A bicistronic DNA vaccine containing apical membrane antigen 1 and merozoite surface protein 4/5 can prime humoral and cellular immune responses and partially protect mice against virulent Plasmodium chabaudi adami DS malaria.
ABSTRACT The ultimate malaria vaccine will require the delivery of multiple antigens from different stages of the complex malaria life cycle. In order to efficiently deliver multiple antigens with use of DNA vaccine technology, new antigen delivery systems must be assessed. This study utilized a bicistronic vector construct, containing an internal ribosome entry site, expressing a combination of malarial candidate antigens: merozoite surface protein 4/5 (MSP4/5) (fused to a monocyte chemotactic protein 3 chemoattractant sequence) and apical membrane antigen 1 (AMA-1) (fused to a tissue plasminogen activator secretion signal). Transfection of COS 7 cells with bicistronic plasmids resulted in production and secretion of both AMA-1 and MSP4/5 in vitro. Vaccination of BALB/c mice via intraepidermal gene gun and intramuscular routes against AMA-1 and MSP4/5 resulted in antibody production and significant in vitro proliferation of splenocytes stimulated by both AMA-1 and MSP4/5. Survival of BALB/c mice vaccinated with bicistronic constructs after lethal Plasmodium chabaudi adami DS erythrocytic-stage challenge was variable, although significant increases in survival and reductions in peak parasitemia were observed in several challenge trials when the vaccine was delivered by the intramuscular route. This study using a murine model demonstrates that the delivery of malarial antigens via bicistronic vectors is feasible. Further experimentation with bicistronic delivery systems is required for the optimization and refinement of DNA vaccines to effectively prime protective immune responses against malaria.
- SourceAvailable from: Mai B Margetts[show abstract] [hide abstract]
ABSTRACT: The Plasmodium merozoite surface antigen apical membrane antigen-1 (AMA-1) has previously been shown to provide partial protection to Saimiri and rhesus monkeys immunised with recombinant Plasmodium fragile or parasite-derived Plasmodium knowlesi AMA-1, respectively. In the study reported here we have used the Plasmodium chabaudi/mouse model system to extend our pre-clinical assessment of an AMA-1 vaccine. We describe here the expression of the full-length Plasmodium chabaudi adami AMA-1 and the P. chabaudi adami AMA-1 ectodomain using both baculovirus and Escherichia coli. The ectodomain expressed in E. coli, which contained an N-terminal hexa-his tag, was purified by Ni-chelate chromatography and refolded in vitro in the presence of oxidised and reduced glutathione to generate intramolecular disulphide bonds. In a series of vaccine trials, in both inbred and outbred mice, highly significant protection was obtained by immunising with the refolded AMA-1 ectodomain. Protection was shown to correlate with antibody response and was dependent on intact disulphide bonds. Passive transfer of antibodies raised in rabbits against the refolded AMA-1 ectodomain was also protective. In view of this demonstration that E. coli expression of a soluble P. chabaudi AMA-1 domain can generate a vaccine that is effective in mice, we are pursuing a similar approach to generating a vaccine against P. falciparum for testing in human volunteers.Vaccine 01/1998; 16(2-3):240-7. · 3.49 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Previous studies of Plasmodium falciparum have identified a region of chromosome 2 in which are clustered three genes for glycosylphosphatidylinositol (GPI)-anchored merozoite surface proteins, MSP2, MSP5, and MSP4, arranged in tandem. MSP4 and MSP5 both encode proteins 272 residues long that contain hydrophobic signal sequences, GPI attachment signals, and a single epidermal growth factor (EGF)-like domain at their carboxyl termini. Nevertheless, the remainder of their protein coding regions are quite dissimilar. The locations and similar structural features of these genes suggest that they have arisen from a gene duplication event. Here we describe the identification of the syntenic region of the genome in the murine malaria parasite, Plasmodium chabaudi adami DS. Only one open reading frame is present in this region, and it encodes a protein with structural features reminiscent of both MSP4 and MSP5, including a single EGF-like domain. Accordingly, the gene has been designated PcMSP4/5. The homologue of the P. falciparum MSP2 gene could not be found in P. chabaudi; however, the amino terminus of the PcMSP4/5 protein shows similarity to that of MSP2. The PcMSP4/5 gene encodes a protein with an apparent molecular mass of 36 kDa, and this protein is detected in mature stages of the parasite. The protein partitions in the detergent-enriched phase after Triton X-114 fractionation and is localized to the surfaces of trophozoites and developing and free merozoites. The PcMSP4/5 gene is transcribed in both ring and trophozoite stages but appears to be spliced in a stage-specific manner such that the central intron is spliced from the mRNA in the parasitic stage in which the protein is expressed.Infection and Immunity 06/1999; 67(5):2075-81. · 4.07 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: The control of Plasmodium falciparum malaria by vaccination will require immunization with multiple parasite antigens effectively formulated in combination. In this regard, proteins expressed on the surface of blood-stage merozoites are attractive as vaccine targets given their functional importance in the invasion of erythrocytes and accessibility to serum antibodies. We have utilized a Plasmodium chabaudi vaccine model to begin to evaluate the efficacy of immunization with combined formulations of apical membrane antigen-1 (AMA-1) and merozoite surface protein-1 (MSP-1). Using a pET/T7 RNA polymerase bacterial expression system, we have expressed, purified and refolded recombinant antigens representing the 54 kDa ectodomain of Pc AMA-1 and the 42 kDa C-terminus of Pc MSP-1. Immunization with recombinant Pc AMA-1+Pc MSP-1(42) induced a high level of protection against P. chabaudi malaria with protective efficacy varying with antigen dose, choice of adjuvant, and immunization protocol. Based on the reduction of P. chabaudi parasitemia, Alum proved effective for use with the combination of Pc AMA-1 and Pc MSP-1(42). The use of Quil A was similarly effective with single or combined antigen immunizations, particularly with low antigen dose. In general, serological analysis of prechallenge sera indicated a dominant IgG1 response. For a given formulation, immunization with the combination of Pc AMA-1 and Pc MSP-1(42) elicited IgG responses comparable to those observed following immunization with each antigen alone. However, prechallenge antibody titers alone were not predictive of protective efficacy. While Pc AMA-1 and Pc MSP-1(42) can be effectively formulated in combination, further study is needed to define measurable parameters of protective T cell and B cell responses induced by Pc AMA-1+Pc MSP-1(42) that are predictive of vaccine efficacy.Vaccine 06/2003; 21(17-18):1843-52. · 3.49 Impact Factor
INFECTION AND IMMUNITY, Oct. 2004, p. 5565–5573
Copyright © 2004, American Society for Microbiology. All Rights Reserved.
Vol. 72, No. 10
A Bicistronic DNA Vaccine Containing Apical Membrane Antigen 1
and Merozoite Surface Protein 4/5 Can Prime Humoral and Cellular
Immune Responses and Partially Protect Mice against Virulent
Plasmodium chabaudi adami DS Malaria
A. Rainczuk,1,2T. Scorza,3T. W. Spithill,3,4* and P. M. Smooker5
Department of Biochemistry and Molecular Biology, Monash University, Clayton,1Department of Biotechnology and
Environmental Biology, RMIT University, Bundoora,5Victoria, and The Cooperative Research Centre for
Vaccine Technology, The Bancroft Centre, Royal Brisbane Hospital, Brisbane, Queensland,2Australia,
and Institute of Parasitology3and FQRNT Centre for Host-Parasite Interactions,4
McGill University, Ste. Anne de Bellevue, Quebec, Canada
Received 24 February 2004/Returned for modification 4 April 2004/Accepted 24 June 2004
The ultimate malaria vaccine will require the delivery of multiple antigens from different stages of the
complex malaria life cycle. In order to efficiently deliver multiple antigens with use of DNA vaccine technology,
new antigen delivery systems must be assessed. This study utilized a bicistronic vector construct, containing an
internal ribosome entry site, expressing a combination of malarial candidate antigens: merozoite surface
protein 4/5 (MSP4/5) (fused to a monocyte chemotactic protein 3 chemoattractant sequence) and apical
membrane antigen 1 (AMA-1) (fused to a tissue plasminogen activator secretion signal). Transfection of COS
7 cells with bicistronic plasmids resulted in production and secretion of both AMA-1 and MSP4/5 in vitro.
Vaccination of BALB/c mice via intraepidermal gene gun and intramuscular routes against AMA-1 and
MSP4/5 resulted in antibody production and significant in vitro proliferation of splenocytes stimulated by both
AMA-1 and MSP4/5. Survival of BALB/c mice vaccinated with bicistronic constructs after lethal Plasmodium
chabaudi adami DS erythrocytic-stage challenge was variable, although significant increases in survival and
reductions in peak parasitemia were observed in several challenge trials when the vaccine was delivered by the
intramuscular route. This study using a murine model demonstrates that the delivery of malarial antigens via
bicistronic vectors is feasible. Further experimentation with bicistronic delivery systems is required for the
optimization and refinement of DNA vaccines to effectively prime protective immune responses against
It is believed that the ultimate malaria vaccine will require
the delivery of multiple antigens from different stages of the
complex malaria life cycle (8, 17). Delivery of combinations of
malarial antigens can evoke enhanced immune responses and
protect to a greater extent than can a single antigen alone, as
well as overcome genetic restrictions in different mouse strains
(3, 9, 15, 16, 32).
Combinations of malarial antigens delivered as malarial
DNA vaccines in primates have also resulted in enhanced
levels of cytotoxic T lymphocytes to pre-erythrocytic-stage vac-
cines (32) and enhanced antibody responses to erythrocytic-
stage malarial vaccines (15). It is believed that “first-genera-
tion” DNA vaccines (i.e., delivery of only a single plasmid-
antigen DNA) are not optimal to protect against malaria and
that immune enhancement strategies for DNA vaccination
alone are required for this method of vaccination to be prac-
tical (reviewed in reference 8).
The use of multivalent DNA vaccine expression systems
such as bicistronic vectors may enable more efficient delivery of
antigen in malaria DNA vaccination and promote synergistic
responses between malarial antigens. Testing of viral bicis-
tronic and polycistronic vectors in cancer gene therapy has
been widely used to obtain synergistic effects with use of com-
binations of antitumor genes (reviewed in reference 6). Exam-
ples of nonviral bicistronic vector use as DNA vaccines include
vaccines against hepatitis B (5) and hepatitis C (4), as well as
vaccination against B-cell lymphoma (28).
Bicistronic plasmids utilize an internal ribosome entry site
(IRES) placed between two coding regions. This allows ribo-
somes to attach to mRNA and translate the downstream cod-
ing sequence, while the upstream sequence is translated by
cap-dependent mechanisms (6). IRES sequences have been
found in viral and eukaryotic mRNA, all differing in primary
sequence, nucleotide length, and secondary structure, although
they do share a hairpin nucleotide structure promoting small
ribosomal subunit binding (reviewed in reference 22). The
nucleotide composition of genes flanking the IRES is also an
important factor in the expression of the genes contained
within bicistronic vectors, both in vivo and in vitro (6, 12).
Bicistronic delivery of malarial DNA vaccines may have the
potential to enhance the ability of first-generation DNA vac-
cines to prime an immune response prior to a malaria infec-
tion. In this study, we examined the ability of a bicistronic
DNA vaccine encoding two malarial erythrocytic-stage candi-
date antigens, apical membrane antigen 1 (AMA-1) and mero-
* Corresponding author. Mailing address: Institute of Parasitology,
McGill University, 21111 Lakeshore Rd., Ste. Anne de Bellevue, Que-
bec, Canada H9X 3V9. Phone: (514) 398-8668. Fax: (514) 398-7857.
zoite surface protein 4/5 (MSP4/5), to express both antigens in
vitro and in vivo and to induce antibody and splenic T-cell
responses after immunization of mice. The effect of bicistronic
immunization of mice on parasitemia after lethal erythrocytic-
stage challenge was also assessed.
MATERIALS AND METHODS
Creation of bicistronic plasmids. (i) Bicistronic vector preparation. A bicis-
tronic pIRES vector backbone was obtained from Clontech (Palo Alto, Calif.).
The pIRES vector was first digested with HpaI and BglII restriction enzymes to
remove the neomycin resistance gene cassette contained within this vector. This
resulted in a 1,920-bp fragment with two multiple cloning sites and an IRES
sequence. The pIRES-CMV vector was kindly provided by Stephen Hobbs (In-
stitute of Cancer Research, London, United Kingdom) (13). This vector was also
digested with HpaI and BglII to produce a 2,791-bp fragment containing an
ampicillin resistance gene and a portion of the simian virus 40 polyadenylation
sequence. This fragment was ligated to the 1,920-bp fragment to produce a
4,711-bp empty bicistronic vector (BCAR1 vector) (Fig. 1A).
(ii) BC construct 1 construction. Bicistronic construct (BC construct) 1 con-
tained the DNA encoding the AMA-1 (Ser 22-to-Gln 479) ectodomain of Plas-
modium chabaudi adami DS fused to a tissue plasminogen activator (TPA)
secretion signal in the first position of the BCAR1 vector and the DNA encoding
MSP4/5 (Met 1 to Ser 190) of P. c. adami DS fused to the monocyte chemotactic
protein 3 (MCP-3) DNA coding sequence in the second position of the vector
The 892-bp MCP-3–MSP4/5 sequence was amplified by PCR from the con-
struct generated as described in reference 27, with use of the oligonucleotides
GAAGTTCTAGAATGAGGATCTCTGCCACG (containing an XbaI site) and
GAAGTGCGGCCGCTTATGAATCTGCACTGAG (containing a NotI site).
The resulting PCR product was digested with XbaI and NotI enzymes and ligated
into position 2 of the BCAR1 vector.
The AMA-1 ectodomain was amplified from P. c. adami DS genomic DNA by
PCR with use of oligonucleotides GGGAAGATCTTCCGAAGGTACAGATA
and GAAGTAGATCTTTACTGATTTATGGACT. The resultant AMA-1 frag-
ment was then digested with BglII and inserted into the BglII site of VR1020
(VICAL, San Diego, Calif.) in frame with the TPA secretion signal. The TPA–
AMA-1 sequence was then amplified by PCR from the VR1020–AMA-1
construct with use of oligonucleotides CGCGGAGCTAGCATGGATGCAAT
FIG. 1. (A) The prepared BCAR1 vector used for inserting AMA-1 and MCP-3–MSP4/5. The source of vector DNA used for the final BCAR1
vector is indicated by lines on the outer edges of the vector maps. The left portion was obtained from the pIRES-CMV vector, while the right
portion was obtained from the pIRES vector (Clontech). The sequences inserted into the BCAR1 vector to produce BC construct 1 and BC
construct 2 were amplified by PCR from previously constructed VR1020–AMA-1 and VR1012–MCP-3–MSP4/5 expression vectors. (B) BC
construct 1 contained a TPA secretion signal fused to AMA-1 in position 1, followed by an IRES sequence, and a secretory MCP-3 chemokine
sequence fused to MSP4/5 in position 2 of the vector. (C) BC construct 2 contained a secretory MCP-3 chemokine sequence fused to AMA-1 in
position 1 and an IRES sequence, followed by the MCP-3–MSP4/5 sequence in position 2 of the vector. SV40, simian virus 40.
5566RAINCZUK ET AL. INFECT. IMMUN.
GAAGAGA and GAAGTGAATTCTTACTGATTTATTGGACT, with the re-
sulting product being digested with NheI and EcoRI. The BCAR1 vector (con-
taining the MCP-3–MSP4/5 sequence) was digested with NheI and EcoRI, and
the TPA–AMA-1 sequence was inserted to complete the construct, which was
designated BC construct 1 (Fig. 1B). The BCAR1 vector (Fig. 1A) was also
produced with TPA–AMA-1 in position 1 alone, or with MCP-3–MSP4/5 in
position 2 alone, with use of the same restriction enzyme positions described
above. Both these single-antigen constructs (within the BCAR1 vector) were
used in control experiments.
(iii) BC construct 2 construction. BC construct 2 contained the AMA-1
ectodomain sequence fused to the MCP-3 sequence in the first position of the
BCAR1 vector and MSP4/5 fused to the MCP-3 sequence in the second position
of the vector. BC construct 1, lacking the TPA–AMA-1 sequence in the first
multiple cloning position but containing MCP-3–MSP4/5 in position 2, was used
as a backbone for BC construct 2 (Fig. 1C).
The AMA-1 ectodomain was amplified by PCR from the VR1020–AMA-1
construct (as described for the construction of BC construct 1) with use of
oligonucleotides ATGATGGGATCCGAAGGTACAGATAAT and GAACTG
GATCCTTACTGATTTATTGGACT, and the product was digested with
BamHI. The PCR product was then inserted into the BamHI site of the
VR1012–MCP-3 vector (27) to produce a VR1012–MCP-3–AMA-1 construct.
The MCP-3–AMA-1 (1,695-bp) sequence was amplified using oligonucleotides
GAAGTGAATTCATGAGGATCTCTGCCACG and GAAGTGAATTCTTA
CTGATTTATTGGACT and digested with EcoRI. The BC construct 1 back-
bone was digested with EcoRI, and the MCP-3–AMA-1 DNA sequence was
inserted into the first multiple cloning site to produce BC construct 2 (Fig. 1C).
Expression and purification of recombinant proteins. (i) Purification of
MSP4/5. Expression and purification of the recombinant MSP4/5 ectodomain
protein were performed as described in reference 2. Briefly, the pTrcHis-A/
MSP4/5 vector was transfected into Escherichia coli BL21(DE3) (Novagen, Mil-
waukee, Wis.) for expression of recombinant MSP4/5 protein. Large-scale puri-
fication of the recombinant protein was performed using Talon metal affinity
resin (Clontech) according to the manufacturer’s instructions.
(ii) Purification of AMA-1. The E. coli strain JPA101, containing the AMA-1
ectodomain sequence in the expression vector pDS56/RBSii, was kindly provided
by Robin Anders (La Trobe University, Melbourne, Australia) (1). A colony of
E. coli containing the plasmid encoding the AMA-1 ectodomain was used to
inoculate 50 ml of Superbroth (3.5% tryptone, 2% yeast extract, 0.5% NaCl) with
50 ?g of ampicillin/ml. The 50-ml culture was grown overnight at 37°C, used to
inoculate 500 ml of Superbroth (with 100 ?g of ampicillin/ml), and incubated for
a further 2 h. Induction of AMA-1 protein expression was performed by addition
of 2 mM isopropyl-?-D-thiogalactopyranoside (IPTG; Progen, Darra, Australia),
and incubation continued for 3 h at 37°C. The culture pellet was collected by
centrifugation for 10 min at 3,000 ? g at 4°C.
The purification of the AMA-1 ectodomain was performed under denaturing
conditions. The culture pellet was resuspended in 20 ml of extraction buffer, pH
8 (50 mM NaH2PO4· 2H2O, 6 M guanidine-HCl [pH 8], 300 mM NaCl) with 1
mM phenylmethylsulfonyl fluoride (Sigma, St. Louis, Mo.). The resuspended
pellet was then applied to a French press for three cycles at a pressure of 4 tons.
The lysate was then centrifuged at 11,000 ? g for 15 min at 4°C, and the pH was
corrected to 8 by addition of 5 M NaOH. The lysate was then applied to 2 ml of
Talon metal affinity resin (Clontech) and incubated with rotation for 1 h at 4°C.
The resin was then washed three times with 50 ml of extraction buffer, pH 8,
before being applied to a column. The final wash step involved the addition of 5
mM imidazole to the extraction buffer before elution. The AMA-1 was then
eluted from the resin with use of 1? elution buffer, pH 7 (45 mM NaH2PO4O ·
2H2O, 5 M guanidine-HCl, 270 mM NaCl, 150 mM imidazole).
To refold the AMA-1 ectodomain, the eluted protein was dialyzed at 4°C. The
dialysis buffer (20 mM Tris-HCl, pH 8) was changed three times over 48 h to
remove any remaining elution buffer. The dialysis tube containing the AMA-1
was then immersed in refolding buffer (1 mM reduced glutathione, 0.2 mM
oxidized glutathione, 20 mM Tris-HCl, pH 8) in a volume 25 times greater than
the AMA-1 solution in the dialysis tube. This mixture was then degassed by
vacuum, sealed under nitrogen, and dialyzed overnight at 4°C. The refolded
AMA-1 protein was then stored at ?80°C in 50% glycerol.
Mammalian cell transfection with bicistronic DNA plasmids. Bicistronic plas-
mid constructs were tested for expression in COS 7 cells prior to use in mice.
Freshly grown COS 7 cells were seeded at 2 ? 105cells per 35-mm-diameter
tissue culture well. Cells were grown in complete RPMI 1640 (Invitrogen, Carls-
bad, Calif.) containing 10% fetal calf serum (FCS), 2 mM glutamine, 100 U of
penicillin/ml, and 100 ?g of streptomycin/ml. COS 7 cells were then incubated in
5% CO2until 80% confluent. Three micrograms of plasmid DNA was used to
transfect COS 7 cells with use of Lipofectamine (Invitrogen) in serum-free RPMI
1640 according to the manufacturer’s instructions. Serum-free RPMI 1640 was
changed to complete RPMI 1640 24 h after transfection. After incubation for a
further 2 days at 37°C the cells were washed with phosphate-buffered saline
(PBS), medium was replaced with serum-free RPMI 1640 (to remove any FCS
that may have masked protein detection by subsequent Western blot analysis),
and cells were grown for a further 24 h. The supernatant was then collected and
subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-
PAGE) and Western blotting.
SDS-PAGE and Western blotting. Protein and COS 7 supernatants were
fractionated by SDS-PAGE on 12% (vol/vol) polyacrylamide gels under reducing
conditions and transferred electrophoretically to nitrocellulose membranes. The
membranes were then blocked in 5% milk powder in PBS and 0.05% Tween 20
(Sigma) (PBS-T) overnight at 4°C. The membranes were probed using an anti-
MSP4/5 rabbit antibody or anti-AMA-1 rabbit antibody, followed by an anti-
rabbit immunoglobulin (Ig) conjugated to horseradish peroxidase (Silenus Lab-
oratories, Melbourne, Australia). The reactive antibodies were then visualized by
enhanced chemiluminescence (Amersham, Piscataway, N.J.).
ELISA. Antibody reactivity after vaccination was tested with recombinant
MSP4/5 or refolded AMA-1 protein and measured by enzyme-linked immu-
nosorbent assay (ELISA). Nunc Maxisorp (Nunc, Roskilde, Denmark) ELISA
plates were coated with 0.1 ml of recombinant MSP4/5 or AMA-1 (1 ?g/ml)/well
overnight at 4°C with carbonate-bicarbonate buffer, pH 9.6. Plates were washed
with PBS-T, followed by blocking overnight at 4°C in 5% skim milk powder and
PBS-T. Plates were again washed, and diluted sera were incubated at 37°C for
2 h. After plates were washed again with PBS-T, total humoral responses were
obtained with horseradish peroxidase-conjugated sheep anti-mouse Ig (Silenus)
diluted 1:2,000 and incubated for 1 h, followed by washing and addition of
substrate. After a final washing, the ELISA product was developed by addition
of the substrate 3,3?,5,5?-tetramethylbenzidine (Sigma). Absorbance was mea-
sured at 450 nm, and titers were defined as the highest dilution required for an
absorbance of 0.2.
Isolation of plasmid DNA and construction of vaccination cartridges. DNA
plasmid constructs were transfected into E. coli DH5? and grown on solid agar
medium containing 50 ?g of ampicillin/ml prior to inoculation into liquid Luria-
Bertani (LB) medium (1% tryptone, 0.5% yeast extract, 1% NaCl) containing 50
?g of ampicillin/ml. Inoculated LB medium (with 50 ?g of ampicillin/ml) was
grown with shaking at 37°C overnight. Plasmid preparation and endotoxin re-
moval were performed using the Qiagen endotoxin-free plasmid Giga kit accord-
ing to the manufacturer’s instructions (Qiagen, Inc., Valencia, Calif.). Purified
DNA was precipitated onto gold microcarriers, and these were attached to
plastic supports per the manufacturer’s recommendations (Bio-Rad Laborato-
ries, Hercules, Calif.). DNA was combined with gold microcarriers at a ratio of
100 ?g of DNA to 50 mg of carriers. Each projectile contained approximately 1
?g of DNA.
Mice and vaccination. All mice were BALB/c, female, and 5 to 6 weeks of age
at the time of first vaccination. DNA-vaccinated mice received three immuniza-
tions at 2-week intervals. For intraepidermal (i.d.) DNA vaccination, the abdom-
inal region was shaved and particles containing 1 ?g of DNA were delivered by
the Helios gene gun (Bio-Rad Laboratories) with a pulse of helium gas at 400
lb/in2. Intramuscular (i.m.) DNA plasmids were delivered into the tibialis ante-
rior muscle (100 ?g total) in PBS.
In vitro spleen cell proliferation. Spleen cell proliferation was performed as
described in reference 26. Briefly, recombinant AMA-1 and MSP4/5 were further
purified using Detoxi-Gel (Pierce, Rockford, Ill.) to remove any endotoxin con-
tamination and reduce nonspecific proliferation according to the manufacturer’s
instructions. The recombinant proteins were used to stimulate splenocytes at a
final concentration of 5 ?g/ml. As a control for cell viability, splenocytes were
stimulated with concanavalin A (Sigma) at a final concentration of 2.5 ?g/ml.
Splenocytes were cultured for 96 h in flat-bottomed microtiter plates in triplicate
at a final concentration of 5 ? 106cells/ml (106cells/well) and pulsed with 1 ?Ci
of [3H]thymidine (Amersham Biosciences Corp.)/well for 18 h before harvesting.
The splenocytes were harvested onto glass fiber filter mats (Saktron Instruments
Inc., Sterling, Va.) with use of an automated cell harvester (Saktron), and
incorporated radioactivity was measured using a liquid scintillation counter (Per-
kin-Elmer Life Sciences, Wellesley, Mass.).
Phagocytosis assays. Phagocytosis assays were performed according to the
method described in reference 25 with the following modifications. Macrophages
were obtained from BALB/c mice by peritoneal lavage with 9 ml of ice-cold 0.34
M sucrose in PBS (pH 7.2). The cells were then centrifuged at 1,000 ? g for 10
min at 4°C. Peritoneal cell exudates were resuspended in complete RPMI 1640
(Invitrogen) supplemented with 10% FCS, 2 mM glutamine (Invitrogen), 0.05
mM 2-mercaptoethanol (Sigma), and penicillin-streptomycin (100 U/ml; Invitro-
gen) to a final concentration of 2 ? 106cells/ml. Eight-well chamber slides
VOL. 72, 2004 BICISTRONIC DNA VACCINATION AGAINST RODENT MALARIA 5567
(Nalge Nunc International Corp., Naperville, Ill.) were used, with 4 ? 105
macrophages added to each well. The macrophages were allowed to adhere for
2 h at 37°C with 5% CO2. After 2 h, nonadherent T cells were removed by careful
washing with 1 ml of 37°C RPMI 1640. Fresh complete RPMI 1640 was added,
and macrophages were left to adhere for a further 2 h. During this time, fresh P.
c. adami DS-infected red blood cells (IRBC; 108IRBC/ml, containing tropho-
zoites and schizonts at approximately 40 to 50% parasitemia) in complete RPMI
1640 were purified on Ficoll gradients (Amersham). The IRBC were washed
twice with complete RPMI 1640. After washing, IRBC pellets were placed into
1.5-ml centrifuge tubes in 15-?l aliquots. Thirty microliters of PBS and 1 ?l of
sera obtained from groups vaccinated with the bicistronic constructs and empty
vector controls were then added to the IRBC pellets and incubated for 1.5 h at
37°C with shaking. After 1.5 h, IRBC and sera were added to adherent macro-
phages and incubated for 3 h at 37°C with 5% CO2. The eight-well slides were
then washed four times with PBS to remove nonadherent macrophages and
noningested IRBC. Noningested (but adherent) IRBC were lysed by incubation
of the slides with cold water for 20 s, followed by washing with PBS. The
eight-well slides were then fixed and stained using Kwik Diff staining solution
(Terno Shandon, Pittsburgh, Pa.). The percentage of macrophages ingesting
IRBC was quantified by examination of 300 cells per individual sample by light
Infection of mice, blood sampling, and parasitemia measurements. Blood
from an infected mouse with a known parasitemia (1 to 10%) was taken and
immediately diluted in PBS to give the required dosage (105infected RBC/dose).
Mice were infected by intraperitoneal injection at day 0, and parasitemia was
assessed from day 6 through the period of crisis until the resolution of para-
sitemia. Infection levels were assessed by Giemsa staining of tail smears. Mean
peak parasitemia levels and days to peak parasitemia were compared with use of
a Student t test.
Analysis of survival curves. Survival curves for vaccinated and control mice
were compared using the Mantel-Haenszel test. Statistical analysis was per-
formed using Prism 3.02 software (GraphPad, San Diego, Calif.).
Expression of protein encoded by bicistronic plasmids in
vitro. The ability of mammalian cells to secrete MSP4/5 or
AMA-1 after transfection with bicistronic DNA vaccine plas-
mids was tested in COS 7 cells. Proteins encoded by the con-
structs were secreted into the culture supernatant in vitro (Fig.
2) and detected by Western blotting after probing with either
anti-AMA-1 or anti-MSP4/5 rabbit sera. BC construct 1 con-
tained the TPA–AMA-1 sequence in the first position, fol-
lowed by MCP-3–MSP4/5 in the second position after the
IRES sequence. AMA-1 protein secretion from COS 7 cells
transfected with BC construct 1 was detected (Fig. 2C), as was
AMA-1 protein from COS 7 cells transfected with the
VR1020–AMA-1 plasmid as a positive control (with AMA-1
secreted by the TPA signal sequence contained in both vectors
[Fig. 2A]). The MCP-3–AMA-1 fusion protein encoded in the
first position of BC construct 2 was also secreted into the cell
supernatant after transfection into COS 7 cells and detected
via Western blotting (Fig. 2E). A VR1012–MCP-3–AMA-1
construct was also included as a positive control and detected
in the COS 7 supernatant via Western blotting (Fig. 2F). The
secretion of the MCP-3–MSP4/5 fusion protein (from the
MSP4/5 sequence contained in the second cloning position
after the IRES sequence) was also detected after transfection
of COS 7 cells with both BC construct 1 and BC construct 2.
However, the signal strength of the secreted MCP-3–MSP4/5
fusion protein was significantly lower with BC construct 1 (con-
taining a TPA leader sequence in position 1 [Fig. 2G]) than
with BC construct 2 (with an MCP-3 leader sequence in posi-
tion 1 [Fig. 2J]) and the VR1012–MCP-3–MSP4/5 construct
expressing only MCP-3–MSP4/5 [Fig. 2I]). Control superna-
tants collected from COS 7 cells after transfection with bicis-
tronic DNA vaccine vectors alone did not react with specific
rabbit sera to each recombinant protein (Fig. 2B, D, and H).
Humoral responses in mice vaccinated with bicistronic DNA
vaccine constructs. The bicistronic constructs were used to
vaccinate mice by either i.m. injection or i.d. injection with the
gene gun. The resulting antibody responses were measured 2
weeks after the third vaccination (week 6) by ELISA with use
of recombinant MSP4/5 protein or recombinant refolded
AMA-1 protein. As shown in Fig. 3, mice vaccinated with BC
construct 1 via i.d. (Fig. 3A) or i.m. (Fig. 3C) routes produced
a detectable IgG response to both AMA-1 and MSP4/5 by
ELISA, with the titer to AMA-1 (untargeted) being signifi-
cantly higher than the titer observed with MSP4/5 by the i.d.
route (Fig. 3A; P ? 0.02). A higher mean IgG antibody titer to
AMA-1 than to MSP4/5 was also detected after vaccination
with BC construct 1 via the i.m. route, although this was not
statistically significant (Fig. 3C; P ? 0.15).
Vaccination of mice by the i.d. route with MCP-3–AMA-1 in
position 1 of BC construct 2 did not promote an enhanced
antibody response to AMA-1, whereas the mean MSP4/5 an-
tibody response was greater (Fig. 3B). The antibody response
to recombinant AMA-1 was low (titer, 1/200) and unable to be
definitively distinguished above the background level of reac-
tivity observed with negative control serum (titer, 1/100). This
is in contrast to the use of a TPA–AMA-1 sequence in the first
position of BC construct 1, which resulted in high antibody
FIG. 2. Western blot of supernatants taken from COS 7 cells trans-
fected with bicistronic and monocistronic DNA vectors containing
antigen-encoding inserts. Western blot analyses were carried out using
equivalent proportional volumes of supernatant loaded per lane. The
Western blot was probed with anti-AMA-1 rabbit sera (A to F) or
anti-MSP4/5 rabbit sera (G to J). Proteins expressed by cells trans-
fected with each construct were secreted into the culture supernatant.
Control vectors (B, D, and H) not containing inserts did not react with
rabbit sera. The sizes of the expressed AMA-1 (80-kDa) and MCP-3–
MSP4/5 (49-kDa) proteins are indicated.
5568 RAINCZUK ET AL.INFECT. IMMUN.
responses to AMA-1 but low responses to MSP4/5 antibodies
when delivered via i.d. or i.m. routes (Fig. 3A and C). There-
fore, BC construct 2 was excluded from future experiments due
to the lack of a detectable antibody response to AMA-1.
As a comparative control experiment, mice (five per group)
were vaccinated with the untargeted monocistronic VR1020–
AMA-1 construct via i.d. (gene gun) or i.m. routes. Control
mice (five per group) vaccinated with the VR1020 vector via
i.m. and i.d. routes showed no antibody response to AMA-1
(data not shown). For this control experiment, sera were
pooled and an ELISA was performed in response to refolded
recombinant AMA-1 (Fig. 3D). Although these data cannot be
directly compared statistically to the data generated by the
bicistronic constructs containing AMA-1 (Fig. 3A and C), the
results suggest that high antibody production to AMA-1 can be
induced by using a bicistronic construct relative to the
VR1020–AMA-1 monocistronic construct (Fig. 3D), i.e., bicis-
tronic expression from position 1 of the construct was not
deleterious to antigenicity. Similar high titers can be induced
by vaccination of mice with a monocistronic MCP-3–MSP4/5
vaccine (27). Vaccination using the empty DNA vaccine
BCAR1 control vector did not produce a detectable IgG re-
sponse to MSP4/5 or AMA-1 protein (data not shown).
Cellular immune responses induced by bicistronic DNA vac-
cination. To evaluate cellular responses to P. c. adami DS
antigens induced by bicistronic vaccination, groups of mice
were vaccinated i.d. or i.m. with BC construct 1 (containing
TPA–AMA-1 and MCP-3–MSP4/5) or empty BCAR1 DNA
control vectors. Ten days after the final vaccination, cell pro-
liferation assays were performed on splenocytes from individ-
ual mice stimulated with recombinant AMA-1 and MSP4/5
protein. Significant levels of proliferation were observed after
splenocytes were stimulated with MSP4/5, previously primed
via the i.m. (Fig. 4A; P ? 0.025) and i.d. (Fig. 4C; P ? 0.004)
routes, relative to the proliferation observed with splenocytes
from mice primed with empty BCAR1 vector control DNA.
Stimulation of splenocytes with AMA-1 also resulted in a sig-
nificant increase in proliferation after vaccination by both the
i.m. (Fig. 4B; P ? 0.012) and i.d. (Fig. 4D; P ? 0.027) routes
over that observed with cells from mice vaccinated with empty
vector DNA. No significant differences in the level of prolifer-
ation were observed between the routes of vaccination.
We were interested to determine whether T-cell responses
were influenced by monocistronic delivery relative to bicis-
tronic vaccination. Figure 5 shows a comparison of prolifera-
tion of splenocytes taken from mice vaccinated i.m. with either
the BCAR1 empty vector, monocistronic TPA–AMA-1 (in po-
sition 1 of BCAR1), monocistronic MCP-3–MSP4/5 (in posi-
tion 2 of BCAR1), or BC construct 1. Vaccination with BC
construct 1 resulted in significantly greater splenocyte prolif-
eration in response to specific antigen compared to that for
both monocistronic constructs and the empty vector control.
With MSP4/5 stimulation, splenocyte proliferation by cells
from mice given BC construct 1 was significantly higher than
that observed with cells from mice vaccinated with monocis-
tronic MCP-3–MSP4/5 (P ? 0.01). With AMA-1 stimulation,
the level of splenocyte proliferation after vaccination with BC
construct 1, relative to vaccination with the monocistronic
TPA–AMA-1, was enhanced, although this was not statistically
significant (P ? 0.06). These results suggest that codelivery of
AMA-1 and MSP4/5 in BC construct 1 leads to a synergistic
T-cell response to both antigens.
FIG. 3. IgG responses of mice vaccinated with bicistronic and
monocistronic DNA vaccine constructs. IgG antibody responses were
measured by ELISA. Vaccines were delivered using either 1 ?g of
DNA i.d. (gene gun) or 100 ?g of DNA i.m. Graphs A, B, and C show
data from sera taken from individual mice reacting to both MSP4/5
and AMA-1 recombinant protein after vaccination with a single bicis-
tronic construct. The mean titer is indicated by a bar. (D) Pooled sera
from five mice vaccinated with VR1020–AMA-1 via both i.d. and i.m.
routes. Mice vaccinated with vectors not containing AMA-1 or MSP4/5
inserts did not mount an antibody response (data not shown).
FIG. 4. Proliferation of splenocytes primed with bicistronic vectors.
Shown is in vitro proliferation of splenocytes from individual BALB/c
mice vaccinated i.d. with the gene gun or i.m. by injection. Mice were
vaccinated with BC construct 1 (BC con 1) or the empty BCAR1 vector
three times at 2-week intervals. Splenocytes were harvested 10 days
after the final vaccination. BC construct 1-primed splenocytes were
stimulated with AMA-1 or MSP4/5 and harvested at 72 h, after
[3H]thymidine was added 18 h previously. [3H]thymidine incorporated
by cells was then measured. Splenocytes from all individual cultures
responded to concanavalin A stimulation (data not shown). The mean
counts per minute (bars) are shown. Statistical analysis was performed
using an unpaired t test.
VOL. 72, 2004 BICISTRONIC DNA VACCINATION AGAINST RODENT MALARIA5569
Phagocytosis of P. c. adami DS-IRBC by macrophages after
incubation with sera from mice vaccinated i.m. with BC con-
struct 1. Opsonization of IRBC and subsequent internalization
and destruction by macrophages have been shown to be a
major factor contributing to a reduction in parasitemia during
crisis in P. chabaudi mouse models (24–26). The opsonizing
capacity of prechallenge sera taken from mice vaccinated i.m.
with BC construct 1 or the empty BCAR1 vector was assessed.
Figure 6 shows that the macrophages incubated with sera from
mice vaccinated i.m. with BC construct 1 ingested significantly
more IRBC than did macrophages incubated with sera from
mice vaccinated with the empty BCAR1 (P ? 0.001) vector.
Sera from mice vaccinated i.d. with BC construct 1 also opso-
nized IRBC (data not shown).
Efficacy of the bicistronic DNA vaccine. (i) Trial 1. Trial 1
contained six female BALB/c mice per group vaccinated with
BC construct 1. Mice were vaccinated i.m. (100 ?g) by injec-
tion or i.d. (1 ?g) with the gene gun three times at 2-week
intervals and challenged with 100,000 P. c. adami DS-IRBC 2
weeks after the final vaccination. This trial represents an ex-
tremely stringent test of a vaccine due to the virulent nature of
the challenge with P. c. adami DS. Figure 7 shows survival
curves of mice challenged with lethal P. c. adami. There were
no significant differences in survival between mice vaccinated
i.m. or i.d. with the BC construct 1 and control mice, with all
mice dying by day 13 (Fig. 7A and B). However, 33% of control
mice died at day 10, compared to the first deaths occurring at
day 11 for mice vaccinated with the BC construct 1 i.m. (Fig.
7A). The parasitemia levels of mice challenged in this trial,
however, were influenced by the i.m. administration of bicis-
tronic vectors, regardless of the lack of survival. Figure 7A.i
shows the percent parasitemia measured from day 6 postinfec-
tion in i.m. vaccinated mice. No significant differences were
found in the peak parasitemia levels between mice vaccinated
with BC construct 1 i.m. and the control vaccinated with
BCAR1 i.m. at day 9 postinfection (Fig. 7A.i). However, four
out of six control mice reached their peak parasitemia at day 9,
with the parasitemia of two remaining mice continuing to rise
FIG. 5. Proliferation of splenocytes primed with bicistronic vectors
via the i.m. route. Shown is in vitro proliferation of splenocytes from
individual BALB/c mice (six per group) vaccinated i.m. by injection
three times at 2-week intervals with either BC construct 1, BCAR1,
monocistronic TPA–AMA-1, or monocistronic MCP-3–MSP4/5 (in
the BCAR1 vector). Splenocytes were harvested 10 days after the final
vaccination. BC construct 1-primed splenocytes were stimulated with
AMA-1 or MSP4/5 and harvested at 72 h, after [3H]thymidine was
added 18 h previously. [3H]thymidine incorporated by cells was then
measured. Splenocytes from all individual cultures responded to con-
canavalin A stimulation (data not shown). The standard errors of mean
counts per minute (stimulated minus unstimulated background counts
per minute) are shown. Statistical analysis was performed using an
unpaired t test.
FIG. 6. Phagocytosis of P. c. adami-IRBC preincubated with sera
from individual mice vaccinated i.m. with BCAR1 or BC construct 1.
The percentage of macrophages containing IRBC was calculated from
a total of 300 macrophages counted for each mouse sample. The
means (solid bars) are shown. Groups were compared using a paired t
test. ?, significantly different from i.m. BCAR1 control vector.
FIG. 7. (A and B) Bicistronic trial 1 survival curves. Six mice per
group were vaccinated i.m. by injection with 100 ?g of DNA or 1 ?g via
gene gun three times at 2-week intervals. Mice were challenged with
100,000 P. c. adami DS-IRBC. There were no significant differences
between survival curves of control mice and those of vaccinates, as
determined by the areas under the curves with use of the Mantel-
Haenszel test for comparing survival curves. (A.i and B.i) Bicistronic
trial 1 parasitemia curves. Smears were taken from individual mice (six
mice per group) from day 6 postinfection, with 300 to 400 cells counted
5570 RAINCZUK ET AL.INFECT. IMMUN.
until day 11 (data not shown). In contrast, there was a sharp
drop in parasitemia by day 12 for the two surviving mice vac-
cinated with BC construct 1 i.m., which outlived BCAR1 i.m.
vaccination control mice and had almost resolved parasitemia
before death at day 13 (Fig. 7A.i). Vaccination with BC con-
struct 1 i.d. did not result in a significant reduction in peak
parasitemia compared to that for BCAR1 vector control ani-
mals (Fig. 7B.i).
(ii) Trials 2 to 4. Figure 8A and B show the survival (and
parasitemia) curves from the second bicistronic vaccine trial.
BCAR1 control mice in trial 2 followed a pattern of survival
similar to that in trial 1, and all mice died by day 11 postchal-
lenge regardless of vaccination route. Mice vaccinated with BC
construct 1 i.m. showed a significantly enhanced survival com-
pared to that of the control group vaccinated with BCAR1,
with 50% of mice surviving (Fig. 8A; P ? 0.03). There was no
significant difference in kinetics of survival of mice between
delivery of BC construct 1 via the i.d. route and the empty
vector (Fig. 8B), although 33% of vaccinated mice survived.
The survival found after BC construct 1 i.d. delivery in this
experiment is comparable to that seen using a VR1012–MCP-
3–MSP4/5 construct (expressing the same sequence as con-
tained in position 2 of BC construct 1) delivered i.d. as a
monocistronic construct (27). Vaccination via the i.d. route
with VR1020–AMA-1 (containing the TPA–AMA-1 sequence
as used in BC construct 1) has no effect on survival (data not
The peak parasitemia of mice vaccinated i.m. with BC con-
struct 1 was significantly reduced by an average of 19.4% com-
pared to that of mice vaccinated i.m. with the empty control
vector (Fig. 8A.i; P ? 0.005). This does not occur when the
antigens are delivered as monocistronic constructs with use of
either VR1020–AMA-1 (data not shown) or MCP-3–MSP4/5
(27). As in trial 1, vaccination via the i.d. route did not have any
effect in reducing the parasitemia, with the BC construct 1
group peaking before the empty vector control group (Fig.
8B.i). In i.d. vaccination with VR1012–MCP-3–MSP4/5 with
use of the P. c. adami DS model, there is no effect on para-
sitemia, although survival is increased relative to that of empty
vector control groups as reported previously (27).
The experiment involving the i.m. route of vaccination with
the BC construct 1 vaccine was repeated a further two times to
confirm the observations from trials 1 and 2 (Fig. 8C and D).
In trial 3, a significant difference in the rate of survival was
observed in mice vaccinated with BC construct 1 (Fig, 8C: P ?
0.03) and the peak parasitemia of vaccinated mice was also
significantly reduced by an average of 10.4% (compared to i.m.
delivery of the empty control vector) (Fig. 8C.i; P ? 0.006). In
the fourth trial, as shown in Fig. 8D and D.i, the BC construct
1 vaccine had a significant effect on survival rate (33% surviv-
ing); however, there was no significant effect on parasitemia
The data from these experiments suggest that codelivery of
AMA-1 and MCP-3–MSP4/5 (with both antigens being se-
creted) in a bicistronic DNA vaccine construct by the i.m. route
appears to significantly increase survival and reduce para-
In this study it has been shown that a bicistronic DNA
vaccine vector can express two candidate malaria vaccine an-
tigens, evoke humoral and cellular immune responses, reduce
parasitemia, and partially protect mice against lethal P. c.
adami DS challenge. The aim of this study was not to directly
compare the efficacies of bicistronic and monocistronic vectors
but to evaluate the potential of bicistronic vectors as a delivery
platform for multivalent vaccines against malaria. To date,
there are no reports evaluating the potential of bicistronic
vectors to deliver malarial antigens, even though a multistage
and multiantigen malarial vaccine is believed to be optimal to
protect against malaria (8, 18).
The bicistronic constructs were shown to express both en-
coded antigens in vitro. The transfection of BC construct 1
(containing a TPA leader secretion signal in position 1 and the
FIG. 8. (A to D) Bicistronic trial 2 to 4 survival curves. Six mice per
group were vaccinated i.m. by injection with 100 ?g of DNA or 1 ?g via
gene gun three times at 2-week intervals. Mice were challenged with
100,000 P. c. adami DS-IRBC. Significant differences between survival
curves of control mice and those of vaccinates were determined by the
areas under the curves with use of the Mantel-Haenszel test for com-
paring survival curves. The percent survival is shown in panels A to D;
there were no survivors in panel C. (A.i to C.i) Bicistronic trial 2 to 4
parasitemia curves. Smears were taken from individual mice (six mice
per group) from day 4 postinfection, with 300 to 400 cells counted per
smear. There were no significant differences in peak parasitemia for
survival curve D (data not shown). Significant differences in survival
and peak parasitemias between controls and vaccinates are indicated
by an asterisk (unpaired t test).
VOL. 72, 2004 BICISTRONIC DNA VACCINATION AGAINST RODENT MALARIA 5571
MCP-3 sequence at position 2) into COS 7 cells resulted in the
secretion of both the AMA-1 and MSP4/5 antigens in vitro.
However, expression of MCP-3–MSP4/5 (position 2 of BC
construct 1) by COS 7 cells was markedly reduced compared to
that of AMA-1 in position 1 of this construct. This was sup-
ported by the observation that, after vaccination of mice with
BC construct 1 (via i.d. and i.m. routes), low MSP4/5 antibody
titers were detected, relative to an enhanced AMA-1 antibody
response. Again, this emphasizes earlier observations that the
dominant determinants of translation efficiency in bicistronic
vectors are the arrangement and nature of coding sequences in
the mRNA (12).
BC construct 2 was aimed at delivering two targeted malarial
candidate antigens with use of the chemoattractant MCP-3,
since we had found that an MCP-3 vector enhanced the sur-
vival of mice with use of MSP4/5 as a vaccine after DNA
vaccination, relative to a VR1020 construct (27). It was hy-
pothesized that the fusion of the MCP-3 sequence to both
AMA-1 (position 1 of BC construct 2) and MSP4/5 (position 2
of BC construct 2) would recruit dendritic cells to the site of
antigen expression and enhance priming of naı ¨ve T cells by
both antigens (reviewed in reference 21). Transfection of BC
construct 2 into COS 7 cells resulted in both antigens fused to
MCP-3 being secreted into the supernatant, and this was de-
tected via Western blotting. However, after vaccination of
mice, a humoral response to MCP-3–AMA-1 could not be
detected in vivo even though secretion was found to be possi-
ble in vitro. This does not exclude the possibility that a cellular
response may have been evoked in the absence of antibody in
vivo (reviewed in reference 7). It has been found that the
nucleotide composition of genes contained within a bicistronic
construct has a significant effect upon IRES-promoted trans-
lation (12). The results from the present study show that the
design of the bicistronic vectors was critical to ensure efficient
coexpression of encoded malaria antigens.
Vaccination with bicistronic vectors containing rodent ma-
laria homologues of human candidate malarial antigens
AMA-1, MSP4, and MSP5 induced immune responses to both
antigens. The route of delivery (i.m. or i.d.) did not have any
significant effects upon antibody titers or the level of T-cell
proliferation when T cells were stimulated with AMA-1 or
MSP4/5 antigens. These results show that the bicistronic ma-
larial DNA vaccines primed the immune system with both
antigens simultaneously, regardless of delivery route. B-cell-
deficient mice can control erythrocytic-stage malaria infections
by limiting parasite growth, emphasizing the importance of
T-cell-mediated immunity in the mouse malaria model (30,
31). The generation of CD4?T-cell responses against Plasmo-
dium falciparum erythrocytic infection is believed to be of
primary importance, by acting as T helper cells for antibody
responses as well as effector cells and by limiting parasite
growth via antibody-independent cell-mediated immunity (re-
viewed in references 10 and 11).
Survival after i.m. bicistronic vaccination following challenge
with lethal P. c. adami DS was variable and did not strictly
correlate with significant reductions in parasitemia in chal-
lenge trials, although there was a trend toward a reduction in
parasitemia compared to that for control animals over the four
trials. Trials 2 to 4 in particular resulted in a significant delay
in death in mice vaccinated with BC construct 1 i.m. In sum-
mary, over the six vaccine trials evaluating the BC construct 1,
the survival rate in vaccinates was 7 of 36 mice, whereas 0 of 36
Several approaches could be used to optimize the bicistronic
vaccination strategy in order to enhance survival and further
reduce parasitemia. The gene sequences used in BC construct
1 (AMA-1 and MSP4/5) were the native P. c. adami DS gene
sequences, which, like the P. falciparum genes, are highly A/T
rich. Hoffman and Doolan (14) have optimized the codon
usage of P. falciparum genes to more closely reflect codon
usage in mammalian genes, resulting in a 5- to 40-fold en-
hancement of in vitro expression in mammalian cells and 5- to
100-fold-higher antibody titers in outbred mice (7, 14). This is
one method of enhancement that could improve BC construct
1 efficacy. The use of a nonlethal strain of rodent malaria (such
as P. c. adami DK) (20) would allow a single parameter, that of
parasitemia, to be measured in the absence of the complica-
tions due to pathology associated with the P. c. adami DS
challenge: this was found to be a key effect in the present study,
and it appears that the P. c. adami DS model is not the optimal
model to evaluate the effects of malaria vaccines on para-
sitemia. Reducing parasitemia is an important aim for a human
malaria vaccine since this would reduce morbidity (and poten-
tial mortality) in human malaria (23).
The ability of bicistronic constructs in the present study to
codeliver malarial antigens and induce immune responses in
vivo after vaccination establishes a new approach to malaria
vaccine design. The inclusion of MCP-3 in the second position
of the bicistronic construct may have increased the presenta-
tion of the untargeted antigen AMA-1 within BC construct 1.
The migration of dendritic cells (due to MCP-3 expression) to
a single site of bicistronic antigen production may have con-
tributed to vaccine efficacy (as both bicistronically expressed
antigens were in the same cellular location), leading to en-
hanced immune responsiveness and significant reductions in
parasitemia. The bicistronic delivery of the cytokine granulo-
cyte-macrophage colony-stimulating factor (GM-CSF) in other
systems has resulted in significant enhancement of responses to
DNA vaccination. By use of a hepatitis C virus bicistronic
DNA vaccine, it has been shown that the delivery of a bicis-
tronic plasmid containing hepatitis virus antigens and GM-
CSF can significantly enhance T-cell proliferation and anti-
body responses, compared to vaccination with two separate
plasmids (4). This has also been shown for hepatitis B virus
DNA vaccination (5). However, the codelivery of GM-CSF
with 3,000 plasmids from the VR1020 P. c. adami DS genomic
library, as separate monocistronic constructs, was not effective
at reducing peak parasitemia after erythrocytic-stage challenge
with P. c. adami DS (29). Monocistronic coadministration of
Plasmodium yoelii circumsporozoite protein DNA plasmids
and plasmids containing GM-CSF has been shown to result in
increased CD4?and CD8?T cells, antibody production, and
protection against sporozoite challenge in murine studies (33).
The use of GM-CSF and MSP-142as a DNA vaccine in rhesus
monkeys resulted in a rapid induction of antibodies after the
first dose but had no effect on the T-cell response (19). The
delivery of cytokines such as GM-CSF, along with malarial
candidate antigens or genomic-cDNA libraries in bicistronic
vectors, may allow for a more efficient vaccine delivery system,
as seen in hepatitis models. However, the efficiency of immune
5572 RAINCZUK ET AL.INFECT. IMMUN.
responses to combinations of gene pairs within bicistronic con-
structs varies markedly between different constructs (12).
Whether the enhancement of DNA vaccines by cytokines can
be applied to bicistronic malarial erythrocytic-stage vaccines
still remains to be tested.
Our results showing the induction of both antibody and
T-cell responses against, as well as the reduction in parasitemia
for, lethal P. c. adami DS challenge in mice demonstrate that
the delivery of malarial antigens via bicistronic vectors is fea-
sible in the murine model. The optimization and refinement of
bicistronic DNA vaccines for malaria will now need to occur.
This may include codon optimization, codelivery of cytokines,
viral boosting, and testing in different murine malaria models.
We thank S. Hobbs (Institute of Cancer Research, London, United
Kingdom) for providing the pIRES-CMV vector and his helpful dis-
cussion in the initial stages of this work. We thank R. Anders (La
Trobe University) for providing the P. c. adami cDNA sequence, rabbit
antisera to AMA-1, and advice on production of recombinant AMA-1
This work was supported by Monash University, the Australia In-
donesia Medical Research Initiative, the Cooperative Research Centre
for Vaccine Technology, McGill University, the McGill Institute of
Parasitology, the Fonds que ´be ´cois de la recherche sur la nature et les
technologies (FQRNT) Centre for Host-Parasite Interactions, and the
Canada Research Chair program. A. Rainczuk is a recipient of an
Australian Postgraduate Award scholarship and a scholarship from the
Cooperative Research Centre for Vaccine Technology. T. Spithill
holds a Canada Research Chair in Immunoparasitology.
1. Anders, R. F., P. E. Crewther, S. Edwards, M. Margetts, M. L. S. M.
Matthew, B. Pollock, and D. Pye. 1998. Immunisation with recombinant
AMA-1 protects mice against infection with Plasmodium chabaudi. Vaccine
2. Black, C. G., L. Wang, A. R. Hibbs, E. Werner, and R. L. Coppel. 1999.
Identification of the Plasmodium chabaudi homologue of merozoite surface
proteins 4 and 5 of Plasmodium falciparum. Infect. Immun. 67:2075–2081.
3. Burns, J. M., Jr., P. R. Flaherty, M. M. Romero, and W. P. Weidanz. 2003.
Immunization against Plasmodium chabaudi malaria using combined formu-
lations of apical membrane antigen-1 and merozoite surface protein-1. Vac-
4. Cho, J. H., S. W. Lee, and Y. C. Sung. 1999. Enhanced cellular immunity to
hepatitis C virus nonstructural proteins by codelivery of granulocyte mac-
rophage-colony stimulating factor gene in intramuscular DNA immuniza-
tion. Vaccine 17:1136–1144.
5. Chow, Y. H., W. L. Huang, W. K. Chi, Y. D. Chu, and M. H. Tao. 1997.
Improvement of hepatitis B virus DNA vaccines by plasmids coexpressing
hepatitis B surface antigen and interleukin-2. J. Virol. 71:169–178.
6. de Felipe, P. 2002. Polycistronic viral vectors. Curr. Gene Ther. 2:355–378.
7. Doolan, D. L., and S. L. Hoffman. 2001. DNA-based vaccines against ma-
laria: status and promise of the multi-stage malaria DNA vaccine operation.
Int. J. Parasitol. 31:753–762.
8. Doolan, D. L., and S. L. Hoffman. 2002. Nucleic acid vaccines against ma-
laria. Chem. Immunol. 80:308–321.
9. Doolan, D. L., M. Sedegah, R. C. Hedstrom, P. Hobart, Y. Charoenvit, and
S. L. Hoffman. 1996. Circumventing genetic restriction of protection against
malaria with multigene DNA immunization: CD8?cell-, interferon gamma-,
and nitric oxide-dependent immunity. J. Exp. Med. 183:1739–1746.
10. Good, M. F. 2001. Towards a blood-stage vaccine for malaria: are we fol-
lowing all the leads? Nat. Rev. Immunol. 1:117–125.
11. Good, M. F., and D. L. Doolan. 1999. Immune effector mechanisms in
malaria. Curr. Opin. Immunol. 11:412–419.
12. Hennecke, M., M. Kwissa, K. Metzger, A. Oumard, A. Kroger, R. Schirm-
beck, J. Reimann, and H. Hauser. 2001. Composition and arrangement of
genes define the strength of IRES-driven translation in bicistronic mRNAs.
Nucleic Acids Res. 29:3327–3334.
13. Hobbs, S., S. Jitrapakdee, and J. C. Wallace. 1998. Development of a
bicistronic vector driven by the human polypeptide chain elongation factor
1? promoter for creation of stable mammalian cell lines that express very
high levels of recombinant proteins. Biochem. Biophys. Res. Commun. 252:
14. Hoffman, S. L., and D. L. Doolan. 2000. Can malaria DNA vaccines on their
own be as immunogenic and protective as prime-boost approaches to im-
munization? Dev. Biol. (Basel) 104:121–132.
15. Jones, T. R., R. A. Gramzinski, J. C. Aguiar, B. K. Sim, D. L. Narum, S. R.
Fuhrmann, S. Kumar, N. Obaldia, and S. L. Hoffman. 2002. Absence of
antigenic competition in Aotus monkeys immunized with Plasmodium falci-
parum DNA vaccines delivered as a mixture. Vaccine 20:1675–1680.
16. Kedzierski, L., C. G. Black, M. W. Goschnick, A. W. Stowers, and R. L.
Coppel. 2002. Immunization with a combination of merozoite surface pro-
teins 4/5 and 1 enhances protection against lethal challenge with Plasmodium
yoelii. Infect. Immun. 70:6606–6613.
17. Kumar, S., J. E. Epstein, and T. L. Richie. 2002. Vaccines against asexual
stage malaria parasites. Chem. Immunol. 80:262–286.
18. Kumar, S., J. E. Epstein, T. L. Richie, F. K. Nkrumah, L. Soisson, D. J.
Carucci, and S. L. Hoffman. 2002. A multilateral effort to develop DNA
vaccines against falciparum malaria. Trends Parasitol. 18:129–135.
19. Kumar, S., F. Villinger, M. Oakley, J. C. Aguiar, T. R. Jones, R. C. Hed-
strom, K. Gowda, J. Chute, A. Stowers, D. C. Kaslow, E. K. Thomas, J. Tine,
D. Klinman, S. L. Hoffman, and W. W. Weiss. 2002. A DNA vaccine encod-
ing the 42 kDa C-terminus of merozoite surface protein 1 of Plasmodium
falciparum induces antibody, interferon-gamma and cytotoxic T cell re-
sponses in rhesus monkeys: immuno-stimulatory effects of granulocyte mac-
rophage-colony stimulating factor. Immunol. Lett. 81:13–24.
20. Langhorne, J., S. J. Quin, and L. A. Sanni. 2002. Mouse models of blood-
stage malaria infections: immune responses and cytokines involved in pro-
tection and pathology. Chem. Immunol. 80:204–228.
21. Mackay, C. R. 2001. Chemokines: immunology’s high impact factors. Nat.
22. Martinez-Salas, E. 1999. Internal ribosome entry site biology and its use in
expression vectors. Curr. Opin. Biotechnol. 10:458–464.
23. Miller, L. H., D. I. Baruch, K. Marsh, and O. K. Doumbo. 2002. The
pathogenic basis of malaria. Nature 415:673–679.
24. Mota, M. M., K. N. Brown, V. E. Do Rosario, A. A. Holder, and W. Jarra.
2001. Antibody recognition of rodent malaria parasite antigens exposed at
the infected erythrocyte surface: specificity of immunity generated in hyper-
immune mice. Infect. Immun. 69:2535–2541.
25. Mota, M. M., K. N. Brown, A. A. Holder, and W. Jarra. 1998. Acute Plas-
modium chabaudi chabaudi malaria infection induces antibodies which bind
to the surfaces of parasitized erythrocytes and promote their phagocytosis by
macrophages in vitro. Infect. Immun. 66:4080–4086.
26. Rainczuk, A., T. Scorza, P. M. Smooker, and T. W. Spithill. 2003. Induction
of specific T-cell responses, opsonizing antibodies, and protection against
Plasmodium chabaudi adami infection in mice vaccinated with genomic ex-
pression libraries expressed in targeted and secretory DNA vectors. Infect.
27. Rainczuk, A., P. M. Smooker, L. Kedzierski, C. G. Black, R. L. Coppel, and
T. W. Spithill. 2003. The protective efficacy of MSP4/5 against lethal Plas-
modium chabaudi adami challenge is dependent on the type of DNA vaccine
vector and vaccination protocol. Vaccine 21:3030–3042.
28. Singh, G., S. Parker, and P. Hobart. 2002. The development of a bicistronic
plasmid DNA vaccine for B-cell lymphoma. Vaccine 20:1400–1411.
29. Smooker, P. M., Y. Y. Setiady, A. Rainczuk, and T. W. Spithill. 2000.
Expression library immunization protects mice against a challenge with vir-
ulent rodent malaria. Vaccine 18:2533–2540.
30. van der Heyde, H. C., D. Huszar, C. Woodhouse, D. D. Manning, and W. P.
Weidanz. 1994. The resolution of acute malaria in a definitive model of B cell
deficiency, the JHD mouse. J. Immunol. 152:4557–4562.
31. von der Weid, T., N. Honarvar, and J. Langhorne. 1996. Gene-targeted mice
lacking B cells are unable to eliminate a blood stage malaria infection.
J. Immunol. 156:2510–2516.
32. Wang, R., D. L. Doolan, Y. Charoenvit, R. C. Hedstrom, M. J. Gardner, P.
Hobart, J. Tine, M. Sedegah, V. Fallarme, J. B. Sacci, Jr., M. Kaur, D. M.
Klinman, S. L. Hoffman, and W. R. Weiss. 1998. Simultaneous induction of
multiple antigen-specific cytotoxic T lymphocytes in nonhuman primates by
immunization with a mixture of four Plasmodium falciparum DNA plasmids.
Infect. Immun. 66:4193–4202.
33. Weiss, W. R., K. J. Ishii, R. C. Hedstrom, M. Sedegah, M. Ichino, K.
Barnhart, D. M. Klinman, and S. L. Hoffman. 1998. A plasmid encoding
murine granulocyte-macrophage colony-stimulating factor increases protec-
tion conferred by a malaria DNA vaccine. J. Immunol. 161:2325–2332.
Editor: W. A. Petri, Jr.
VOL. 72, 2004BICISTRONIC DNA VACCINATION AGAINST RODENT MALARIA 5573