Vaccine 28 (2010) 36793687

Contents lists available at ScienceDirect

Vaccine
journal homepage: www.elsevier.com/locate/vaccine

Reactogenicity and immunogenicity of live attenuated Salmonella enterica

serovar Paratyphi A enteric fever vaccine candidates
Kenneth L. Roland , Steven A. Tinge 1 , Sims K. Kochi 2 , Lawrence J. Thomas, Kevin P. Killeen 3
Celldex Therapeutics, Inc., 119 Fourth Avenue, Needham, MA 02494, United States

a r t i c l e i n f o a b s t r a c t

Article history: Eight Salmonella enterica serovar Paratyphi A strains were screened as candidates to create a live attenu-
Received 6 August 2009 ated paratyphoid vaccine. Based on biochemical and phenotypic criteria, four strains, RKS2900, MGN9772,
Received in revised form 1 March 2010 MGN9773 and MGN9779, were selected as progenitors for the construction of phoPQ mutant deriva-
Accepted 10 March 2010
tives. All strains were evaluated in vitro for auxotrophic phenotypes and sensitivity to deoxycholate and
Available online 23 March 2010
polymyxin B. All phoPQ mutants were more sensitive to deoxycholate and polymyxin B than their
wild-type progenitors, however MGN10028, MGN10044 and MGN10048, required exogenous purine for
Keywords:
optimal growth. Purine requiring strains had acquired point mutations in purB during strain construction.
Live Salmonella vaccine
Paratyphi
All four mutants were evaluated for reactogenicity and immunogenicity in an oral rabbit model. Three
Reactogenicity strains were reactogenic in a dose-dependent manner, while one strain, MGN10028, was well-tolerated
Rabbit model at all doses administered. All phoPQ strains were immunogenic following a single oral dose. The in
vitro prole coupled with the favorable reactogenicity and immunogenicity proles render MGN10028
a suitable live attenuated Paratyphi A vaccine candidate.
2010 Elsevier Ltd. All rights reserved.

1. Introduction
Salmonella enterica serovar Typhi and Salmonella enterica
serovars Paratyphi A and B are the causative agents of enteric
fever. Estimates of the worldwide incidence of paratyphoid fever
range from about one-tenth to one-fourth that of typhoid fever
amounting to approximately 5 million cases a year [1]. However,
there has been a recent sharp increase in the incidence of paraty-
phoid fever in a number of Asian countries [2,3], in particular China
[4] and India [5]. Although typhoid and paratyphoid infections
occur worldwide, they are considered rare in the developed world.
The majority of infections in the developed world are in travelers
returning from areas endemic with enteric fever [68]. Recently,
a rise in the proportion of drug resistant S. Paratyphi strains has
been reported [2,9,10]. Given the emergence of antibiotic resistant
strains and an increase in the frequency of enteric fever caused by S.
Corresponding author. Present address: Center for Infectious Diseases and Vacci-
nology, Arizona State University, 1001 S McAllister, Tempe, AZ 85287, United States.
Tel.: +1 480 727 9992.
E-mail address: kenneth.roland@asu.edu (K.L. Roland).
1
Present address: Research Compliance, Saint Louis University, St. Louis, MO
63104, United States.
2
Present address: Department of Microbiology and Molecular Genetics, Harvard
Medical School, Boston, MA 02115, United States.
3
Present address: Matrivax Research & Development Corp., Boston, MA 02118,
United States.
Paratyphi A, the need for an effective vaccine has become apparent
[11].
There is currently no licensed vaccine to protect against S.
Paratyphi A despite an obvious and growing need. A conjugated O-
antigen vaccine candidate was tested in Phase 1 and Phase 2 studies
and was found to be safe and immunogenic following one or two
injections [12]. Interestingly, this conjugated O-antigen vaccine
approach failed to elicit a booster response [12]. Although these
results are encouraging, the preferred vaccine formulation in devel-
oping countries is oral, single-dose, for logistical reasons including
ease of administration, acceptability, and compliance [13]. Orally
delivered live attenuated vaccines are consistent with this goal
and provide additional advantages, including inexpensive manu-
facture and stimulation of both systemic and local mucosal immune
responses [14].
Live attenuated phoP and phoPQ Salmonella mutants have been
extensively studied for their vaccine potential [15,16]. In vitro
studies demonstrate that the PhoP/PhoQ proteins regulate the
expression of more than 40 genes [17] including those involved
in macrophage survival, as shown for Salmonella enterica serovar
Typhimurium [18] and S. enterica serovar Typhi [19]. In addition,
phoP-regulated genes affect sensitivity of Salmonella to a variety of
antibacterial agents such as bile salts [20] and antimicrobial pep-
tides including melittin [21] and polymyxin B [22]. PhoP is essential
for S. Typhimurium virulence in the mouse typhoid fever model and
S. Typhimurium phoP mutants are avirulent, immunogenic and pro-
tective against wild-type challenge [15,18]. It has also been shown

0264-410X/$ see front matter 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.vaccine.2010.03.019
3680 K.L. Roland et al. / Vaccine 28 (2010) 36793687
Fig. 1. Schematic diagram of the phoPQ1209 deletion in S. Typhi strain Ty800. Arrows
indicate the sites of primer binding for PCR cloning. The DNA between the arrows
was cloned from Ty800 into plasmid pMEG-375 to create plasmid pMEG-2575.
that a phoP mutant of the swine pathogen, Salmonella enterica
serovar Choleraesuis is avirulent in pigs [23]. The phoPQ genes have
also been identied as virulence regulators in a number of other
bacteria [24,25].
Data obtained from preclinical studies, such as those described
above, formed the basis for developing phoPQ Salmonella strains
for use as vaccines in humans. The typhoid vaccine candidate Ty800
is a derivative of S. Typhi strain Ty2 with a 1209-bp deletion remov-
ing all of phoP and part of phoQ (Fig. 1). In a Phase 1 clinical study, S.
Typhi strain, Ty800, was safe and immunogenic following a single
oral dose [26] and elicited anti-S. Typhi specic antibody responses
in a higher percentage of volunteers than four oral doses of the
licensed live attenuated typhoid fever vaccine, Ty21a [26]. Ty800
was recently evaluated in a larger Phase 1/2 trial in which 120 vol-
unteers received escalating doses of vaccine [27] (Mitchell Cohen,
personal communication). A phoPQ S. Typhimurium vector strain,
LH1160, has also been tested in volunteers and shown to be well-
tolerated and immunogenic [28].
Given the clinical success of Ty800 and LH1160, we hypothe-
sized that it is likely the phoPQ attenuation strategy will also be
applicable to S. Paratyphi A. S. Paratyphi and S. Typhi cause sim-
ilar diseases and interact with their human host through similar
pathways. In addition, multi-locus enzyme analysis performed on
the host-adapted serovars revealed a general tendency for clones of
the host-restricted serovars, such as S. Typhi and S. Paratyphi A, to
be less diverse in multi-locus genotype than the broad host range
serovars [29]. The identication of unique genetic polymorphisms
present in both S. Typhi and S. Paratyphi A [30] further supports
the similarities observed in the epidemiology and pathogenicity of
these serovars and supports the premise that vaccine development
based on strategies that have been successful with S. Typhi will
extend to S. Paratyphi A.
We have been developing a preclinical rabbit model to evaluate
the reactogenicity and immunogenicity of orally administered live
Table 1
Preliminary characterization of S. Paratyphi A wild-type strainsa .
Strain # LPS Biochemical prole
RKS2900 Smooth Typical
MGN9772 Smooth Typical
MGN9773 Smooth Typical
MGN9774 Smooth Atypical
MGN9775 Atypical Typical
MGN9778 Not done Atypical
MGN9779 Smooth Typical
MGN9780 Rough Atypical
a
All strains were positive for group A LPS by agglutination with anti-group A antiserum.
b
Strains were evaluated for presence of both large and small plasmids.
c
Antibiotics tested are listed in Section 2.2.
d
trp indicates tryptophan is required for growth.
e
Undened refers to an unknown auxotrophic requirement.
enteric disease vaccines derived from host-restricted pathogens
including the cholera vaccine candidate Vibrio cholerae strain Peru-
15 [31], its derivative, Peru-15pCTB [32] and the S. Typhi vaccine
candidate, Ty800 [33]. Although the intranasal mouse model is
well-established and is often used for evaluating the immunogenic-
ity of S. Typhi vaccine candidates [34,35], the rabbit model involves
oral dosing (i.e. the same route of administration used in humans)
and may therefore generate a more analogous immune response
and provide more relevant insights into vaccine/host interactions.
In this work, we characterized a series of wild-type S. Paratyphi
A isolates to determine their innate suitability as to serve as vac-
cine parental strains. We described the introduction of the phoPQ
mutation into several parental strains, their subsequent characteri-
zation in vitro and preclinical evaluation of their reactogenicity and
immunogenicity in an oral rabbit model.
2. Materials and methods
2.1. Bacterial strains, plasmids, antibiotics and growth media
The S. Paratyphi A parent strains used in this study (Table 1) were
obtained from the Salmonella Stock collection at the University of
Calgary, Alberta Canada. Mutant phoPQ derivatives are described
in Table 2. S. Typhi strain Ty800 is a phoPQ1209 derivative of Ty2
[26]. The S. Typhi strain Ty21a is a live attenuated typhoid vac-
cine [36]. Escherichia coli strain MGN617 was used as the donor
strain for all conjugal transfer procedures [37]. Plasmid pMEG-375
is an R6K pir dependent suicide plasmid that encodes the cat and
sacB genes [38]. Bacteria were routinely grown at 37 C on LB-V, a
vegetable-based rich medium, containing 20 g of HiVeg Luria Broth
(HiMedia Laboratories, PVT. Ltd., Mumbai, India) and 5 g NaCl per
liter. Chloramphenicol was added to 20 g/ml when required. Agar
(Difco, Detroit, MI) was added to a nal concentration of 1.5% for
making solid plate media. M9 minimal medium (Difco) contain-
ing 1% glucose and appropriate supplements was used to evaluate
auxotrophic phenotypes. Adenine was added to a nal concentra-
tion of 5 mM where indicated. MuellerHinton agar was purchased
from Difco. Vaccine strains were grown in LB-V broth containing
5 mM adenine to an optical density at 600 nm (OD600 ) of 1.0 prior to
immunization of rabbits. Cultures were centrifuged, resuspended
in phosphate-buffered saline, pH 7.0 (PBS), adjusted to the desired
dose and administered on the same day. In one experiment, S. Typhi
strain Ty800 cells were heat-killed by incubation at 56 C for 1 h
prior to administration to rabbits.
2.2. In vitro characterization of strains
Assays to determine sensitivity to deoxycholate [20] and
polymyxin B [39] were performed essentially as described.
Lipopolysaccharide proles were assessed by SDS-PAGE [40].
Plasmidb Antibiotic resistancec Auxotrophy
None None trpd
None None None
None Streptomycin trp
3.5 kb None Undenede
None None None
None None Undened
3.5 kb None None
None Streptomycin None
 K.L. Roland et al. / Vaccine 28 (2010) 36793687
Table 2
In vitro characteristics of S. Paratyphi A phoPQ1209 vaccine candidates and parental strains.
Strain Parent
RKS2900
MGN9720 phoPQ RKS2900
MGN9772
MGN10028 phoPQ MGN9772
MGN9773
MGN10044 phoPQ MGN9773
MGN9779
MGN10048 phoPQ MGN9779
Ty2
Ty800 phoPQ Ty2
a
MIC = minimum inhibitory concentration.
b
MBC = minimum bactericidal concentration.
c
ND = not determined.
API20E strips (BioMrieux, Inc., Marcy lEtoile, France) were used
according to the manufacturers instructions. Antibiotic resistance
was determined on MuellerHinton plates using the KirbyBauer
method [41]. For the KirbyBauer test, standard commercial antibi-
otic disks were used, each containing one of the following:
ampicillin, amikacin, amoxicillin, aztreonam, ceftaxime, cefotetan,
ceftazidime, cefalothin, chloramphenicol, ciprooxacin, colistin,
gentamicin, imipenem, kanamycin, nitrofurantoin, novobiocin,
piperacillin, tetracycline, tobramycin, trimethoprim, mezlocillin,
ceftriaxone, metronidazole, streptomycin, azithromycin. Plasmid
preparations were performed with QIAGEN mini-prep kits (QIA-
GEN, Valencia, CA) according to the manufacturers instructions.
The presence of large plasmids was determined by the method
of Kado and Liu [42]. Auxotrophic requirements were determined
by growth on M9 medium spread with various nutritional supple-
ments as described [43]. Growth curves in LB-V were performed
by growing overnight cultures of each strain in LB-V supplemented
with 5 mM adenine. The overnight cultures were used to inoculate
duplicate pairs of asks containing LB-V broth that were then incu-
bated at 37 C at 200 rpm for 5 h in the absence of adenine. After 5 h,
5 mM adenine was added to one of the duplicate asks and growth
continued. Growth was monitored by measuring the OD600 for each
culture at various times.
2.3. Construction of phoPQ deletion mutations in S. Paratyphi A
strains
The phoPQ1209 allele was cloned from Ty800 by PCR using
primers phoPQ1186-1209-Bam (GAAGGGATCCTCGGTCTGTC-
TAAC) and phoPQ2129-2108-XbaI (CGGTCTAGATTTGCTGACGCTC),
yielding a 944-bp DNA fragment containing 537-bp of upstream
sequences, including a portion of purB and 387-bp of phoQ
sequences downstream of the deletion. The PCR product contain-
ing the phoPQ1209 deletion was then digested with restriction
enzymes BamHI and XbaI and ligated into the pMEG-375 digested
with the same enzymes, to produce the phoPQ1209 suicide
plasmid pMEG-2575. Allelic replacement of the wild-type phoPQ
genes was obtained by conjugal transfer of the phoPQ1209
suicide plasmid pMEG-2575 into a selected wild-type S. Paratyphi
A parental strain. Transconjugants were selected on LB-V agar
containing 20 g/ml chloramphenicol. Selection of mutants that
had undergone the second crossover event to replace the wild-
type phoPQ alleles with the phoPQ1209 deletion was performed
on LB-V Agar containing 10% sucrose and no sodium chloride at
25 C [44]. Chloramphenicol-sensitive isolates were screened for
loss of acid phosphatase activity using the agar overlay method
[45]. A PhoP isolate from each parental strain was selected. The
phoPQ1209 deletion was conrmed by PCR amplication of the
3681
Auxotrophy MICa deoxycholate (g/ml) MBCb polymyxin B (g/ml)
c
trp ND 15
trp 1.9 4
None 7.5 15
pur 0.5 4
trp 15 15
trp, pur 0.1 4
trp 7.5 15
trp, pur 0.1 4
trp 15 ND
trp 0.9 ND
mutated region with anking PCR primers phoPQ1186-1209 and
phoPQ2129-2108. Wild-type phoPQ yielded a 2252-bp fragment
and phoPQ1209 yielded a 944-bp fragment, allowing the two
alleles to be easily distinguished. In each case, the resulting PCR
yielded products of the expected size.
2.4. Isolation of S. Paratyphi A LPS
The Darveau and Hancock procedure for isolating bacterial
lipopolysaccharide (LPS) was used to purify S. Paratyphi A LPS from
strain MGN10028 [46]. The puried LPS was resuspended in dis-
tilled water and stored in aliquots at 80 C. The concentration and
purity of sample preparations were determined by LAL assay and
BCA, respectively.
2.5. Immunization regimen and reactogenicity of phoPQ S.
Paratyphi A isolates and Ty800 in rabbits
New Zealand White rabbits (Milbrook Breeding Labs, Amherst,
MA) were deprived of food and water overnight prior to
immunization. At time zero (t = 0), anesthesia was administered
intramuscularly with a cocktail consisting of 50 mg/kg ketamine
and 3 mg/kg xylazine. At t = 20 min, cimetidine (50 mg/kg) was
administered via i.v. injection. Cimetidine was diluted to a 5 ml
administered volume using sterile saline. At t = 35 min, each rab-
bit was given 10 ml 0.5 M NaHCO3 (pH 8) via oral gavage using a
disposable plastic animal feeding tube (Tyco Healthcare, Kendall
Company #155722, feeding tube, 8fr 15 , sterile). At t = 50 min,
another 10 ml bicarbonate buffer was given immediately followed
by a 10 ml dose of freshly grown bacteria in bicarbonate buffer. At
t = 80 min, morphine (5 mg/kg) was administered i.p., diluted to a
5 ml administered volume in sterile saline.
Three health status indicators (lethargy, water and food con-
sumption) were employed using a scoring system in an attempt to
quantify the overall level of reactogenicity. A score of either 0 or 2
was assigned if the rabbit exhibited no or any lethargy, respectively.
A score of either 0 or 4 was assigned if the rabbit exhibited normal or
reduced water consumption, respectively. The third health status
indicator was based on the level of consumption of the approxi-
mately 125 g of food the rabbits received each day and was graded
as follows. If an animal consumed no food, it received a score of 4 for
that day. The remaining scores were based on the approximate per-
centage of available food consumed. If the animal consumed only
up to 25% of its food, it received a score of 3, 2650% it received a
score of 2, 5175% it received a score of 1, and 76100% it received
a score of 0. A total score for the day was assigned to each rabbit
as a sum of the scores of the three health status indicators, and the
sum number for the rst 4 days following inoculation was deter-
3682 K.L. Roland et al. / Vaccine 28 (2010) 36793687
mined. The group average was then calculated for a given vaccine
strain and dose, to normalize the data for the number of animals in
a given group. Rectal temperatures were monitored daily.
Each S. Paratyphi A vaccine candidate was evaluated at 2 108
(low dose), 2 109 (medium dose), and 2 1010 (high dose) CFU
per rabbit, administered by inoculation on day 0 and a boost on
day 43 or 44. S. Typhi strain Ty800 was evaluated in a comparable
vaccine regimen. However, animals receiving S. Typhi strain Ty21a
were evaluated at a dose of 7 108 CFU and boosted on days 2, 4,
and 6 (similar to the clinical protocol). A reactogenicity score for
Ty21a was calculated only for the 4 days following the day 6 boost.
Finally, for the subset of MGN10028-vaccinated rabbits that were
each challenged with 2 1010 CFU of MGN9720 on approximately
day 92, reactogenicity scores were calculated for the 4 days post-
challenge.
2.6. S. Paratyphi A, S. Typhi Ty800 and S. Typhi Ty21a
immunogenicity in rabbits
Blood samples were collected before dosing and then on days
19, 29, 42 (pre-boost), 57, 78 and 92. Due to scheduling issues,
some rabbits were bled 1 or 2 days earlier or later than indicated.
Serum antibody titers against S. Paratyphi A or S. Typhi LPS were
determined by ELISA. S. Typhi LPS was purchased from Sigma (St.
Louis, MO). Briey, Immulon 1B microtiter plates (ThermoLabsys-
tems, Waltham, MA) were coated overnight at 4 C, 100 l/well
with 6.8 g/ml S. Paratyphi A LPS or 10 g/ml S. Typhi LPS diluted
in PBS. After coating the wells, unbound antigen was decanted from
plates, and the plates were blocked for 1 h at 37 C by the addition
of 300 l/well of a casein blocking buffer. The blocking buffer was
decanted from plates and positive control samples or rabbit serum
samples, diluted 1:100 in casein buffer, were serially diluted 2-fold
(100 l/well), and incubated 1 h at 37 C. Following this incuba-
tion, plates were washed 3 times with Tris buffered saline (pH 8.4)
containing 0.1% Tween 20. A 1:20,000 dilution of goat anti-rabbit
IgG horseradish peroxidase conjugate (Southern Biotech, Birming-
ham, Alabama) was added in 100 l to each well and plates were
incubated 1 h at 37 C. Following a wash, TMB substrate mix (KPL,
Gaithersburg, Maryland) was added to each well and incubation
continued for approximately 10 min. The reaction was stopped by
the addition of 2N sulfuric acid and each well was read at A450 .
Antibody titers were determined to be the inverse of the great-
est average sample dilution that was 2-times the plate blank
background. The plate blank contained all reagents except rabbit
sera.
3. Results
3.1. Characterization of S. Paratyphi A wild-type strains
Eight wild-type S. Paratyphi A strains in our collection were
screened towards the identication of suitable parental strains
(Table 1). These strains were screened for a variety of pheno-
types including smooth or rough LPS, biochemical prole on
API20E strips, presence of plasmids, antibiotic resistances and aux-
otrophies. Four strains were eliminated because they possessed
atypical or rough LPS proles (MGN9775, MGN9780), did not
have the expected biochemical prole on API strips (MGN9775,
MGN9778, MGN9780) or had an undened auxotrophy (MGN9774,
MGN9778). trp auxotrophs were not excluded because Ty2-based
S. Typhi vaccine strains, e.g. Ty800 and Ty21a, also require trypto-
phan for growth, and are both well-tolerated and immunogenic
in humans. The remaining four strains, RKS2900, MGN9772,
MGN9773 and MGN9779, were advanced and used as parents for
the construction of phoPQ mutants.
Fig. 2. Growth of phoPQ S. Paratyphi A mutants in LB-V broth with or without the
addition of adenine. Overnight cultures were diluted 1:50 into fresh LB-V broth and
grown with aeration at 37 C. Cultures were grown for 5 h in duplicate asks. Adenine
was added to one of the asks to a nal concentration of 5 mM and incubation was
continued. The arrow in the gure indicates the point at which adenine was added.
Closed symbols represent cultures with adenine.
3.2. Construction and characterization of phoPQ mutations in
selected S. Paratyphi A parental strains
The live attenuated typhoid fever vaccine candidate S. Typhi
strain Ty800 carries the phoPQ1209 allele, a dened 1209-bp dele-
tion that removes all of phoP and a portion of phoQ [26] (Fig. 1). We
subcloned the allele from the Ty800 chromosome into suicide plas-
mid pMEG-375. The resulting plasmid, pMEG-2575, was mobilized
into each S. Paratyphi A strain by conjugation. After screening for
double crossover events on sucrose plates, we obtained phoPQ
derivatives of each suitable parent (Table 2). The presence of the
phoPQ1209 allele was conrmed by PCR and an acid phosphatase
negative phenotype as described in Section 2.3. The mutant strains
produced by this procedure carried both S. Typhi and S. Paratyphi
A DNA sequences anking the phoPQ deletion site.
3.3. Characterization of S. Paratyphi A vaccine candidates
requirement for purine
We evaluated all of the S. Paratyphi A phoPQ1209 mutants
for growth on minimal medium. Unexpectedly, we discovered that
three of the four mutants, MGN10028, MGN10044, and MGN10048,
required purine for optimal growth (Table 2). To evaluate the extent
of the purine requirements, strains were grown in duplicate pairs
of asks containing LB-V broth for 5 h with no exogenous adenine,
at which time adenine was spiked into one of the duplicate asks to
5 mM with growth continuing for an additional 4 h. S. Typhimurium
strain LH1160, a strict purine auxotroph, was included as a con-
trol [28]. Our results indicate that all phoPQ1209 mutants were
able to grow without a purine supplement (Fig. 2); however the
growth of strains MGN10028, MGN10044, and MGN10048 was sig-
nicantly enhanced by the addition of adenine. Once adenine was
added the growth of these strains was similar to the growth of strain
MGN9720, whose rate of growth was not inuenced by the addition
of adenine (Fig. 2). As expected, LH1160 did not grow until adenine
was added.
The purB gene is located immediately upstream of the phoPQ
genes (Fig. 1) [47]. We determined the DNA sequence of the purB
genes in each of our vaccine candidate strains. The purB gene in
strain MGN9720 matched the expected sequence from wild-type
S. Paratyphi A [47], consistent with its phenotype as a purine pro-
totroph (Table 2; Fig. 2). However, DNA sequence analysis revealed
 K.L. Roland et al. / Vaccine 28 (2010) 36793687
mutations that led to three amino acid substitutions in the purB
genes of three of the phoPQ1209 strains, MGN10028, MGN10044
and MGN10048. Two of the mutations, a T to C transition at
nucleotide 1264 and a C to A transversion at nucleotide 1349, lead
to F422L and T450N substitutions, respectively, in PurB. The third
mutation was a T to C transition at nucleotide 1151, leading to
substitution of a highly conserved valine residue at a 384 with an
alanine residue (V384A).
An important consideration for live attenuated vaccines is the
stability of the attenuating mutation(s). Thus, we passaged each
of the four S. Paratyphi A phoPQ mutants in LB-V broth supple-
mented with 5 mM adenine for over 40 generations and observed
no reversion to a wild-type PhoPQ+ phenotype (data not shown).
3.4. Characterization of S. Paratyphi A vaccine candidates
sensitivity to deoxycholate
Pathogenic salmonellae are generally resistant to bile salts
and resistance is enhanced in strains in which the phoP regu-
lon is activated. Previous investigators showed that phoP mutants
of S. Typhimurium are sensitive to bile salts [20]. Therefore we
evaluated the minimum inhibitory concentration (MIC) of the
phoPQ strains and their respective parents to deoxycholate, an
unconjugated bile salt. The S. Paratyphi A parental strains were
inherently more resistant to deoxycholate than their respective
phoPQ derivatives (Table 2). MGN9773 was the most resistant
strain with a MIC of 15 g/ml; its phoPQ derivative, MGN10044,
was 150-fold more sensitive. Two other parental strains, MGN9772
and MGN9779, were each 75-fold more resistant to the bile salt
than their respective phoPQ derivatives. When S. Typhi strain
Ty800 and its parent, Ty2 were compared, we observed a 16-fold
difference in deoxycholate sensitivity. Strain MGN9720 phoPQ,
was more resistant to deoxycholate than the other three mutants
(Table 2).
3.5. Characterization of S. Paratyphi A vaccine candidates
sensitivity to polymyxin B
S. Typhimurium phoP mutants are also more sensitive to
the cationic peptide, polymyxin B than PhoPQ+ strains [22]. We
discovered that each of the phoPQ mutants evaluated was 3.75-
fold more sensitive to polymyxin B than their wild-type parents
(Table 2). Interestingly, there was no difference between mutants
in the degree of sensitivity to polymyxin B nor was there any differ-
ence in the level of resistance between parental strains. That is, all
parents possessed the same innate MBC (15 g/ml) and all mutant
derivatives possessed the same MBC (4 g/ml). These results were
similar to what has been reported previously for phoP strains of S.
Typhimurium [22].
3.6. Reactogenicity of phoPQ S. Typhi and S. Paratyphi A strains
in orally immunized rabbits
We evaluated the reactogenicity of phoPQ S. Paratyphi A
mutants by inoculating rabbits as described in Section 2.5. Strains
MGN9720, MGN10028, MGN10044 and MGN10048 were admin-
istered to groups of rabbits (3 rabbits per dose) using doses of
2 108 (low dose), 2 109 (medium dose), and 2 1010 (high dose)
CFU on days 1 and 43 or 44. Animals were monitored for clinical
signs for at least 4 days following each immunization. Following
the initial 2 1010 CFU, high priming dose of MGN9720, all animals
experienced clinical symptoms such as anorexia, low water intake
and lethargy (Table 3). Animals receiving the medium, 2 109 CFU,
priming dose and the low, 2 108 CFU, priming dose of MGN9720
also developed similar symptoms, however the severity and dura-
tion was reduced relative to the high dose animals and the severity
3683
generally titrated with the dose. Following the booster immuniza-
tions of MGN9720 on day 43, all animals exhibited minimal clinical
symptoms generally limited to the 24 h following immunization.
No signicant changes or spikes in temperature were observed
during these studies.
Following the primary dose of MGN10028, MGN10044 or
MGN10048, some animals experienced symptoms manifested as
minor anorexia, low water intake and lethargy for the rst 2448 h
(Table 3). Overall, the severity and duration of the observations
were less than those animals immunized with MGN9720, and
generally dose-dependent. Most animals exhibited limited symp-
toms within the rst 24 h of administration following the day
43/44 immunization, but none thereafter. In general, MGN10044
and MGN10048 had comparable reactogenicity proles whereas
MGN10028 was well-tolerated, with no marked reactogenicity
observed at all doses tested. None of the strains induced either a
shift in body temperature of greater than approximately 1 C or
diarrhea in inoculated animals.
To establish a baseline level of reactogenicity for the phoPQ
S. Paratyphi A mutants, rabbits were immunized with Salmonella
vaccine possessing a known safety prole in humans. Using the
same dosing regimen described above, three groups of rabbits
(n = 3/group) were inoculated with the phoPQ S. Typhi strain
Ty800. One of three animals that received the highest dose,
2 1010 CFU, exhibited symptoms that resolved after 2 days. At the
two lower doses, 2 109 and 2 108 CFU, rabbits exhibited mild or
no symptoms. As an additional control, six rabbits were inoculated
with a 7 108 CFU dose of S. Typhi strain Ty21a, the licensed, live
attenuated typhoid fever vaccine that is well-tolerated in humans.
The reactogenicity score we obtained for Ty21a, 10.3, was adopted
as a benchmark against which the other scores were compared. For
the purpose of this study, any strain or dose that induced a score
of 11.0 or greater was considered to be reactogenic and any strain
that yielded a score less than 11.0 was considered to be generally
well-tolerated.
3.7. Immunogenicity of phoPQ S. Paratyphi A strains and S.
Typhi strains Ty800 and Ty21a in the oral rabbit model
To evaluate the validity of the oral rabbit model for determining
the immunogenicity of host-restricted Salmonella vaccine candi-
dates, we orally inoculated rabbits with a 1 1010 CFU dose of
either live or heat-killed phoPQ S. Typhi strain Ty800 and mon-
itored the serum immune responses to S. Typhi LPS by ELISA.
Our results (Fig. 3) showed that a strong anti-LPS serum immune
response developed in rabbits immunized with live Ty800, con-
sistent with our previous ndings [33]. However, rabbits that were
orally administered heat-killed Ty800 failed to develop a detectable
anti-LPS antibody response. These data indicate that replication is
a pre-requisite for immunogenicity of a single-dose typhoid fever
vaccine.
Subsequently we orally immunized rabbits with phoPQ S.
Paratyphi A vaccine candidates and analyzed the serum IgG
responses against S. Paratyphi A LPS by ELISA. The results
(Fig. 4) demonstrated that immunization with low (2 108 CFU)
or medium (2 109 CFU) doses of each phoPQ S. Paratyphi A
mutant except MGN10028, yielded high anti-LPS titers by about
day 19, indicating that high titers can be achieved with a single
oral dose. Peak reciprocal geometric mean antibody titers (GMT)
in the range of 100,000 or greater developed by about day 57. The
anti-LPS serum IgG titers in animals immunized with MGN10028
were generally both slower to develop and lower in magnitude
compared to those induced by the other S. Paratyphi A vaccine can-
didates. However, in animals inoculated with the highest dose of
MGN10028, 2 1010 CFU, titers greater than 10,000 were achieved
by day 42, with a peak reciprocal GMT of 100,000 (1213 days
3684 K.L. Roland et al. / Vaccine 28 (2010) 36793687

Table 3
Reactogenicity of S. Paratyphi A phoPQ vaccine candidates compared to S. Typhi vaccines Ty21a and Ty800a .

Dose (CFU) MGN9720 MGN10028a MGN10044 MGN10048 Ty800 Ty21a

2 1010
17.3 (3) 3.6 (10) 18.7 (6) 10.9 (8) 7.7 (6) b
2 109 21.7 (3) 3.2 (6) 19.7 (6) 17.0 (6) 6.2 (6) 10.3 (6)c
2 108 8.7 (3) 1.2 (6) 6.7 (6) 5.8 (6) 5.2 (6)
a
Reactogenicity scores were determined using the scoring system described in Section 2 by averaging scores for all the animals in each group over the rst 4 days post-
inoculation. The number in parenthesis represents the number of animals in each group. A dose was considered to be reactogenic if it resulted in an average score of 11 or
greater.
b
not determined at this dose.
c
Rabbits received three doses of Ty21a at 7 108 CFU/dose.

after boosting) by day 56. These immune response titers and kinet-
ics are comparable to those achieved in rabbits immunized with
Ty800 (Fig. 3). Rabbits immunized with the two highest doses of
MGN10028 developed higher titers than rabbits immunized with
the lowest dose, indicating that the immune response was dose-
dependent. Rabbits immunized with four doses of Ty21a were
evaluated for serum IgG responses against S. Typhi LPS (Fig. 4C).
Largely, the anti-LPS antibody reciprocal GMT after four doses was
<200 throughout the time course of the experiment, although one
rabbit reached a titer of 1600 by day 78.
4. Discussion
Enteric fever caused by S. Paratyphi A is becoming of increasing
concern in many parts of the world [48] and it is often endemic in
areas that are also endemic for typhoid fever. Populations at risk
for typhoid fever are being vaccinated and are therefore protected
against S. Typhi infections. One reason for the recent emergence of S.
Paratyphi A is likely due to the fact that vaccines that protect against

3.8. Immunization of rabbits with vaccine candidate S. paratyphi

A MGN10028 protects against challenge with S. paratyphi A
MGN9720

MGN10028 was the least reactogenic S. Paratyphi A mutant we

evaluated (Table 3). Moreover, its immunogenicity prole in rab-
bits was comparable to that of Ty800. Therefore we wanted to
evaluate the protective capacity of vaccine candidate MGN10028.
Three rabbits were orally immunized with either a low, medium
or high dose of MGN10028 on day 1 and boosted on day 43 with
the same dose of MGN10028. Animals were challenged on approx-
imately day 92 with a dose of 2 1010 CFU of MGN9720. Strain
MGN9720 was highly reactogenic when previously tested at this
dose, evoking clinical symptoms including anorexia, low water
intake and lethargy that lasted up to 5 days (Table 3). Immunized
animals exhibited only mild symptoms for the rst 24 h follow-
ing the challenge and little or no symptoms were observed for the
following 4 days (Table 4). These results indicate that immuniza-
tion with MGN10028 can induce an immune response that protects
against the clinical manifestations of a reactogenic S. Paratyphi A
strain.

Fig. 3. Reciprocal anti-LPS serum IgG titers in rabbits orally immunized with live or
heat-killed Ty800. IgG titers were measured by ELISA.
Fig. 4. Reciprocal anti-LPS serum IgG geometric mean titers in rabbits orally immu-
nized with various doses of phoPQ S. Paratyphi A mutants. IgG titers were
measured by ELISA. (A) 2 108 CFU dose, (B) 2 109 CFU dose, (C) 2 1010 CFU dose.
Rabbits were boosted with the same dose of the same strain on days 43 or 44. Rabbits
immunized with Ty21a (C) received 4 doses of 2 107 CFU over a 6-day period.
 K.L. Roland et al. / Vaccine 28 (2010) 36793687
Table 4
Inoculation with S. Paratyphi A phoPQ strain MGN10028 protects against the reac-
togenic effects of strain MGN9720a .
Dose (CFU) Primary Boost Challenge
2 1010 3.6 (10) 2.0 (10) 6.1 (7)
2 109 3.2 (6) 2.8 (6) 7.8 (6)
2 108 1.2 (6) 0.3 (6) 6.1 (6)
a
Rabbits were orally inoculated with the indicated doses of MGN10028 on days
1 and 43. They were challenged on approximately day 92 with 2 1010 CFU of
MGN9720 and scored for 4 days thereafter. The number in parenthesis represents
the number of animals in each group.
S. Typhi are not protective against S. Paratyphi A, although some
protection against S. Paratyphi B has been reported [49]. Therefore,
there is a growing need for an effective paratyphoid fever vaccine
against S. Paratyphi A.
In this work, we report the construction and evaluation of four
phoPQ mutants, each derived from a different S. Paratyphi A pro-
genitor. In light of the current lack of published protocols with
which to characterize and identify attenuated strains of S. Paraty-
phi A, we sought to utilize established in vitro assays that could
be adapted to detect phenotypic markers of virulence attenua-
tion. These techniques were originally developed for use with S.
Typhimurium and validated in murine models of infection. Similar
techniques were successfully employed to develop typhoid fever
vaccines including Ty21a and the candidate vaccine Ty800.
The S. Paratyphi A mutants we constructed have the expected
phenotypes associated with phoP or phoPQ mutants of S.
Typhimurium and S. Typhi, including loss of acid phosphatase activ-
ity and increased susceptibility to deoxycholate and polymyxin B
(Table 1). In addition, the sensitivity of these mutants in vitro to
deoxycholate is similar to that observed in a phoP mutant of S.
Typhimurium shown to be attenuated in mice [18], where a 66-fold
difference in sensitivity was observed between parent and mutant
[20]. The sensitivity of the S. Paratyphi A mutants to polymyxin B is
also consistent with the observation that Salmonella phoP mutants
are sensitive to polymyxin, due to the fact that the products of PhoP-
regulated genes are responsible for modications to the Salmonella
lipid A necessary for polymyxin resistance [22].
The purB gene encodes adenylosuccinate lyase, an enzyme that
catalyzes an essential step in the de novo synthesis of AMP [50,51]
and 430 bp of purB was encoded in our allelic exchange plasmid
pMEG-2575 (Fig. 1). During introduction of the phoPQ deletion,
three of these mutants, MGN10028, MGN10044, and MGN10048,
developed a partial requirement for purines (Fig. 1), presumably
due to one or more of the amino acid changes in PurB leading to a
deciency in adenylosuccinate lyase activity. Two of these muta-
tions, F422L and T450N, are identical to the amino acid sequence
of S. Typhi PurB [52] and appear to have occurred as a result of the
fact that S. Typhi DNA was used as the allelic donor for recombi-
nation. Therefore, it is unlikely that these changes are responsible
for the observed phenotype, although the effects of these substitu-
tions in the Paratyphi/Typhi hybrid protein cannot be completely
ruled out. The third mutation, V384A, changes a highly conserved
valine residue to alanine. We performed a BLAST search using the
PurB amino acid sequence and, compared 100 proteins of simi-
lar sequence. We found that all proteins that were homologous
to S. Paratyphi A PurB had either a valine or a leucine at that posi-
tion. Thus, it seems likely that the V384A amino acid substitution is
responsible for the observed phenotype, although we cannot rule
out the possibility that some other undetected change or com-
bination of changes could be responsible. The purine auxotrophy
of these mutants is not complete, as all three strains grow in the
absence of adenine (Fig. 2). Three of the four isolates (MGN10028,
MGN10044, MGN10048) had reduced growth rates in LB-V broth
compared to MGN9720, whose rate of growth was independent of
3685
adenine supplementation. The addition of adenine to the growth
medium resulted in growth rates that were similar to MGN9720,
indicating that normal growth was restored. Thus, it is likely that
the slow growth phenotype of these strains is due to an inability to
produce adequate amounts of purines.
Each of the S. Paratyphi A phoPQ strains were evaluated for
reactogenicity and immunogenicity in an oral rabbit immunization
model. Interestingly, all of the purB mutants were less reactogenic
than the Pur+ strain MGN9720 (Table 3), although we cannot rule
out the possibility that these differences are unrelated to the purB
genotype and are solely due to the parental strains used. It is clear
that the parental background of the vaccine strain strongly inu-
ences reactogenicity. Of note, while all three purB mutants had
identical mutations, immunization with MGN10028 at any dose
tested was associated with mild symptoms that generally resolved
after 24 h. In contrast, inoculation with MGN10044 and MGN10048
led to more severe symptoms that often lasted more than a day
after immunization. We infer that other, unidentied MGN10028
parental strain phenotypes contribute to this improved safety pro-
le. Based on these reactogenicity criteria, strain MGN10028 was
clearly the best tolerated of the four phoPQ mutants evaluated in
the rabbit model.
Previous results with S. Typhimurium and S. Typhi mutants
carrying deletions or non-leaky mutations in purA or purB were
found to be completely safe, but poorly immunogenic [53,54].
While it appears that the purB mutation in our strains may
be associated with reduced reactogenicity, there was no signif-
icant effect on immunogenicity, as MGN10044 and MGN10048
induced serum IgG responses similar to the Pur+ strain MGN9720
(Fig. 2). Strain MGN10028, which was not notably reactogenic,
elicited the weakest immune response, particularly at the two
lowest doses. This type of phenomenon has been reported in clin-
ical trials of some attenuated S. Typhi strains, where induction
of a strong humoral immune response is often accompanied by
reactogenic side effects [5557]. However, at the highest dose eval-
uated, MGN10028 elicited titers approaching those of the other
strains attained at day 57 post-immunization. Since the purB DNA
sequence for MGN10028, MGN10044 and MGN10048 are identi-
cal, the differences between them with regard to immunogenicity
and reactogenicity are not due to the purB mutation, but must
be due to other inherent genetic and phenotypic differences con-
ferred by their respective parental strains. It is interesting to note
that the three most immunogenic strains are trp auxotrophs, while
MGN10028 displayed no auxotrophy other than the partial require-
ment for purines.
The rabbit immunogenicity model has been a reliable model
at predicting immune responses to the Peru-15 cholera vaccine
[3133]. In this study, the magnitude of anti-LPS serum responses
observed in rabbits immunized with Ty800 were signicantly
greater than what is observed in human vaccinees; serum IgG
responses in patients given the live attenuated typhoid fever vac-
cine Vivotif [Ty21a] or typhoid vaccine candidates such as Ty800
[26] or M01ZH09 [58] have typically been modest as have titers in
patients with nonsevere typhoid [59]. It is more difcult to compare
published, serum anti-LPS responses in paratyphoid fever patients
to our preclinical results as there is a relative lack of published
data. This is likely due to the low number of cases of paratyphoid
fever compared to typhoid fever and the fact that in endemic areas
relatively simple and inexpensive methods are used for analysis
e.g. stool/blood culture. In instances where serological analysis has
been conducted, Widal (O-antigen) or agellar (H-antigen) aggluti-
nation tests are often the preferred method for quantifying serum
titers. In these cases, titers from patients with paratyphoid fever
are again modest [60]. The capacity of Salmonella vaccines to elicit
such elevated anti-LPS titers in rabbits may reect an ongoing expo-
sure to the vaccine organism, perhaps through coprophagia that is
3686 K.L. Roland et al. / Vaccine 28 (2010) 36793687
reseeding the animals intestinal tract. If reingestion did occur, this
could contribute to the overall enhanced immunogenicity of the
vaccine candidates. On the other hand, we have no evidence that
rabbits shed S. Paratyphi A in their feces, and phoPQ Salmonella
strains are typically hypersensitive to low pH, making it unlikely
that more than a few bacteria would survive repeated passage
through a rabbits stomach [61].
The lack of established animal models has hampered evalua-
tion of live vaccines derived from S. Typhi and other host-restricted
bacteria. We have described here an oral rabbit model for evaluat-
ing the reactogenicity and immunogenicity of vaccine candidates
derived from host-restricted species of Salmonella. The potential
utility of this model was supported by our results. Bacterial repli-
cation in the animal was required for the immunogenicity of S.
Typhi Ty800, as evidenced by the lack of immune response seen
in rabbits immunized with heat-killed bacteria (Fig. 3). In addition,
S. Typhi Ty800, which was well-tolerated and highly immunogenic
in clinical studies was also well-tolerated and highly immunogenic
in this model, inducing high titer anti-LPS serum IgG antibod-
ies (Fig. 3). Finally, and perhaps more importantly, immunization
with MGN10028 elicited an immune response that protected rab-
bits against the clinical manifestations of a reactogenic dose of
S. Paratyphi A strain MGN9720. While titers elicited by vaccine
strains of different serotypes are not always directly comparable, it
is worth noting that the magnitude and the kinetics of the immune
responses were similar in rabbits immunized with the phoPQ S.
Paratyphi A mutants and Ty800. The low anti-LPS antibody titers
observed in rabbits immunized with strain Ty21a is consistent
with the fact that this strain is a galE mutant and does not pro-
duce O-antigen in vivo [36]. This oral rabbit immunization model
is also promising model to evaluate vaccine reactogenicity. This
was demonstrated by the relatively low reactogenicity of Ty21a
and Ty800, two clinically evaluated vaccines compared to several of
the highly reactogenic S. Paratyphi A vaccine candidates. However,
additional studies will need to be performed before validating the
rabbit as a reliable S. Paratyphi A vaccine model. Previous experi-
ence has shown that live bacterial strains that are highly attenuated
in animals may still be reactogenic in humans [62,63].
The potential of live, attenuated bacteria (e.g. BCG) as safe, ef-
cacious vaccines has long been recognized. Live attenuated vaccines
offer a number of advantages in terms of dosing convenience and
the induction of strong humoral, mucosal and cellular immunity
compared to multi-dose inactivated organisms or antigen-subunit
based vaccines. This report is, to our knowledge, the rst to describe
an orally administered, live attenuated S. Paratyphi A vaccine
candidate that is based upon a well-studied means of virulence
attenuation. And our results suggest that, phoPQ (purB) S. Paraty-
phi A strain MGN10028 merits consideration as a paratyphoid fever
vaccine candidate for human clinical studies.
Acknowledgements
We thank Cheryl Cloninger, James Storey, Yoeuy Chhung, Eric
Forsberg, Brad Karalius, Kathleen Borrelli, James Boyer, Kristen
Jones and Jacquelyn Caissie for their expert technical assistance.
References
[1] Crump J, Luby S, Mintz E. The global burden of typhoid fever. Bull World Health
Organ 2004;82:34653.
[2] DeRoeck D, Jodar L, Clemens J. Putting typhoid vaccination on the global health
agenda. N Engl J Med 2007;357(September (11)):106971.
[3] Ochiai RL, Wang X, von Seidlein L, Yang J, Bhutta ZA, Bhattacharya SK, et
al. Salmonella paratyphi A rates, Asia. Emerg Infect Dis 2005;11(November
(11)):17646.
[4] Yang J, Dong B, Wang M, Tang Z, Gong J, Li C. Analysis of prevalent status of
paratyphi A and typhi in Guangxi autonomous region in 19942002. Chin Trop
Med 2004;4:17780.
[5] Sood S, Kapil A, Dash N, Das BK, Goel V, Seth P. Paratyphoid fever in India: an
emerging problem. Emerg Infect Dis 1999;5(MayJune (3)):4834.
[6] Caumes E, Ehya N, Nguyen J, Bricaire F. Typhoid and paratyphoid fever: a
10-year retrospective study of 41 cases in a Parisian hospital. J Travel Med
2001;8(NovemberDecember (6)):2937.
[7] Freedman DO, Weld LH, Kozarsky PE, Fisk T, Robins R, von Sonnenburg F, et
al. Spectrum of disease and relation to place of exposure among ill returned
travelers. N Engl J Med 2006;354(January (2)):11930.
[8] Wilde H. Enteric fever due to Salmonella typhi and paratyphi A a neglected and
emerging problem. Vaccine 2007;25(July (29)):52467.
[9] Hasan R, Zafar A, Abbas Z, Mahraj V, Malik F, Zaidi A. Antibiotic resistance among
Salmonella enterica serovars Typhi and Paratyphi A in Pakistan (20012006). J
Infect Dev Countries 2008;2:28994.
[10] Chandel DS, Chaudhry R, Dhawan B, Pandey A, Dey AB. Drug-resistant
Salmonella enterica serotype Paratyphi A in India. Emerg Infect Dis
2000;6(JulyAugust (4)):4201.
[11] Arya SC, Sharma KB. Urgent need for effective vaccine against Salmonella paraty-
phi A, B and C. Vaccine 1995;13(December (17)):17278.
[12] Konadu EY, Lin FY, Ho VA, Thuy NT, Van Bay P, Thanh TC, et al. Phase 1 and phase
2 studies of Salmonella enterica serovar Paratyphi A O-specic polysaccharide-
tetanus toxoid conjugates in adults, teenagers, and 2- to 4-year-old children in
Vietnam. Infect Immun 2000;68(March (3)):152934.
[13] Levine MM. Can needle-free administration of vaccines become the norm in
global immunization? Nat Med 2003;9(January (1)):99103.
[14] Guzman CA, Borsutzky S, Griot-Wenk M, Metcalfe IC, Pearman J, Collioud A, et
al. Vaccines against typhoid fever. Vaccine 2006;24(May (18)):380411.
[15] Galan JE, Curtiss III R. Virulence and vaccine potential of phoP mutants of
Salmonella typhimurium. Microb Pathog 1989;6(6):43343.
[16] Curtiss III R, Wanda SY, Gunn BM, Zhang X, Tinge SA, Ananthnarayan V, et al.
Salmonella strains with regulated delayed attenuation in vivo. Infect Immun
2009;(December).
[17] Belden WJ, Miller SI. Further characterization of the PhoP regulon: identica-
tion of new PhoP-activated virulence loci. Infect Immun 1994;62(November
(11)):5095101.
[18] Miller SI, Kukral AM, Mekalanos JJ. A two-component regulatory system (phoP
phoQ) controls Salmonella typhimurium virulence. Proc Natl Acad Sci USA
1989;86(13):50548.
[19] Baker SJ, Gunn JS, Morona R. The Salmonella typhi melittin resistance gene
pqaB affects intracellular growth in PMA-differentiated U937 cells, polymyxin
B resistance and lipopolysaccharide. Microbiology 1999;145(February (Pt
2)):36778.
[20] van Velkinburgh JC, Gunn JS. PhoP-PhoQ-regulated loci are required for
enhanced bile resistance in Salmonella spp. Infect Immun 1999;67(April
(4)):161422.
[21] Baker SJ, Daniels C, Morona R. PhoP/Q regulated genes in Salmonella typhi
identication of melittin sensitive mutants. Microb Pathog 1997;22(March
(3)):16579.
[22] Gunn JS, Miller SI. PhoP-PhoQ activates transcription of pmrAB, encoding
a two-component regulatory system involved in Salmonella typhimurium
antimicrobial peptide resistance. J Bacteriol 1996;178(December (23)):
685764.
[23] Dominguez-Bernal G, Tierrez A, Bartolome A, Martinez-Pulgarin S, Salguero
FJ, Antonio Orden J, et al. Salmonella enterica serovar Choleraesuis derivatives
harbouring deletions in rpoS and phoP regulatory genes are attenuated in pigs,
and survive and multiply in porcine intestinal macrophages and broblasts,
respectively. Vet Microbiol 2008;130(August (34)):298311.
[24] Moss JE, Fisher PE, Vick B, Groisman EA, Zychlinsky A. The regulatory pro-
tein PhoP controls susceptibility to the host inammatory response in Shigella
exneri. Cell Microbiol 2000;2(December (6)):44352.
[25] Perez E, Samper S, Bordas Y, Guilhot C, Gicquel B, Martin C. An essential role
for phoP in Mycobacterium tuberculosis virulence. Mol Microbiol 2001;41(July
(1)):17987.
[26] Hohmann EL, Oletta CA, Killeen KP, Miller SI. phoP/phoQ-deleted Salmonella
typhi (Ty800) is a safe and immunogenic single-dose typhoid fever vaccine in
volunteers. J Infect Dis 1996;173(6):140814.
[27] Anonymous. AVANTs single-dose oral Ty800 vaccine meets primary endpoints
in Phase 2 study. Med N Today 2008.
[28] Angelakopoulos H, Hohmann EL. Pilot study of phoP/phoQ-deleted Salmonella
enterica serovar Typhimurium expressing Helicobacter pylori urease in adult
volunteers. Infect Immun 2000;68(4):213541.
[29] Selander RK, Beltran P, Smith NH, Helmuth R, Rubin FA, Kopecko DJ, et al. Evolu-
tionary genetic relationships of clones of Salmonella serovars that cause human
typhoid and other enteric fevers. Infect Immun 1990;58(7):226275.
[30] Tracz DM, Tabor H, Jerome M, Ng LK, Gilmour MW. Genetic determinants and
polymorphisms specic for human-adapted serovars of Salmonella enterica that
cause enteric fever. J Clin Microbiol 2006;44(June (6)):200718.
[31] Linnetz EG, Borrelli KM, Mazzola AM, Ryan GS, Sikes KB, Kearney MA, et al.
Optimization of a rabbit model of attenuated Vibrio cholerae immunogenicity.
Contemp Top 2004;43(4):58.
[32] Roland KL, Cloninger C, Kochi SK, Thomas LJ, Tinge SA, Rouskey C, et al. Con-
struction and preclinical evaluation of recombinant Peru-15 expressing high
levels of the cholera toxin B subunit as a vaccine against enterotoxigenic
Escherichia coli. Vaccine 2007;25(December (51)):857484.
[33] Thomas LJ, Forsberg E, Karalius B, Borrelli KM, Marsh H. A rabbit model to
assess the immunogenicity of an attenuated Salmonella typhi oral vaccine. In:
The 27th Annual Meeting of the American College of Toxicology, 2007. 2007.
 K.L. Roland et al. / Vaccine 28 (2010) 36793687
[34] Galen JE, Gomez-Duarte OG, Losonsky GA, Halpern JL, Lauderbaugh CS, Kaintuck
S, et al. A murine model of intranasal immunization to assess the immunogenic-
ity of attenuated Salmonella typhi live vector vaccines in stimulating serum
antibody responses to expressed foreign antigens. Vaccine 1997;15(AprilMay
(67)):7008.
[35] Pasetti MF, Levine MM, Sztein MB. Animal models paving the way for clinical
trials of attenuated Salmonella enterica serovar Typhi live oral vaccines and live
vectors. Vaccine 2003;21(January (56)):40118.
[36] Germanier R, Frer E. Isolation and characterization of galE mutant Ty21a of
Salmonella typhi: a candidate strain for a live, oral typhoid vaccine. J Infect Dis
1975;131(May (5)):5538.
[37] Roland K, Curtiss III R, Sizemore D. Construction and evaluation of a cya
crp Salmonella typhimurium strain expressing avian pathogenic Escherichia
coli O78 LPS as a vaccine to prevent airsacculitis in chickens. Avian Dis
1999;43(JulySeptember (3)):42941.
[38] Dozois CM, Dho-Moulin M, Bree A, Fairbrother JM, Desautels C, Curtiss III
R. Relationship between the Tsh autotransporter and pathogenicity of avian
Escherichia coli and localization and analysis of the Tsh genetic region. Infect
Immun 2000;68(July (7)):414554.
[39] Roland KL, Martin LE, Esther CR, Spitznagel JK. Spontaneous pmrA mutants of
Salmonella typhimurium LT2 dene a new two-component regulatory system
with a possible role in virulence. J Bacteriol 1993;175(July (13)):415464.
[40] Hitchcock PJ, Brown TM. Morphological heterogeneity among Salmonella
lipopolysaccharide chemotypes in silver-stained polyacrylamide gels. J Bac-
teriol 1983;154(April (1)):26977.
[41] Bauer AW, Kirby WM, Sherris JC, Turck M. Antibiotic susceptibility testing by a
standardized single disk method. Am J Clin Pathol 1966;45(April (4)):4936.
[42] Kado CI, Liu ST. Rapid procedure for detection and isolation of large and small
plasmids. J Bacteriol 1981;145(March (3)):136573.
[43] Davis R, Botstein D, Roth J. Advanced bacterial genetics. Cold Spring Harbor,
NY: Cold Spring Harbor Laboratory; 1980.
[44] Blomeld IC, Vaughn V, Rest RF, Eisenstein BI. Allelic exchange in Escherichia
coli using the Bacillus subtilis sacB gene and a temperature-sensitive pSC101
replicon. Mol Microbiol 1991;5(June (6)):144757.
[45] Kier LD, Weppelman RM, Ames BN. Regulation of nonspecic acid phos-
phatase in Salmonella: phoN and phoP genes. J Bacteriol 1979;138(April (1)):
15561.
[46] Darveau RP, Hancock RE. Procedure for isolation of bacterial lipopolysaccha-
rides from both smooth and rough Pseudomonas aeruginosa and Salmonella
typhimurium strains. J Bacteriol 1983;155(August (2)):8318.
[47] McClelland M, Sanderson KE, Clifton SW, Latreille P, Porwollik S, Sabo A, et al.
Comparison of genome degradation in Paratyphi A and Typhi, human-restricted
serovars of Salmonella enterica that cause typhoid. Nat Genet 2004;36(Decem-
ber (12)):126874.
[48] Kothari A, Pruthi A, Chugh T. The burden of enteric fever. J Infect Dev Countries
2008;2(4):2539.
3687
[49] Simanjuntak CH, Paleologo FP, Punjabi NH, Darmowigoto R, Soeprawoto, Toto-
sudirjo H, et al. Oral immunisation against typhoid fever in Indonesia with
Ty21a vaccine. Lancet 1991;338(October (8774)):10559.
[50] Green SM, Malik T, Giles IG, Drabble WT. The purB gene of Escherichia coli K-12
is located in an operon. Microbiology 1996;142(November (Pt 11)):321930.
[51] He B, Smith JM, Zalkin H. Escherichia coli purB gene: cloning, nucleotide
sequence, and regulation by purR. J Bacteriol 1992;174(January (1)):1306.
[52] Deng W, Liou SR, Plunkett 3rd G, Mayhew GF, Rose DJ, Burland V, et al. Com-
parative genomics of Salmonella enterica serovar Typhi strains Ty2 and CT18. J
Bacteriol 2003;185(April (7)):23307.
[53] Levine MM, Herrington D, Murphy JR, Morris JG, Losonsky G, Tall B, et al. Safety,
infectivity, immunogenicity, and in vivo stability of two attenuated auxotrophic
mutant strains of Salmonella typhi, 541Ty and 543Ty, as live oral vaccines in
humans. J Clin Invest 1987;79(March (3)):888902.
[54] OCallaghan D, Maskell D, Liew FY, Easmon CS, Dougan G. Characterization of
aromatic- and purine-dependent Salmonella typhimurium: attention, persis-
tence, and ability to induce protective immunity in BALB/c mice. Infect Immun
1988;56(February (2)):41923.
[55] Frey SE, Bollen W, Sizemore D, Campbell M, Curtiss III R. Bacteremia associ-
ated with live attenuated 8110 Salmonella enterica serovar Typhi ISP1820 in
healthy adult volunteers. Clin Immunol 2001;101(October (1)):327.
[56] Hone DM, Tacket CO, Harris AM, Kay B, Losonsky G, Levine MM. Evaluation in
volunteers of a candidate live oral attenuated Salmonella typhi vector vaccine.
J Clin Invest 1992;90(2):41220.
[57] Nardelli-Haeiger D, Kraehenbuhl JP, Curtiss III R, Schdel F, Potts A, Kelly
S, et al. Oral and rectal immunization of adult female volunteers with
a recombinant attenuated Salmonella typhi vaccine strain. Infect Immun
1996;64(12):521924.
[58] Kirkpatrick BD, Tenney KM, Larsson CJ, ONeill JP, Ventrone C, Bentley M, et al.
The novel oral typhoid vaccine M01ZH09 is well tolerated and highly immuno-
genic in 2 vaccine presentations. J Infect Dis 2005;192(August (3)):3606.
[59] House D, Chinh NT, Diep TS, Parry CM, Wain J, Dougan G, et al. Use of
paired serum samples for serodiagnosis of typhoid fever. J Clin Microbiol
2005;43(September (9)):488990.
[60] Pokhrel BM, Karmacharya R, Mishra SK, Koirala J. Distribution of antibody titer
against Salmonella enterica among healthy individuals in Nepal. Ann Clin Micro-
biol Antimicrob 2009;8:1.
[61] Foster JW, Hall HK. Adaptive acidication tolerance response of Salmonella
typhimurium. J Bacteriol 1990;172(February (2)):7718.
[62] Hone DM, Attridge SR, Forrest B, Morona R, Daniels D, LaBrooy JT, et al. A galE
via (Vi antigen-negative) mutant of Salmonella typhi Ty2 retains virulence in
humans. Infect Immun 1988;56(May (5)):132633.
[63] Noriega FR, Wang JY, Losonsky G, Maneval DR, Hone DM, Levine MM. Construc-
tion and characterization of attenuated aroA virG Shigella exneri 2a strain
CVD 1203, a prototype live oral vaccine. Infect Immun 1994;62(November
(11)):516872.