDAC in the reaction mixture. Using half the
concentration of EDAC, the CRMADH was less derivatized containing 2–3 linkers per mole
of protein (Fig. 3B).
TT was derivatized using the same conditions described for CRM197, with a recovery of
86%. The TTADH SDS-PAGE pattern and SEC profile were similar to native TT. Molar
ratio of ADH to TT was measured by MALDI-TOF mass spectrometry and had an average
of 11 linkers per TT (data not shown).
3.4. Vi-CRM197 and Vi-TT

Sephacryl S-1000 gel filtration profiles of Vi-CRM197 and Vi-TT mixtures showed peaks
containing protein (as judged by A214 and A280) at higher MW than either unconjugated Vi
or the unconjugated carrier protein, CRM197 or TT (Fig. 4). Analysis of the conjugates
showed that they contained both protein and Vi (Table 4). Consistent with the large size of
the conjugate, close to the void volume of the Sephacryl S-1000 column, the conjugate was
unable to migrate into the SDS-PAGE gel and protein stained band was observed in the well
[27].
Conjugates were easily separated from unconjugated CRM197 or TT but, due to the
polydisperse nature of Vi and Vi conjugates and to the polysaccharide large size compared
with the protein, the elution of conjugate and free Vi overlapped (Fig.4).Fractions pooled
from the first half of the conjugate peak (pool 1)were characterized by a Vi to protein ratio
lower than that in fractions from the second half (pool 2). As pool 1 was collected at elution
times that did not overlap with free Vi, we assume that this material did not contain
unconjugated Vi, while the presence of unconjugated Vi could not be ruled out for pool 2.
Similar results were obtained with analysis of the Vi-TT conjugates.

Separation on Sephacryl S-1000 was subsequently optimized and elution with 10 mM

NaH2PO4 pH 7.0 with 5 mM NaCl gave a better separation of conjugate and free Vi than
with 200 mM NaCl. Vi-CRM197 Lots 7–11 were purified in this way. A better separation
between conjugate and free Vi was obtained and enabled also the purification of lower
molecular weight conjugates prepared by using a higher ratio of Vi to CRM197 (Fig. 5).
Reproducibility of conjugate formation—Three lots of Vi-CRM197 using the same

batch of CRMADH were synthesized to test the reproducibility of the process on small scale.
Conjugations were performed on 4.6 mg of Vi (1.53 mg/mL) and a ratio Vi/CRM197 of 0.5:1
(w/w). EDAC was used at 1.47 mg/mL. As shown in Table 5, the three independent
conjugate lots gave similar recoveries.
Varying the ratio of Vi to CRM197—Additional lots of Vi-CRM197 were conjugated

varying the initial ratio of Vi to CRM197 (Table 2, Lots 7–11). In each case as judged by
profile on gel filtration, conjugates were successfully formed. As the initial ratio of Vi to
CRMADH was increased, the peak elution time increased, consistent with fewer CRM197
molecules attached to polysaccharide. As expected the Vi to protein ratio also increased
(Table 6).
3.5. Immunogenicity of Vi conjugates in mice

The immunogenicity of unconjugated Vi, Vi conjugates (with or without adjuvant) and
mixtures of Vi with CRMADH or TTADH was tested in mice (Table 1). Tested conjugates
differed by the ratio of Vi to protein. Furthermore, Vi-CRM197 Lot 1 and Vi-TT Lot 1 pool
1 did not contain free polysaccharide, while Vi-CRM197 Lot 3 pool 2 and Vi-TT Lot 1 pool
2 likely contained unconjugated Vi after purification by Sephacryl S-1000. Anti-Vi IgG
were measured by ELISA (Fig. 6A) and compared with anti Vi levels in mice serum
obtained after Vi-rEPA immunization (serum provided by NIH). The anti-CRM197
responses for the groups of mice receiving free CRMADH or Vi-CRM197 conjugates are
shown in Fig. 6B. Unconjugated Vi, either alone (Fig. 6A, group 2) or mixed with CRMADH
or TTADH (Fig. 6A, groups 3 and 4, respectively), failed to generate a detectable antibody
response, even after three injections. By contrast all mice receiving Vi conjugates, with
either CRM197 or TT as the carrier protein, showed a significant increase in Vi antibody
levels even after the first injection and in the absence of adjuvant (Fig. 6A, groups 5–8). The
anti-Vi response peaked after the second injection, while the anti-CRM197 antibody levels
increased also after the third injection (Fig. 6B).
No significant difference in anti-Vi antibody levels was detected in sera of mice immunized
with Vi-conjugates (Fig. 6A, groups 5–14) when compared on a given bleed day (Kruskal–
Wallis ANOVA, p > 0.05). Additionally, no differences in anti-Vi antibody levels were
observed between 2.5 µg and 10 µg doses (Fig. 6A, group 5 in comparison with 13 and 14),
or between the unadjuvanted or adjuvanted formulations (Fig. 6A, group 5 in comparison to
9, 13 and 14; group 6 in comparion to 10; groups 7 and 8 in comparison to 11 and 12,
respectively). Finally, animals immunized with Vi-CRM197 or Vi-TT produced anti-Vi
antibody levels similar to that detected in NIH mice serum (mice immunized with Vi-rEPA
conjugate). Differences in Vi/CRM197 ratios (Fig. 6A, groups 5 and 6) or possible presence

of free Vi (Fig. 6A, groups 6 and 8 vaccinated with pool 2) did not produce significant
variations of anti-Vi antibody responses.
4. Discussion
Surface polysaccharides from bacteria have been widely used in vaccines, being both
essential virulence factor and protective antigens. As saccharides are T-independent
antigens, they are poorly immunogenic, do not induce immunological memory and are not
effective in infants. Re-vaccination is required at regular intervals as antibody levels decline,
but reinjection at any age does not elicite a booster effect [17,35,36]. It is well known that
conjugation to a carrier protein converts T-independent antigens into T-dependent ones,
thereby providing a long lasting protection and enhancing memory responses. Antibody
response is boosted by repeated immunization and glycoconjugate vaccines confer
protection in children younger than 2 years of age [17,35,37].
The first conjugate vaccine was developed against Haemophilus influenzae type b (Hib)
[38,39]. With the success of the Hib conjugate vaccine, glycoconjugate vaccines against
Neisseria meningitidis and Streptococcus pneumoniae have been developed and licensed
[22–24,35].
In addition to native purified polysaccharides, oligosaccharides obtained by polysaccharide

degradation or through organic synthesis can be used for the preparation of glycoconjugate
vaccines [37]. In general, native polysaccharides are widely polydisperse in molecular
weight as are their corresponding conjugates. Whereas, the use of oligosaccharides allows
for well defined vaccines in a more reproducible way [37]. Saccharide chain length as well
as saccharide to protein ratio are important parameters that can influence the
immunogenicity of a glycoconjugate.
As shown for other polysaccharides, the introduction of Vi vaccine into routine

immunization programs has been limited by its age-related immunogenicity and T-cell-
independent properties. Conjugation to a carrier protein conferred T-cell dependence and
increased immunogenicity [28,40,41]. Several proteins, such as diphtheria toxoid (DT),
tetanus toxoid (TT), cholera toxins (CT), the B subunit of the heat-labile toxin (LT-B) of
Escherichia coli, recombinant OMP of Klebsiella pneumoniae (rP40) and iron-regulated
OMP of S. Typhi have been tested as carriers for Vi polysaccharide [42]. In particular,
clinical trials of Vi-rEPA conjugate conferred 89% protection over 46 months in
Vietnamese children 2–5 years old [19,20]. However, the lack of regulatory precedent for
rEPA in licensed vaccine hinders its use in areas where a vaccine is more needed. Recently a
Vi-TT conjugate has been licensed for local distribution in India [43] and the International
Vaccine Institute (IVI) is planning to perform clinical evaluation of the safety and
immunogenicity of a Vi-DT vaccine [42]. In the present study, we described the preparation,
characterization and immunological properties of an anti-typhoid fever conjugate vaccine
composed of the C.freundii capsular polysaccharide covalenty coupled to the non toxic
mutant of diphtheria toxin CRM197.

We decided to use the high molecular weight native polysaccharide since it has been
reported that conjugates obtained after reduction of Vi molecular size elicited lower levels of
antibodies than those prepared with native Vi [41]. Furthermore, Vi polysaccharide is
resistant to acid hydrolysis and its depolymerisation is unlikely in conditions that would
maintain its structure substantially unchanged.
CRM197 was derivatized with adipic acid dihydrazide and conjugated to Vi polysaccharide
by carbodiimide chemistry. As there are multiple activation points within each Vi chain and
multiple linkage points on each protein, crosslinked network of very high molecular weight
was formed. The Vi-CRM197 conjugates were fully characterized, including a newly
developed method for total sugar quantification based on HPAEC-PAD.
Manufacture of vaccines requires good characterization and quality control of all its
components. The standardization of vaccines composed of conjugated or unconjugated Vi
requires a method for quantifying Vi [44]. Colorimetric methods for measuring amino sugar
residues or uronic acids do not work for Vi because of its resistance to acid hydrolysis. This
also prevented the application of traditional HPAEC-PAD methods [45–48] for Vi
quantification. Our new method is based on alkaline hydrolysis with NaOH, in conditions
that cause both complete de-acetylation and hydrolysis of the Vi. In this way quantification
of the released species by HPAEC-PAD becomes possible. The same procedure can be used
to quantify Vi in conjugates. The optimized procedure is reproducible, simple and precise
(CV < 2% both for Vi and Vi conjugate samples) and more sensitive than other methods
previously employed [29,32] (detection of ≥ 1 µg/mL Vi concentrations). Furthermore, it is
suitable for quantifying Vi in complex matrixes and for analysis of formulated conjugates.
By modifying the reaction conditions described in published Vi conjugation methods, we

have generated conjugates with carbohydrate/protein ratio ranging from 0.5:1 to 10:1 (Table
2). Vi-CRM197 conjugates with a ratio 1.15 and 0.71 were tested in this preliminary study
and their immunogenicity was compared to those induced by Vi-TT conjugates containing a
similar ratio of 0.41 and 0.67 (Table 1). Similar levels of serum IgG Vi antibodies were
elicited in mice by all Vi conjugates tested and were significantly higher than those elicited
either by Vi alone or its physical mixture with carrier proteins (Fig. 6A). Conjugation ratio
or dose within range tested had little impact on immunogenicity. It would be interesting to
test other conjugates with higher Vi to protein ratios. Furthermore, anti-Vi IgG levels were
similarly elicited by Vi-CRM197 and Vi-TT conjugates and were also comparable to those
elicited by Vi-rEPA at NIH. The nature of the carrier protein used (i.e., CRM197, TT or
rEPA) did not impact on immunogenicity, and confirms the data of Cui et al. [42] where Vi-
DT was compared with Vi-rEPA produced at NIH. Importantly, a powerful adjuvant
(Freunds complete followed by incomplete) did not increase antibody responses.
Based on these preliminary results, additional immunogenicity studies of Vi-CRM197 have

been conducted (Rondini et al., manuscript submitted). Vi-CRM197 appears to be a suitable
candidate for the development of a commercially viable, effective typhoid vaccine,
especially for young children in low income countries.

Acknowledgments
The authors would like to thank Nathalie Norais for performing MALDI analysis; Mariagrazia Pizza and Vega
Masignani for their comments and suggestions on this manuscript; and the Novartis Animal Resources Center for
conducting in vivo studies. This study was funded in part by grants received by Associazione A. Sclavo from
Regione Toscana and Fondazione Monte Dei Paschi di Siena. This manuscript is dedicated to the memory of
Angela Bardotti who provided exceptional contribution in the HPAEC-PAD method development.
References
1. Crump JA, Luby SP, Minta ED. The global burden of typhoid fever. Bull WHO. 2004; 82(5):346–
353. [PubMed: 15298225]
2. Khan MI, Ochiai RL, Clemens JD. Population impact of Vi capsular polysaccharide vaccine. Expert
Rev Vaccines. 2010; 9(5):485–496. [PubMed: 20450323]
3. Ackers ML, Puhr ND, Tauxe RV, Mintz ED. Laboratory-based surveillance of Salmonella serotype
Typhi infections in the United States: antimicrobial resistance on the rise. JAMA. 2000; 283(20):
2668–2673. [PubMed: 10819949]
4. Lawrence, J.; Jones, J. Final report: pilot of enhanced enteric fever surveillance in England, Wales,
and Northern Ireland 2008. London, UK: Health Protection Agency; 2008.
5. Saha SK, Baqui AH, Hanif M, Darmstadt GL, Ruhulamin M, Nagatake T, et al. Typhoid fever in
Bangladesh: implications for vaccination policy. Pediatr Infect Dis J. 2001; 20(5):521–524.
[PubMed: 11368111]
6. Lin F-YC, Ho VA, Bay PV, Thuy NTT, Bryla D, Thanh CT, et al. The epidemiology of typhoid
fever in the Dong Thap Province, MeKong Delta region of Vietnam. Am J Trop Med Hyg. 2000;
62:644–648. [PubMed: 11289678]
7. Sinha A, Sazawal S, Kumar R, Sood S, Reddaiah VP, Singh B, et al. Typhoid fever in children aged
less than 5 years. Lancet. 1999; 354:734–737. [PubMed: 10475185]
8. Brooks WA, Hossain A, Goswami D, Sharmeen AT, Nahar K, Alam K, et al. Bacteremic typhoid
fever in children in an urban slum, Bangladesh. Emerg Infect Dis. 2005; 11:326–329. [PubMed:
15752457]
9. Sur D, Ochiai RL, Bhattacharya SK, Ganguly NK, Ali M, Manna B, et al. A cluster-randomized
effectiveness trial of Vi typhoid vaccine in India. Engl J Med. 2009; 361(4):335–344.
10. Szu SC, Li X, Stone AL, Robbins JB. Relation between structure and immunologic properties of
the Vi capsular polysaccharide. Infect Immun. 1991; 59(12):4555–4561. [PubMed: 1937814]
11. DeRoeck D, Ochiai RL, Yang J, Anh DD, Alag V, Clemens JD. Typhoid vaccination: the Asian
experience. Expert Rev Vaccines. 2008; 7(5):547–560. [PubMed: 18564010]
12. Guzman CA, Borsutzky S, Griot-Wenk M, Metcalfe IC, Pearman J, Col-lioud A, et al. Vaccines
against typhoid fever. Vaccine. 2006; 24(18):3804–3811. [PubMed: 16278037]
13. Fraser A, Goldberg E, Acosta CJ, Paul M, Leibovici L. Vaccines for preventing typhoid fever.
Cochrane Database of Syst Rev. 2007; (3) Art. No.: CD001261.
14. Fraser A, Paul M, Goldberg E, Acosta CJ, Leibovici L. Typhoid fever vaccines: systematic review
and meta-analysis of randomised controlled trials. Vaccine. 2007; 25(45):7848–7857. [PubMed:
17928109]
15. World Health Organization. Typhoid vaccines: WHO position paper. Wkly Epidemiol Rec. 2008;
83(6):49–60. [PubMed: 18260212]
16. Ada G, Isaacs D. Carbohydrate–protein conjugate vaccines. Clin Microbiol Infect. 2003; 9:79–85.
[PubMed: 12588327]
17. Jones C. Vaccines based on the cell surface carbohydrates of pathogenic bacteria. An Acad Bras
Cienc. 2005; 77(2):293–324. [PubMed: 15895165]
18. Szu, SC.; Robbins, JB.; Schneerson, R.; Lin, FY. Polysaccharide-based conjugate vaccines for
enteric bacterial infections: typhoid fever, non-typhoidal Salmonella and Escherichia coli
O157:H7. In: Levine, MM., editor. New generation vaccines. 4th ed. Vol. 47. 2009. p. 1040

19. Canh DG, Lin F, Thiem VD, Trach DD, Trong ND, Mao ND, et al. Effect of dosage on
immunogenicity of a Vi conjugate vaccine injected twice into 2- to 5-year-old Vietnamese
children. Infect Immun. 2004; 72(11):6586–6588. [PubMed: 15501790]
20. Lin FY, Ho VA, Khiem HB, Trach DD, Bay PV, Thanh TC. The efficacy of a Salmonella typhi Vi
conjugate vaccine in two-to-five-year-old children. N Engl J Med. 2001; 344(17):1263–1269.
[PubMed: 11320385]
21. Croxtall JD, Keating GM. Pneumococcal polysaccharide protein D-conjugate vaccine (Synflorix;
PHiD-CV). Paediatr Drugs. 2009; 11(5):349–357. [PubMed: 19725600]
22. Rennels MB, Edwards KM, Keyserling HL, Reisinger KS, Hogerman DA, Madore DV, Chang I,
Paradiso PR, Malinoski FJ, Kimura A. Safety and immunogenicity of heptavalent pneumococcal
vaccine conjugated to CRM197 in United States infants. Pediatrics. 1998; 101(4 Pt 1):604–611.
[PubMed: 9521941]
23. Shinefield HR, Black S, Ray P, Chang I, Lewis N, Fireman B, et al. Safety and immunogenicity of
heptavalent pneumococcal CRM197 conjugate vaccine in infants and toddlers. Pediatr Infect Dis J.
1999; 18(9):757–763. [PubMed: 10493334]
24. Snape MD, Perrett KP, Ford KJ, John TM, Pace D, Yu LM, et al. Immunogenicity of a tetravalent
meningococcal glycoconjugate vaccine in infants: a randomized controlled trial. JAMA. 2008;
299(2):173. [PubMed: 18182599]
25. Daniels EM, Schneerson R, Egan WM, Szu SC, Robbins JB. Characterization of the Salmonella
paratyphi C Vi polysaccharide. Infect Immun. 1989; 57(10):3159–3164. [PubMed: 2506132]
26. Hale C, Bowe F, Pickard D, Clare S, Haeuw JF, Powers U, et al. Evaluation of a novel Vi
conjugate vaccine in a murine model of salmonellosis. Vaccine. 2006; 24:4312–4320. [PubMed:

16600446]
27. Kossaczka Z, Bystricky S, Bryla DA, Shiloach J, Robbins JB, Szu SC. Synthesis and
immunological properties of Vi and di-O-acetyl pectin protein conjugates with adipic acid
dihydrazide as the linker. Infect Immun. 1997; 65(6):2088–2093. [PubMed: 9169736]
28. Szu SC, Li XR, Schneerson R, Vickers JH, Bryla D, Robbins JB. Comparative immunogenicities
of Vi polysaccharide–protein conjugates composed of cholera toxin or its B subunit as a carrier
bound to highor lower-molecular-weight Vi. Infect Immun. 1989; 57:3823–3827. [PubMed:
2807549]
29. Hestrin S. The reaction of acetylcholine and other carboxylic acid derivatives with hydroxylamine,
and its analytical application. J Biol Chem. 1949; 180:249–61. [PubMed: 18133390]
30. Atomic absorption spectrometry (2.2.23). Ph. Eur. 2006; 5.0
31. Lemercinier X, Martinez-Cabrera I, Jones C. Use and validation of an NMR test for the identity
and O-acetyl content of the Salmonella typhi Vi capsular polysaccharide vaccine. Biologicals.
2000; 28(1):17–24. [PubMed: 10799051]
32. Stone AL, Szu SC. Application of optical properties of the Vi Capsular polysaccharide for
quantitation of the Vi antigen in vaccines for typhoid fever. J Clin Microbiol. 1988; 26(4):719–
725. [PubMed: 3366868]
33. Kao G, Tsai C. Quantification of O-acetyl, N-acetyl and phosphate groups and determination of the
extent of O-acetylation in bacterial vaccine polysaccharides by high-performance anion-exchange

chromatography with conductivity detection (HPAEC-CD). Vaccine. 2004; 22:335–344.
[PubMed: 14670314]
34. Stone AL, Bradley DF. Aggregation of cationic dyes on acid polysaccharides I.
Spectrophotometric titration with acridine orange and other metachromatic dyes. Biochim Biophys
Acta. 1967; 148:172–192. [PubMed: 4229557]
35. Finn A. Bacterial polysaccharide–protein conjugate vaccines. Br Med Bull. 2004; 70:1–14.
[PubMed: 15339854]
36. Lucas AH, Apicella MA, Taylor CE. Carbohydrate moietiesas vaccine candidates. Clin Infect Dis.
2005; 41:705–712. [PubMed: 16080094]
37. Vliegenthart JFG. Carbohydrate based vaccines. FEBS Lett. 2006; 580:2945–2950. [PubMed:
16630616]

38. Robbins JB, Schneerson R, Anderson P, Smith DH. Prevention of systemic infections, especially
meningitis, caused by Haemophilus influenzae type b: impact on public health and implication for
other polysaccharide-based vaccines. JAMA. 1996; 276:1181–1185. [PubMed: 8827975]
39. Chu C, Schneerson R, Robbins JB, Rastogi SC. Further studies on the immunogenicity of
Haemophilus influenzae type b and pneumococcal type 6A polysaccharide–protein conjugates.
Infect Immun. 1983; 40:245–256. [PubMed: 6601061]
40. Szu SC, Stone AL, Robbins JD, Schneerson R, Robbins JB. Vi capsular polysaccharide–protein
conjugates for prevention of typhoid fever. Preparation, characterization, and immunogenicity in
laboratory animals. J Exp Med. 1987; 166:1510–1524. [PubMed: 3681191]
41. Szu SC, Taylor DN, Trofa AC, Clemens JD, Shiloach J, Sadoff JC, et al. Laboratory and
preliminary clinical characterization of Vi capsular polysaccharide–protein conjugate vaccines.
Infect Immun. 1994; 62:4440–4444. [PubMed: 7927707]
42. Cui C, Carbis R, An SJ, Jang H, Czerkinsky C, Szu SC, et al. Physical and chemical
characterization and immunologic properties of Salmonella enterica Serovar Typhi capsular
polysaccharide–diphtheria toxoid conjugates. Clin Vac Immunol. 2010; 17(1):73–79.
43. Singhal, T.; Amdekar, Y.; Agarwal, R. IAP guidebook on immunization: IAP committee on
immunization 2007–2008. New Delhi, India: Jaypee; 2009. p. 57
44. WHO TRS. Requirements for Vi polysaccharide typhoid vaccine. Adopted 1992, TRS No. 840,
Annex 1.
45. Yu Ip CC, Manam V, Hepler R, Hennessey JP Jr. Carbohydrate composition analysis of bacterial
polysaccharides: optimized acid hydrolysis conditions for HPAEC-PAD analysis. Anal Biochem.
1992; 201:343–349. [PubMed: 1632523]

46. Talaga P, Vialle S, Moreau M. Development of a high-performance anion-exchange
chromatography with pulsed-amperometric detection based quantification assay for pneumococcal
polysaccharides and conjugates. Vaccine. 2002; 20:2474–2484. [PubMed: 12057602]
47. Tsai CM, Gu XX, Byrd A. Quantification of polysaccharide in Haemophilus influenzae type b
conjugate and polysaccharide vaccines by high-performance anion-exchange chromatography with
pulsed amperometric detection. Vaccine. 1994; 12:700–706. [PubMed: 8091847]
48. Ricci S, Bardotti A, D’Ascenzi S, Ravenscroft N. Development of a new method for the
quantitative analysis of the extracellular polysaccharide of Neisseria meningitidis serogroup A by
use of high-performance anion-exchange chromatography with pulsed-amperometric detection.
Vaccine. 2001; 19:1989–1997. [PubMed: 11228369]

Fig. 1.
TFA and NaOH hydrolysis of Vi detected by HPAEC-PAD. HPAEC-PAD profiles were
obtained with CarboPac PA1 column using PA1 guard-column and a flow rate of 1 mL/min.
Vi treatment with (A) 4 M TFA at 120 °C for 2h resulting in a mixture of oligosaccharides,
analysed by elution with 18 mM NaOH for 15 min and then reaching 40 mM NaOH, 10 mM
NaNO3 in 5 min and then 100 mM NaOH, 80 mM NaNO3 in 10 min and finally 100 mM
NaOH, 200 mM NaNO3 in 30 min. (B) 2 M NaOH at 110 °C for4h resulting in a single
peak, analysed by elution in gradient from 40 mM to 150 mM NaNO3 in 100 mM NaOH in
22 min.

Fig. 2.
HPLC-SEC analysis of underivitized CRM197 and CRMADH. Samples were run on TSK gel
3000 PW column and eluted with 100 mM NaH2PO4, 100 mM NaCl, 5% CH3CN pH 7.2
with UV detection at 214 nm. Derivatized CRM197 (CRMAHD) showed similar retention
time on HPLC-SEC as the underivatized protein.

Fig. 3.
Impact of EDAC/CRM197 ratio on derivitization of CRMADH as detected by MS Q-TOF.
Derivatized CRMADH was prepared using (A) EDAC/CRM197 ratio of 0.15 (w/w) or (B)
half amount of EDAC. The result is the formation of several products characterized by the
presence of a different number of linkers bound to the protein: (A) 3–10 linkers per mol
protein with the principle product containing6 linkers and (B) 2–3 linkers per mole of
protein.

Fig. 4.
Sephacryl S-1000 profile of Vi, CRMADH and Vi-CRM197. Vi-CRM197 Lot 3 was purified
on Sephacryl S-1000 (1.6 cm × 90 cm) eluting with 10 mM NaH2PO4, 200 mM NaCl pH7.0
at a flow rate of 0.2 mL/min. Conjugate reaction mixture profile is in comparison with
unconjugated Vi run on the column in the same conditions; black arrow indicates the peak
corresponding to unconjugated protein.

Fig. 5.
Sephacryl S-1000 purification eluted with 5 mM NaCl instead of 200 mM NaCl gives better
separation of Vi and Vi-CRM197. Vi-CRM197 Lot 11 purification on Sephacryl S-1000 in
comparison to free Vi. Sephacryl S-1000 (1.6 cm × 90 cm column); eluent: 10 mM
NaH2PO4, 5 mM NaCl, pH 7.0; flow: 0.2 mL/min. Separation conditions on Sephacryl

S-1000 were optimized, finding that elution with this buffer containing lower NaCl gave a
better separation between conjugate and free Vi, also for lots synthesized using a higher
ratio of Vito CRM197 and so producing a conjugate shifted at lower MW.

Fig. 6.
Conjugation of Vi to carrier protein is required to generate anti-Vi antibodies. Fourteen
groups of BALB/c female mice (N = 8 per group) were subcutaneously immunized on days
0,14 and 28 as detailed in Table 1. Sera were assessed by ELISA for (A) anti-Vi IgG
antibodies and (B) anti-CRM197 IgG antibodies (measured just in those groups receiving
CRM197). Bars represent the geometric mean ELISA units of the group, individual animals
are represented by the scatter plots. Comparisons are made to serum from mice vaccinated

with Vi-rEPA (serum provided by PDMI, NIH, obtained from mice immunized with 2.5 µg
of the Vi-rEPA conjugate used in human trials in Vietnam).

Table 1
Immunogenicity study.
Group number Vaccine Vi/protein ratio

in conjugate
(w/w)
1 PBS
2 Vi
3 Vi+CRMADH
4 Vi+TTADH
5 Vi-CRM197 Lot 1 1.15
6 Vi-CRM197 Lot 3 pool 2 0.71
7 Vi-TT Lot 1 pool 1 0.41

8 Vi-TT Lot 1 pool 2 0.67
9 Alum formulated Vi-CRM197 Lot 1 1.15
10 Alum formulated Vi-CRM197 Lot 3 pool 2 0.71
11 Alum formulated Vi-TT Lot 1 pool 1 0.41

12 Alum formulated Vi-TT Lot 1 pool 2 0.67
13 CFA/IFA formulated Vi-CRM197 Lot 1 1.15
14 CFA/IFA formulated Vi-CRM197 Lot 1 1.15
Vi dose per vaccination was 2.5 µg for all the groups with the exception of groups 13 and 14 that received 10 µg (both groups were identical).

Table 2
Conjugation conditions used for generation of different Vi-CRM197 lots.

Conjugate Vi Vi/CRM197 ratio EDAC/Vi ratio

concentration (w/w) (w/w)
(mg/mL)
Lot 1 2.1 1:1 0.88
Lot 3 2.1 0.5:1 0.88
Lots 4–6 1.5 0.5:1 0.96
Lot 7 2.2 0.7:1 0.69
Lot 8 2.2 1:1 0.69
Lot 9 2.2 2:1 0.69
Lot 10 2.2 5:1 0.67
Lot 11 2.2 10:1 0.67

Table 3
Vi concentrations in unconjugated Vi and Vi-CRM197 samples determined by HPAEC-PAD after alkaline

hydrolysis and by acridine orange method.
Sample Vi concentration (µg/mL) determined by
HPAEC-PAD acridine orange

Unconjugated 30.0 34.6
Vi 50.0 55.8
100.0 122.0
Vi- 23.2 26.0
CRM197 99.6 97.5
116.0 115.2

Table 4
Vi-CRM197 and Vi-TT conjugate characterization for protein and saccharide content.
Conjugate Vi contenta µg/mL (HPAEC- Vi contentb µg/mL(acridine Protein content µg/mL Vi/protein ratio (w/w)
PAD) orange) (micro BCA)
Vi-CRM197 70.0 60.9 53.0 1.32a

Lot 1 1.15b
Vi-CRM197 23.2 25.9 41.9 0.55a
Lot 3 pool 1 0.62b
Vi-CRM197 99.6 97.6 137.6 0.72a
Lot 3 pool 2 0.71b
Vi-TT nd 26.8 64.64 0.41b
Lot 1 pool 1
Vi-TT 84.7 82.8 123.68 0.68a
Lot 1 pool 2 0.67b
a
Vi quantified by HPAEC-PAD analysis.

b
Vi quantified by acridine orange colorimetric assay.
CV< 2% for HPAEC-PAD and < 3% for micro BCA on analysed conjugates. Acridine orange measurements were more variable (CV from 3 to
12%).

NIH-PA Author Manuscript NIH-PA Author Manuscript NIH-PA Author Manuscript
Table 5
Reproducibility of Vi-CRM197 conjugation and purification.
Conjugate Vi conc. µg/mL (HPAEC-PAD) CRM197 conc. µg/mL (micro BCA) Vi/CRM197 ratio (w/w) % yield
Micoli et al.
Vi CRM197
Lot 4 pool 1 24.37 52.06 0.47

84.0 64.9
Lot 4 pool 2 78.36 106.1 0.74
Lot 5 pool 1 34.51 63.17 0.55
86.8 63.9
Lot 5 pool 2 71.39 93.42 0.76
Lot 6 pool 1 39.12 59.1 0.66
86.2 56.9
Lot 6 pool 2 67.73 82.41 0.82

Page 27
NIH-PA Author Manuscript NIH-PA Author Manuscript NIH-PA Author Manuscript
Table 6
Characterization of Vi-CRM197 produced by varying the ratio of Vi to CRM197
Conjugate Nominal Vi/CRM197 ratio in reaction Vi conc. µg/mL CRM197 conc. Actual Retention time
Micoli et al.
(HPAEC-PAD) µg/mL (micro Vi/CRM197 ratio of the major

BCA) (w/w) in peak on
purified Sephacryl
conjugate S-1000 (mL)
Lot 7 0.7:1 18.6 27.4 0.68 100.33

Lot 8 1:1 14.3 16.6 0.86 102.57
Lot 9 2:1 22.4 11.9 1.88 106.79
Lot 10 5:1 49.5 8.4 5.90 107.14
Lot 11 10:1 32.9 5.0 6.58 107.47

Page 28

VI Polysaccharide

Uploaded by

Copyright:

Available Formats

VI Polysaccharide

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

VI Polysaccharide

Uploaded by

Copyright:

Available Formats

NIH Public Access

Vaccine. 2011 January 17; 29(4): 712–720. doi:10.1016/j.vaccine.2010.11.022.

Vi-CRM197 as a new conjugate vaccine against Salmonella Typhi

© 2010 Elsevier Ltd. All rights reserved.

recent vaccine trial in Kolkuta, India [9].

The capsular polysaccharide of Salmonella Typhi (Vi) is a linear homopolymer of α1,4-N-

A recent meta-analysis of both Ty21a (oral) and Vi polysaccharide (parenteral) vaccines

As observed with other polysaccharides [16,17], conjugation of Vi to a carrier protein

Vaccine. Author manuscript; available in PMC 2014 September 15.

Citrobacter freundii WR7011 has been chosen as source of Vi instead of Salmonella

2. Materials and methods

Reagents and materials

dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride (EDAC), formic acid, 4-

2.1. Analytical methods

Vaccine. Author manuscript; available in PMC 2014 September 15.

Derivatized proteins and conjugates were examined by sodium dodecyl sulfate-

Nucleic acid analysis—Nucleic acid content was determined by UV spectroscopy at a

Thermogravimetric analysis—Vi sample was analysed in duplicate for volatile content

Mass spectrometry—Protein samples were analysed by MS Q-TOF and MALDI-TOF.

Measurements were recorded on an Ultraflex (Bruker Daltonics) MALDI-TOF/TOF MS in

1H NMR spectroscopy—Nuclear Magnetic Resonance (NMR) experiments were

Vaccine. Author manuscript; available in PMC 2014 September 15.

Dried Vi sample was solubilized in deuterated water (Aldrich) and transferred to 5 mm

HPLC-SEC—Analytical size exclusion high pressure liquid chromatography (HPLC-SEC)

HPAEC-PAD—After evaluating different conditions of hydrolysis, Vi samples were

Vaccine. Author manuscript; available in PMC 2014 September 15.

2.2. Immunogenicity in mice

Statistical methods—Kruskal–Wallis ANOVA was used to compare the anti-Vi ELISA

Vaccine. Author manuscript; available in PMC 2014 September 15.

2.3. Protein derivatization with ADH

Vi conjugation to derivatized CRM197—Multiple Vi-CRM197 lots were synthesized.

Vi conjugation to derivatized tetanus toxoid—A similar procedure was used to

Vaccine. Author manuscript; available in PMC 2014 September 15.

3.2. Assay for unconjugated and conjugated Vi

To measure the concentration of Vi in solution or in glycoconjugate samples, we developed

Hydrolysis conditions—Hydrolysis of Vi in 4 M TFA at 120 °C for periods up to 2 h

Range of detection—Calibration curves generated from hydrolysis of purified Vi and

Vaccine. Author manuscript; available in PMC 2014 September 15.

purification using acridine orange method failed, as did the quantification of Vi in

3.3. Protein derivatization with ADH

un-derivatized CRM197. As shown by Kossaczka et al. [27], the level of derivatization

3.4. Vi-CRM197 and Vi-TT

Vaccine. Author manuscript; available in PMC 2014 September 15.

Separation on Sephacryl S-1000 was subsequently optimized and elution with 10 mM

Reproducibility of conjugate formation—Three lots of Vi-CRM197 using the same

Varying the ratio of Vi to CRM197—Additional lots of Vi-CRM197 were conjugated

3.5. Immunogenicity of Vi conjugates in mice

Vaccine. Author manuscript; available in PMC 2014 September 15.

In addition to native purified polysaccharides, oligosaccharides obtained by polysaccharide

As shown for other polysaccharides, the introduction of Vi vaccine into routine

Vaccine. Author manuscript; available in PMC 2014 September 15.

By modifying the reaction conditions described in published Vi conjugation methods, we

Based on these preliminary results, additional immunogenicity studies of Vi-CRM197 have

Vaccine. Author manuscript; available in PMC 2014 September 15.

Vaccine. Author manuscript; available in PMC 2014 September 15.

conjugate vaccine in a murine model of salmonellosis. Vaccine. 2006; 24:4312–4320. [PubMed:

extent of O-acetylation in bacterial vaccine polysaccharides by high-performance anion-exchange

Vaccine. Author manuscript; available in PMC 2014 September 15.

1992; 201:343–349. [PubMed: 1632523]