Development of a candidate DNA vaccine against Maedi-Visna virus

www.elsevier.com/locate/vetimm

Veterinary Immunology and Immunopathology 119 (2007) 222–232

Development of a candidate DNA vaccine against

Maedi-Visna virus

Ana M. Henriques a,b, Miguel Fevereiro b, Duarte M.F. Prazeres a,Gabriel A. Monteiro a,*

a Institute for Biotechnology and Bioengineering, Centre for Biological and Chemical Engineering,

Instituto Superior Tecnico, Av. Rovisco Pais, 1049–001 Lisboa, Portugalb Laboratorio Nacional de Investigacao Veterinaria, Lisboa, Portugal

Received 2 March 2007; received in revised form 18 May 2007; accepted 29 May 2007

Abstract

DNA vaccine candidates against Maedi-Visna virus (MVV) infection in ovines were developed as an alternative to conventional

vaccines. Candidates were constructed by cloning genes encoding the MVV gag polyprotein and gag proteins p16 and p25 fused to a

b-galactosidase reporter in a plasmid backbone. Transfection of different ovine cells showed a higher protein expression with

plasmid lacZp16, which was hence further optimised by (i) removing a putative inhibitory sequence via reduction of the AU-content

in the p16 gene or by (ii) introducing a secretory signal (Sc) to promote antigen secretion and increase its presentation to APCs.

Unexpectedly, plasmids constructed on the basis of the first strategy by mutagenesis of lacZp16 (lacZp16mut24), led to a reduction

in the expression of the antigen/reporter fusion in cultured ovine cells. This indicates that the high AU content in MVV does not

inhibit protein expression. However, mice primed with lacZp16mut24 and boosted with MVV protein displayed higher humoral

response when compared with control lacZp16. The addition of the Sc signal (Sc-p16) led to lower amounts of intracellular antigen/

reporter fusion in transfected ovine cells, thus confirming secretion. These findings correlate with in vivo experiments, which

showed that mice primed with Sc-p16 and boosted with MVV exhibited stronger antibody responses when compared with control

mice primed with lacZp16 and boosted with MVV. Stronger humoral responses were recorded by immunising mice with (i) Sc-p16

and lacZp16mut24 plasmids together or with (ii) one plasmid containing both the mutations and the Sc signal.

# 2007 Elsevier B.V. All rights reserved.

Keywords: DNA vaccine; Maedi-Visna virus; MVV; Lentivirus; Small ruminant lentivirus

1. Introduction

Maedi-Visna virus (MVV) is the prototype of the

small ruminant lentiviruses (SRLV) that causes a slowly

progressive disease in sheep, characterised by a

relatively long asymptomatic period in which virus

persists in the presence of strong humoral and cellular

* Corresponding author. Tel.: +351 218419195;

fax: +351 218419062.

E-mail address: [email protected] (G.A. Monteiro).

0165-2427/$ – see front matter # 2007 Elsevier B.V. All rights reserved.

doi:10.1016/j.vetimm.2007.05.004

responses (Clements and Zink, 1996). The virus derives

its Icelandic name from the symptoms of the disease it

causes in sheep. Maedi means dyspnea and is used to

describe a slow progressive interstitial pneumonia.

Visna (wasting) is a condition resulting from a slow

progressive inflammatory disease of the central nervous

system, which causes paralysis. MVV infection is

spread in North America, Africa, Asia and Europe, but

has never occurred in Australia or New Zealand. In

Portugal, the prevalence of MVV is particularly high,

with almost 82% of the flocks infected (Fevereiro,

1995).

mailto:[email protected]

http://dx.doi.org/10.1016/j.vetimm.2007.05.004

A.M. Henriques et al. / Veterinary Immunology and Immunopathology 119 (2007) 222–232 223

The 9.2 kb genome of MVV has three major genes,

gag, pol and env (Sonigo et al., 1985). The gag gene

encodes for a polyprotein precursor, which is cleaved

into three nucleoproteins: a 25 kDa capsid protein

(p25), a 16 kDa matrix protein (p16) and a 14 kDa

nucleic acid-binding protein (p14). Viral protease,

reverse transcriptase and endonuclease/integrase are

codified by the pol gene (Sonigo et al., 1985). The env

region contains the sequence for the envelope glyco-

protein, gp105, which is subsequently processed into a

40 kDa transmembrane (TM) portion and a 70 kDa

external surface glycoprotein (SU) (Power et al., 1995).

Regulatory genes tat and rev, and the accessory gene vif

are also found in the MVV genome. Long terminal

repeats (LTR) are located in both ends of the genome,

which contain important controlling elements for MVV

gene expression (Haase, 1986).

So far, conventional vaccines have failed to induce

an effective immune response in sheep. Although

attenuated (Petursson et al., 2005) and inactivated

(Cutlip et al., 1987) MVV vaccines readily stimulate

production of precipitating antibodies in sheep, these

are not able to prevent MVV infection.

DNA vaccination may be pursued as an alternative to

conventional vaccines. For instances, both Gag-specific

antibody and CTL responses have been achieved in

mice after DNA vaccination against human immuno-

deficiency virus (HIV), a virus related to MVV (Qiu

et al., 1999). The first attempt to immunise sheep from a

naturally infected flock by gene gun administration of

the env gene demonstrated an early protective effect

against MVV infection that restricts the virus replica-

tion following challenge, even though in the absence of

neutralizing antibody production (Gonzalez et al.,

2005).

A possible drawback of DNA vaccines is that the

encoded immunogen expressed in the transfected cells

may not enter the MHC II antigen processing and

presentation pathway, which conventionally operates

only in professional antigen presenting cells (APCs)

(Ruff et al., 1997). The most likely pathway for antigen

presentation involves uptake of the DNA vaccine by a

cell (non APC), followed by expression and transmis-

sion of the antigen to dendritic cells for presentation

(Qiu et al., 2000). If this is the dominant pathway, then

the effectiveness of DNA vaccines to elicit immune

responses could be improved by using secretory signals

to target the antigen to the extracellular medium, thus

increasing its exposition to APCs.

The use of heterologous prime-boost immunisation

with DNA and protein has proved useful in most cases

(Amara et al., 2002). This strategy has the potential for

inducing a stronger immune response, probably because

boosting with a formulation containing only the relevant

epitope in common with the prime immunisation may

allow preferential expansion of pre-existing memory T-

cells to the epitope of interest (Dunachie and Hill,

2003).

Studies carried out with HIV lentivirus showed that

expression of the gag gene depends on the presence of

the viral Rev protein, which binds to its RNA target,

named Rev-responsive element (RRE). Rev protein acts

by promoting the nuclear export and increasing the

stability of the RRE-containing mRNAs (Schwartz

et al., 1992). Since RRE-containing mRNAs are not

expressed in the absence of Rev, it was postulated that

they are defective due to the presence of inhibitory

sequences (INS) that prevent their expression (Schnei-

der et al., 1997). In the gag gene, the INS is

characterised by the high AU-content and the presence

of the pentanucleotide AUUUA (Kotsopoulou et al.,

2000). High percentages of AU in human mRNAs have

shown to result in instability, increased turnover, and

low expression levels (zur Megede et al., 2000).

In this work, a p16-based DNA vaccine candidate

against Maedi-Visna Virus was selected and further

modified avoiding INS and promoting accessibility of

the expressed protein to antigen presenting cells in order

to increase the immune response.

2. Materials and methods

2.1. Plasmids construction

The candidate genes were cloned in the backbone

vector pVAX1lacZ (Invitrogen) upstream the b-galacto-

sidase (b-Gal) reporter gene and under the control of the

CMV promoter. The MVV gag gene and sequences

encoding proteins p16 and p25 were amplified from a

Portuguese isolate of MVV (P1OLV) (Barros et al., 2004)

by PCR, using primers containing NheI and AflII

restriction sites respectively at 50 and 30 ends and two

additional nucleotides were designed to remove the stop

codon of gag gene and bring the MVV cloned sequences

and lacZ gene to the same ORF (Table 1). Since the MVV

genes were cloned upstream of lacZ gene, any reporter

activity confirms the correct expression of the MVV/b-

Gal fusion protein.

Each reaction was performed with 500 ng of

template (genomic DNA extracted from lung cells of

a naturally infected sheep), 100 pmol of each primer,

400 mM of each dNTP, 2 mM MgSO4, 10 ml of reaction

buffer 10x (100 mM (NH4)2SO4, 200 mM Tris-acetate

pH 8.9) and 5 U of Tfl DNA polymerase (Promega) in a

A.M. Henriques et al. / Veterinary Immunology and Immunopathology 119 (2007) 222–232224

Table 1

Nucleotide sequences of primers used in amplification reactions

Fragment Primers Annealing temp. (8C) Extension time (s)

lacZp16 GCGCGCTAGCATGGCGAAGCAAGGCTCAAAGGAG

GCGCCTTAAGCCGTAGACCTCCTTATGTGTCTC

50 90

lacZp25 GCGCGCTAGCATGGCCATAGTAAATTTACAAGCAG

GATACTTAAGCCCAATTGCATTTTAAATCCTTCTG

50 90

lacZgag GCGCGCTAGCATGGCGAAGCAAGGCTCAAAGGAG

GCGCCTTAAGCCCAACATAGGGGGTGCGGACGGC

60 180

NheI and AflII restriction sites are in italic and the start codon is shown in bold. Also annealing temperature and extension time used in PCR reaction

are shown.

final volume of 100 ml. The amplification program for

p16 and p25 encoding sequences included an initial

denaturation for 2 min at 94 8C, followed by 35 cycles

of 30 s denaturation at 94 8C, 30 s annealing at 50 8Cand 90 s extension at 72 8C. For gag gene, the

amplification program was similar, but with an

annealing temperature of 60 8C and 180 s for extension.

A final extension was performed at 72 8C for 10 min.

Once amplified, fragments were digested with restric-

tion enzymes NheI and AflII and cloned into pVAX1-

lacZ. The resulting plasmids, named lacZp16, lacZp25

and lacZgag were used to transform Escherichia coli

DH5a competent cells. The obtained transformants

were firstly screened by restriction analysis and the

nucleotide sequence of the inserts was confirmed by

automatic sequencing using ALFexpressII (Amersham

Pharmacia Biotech).

2.2. Mutagenesis

In order to eliminate inhibitory sequences, point

mutations were introduced in the sequence encoding

protein p16 to reduce the AU content and modify the

AUUUA pentanucleotide. These point mutations, in a

total of 24 spanning a region of 353 nucleotides, were

Table 2

Primers used to remove the inhibitory sequence

Fragment P

lacZp16mut3 C

lacZp16mut6 C

lacZp16mut9 G

lacZp16mut12 G

lacZp16mut14 G

lacZp16mut17 G

lacZp16mut20 G

lacZp16mut24 G

Only forward primers are shown and the mutated nucleotides are in italic. The

the plasmid name.

introduced by sequential PCR-based site-directed

mutagenesis.

A linear double chain fragment complementary to

the entire plasmid was created in a PCR reaction, using

Pfu DNA polymerase. The mutagenesis product was

digested with DpnI, a restriction enzyme that only

recognizes methylated DNA and thus destroys the

original methylated plasmid. Transformation of DH5a

competent cells with this mutated fragment was then

necessary to close and restore the plasmid. In order to

introduce several mutations along the gene, eight

sequential reactions were carried out using eight

different pairs of primers (Table 2).

Each reaction was performed in a total volume of

50 ml, containing 25 ng of plasmid DNA, 125 ng of

each primer, 25 mM of each dNTP, 5 ml reaction buffer

(100 mM KCl, 100 mM (NH4)2SO4, 200 mM Tris-Cl

pH 8.75, 20 mM MgSO4, 1% Triton1 X-100, 1 mg/ml

BSA) and 2.5 U of Pfu DNA polymerase (Stratagene).

The amplification program comprised an initial

denaturation for 5 min at 95 8C and 20 cycles of 30 s

at 95 8C, 1 min at 52 8C and 13 min at 72 8C (about

2 min for each 1000 bp). Automatic sequencing was

performed to confirm the correct introduction of

mutations.

rimers

TCAAAGAGGTAATCAAAGCAACTTGCAAAATAAAGGTTG

TCAAAGAGGTGATCAAGGCAACTTGCAAAATCAAGGTTG

GGCATTGAAAACTATAGACTTCATATTTGAAGATATCAAGGC

GGCACTGAAGACTATAGACTTCATATTCGAAGATATCAAGGC

GGAATAATCAGTATGAAAGGAGGGCTTCAGGAAAAAC

GGAATCATCAGTATGAAGGGAGGGCTTCAGGAGAAAC

AAAGAAAAGAAAATAGAGCAATTGTATCCCAATCTAGAG

AAGGAAAAGAAGATAGAGCAGTTGTATCCCAACCTAGAG

number of mutations introduced is indicated by the superscript close to


2.3. Secretion targeting sequence

A secretory signal (Sc) that promotes secretion of

synthesised antigens was introduced in the lacZp16

plasmid by cloning the first 23 amino acids of human

tissue plasminogen activator (t-PA) (Qiu et al., 2000)

upstream of p16 encoding sequence: MDAMKRGL-

CCVLLLCGAVFVSAR.

Two oligonucleotides were used to introduce the

targeting sequence, one containing the 50 end of the

targeting sequence and the other containing its 30 end.

Both oligonucleotides have the central region of the

targeting sequence, so they can anneal to each other,

during the annealing step of a PCR reaction. The region

of the oligonucleotides that do not anneal (50 end and 30

end of targeting sequence), acts as templates for

incorporation of nucleotides by DNA polymerase

during the extension step. The PCR reaction completed

the sequence in both directions.

Synthetic oligonucleotides also comprise BstEII and

NheI restriction sites. The sequences of synthetic

oligonucleotides were 50–GGGGGTGACCATGGACG-

CCATGAAGCGCGGCCTGTGCTGCGTGCTGCT-

GCTGTG–30 (BstEII site in italics and sequence

complementary to oligo 2 in bold) and 50–CCCGCT-

AGCGCGGGCGCTCACGAACACGGCGCCGCACA-

GCAGCAGCACGCAGCAC–30 (NheI site in italics

and sequence complementary to oligo 1 in bold).

A BstEII restriction site was created in plasmid

lacZp16 by site-direct mutagenesis 7 bp upstream NheI

restriction site. The reaction was carried out with

100 pmol of each primer, 400 mM of each dNTP, 5 ml

of reaction buffer (100 mM KCl, 100 mM (NH4)2SO4,

200 mM Tris-Cl pH 8.75, 20 mM MgSO4, 1% Triton1

X-100, 1 mg/ml BSA) and 2.5 U of Pfu DNA

polymerase (Stratagene), in a final volume of 50 ml.

After an initial denaturation at 94 8C for 2 min, 35

cycles of 30 s at 94 8C, 30 s at 65 8C and 2 min at 72 8Cwere performed. The amplification program ended with

a final extension at 72 8C for 7 min. Plasmid and

fragments obtained were digested with BstEII and NheI,

ligated and used to transform competent DH5a cells.

The introduction of the correct targeting sequence (Sc-

p16 plasmid) was verified by automatic sequencing.

2.4. Construction of Sc-p16mut

Sc-p16mut was constructed from Sc-p16 and

lacZp16mut24, both previously double-digested with

NheI/AflII enzymes. The fragment containing the

backbone vector and the targeting sequence (Sc-p16

plasmid without p16 encoding sequence) and the

p16mut24 fragment (from lacZp16mut24 plasmid) were

ligated by T4 DNA ligase and used to transform E. coli

DH5a competent cells. The obtained plasmid was

automatically sequenced in order to confirm the

nucleotide sequence.

2.5. Cell transfection

Primary fibroblastic like sheep choroid plexus

(SCP), ovine synovial membrane (OSM), ovine testicle

(TO) and ovine skin cells (OSk) were available in our

laboratory and were maintained in Dulbecco’s modified

Eagle’s medium (DMEM) supplemented with 10%

foetal bovine serum, 100 units/ml of penicillin, 100 mg/

ml of streptomycin and 50 mg of gentamicin/ml until

confluence. Chinese hamster ovary (CHO) cells were

grown in F-12 nutrient mixtures (GibcoTM) supple-

mented with the same additives. Cells were trypsinised,

pelleted, resuspended in 5 ml of PBS buffer (0.9%

NaCl, 10 mM sodium phosphate, pH 7.2) and counted.

The amount of cells needed (5 � 105 cells) per

transfection was centrifuged and pellets were resus-

pended in PB-sucrose buffer (272 mM sucrose, 1 mM

MgCl2, 7 mM Na2HPO4, pH 7.4). The cells (200 ml)

were mixed with 2 mg of DNA and put into an

electroporation cuvette. The electroporation was carried

out using 10 bursts of 200 V, 100% modulation, 40 kHz,

with 7 ms duration and 1 ms burst interval in a Bio-Rad

Gene Pulser II RF Module.

After electroporation, cells were transferred to a 24-

well plate with 500 ml complete medium and incubated

at 37 8C in a 5% CO2 atmosphere for 24 h.

2.6. Protein expression

Protein expression was monitored by detecting

the MVV-b-Gal fusion protein. b-Gal was detected

qualitatively using the b-Gal Staining Set (Roche)

and quantitatively through b-Gal ELISA (Roche).

Firstly, cells were fixed after transfection with 2%

formaldehyde and 0.2% glutheraldehyde in PBS

for 15 min. After washing with PBS, X-Gal, the

substrate that is hydrolysed by b-Gal yielding a

blue precipitate was added in a solution 1:20 in iron

buffer. Cells were incubated at 37 8C for 2 h and

detection of blue cells was performed by microscope

observation.

For the quantitative detection by b-Gal ELISA, 96-

well plates coated with anti-b-Gal were used. The

ELISA reaction was performed according to the

manufacturer’s description and the absorbance was

measured at l = 405 nm. The concentration of b-Gal


was determined in relation to a calibration curve

prepared with adequate standards.

2.7. Vaccine production and purification

The candidate vaccines were propagated in 1 L of E.

coli DH5a and were purified by a process based on

hydrophobic interaction chromatography (Diogo et al.,

2005; Diogo et al., 2001). Briefly, cell pellets were

resuspended in 8 ml of 10 mM Tris-HCl, 50 mM glucose

and 10 mM EDTA (pH 8.0). Alkaline lysis was

performed by adding 8 ml of a 200 mM NaOH,

1% (w/v) sodium dodecyl sulphate solution (10 min,

room temperature). Cellular debris, genomic DNA and

proteins were precipitated (10 min, on ice) by adding

8 ml of pre-chilled 3 M potassium acetate (pH 5.5). The

precipitate was removed by centrifugation at 12,000 g

(30 min, 4 8C). Plasmid DNA was then precipitated by

adding 0.7 volumes of isopropanol to the clarified lysate

and the precipitate was separated by centrifugation

(12,000 g, 4 8C) for 30 min. The plasmid-containing

pellet was then washed with 70% ethanol and centrifuged

(12,000 g, 4 8C) for 10 min. Supernatant was discarded

and the pellet was resuspended in 0.5 ml of 10 mM Tris-

Cl buffer (pH 8.0). After that, solid ammonium sulphate

was dissolved in the plasmid solution up to a

concentration of 2.5 M, followed by incubation on ice

for 15 min. Precipitated impurities were removed by

centrifugation (12,000 g, 4 8C) for 15 min. The plasmid-

containing supernatant (approximately 0.5 ml) was

loaded directly on a gravity-operated hydrophobic

interaction chromatography column, packed with

10 ml of phenyl sepharose 6 fast flow (GE Healthcare)

and equilibrated with 1.5 M ammonium sulphate.

Plasmid was recovered in a 2-ml, salt-rich fraction that

was finally desalted in gravity-operated size exclusion

column using PBS as the elution buffer.

2.8. Mice immunisation

Female BALB/c mice, 6–8 weeks old were

immunised subcutaneously or intramuscularly with

the DNA vaccine prototypes lacZp16, lacZp16mut12,

lacZp16mut24, Sc-p16, Sc-p16mut, and pVAX1lacZ

(lacZ). The first immunisation with 50 mg of plasmid

DNA in 100 ml of PBS was followed by two DNA boost

vaccinations. Since MVV does not infect mice, 5 mg of

purified MVV were administered as protein boost. The

negative control group of mice received 100 ml PBS and

purified virus.

Mice were regularly bled up to the 197th day by

facial venipuncture and blood was collected to

MicrotainerTM (Becton Dickinson). The blood was

centrifuged at 2000 g for 5 min and the serum fraction

was collected. Pools of sera from each group were

prepared by adding the same volume of each mice

serum. All samples were stored at –20 8C.

To obtain viral antigen, MVV was grown in SCP

cells and the supernatants of cultures showing 80% CPE

were clarified by centrifugation. The virus was

precipitated by the addition of PEG and purified in a

discontinuous sucrose gradient as previously described

(Fevereiro et al., 1999)

2.9. Antibody response

The mice antibody response was monitored by

ELISA as described before (Fevereiro et al., 1999), with

minor modifications. The MVV antigen (0.45 mg/ml

MVV) prepared as described above (Section 2.8) was

pre-treated (1:1) with 2% octyl in PBS and incubated for

15 min at 4 8C in order to prevent clump formation. This

solution was diluted 1:500 in carbonate/bicarbonate

buffer (15 mM Na2CO3, 35 mM NaHCO3, pH 9.6) and

50 ml per well were used to coat a 96-well ELISA plate

that was subsequently incubated overnight at 4 8C.

After incubation, the plate was washed four times with

washing solution (0.05% Tween 20 in H2O). Serum

samples were prepared in four serial dilutions from

1:100 to 1:800 in serum dilution buffer (2.5 mM

NaH2PO4, 7.5 mM Na2HPO4, 500 mM NaCl, 0.05%

Tween 80). Each dilution (50 ml) was added to the

coated plate, which was then incubated at 37 8C for 1 h.

Horseradish peroxidase-conjugated rabbit anti-mouse

immunoglobulin (1.3 g/l) was diluted 1:1000 in serum

dilution buffer with 5% foetal ovine serum and 50 ml

were added to each well after washing the plate. The

plate was incubated at room temperature for 1 h and

then washed. A 10 mg OPD (o-phenylenediamine

dihydrochloride) tablet was dissolved in 25 ml

citrate/phosphate buffer (35 mM citric acid, 67 mM

Na2HPO4) with 10 ml H2O2. This OPD solution

(100 ml) was added to each well and plate was

incubated for 1 h in the dark, at room temperature.

The reaction was stopped by adding 100 ml of 10%

sulphuric acid and absorbance was measured at 492 nm

in a Sunrise Tecan plate reader.

2.10. Data analysis

Absorbance values measured in the ELISA were

represented in a chart as a function of the logarithm of

dilution. The obtained values gave rise to a linear

representation whose equation was determined. Theo-


retically, interception corresponds to the absorbance

value of the undiluted serum. Therefore, each inter-

ception value of each mice group was used as a measure

of antibody response and was represented in another

chart as function of time after immunisation. In order to

normalise the different mice group representations,

each point was divided by the correspondent value at

Day 0 (pre immunisation titre).

3. Results

3.1. Antigen expression in vitro

3.1.1. Gene selection for DNA vaccine candidate

A first series of DNA vaccine candidates against

MVV were constructed by cloning the gag gene and

sequences encoding proteins p16 and p25. These genes

were fused to the lacZ gene and were chosen because

they are more conserved than env genes. MVV genes

were amplified by PCR and cloned into the pVAX1lacZ

expression vector. The ligation products were used to

transform E. coli DH5a cells and recombinants were

screened by restriction analysis. The nucleotide

sequences of the inserts were confirmed by automatic

sequencing.

CHO, SCP, OSM, TO and OSk cells were transfected

with plasmids lacZp16, lacZp25 and lacZgag by

Fig. 1. Expression of the antigen-b-Gal fusion protein following electroporat

vaccines lacZp16, lacZp25 and lacZgag. b-Gal expressed from each plasmid

cell line. Three independent assays, with three replicates each, were perform

standard deviation.

electroporation. Preliminary experiments were carried

out with CHO cells by using increasing amounts of

plasmid DNA. A direct correspondence was achieved

up to the maximum amount tested (5 mg). Nevertheless

2 mg of DNA were used in all subsequent experiments,

since it gave the best results in terms of efficiency and

toxicity to the cells.

Protein expression was assessed through the fused b-

Gal reporter protein, which was analysed qualitatively

by incubating cells with the substrate X-Gal. In all

cases, blue cells characteristic of b-Gal hydrolysis of X-

Gal were observed, indicating that clones were correctly

constructed and that all fusion proteins were being

expressed.

A b-Gal ELISA was performed in order to

quantitatively determine the amount of b-Gal produced.

Independently of the cell line used the b-Gal fusion

expression was always higher with lacZp16 when

compared with the lacZp25 and lacZgag constructs

(Fig. 1). Therefore, the lacZp16 construct was selected

for further optimisation.

3.1.2. Effect of INS removal

The inhibitory sequence (INS) present in the

sequence encoding the p16 protein was removed by a

site-direct mutagenesis procedure that sequentially

introduced mutations up to a total of 24 in a region

ion of CHO, SCP, OSM, TO and OSk cell lines with the candidate DNA

is shown as the relative percentage to the lacZp16 plasmid in the same

ed with all plasmid/cell line combinations. Vertical bars represent the


of 353 nucleotides. These mutations were carried out to

reduce the AU content and remove the AUUUA

pentanucleotide, according to sheep codon-usage

(sheep codon-usage at www.kazusa.or.jp/codon), while

maintaining the amino acid sequence of the protein. The

expected mutations were confirmed by automatic

sequencing.

To check if these mutated lacZp16 plasmids were

more expressed when compared with the non-mutated,

ovine and CHO cell lines were transfected by

electroporation with the lacZp16 control and, lacZp16-

mut12 and lacZp16mut24 which respectively, comprise

the first twelve and twenty four mutations. The mutated

Fig. 2. Effect of secretion targeting sequence and removal of inhibitory se

detected after electroporation of CHO, SCP, OSM, TO and OSk cell lines wi

Sc-p16. b-Gal expressed from each plasmid is shown as the relative perc

independent assays, with three replicates each, were performed with all

deviation.

lacZp16mut12 and lacZp16mut24 plasmids gave rise to

blue cells after addition of X-Gal, indicating that they

maintained the ability to express the p16/b-Gal fusion

protein in spite of the introduced mutations. However,

unexpectedly, b-Gal expression decayed as the number

of mutated nucleotides increased in all cell lines tested,

except OSk (Fig. 2). In CHO, SCP and TO cells, the

expression of lacZp16mut12 decayed approximately

40% when compared with the expression of lacZp16,

whereas in OSM cells the decay was only 8%. The

expression of lacZp16mut24 decreased 80–90% when

compared with the expression of lacZp16 in all cells

(except OSk). In OSk cells, the amount of b-Gal

quence on the expression of antigen-b-Gal fusion protein. b-Gal was

th candidate DNA vaccines lacZp16, lacZp16mut12, lacZp16mut24 and

entage to the most expressed plasmid in the same cell line. Three

plasmid/cell line combinations. Vertical bars represent the standard

http://www.kazusa.or.jp/codon


Fig. 3. Antibody production against MVV in mice immunised sub-

cutaneously with 50 mg of lacZp16 (^), lacZp16mut12 (^), lacZp16-

mut24 (&), Sc-p16 (&) and the empty vector lacZ (4). As negative

control, mice were inoculated with PBS (�). Immunisations with the

DNA vaccines were carried out at Days 0, 28 and 56 (~). Protein

boost consisting of 5 mg of purified MVV was given at Day 84 ( ).

The antibody titre in each data point represents a group with 5 mice,

and was measured by ELISA in triplicate.

produced was higher with plasmid lacZp16mut12, with

expression decreasing 30 and 50% with lacZp16 and

lacZp16mut24, respectively. Overall, the results suggest

that the high AU content in the p16 gene of MVV is not

associated with an in vitro inhibition of expression in the

cell lines tested in this work.

3.1.3. Effect of antigen secretion

To enhance the immune response against MVV p16,

a sequence containing a secretion signal (Sc) from

human tissue plasminogen activator (t-PA) was cloned

upstream of p16 encoding sequence of lacZp16 (see

Section 2.3). This targeting sequence promotes secre-

tion of the p16 antigen, rendering it more susceptible to

recognition by APCs. This sequence contains a signal

that directs the expressed protein into the endoplasmic

reticulum lumen. The protein is excreted following

transport across the endoplasmic reticulum and Golgi

apparatus (Qiu et al., 2000).

The effect of the Sc targeting sequence in expression

was studied by electroporating CHO and ovine cell lines

with Sc-p16. The amount of intracellular b-Gal was

determined by b-Gal ELISA (Fig. 2). The results show

that when the targeting sequence is present, the amount

of intracellular p16/b-Gal fusion protein detected

decreased more than 90% in all cell lines tested. This

is in accordance with expectations and reveals efficient

secretion, since the Sc signal targets the protein to the

outside of the cells, concomitantly decreasing their

intracellular concentration.

3.2. Immunisation experiments

3.2.1. Effect of INS removal and of antigen

secretion

Mice were immunised subcutaneously with 50 mg of

each plasmid construct, lacZp16, lacZp16mut12,

lacZp16mut24, and Sc-p16. Groups of negative control

mice were inoculated with the empty vector lacZ and

with PBS. The schedule used comprised a prime

administration at Day 0, followed by two boosts at Days

28 and 56, and a heterologous boost of 5 mg of purified

MVV at Day 84. Blood was collected regularly and

mice antibody response was measured by ELISA. The

results were analysed as described in Materials and

Methods.

The injection of lacZp16, lacZp16mut12 and

lacZp16mut24 has shown a slight increase (around 2

fold) of the antibody titre while the response for both

controls, lacZ and PBS, was null until MVV boost

(Fig. 3). After administration of the purified MVV

(mice were not challenged by MVV) the antibody

response doubled in mice immunised previously with

lacZp16mut12 and lacZp16. However, mice primed with

lacZp16mut24, which had shown a similar behaviour to

lacZp16 and lacZp16mut12, increased significantly the

antibody titre after MVV boost (Fig. 3). The secondary

antibody response exhibited by these mice was

maintained within high values until the end of the

experiment, showing that this plasmid may have

stimulated memory cells.

Experiments were carried out in order to evaluate

antibody response after immunising mice with plasmid

carrying a secretion targeting sequence (see Section

2.3).

The antibody titre exhibited by mice inoculated with

plasmids with and without the targeting sequence was

quite similar with a small increase after each DNA

administration (Fig. 3). Even so, after inoculation of

MVV antigen, the antibody titre in mice previously

immunised with Sc-p16 was much higher when

compared with the titre of mice primed with lacZp16.

Moreover, this immune response was maintained within

high values until the end of the experiment, while in

mice firstly inoculated with lacZp16 the antibody titre

decreased to the initial levels.

3.2.2. Effect of a combined vaccine

The previous results have shown that plasmids

lacZp16mut24 and Sc-p16 were more effective in

eliciting a humoral response when compared with the

parental non-modified lacZp16 plasmid. In order to

enhance even more this humoral response, an experi-


Fig. 4. Antibody production against MVV in mice immunised intra-

muscularly with 50 mg of lacZp16mut24 (&) and Sc-p16 (&), indi-

vidually and simultaneously (*). As negative control, mice were

inoculated with PBS (�). Immunisations with the DNA vaccines were

carried out at Days 0 and 21 (~). Protein boost consisting of 5 mg of

purified MVV was given at Day 65 ( ). The antibody titre in each data

point represents a group with 7 or 6 (negative control) mice, and was

measured by ELISA in triplicate.

Fig. 5. Antibody production against MVV in mice immunised intra-

muscularly with 50 mg of Sc-p16mut24 (*) and with 25 mg of each

plasmid lacZp16mut24 and Sc-p16 in simultaneous (*). As negative

control, mice were inoculated with PBS (�). Immunisations with the

DNA vaccines were carried out at Days 0 and 21 (~). Protein boost

consisting of 5 mg of purified MVV was given at Day 65 ( ). The

antibody titre in each data point is representative of a group with 7

mice or 8 (negative control) mice, and was measured by ELISA in

triplicate.

ment was performed by intramuscular administration

of lacZp16mut24 and Sc-p16 plasmids to the same

mice simultaneously. The antibody titres are shown

in Fig. 4. All groups, except the negative control,

showed a humoral response after each administration

of DNA. The administration of lacZp16mut24 and Sc-

p16 in separate, followed by the MVV boost, gave rise

to similar antibody titres. However, the group that

received both plasmids simultaneously showed a higher

humoral response, which was more evident after MVV

boost.

3.2.3. Effect of the optimised vaccine Sc-p16mut

To assess the improved contribution to humoral

response of both of the previous optimisations a new

vaccine candidate containing a secretable antigen

devoid of INS was constructed (Sc-p16mut). The

antibody titre obtained by ELISA (Fig. 5) have shown a

small increase in the humoral response after each DNA

administration, both in mice immunised with the two

plasmids and in mice inoculated with Sc-p16mut. Even

though, in this later group, the humoral response was

slightly higher. The difference between the two groups

was enhanced after protein boost, with mice immunised

with Sc-p16mut exhibiting a higher antibody response.

In the negative control group, the antibody titre was

maintained almost constant until MVV boost. Then the

antibody titre increased, but was kept lower when

compared with titres in the other groups. The optimised

plasmid, Sc-p16mut, is thus a promising DNA vaccine

against Maedi-Visna virus.

4. Discussion

This work describes the initial stages of the

development of a DNA vaccine against Maedi-Visna

virus. A series of vaccine prototypes were constructed

by cloning the MVV gag gene and the sequences

encoding the core proteins p16 and p25, upstream of the

reporter lacZ gene and under the control of CMV

promoter. These proteins were chosen because they are

the most conserved and immunogenic proteins of MVV,

minimising the large variability observed between

MVV strains and favouring cross-protection of vacci-

nated animals.

The candidate DNA vaccines were evaluated in vitro

by electroporating CHO and four ovine cell lines (SCP,

OSM, TO and OSk). The latter were available in our

laboratory and were chosen because we wanted to

determine the expression level of the constructs in cells

originated from sheep, therefore better mimicking what

would occur in the natural host. The reporter activity

was estimated both qualitatively, by the addition of X-

Gal, and quantitatively, by b-Gal ELISA. All plasmid

constructions tested originated positive results (blue

cells), indicating that candidate DNA vaccines were

correctly constructed. Although the type of cells used

had an effect on antigen expression, plasmid lacZp16

consistently gave the highest expression, followed by

plasmids lacZp25 and lacZgag. This can be attributed to

the following effects, each one per se, or conjugated: (i)

a more efficient internalisation of plasmid lacZp16,

which is the smallest of the three plasmids; (ii) a more


stable mRNA and/or higher level of transcription, as

well as (iii) a more stable expressed protein by lacZp16

construct.

Although mice are not susceptible to MVV infection,

the ability of the DNA vaccine candidates to elicit a

humoral response was tested by mice immunisation.

The obtained result (data not shown) was slightly

different from that obtained in vitro, since lacZp16 and

lacZp25 elicited similar humoral responses though the

level of p25 expression was much lower. This may be

related to the fact that p25 is a more immunogenic

protein (Singh et al., 2006). The mice immunised with

lacZgag did not develop significant levels of antibody

response.

The high AU-content and the presence of the

pentanucleotide AUUUA in the p16 sequence of the

MVV gag gene is characteristic of an inhibitory

sequence (INS). The presence of such sequences could

potentially hamper or prevent protein expression

(Kotsopoulou et al., 2000; Schneider et al., 1997).

Thus, in order to reduce the AU content, point mutations

were introduced by site-direct mutagenesis according to

sheep codon-usage, maintaining the amino acid

sequence of the encoded p16 protein. Results obtained

by in vitro assays showed that, by contrast to HIV (zur

Megede et al., 2000), the high AU content and the

pentanucleotide are not responsible for the inhibition of

p16 expression. On the contrary, the amount of b-Gal

expressed was lower with mutated plasmids in all cell

lines tested, with the exception of OSk cells. Since all

variants of plasmid lacZp16 have the same size and the

expressed protein is the same, the decrease in protein

expression may be imputable to different transcriptional

rates and/or different mRNA half-lives.

The DNA vaccine candidates were then tested by

immunising groups of mice subcutaneously with

3 � 50 mg of each construct, followed by a boost with

5 mg of MVV. After each DNA administration, the

antibody titre was slightly higher in mice immunised

with plasmid lacZp16mut24, when compared with mice

immunised with plasmids lacZp16mut12 or lacZp16.

However, the antibody response in mice primed with

plasmid lacZp16mut24 increased significantly after the

inoculation of MVV antigen. Furthermore, the level of

the immune response was maintained high almost until

the end of the experiment. Thus, and although the

expression of p16 in vitro was higher with the original

lacZp16 plasmid, an enhanced immune response was

elicited with plasmids mutated at the INS after the use

of a combined vaccination with lacZp16mut24/MVV

antigen. These contradictory results could be due to

different experimental conditions, including the type of

cells, transfection method and cellular environment

conditions that may result in different transcription or

expression efficiencies.

Another strategy attempted to increase mice immune

response was the inclusion of a targeting sequence (Sc)

to promote antigen secretion. The in vitro analysis

showed that the amount of intracellular antigen/b-Gal

fusion protein was lower when the Sc sequence was

present. This confirms that the secretory signal leads to

protein secretion, therefore resulting in lower levels of

protein inside the cells. These findings correlate with in

vivo experiments, which showed that mice immunised

with plasmid Sc-p16 exhibited stronger antibody

responses when compared to mice immunised with

lacZp16 (and lacZ and PBS). Thus, after DNA uptake

by transfected cells the expressed antigen is secreted

becoming available for presentation by APCs cells.

Since both the elimination of the putative inhibitory

sequence and the introduction of the secretory signal

resulted in an enhancement of the efficacy of the

vaccine candidate (when measured in terms of

antibodies generated), another experiment was per-

formed in which mice were immunised simultaneously

with the Sc-p16 and lacZp16mut24 constructions. The

corresponding antibody titre was higher than the titre

obtained when mice were inoculated with either

plasmid individually. A combined plasmid, Sc-

p16mut24, was then constructed, in which both the

secretory signal and the mutated p16 gene were present.

The antibody titres obtained after immunisation of mice

with this new construct showed that the humoral

response was even higher when compared with the

simultaneous administration of Sc-p16 and lacZp16-

mut24.

In the experiments the negative control mice

receiving PBS and MVV by intramuscular inoculation

showed a stronger immune response than those

inoculated subcutaneously. The result may be due to

different batches of virus that may not be in the same

conditions or that in the muscle MVV has been more

rapidly and effectively presented to the immune system.

In conclusion, several candidate DNA vaccines

against Maedi-Visna Virus were constructed and the

one carrying the most promising immunogen gene

(p16) was further optimised by reducing the AU-content

and by introducing a secretory signal. Both strategies

originated plasmids (lacZp16mut24 and Sc-p16) that led

to strong enhancements in the humoral response of

mice, when injected independently or simultaneously.

The humoral response was even higher when a plasmid

containing both the secretory signal and a reduced AU

content was used. Immunisation of sheep will be


required to check if these candidate vaccines can protect

the natural host against MVV infection. Different doses,

adjuvants and routes of administration will be assayed

in sheep with plasmid Sc-p16mut24. To improve the

effectiveness of the immunisation this plasmid will be

combined with other constructs, including MVV-env

and -gag genes.

Acknowledgements

The authors thank the Portuguese Fundacao para a

Ciencia e a Tecnologia for financial support (POCI/

CVT/47260/2002, and a PhD grant SFRH/BD/2904/

2000 to A.M. Henriques).

References

Amara, R.R., Villinger, F., Altman, J.D., Lydy, S.L., O’Neil, S.P.,

Staprans, S.I., Montefiori, D.C., Xu, Y., Herndon, J.G., Wyatt,

L.S., Candido, M.A., Kozyr, N.L., Earl, P.L., Smith, J.M., Ma,

H.L., Grimm, B.D., Hulsey, M.L., McClure, H.M., McNicholl,

J.M., Moss, B., Robinson, H.L., 2002. Control of a mucosal

challenge and prevention of AIDS by a multiprotein DNA/

MVA vaccine. Vaccine 20, 1949–1955.

Barros, S.C., Ramos, F., Duarte, M., Fagulha, T., Cruz, B., Fevereiro,

M., 2004. Genomic characterization of a slow/low Maedi Visna

virus. Virus Genes 29, 199–210.

Clements, J.E., Zink, M.C., 1996. Molecular biology and pathogenesis

of animal lentivirus infections. Clin. Microbiol. Rev. 9, 100–117.

Cutlip, R.C., Lehmkuhl, H.D., Brogden, K.A., Schmerr, M.J., 1987.

Failure of experimental vaccines to prevent against infection with

ovine progressive pneumonia (maedi-visna) virus. Vet. Microbiol.

13, 201–204.

Diogo, M.M., Queiroz, J.A., Prazeres, D.M.F., 2005. Purification of

plasmid DNA vectors produced in Escherichia coli for gene

therapy and DNA vaccination applications. In: Barredo, J.-L.

(Ed.), Microbial Processes and Products. Humana Press,

Totowa, pp. 165–278.

Diogo, M.M., Ribeiro, S.C., Queiroz, J.A., Monteiro, G.A., Tordo, N.,

Perrin, P., Prazeres, D.M., 2001. Production, purification and

analysis of an experimental DNA vaccine against rabies. J. Gene

Med. 3, 577–584.

Dunachie, S.J., Hill, A.V., 2003. Prime-boost strategies for malaria

vaccine development. J. Exp. Biol. 206, 3771–3779.

Fevereiro, M., 1995, Prevalence of Maedi-Visna infection in sheep in

Portugal. Revista Portuguesa de Ciencias Veterinarias XC, 66–70.

Fevereiro, M., Barros, S., Fagulha, T., 1999. Development of a

monoclonal antibody blocking-ELISA for detection of antibodies

against Maedi-Visna virus. J. Virol. Methods 81, 101–108.

Gonzalez, B., Reina, R., Garcia, I., Andres, S., Glaria, I., Alzueta, M.,

Mora, M.I., Jugo, B.M., Arrieta-Aguirre, I., de la Lastra, J.M.,

Rodriguez, D., Rodriguez, J.R., Esteban, M., Grillo, M.J., Black-

laws, B.A., Harkiss, G.D., Chebloune, Y., Lujan, L., de Andres, D.,

Amorena, B., 2005. Mucosal immunization of sheep with a Maedi-

Visna virus (MVV) env DNA vaccine protects against early MVV

productive infection. Vaccine 23, 4342–4352.

Haase, A., 1986. Pathogenesis of lentivirus infections. Nature 322,

130–136.

Kotsopoulou, E., Kim, V.N., Kingsman, A.J., Kingsman, S.M., Mitro-

phanous, K.A., 2000. A Rev-independent human immunodefi-

ciency virus type 1 (HIV-1)-based vector that exploits a codon-

optimized HIV-1 gag-pol gene. J. Virol. 74, 4839–4852.

Petursson, G., Matthiasdottir, S., Svansson, V., Andresdottir, V.,

Georgsson, G., Martin, A.H., Agnarsdottir, G., Gisladottir, E.,

Arnadottir, S., Hognadottir, S., Jonsson, S.R., Andresson, O.S.,

Torsteinsdottir, S., 2005. Mucosal vaccination with an attenuated

Maedi-Visna virus clone. Vaccine 23, 3223–3228.

Power, C., Richardson, S., Briscoe, M., Pasick, J., 1995. Evaluation

of two recombinant Maedi-visna virus proteins for use in an

enzyme-linked immunosorbent assay for the detection of serum

antibodies to ovine lentiviruses. Clin. Diagn. Lab. Immunol. 2,

631–633.

Qiu, J.T., Liu, B., Tian, C., Pavlakis, G.N., Yu, X.F., 2000. Enhance-

ment of primary and secondary cellular immune responses against

human immunodeficiency virus type 1 gag by using DNA expres-

sion vectors that target Gag antigen to the secretory pathway. J.

Virol. 74, 5997–6005.

Qiu, J.T., Song, R., Dettenhofer, M., Tian, C., August, T., Felber, B.K.,

Pavlakis, G.N., Yu, X.F., 1999. Evaluation of novel human immu-

nodeficiency virus type 1 Gag DNA vaccines for protein expres-

sion in mammalian cells and induction of immune responses. J.

Virol. 73, 9145–9152.

Ruff, A.L., Guarnieri, F.G., Staveley-O’Carroll, K., Siliciano, R.F.,

August, J.T., 1997. The enhanced immune response to the HIV

gp160/LAMP chimeric gene product targeted to the lysosome

membrane protein trafficking pathway. J. Biol. Chem. 272, 8671–

8678.

Schneider, R., Campbell, M., Nasioulas, G., Felber, B.K., Pavlakis,

G.N., 1997. Inactivation of the human immunodeficiency virus

type 1 inhibitory elements allows Rev-independent expression of

Gag and Gag/protease and particle formation. J. Virol. 71, 4892–

4903.

Schwartz, S., Campbell, M., Nasioulas, G., Harrison, J., Felber, B.K.,

Pavlakis, G.N., 1992. Mutational inactivation of an inhibitory

sequence in human immunodeficiency virus type 1 results in

Rev-independent gag expression. J. Virol. 66, 7176–7182.

Singh, I., McConnell, I., Blacklaws, B., 2006. Immune response to

individual maedi-visna virus gag antigens. J Virol. 80, 912–919.

Sonigo, P., Alizon, M., Staskus, K., Klatzmann, D., Cole, S., Danos,

O., Retzel, E., Tiollais, P., Haase, A., Wain-Hobson, S., 1985.

Nucleotide sequence of the visna lentivirus: relationship to the

AIDS virus. Cell 42, 369–382.

zur Megede, J., Chen, M.C., Doe, B., Schaefer, M., Greer, C.E., Selby,

M., Otten, G.R., Barnett, S.W., 2000. Increased expression and

immunogenicity of sequence-modified human immunodeficiency

virus type 1 gag gene. J. Virol. 74, 2628–2635.

Documents

Development of a candidate DNA vaccine against Maedi-Visna virus