A

aplavao1sa©

K

1

nIecdTi[vdn

Fhf

0d

Vaccine 25 (2007) 4540–4553

Prophylactic DNA immunization against multiple papillomavirus types

Maja Gasparic, Ivonne Rubio, Nadja Thones, Lutz Gissmann, Martin Muller ∗

Deutsches Krebsforschungszentrum, Forschungsschwerpunkt Infektionen und Krebs, Im Neuenheimer Feld 242, 69120 Heidelberg, Germany

Received 24 October 2006; received in revised form 13 February 2007; accepted 2 April 2007Available online 20 April 2007

bstract

At least 15 different papillomavirus types are causatively associated with the development of tumors in humans. Since the middle of 2006protective, virus-like particle based vaccine against the tumor-related HPV types 16 and 18 is commercially available. We investigated theossibility of applying DNA vaccination to obtain protective antibody responses against multiple papillomavirus types. Our data indicate thatow amounts of DNA were sufficient to induce neutralizing antibodies in mice although a DNA dose-dependency in respect to the L1-specificntibody titers was observed. Furthermore, we found that immune responses against different PV types could be induced by simultaneous DNAaccination with a mixture of expression vectors encoding L1 proteins of different papillomavirus types. However, we observed that there wasstrong interference when plasmids encoding different L1 genes were used together. HPV 16 responses were repressed by co-administrationf HPV 11 and/or BPV 1 L1 expression constructs. Likewise, BPV 1 responses were repressed by co-administration of HPV 16 or HPV

1 L1 plasmids. This interference could be overcome by administration of the different constructs into different sites of the animals or byequential immunization. Thus, our results suggest that the mode of repression was due to interference with L1 particle assembly and was notconsequence of immunodominance of certain L1 proteins.2007 Elsevier Ltd. All rights reserved.

dvctoeifsrc

eywords: HPV; Papillomavirus; DNA vaccine; L1; VLPs

. Introduction

The human papillomaviruses comprise a highly heteroge-ous group of more than 120 epitheliotropic viruses [1].nfection by certain types of human papillomavirus is consid-red to be the main risk factor in the development of cervicalancer [2–4]. Every year, about half a million of womenevelop cervical cancer of which 99% are HPV positive.he most prevalent HPV type associated with malignancy

s HPV 16 which can be found in about 50% of the tumors5]. Prophylactic vaccination against HPV was shown to pre-

ent infection and cancer precursors and thus most likely theevelopment of malignant disease [6,7]. Protective immu-ity is achieved by inducing virus-neutralizing antibodies

∗ Corresponding author at: Deutsches Krebsforschungszentrum,orschungsschwerpunkt Angewandte Tumorvirologie, F035, Im Neuen-eimer Feld 242, 69120 Heidelberg, Germany. Tel.: +49 6221 424628;ax: +49 6221 424932.

E-mail address: Martin.Mueller@dkfz.de (M. Muller).

tivigp

tb

264-410X/$ – see front matter © 2007 Elsevier Ltd. All rights reserved.oi:10.1016/j.vaccine.2007.04.001

irected against the major capsid protein L1 in the form ofirus-like particles or capsomeres. In fact, a commercial vac-ine containing L1 VLPs of the cancer-related papillomavirusypes 16 and 18 has been licensed by mid-2006 and anotherne is expected within a few months [6–9]. Although highlyfficient, these vaccines induce protective immunity predom-nantly against the vaccine types. This is mainly due to theact that each papillomavirus type is considered a separateerotype and that neutralizing antibodies are highly type-estricted [10–13]. Although the L1 capsid protein is the bestonserved viral protein, it displays heterogenous loops onhe surface of intact virions which are targets for neutraliz-ng antibodies [14]. The development of second generationaccines is aimed to provide either broader protection byncluding more HPV types into a vaccine or to develop anti-ens that induce cross-protective immunity against multiple

apillomavirus types.

A number of studies have shown that as an alternativeo protein-based vaccines, protective humoral responses cane achieved by DNA vaccination using codon-modified L1

ccine 2

evccltwdciisosa

2

2

fas(odrscSstmTdcc

2

fs

asDdoLFaiI

aecceiDoDw

cl1t

2

aVi

HaTmwlAps1ihibaNa3T

2

fh(pp

M. Gasparic et al. / Va

xpression constructs [15–19]. It is an advantage of DNAaccines that they can be manufactured under standardizedonditions although wide spread application of DNA vac-ination in humans is still generally hampered by ratherow immunogenicity. Hence, further improvements, e.g. byhe use of viral delivery systems [16] are required. Heree investigate combined DNA vaccination with L1 geneserived from different papillomavirus types. Our results indi-ate that small amounts of expression vectors are sufficient tonduce anti-L1 responses. However, we observed that theres a strong interference between different L1s when appliedimultaneously. Our results seem to rule out the possibilityf an immunodominance of one L1 over another but ratheruggests that the interference is due to an inhibition of L1ssembly.

. Material and methods

.1. DNA constructs

DNA encoding for humanized L1 (L1h) antigen of dif-erent papillomavirus types (HPV 11, HPV 16, HPV 18nd BPV 1) were subcloned into the mammalian expres-ion vector pUF3 under the control of the cytomegalovirusCMV) immediate early promoter [23]. The L1h sequencesf HPV 11, HPV 16 and BPV 1 [18,24,25] were insertedirectly after the CMV promoter at the restriction site NotI,eplacing the GFP gene, whereas the HPV 18 L1h DNAequence was cloned at the restriction site HincII. By sub-loning the HPV 18 L1h DNA into the HincII site, theD/SA intron was disrupted. Consequently, the HPV 18 con-truct might not be fully comparable to the constructs ofhe other L1h genes, as the disruption of the SD/SA intron

ight affect the expression level of the HPV 18 L1h gene.he empty pUF3 vector was created by HincII restrictionigest to excise the GFP gene and the neomycin expressionassette, whereas the integrity of the CMV promoter wasonserved.

.2. Immunization

The plasmid DNA used for immunization was isolatedrom E. coli DH5α by CsCl gradient centrifugation and dis-olved in phosphate buffered saline (PBS).

The concentration of DNA was determined visually fromgarose gels by comparing the band intensity with the inten-ity of the marker bands, which contain known amounts ofNA. Upon linearization of plasmid DNA by restrictionigest, different amounts of linearized DNA were loadedn the gel (10, 5, 2.5, 1.25 and 0.625 �l). HindIII digestedambda marker (New England Biolabs) and SM mass (MBI

ermentas) ruler were used as a reference with known DNAmounts in particular bands to produce ng-to-intensity cal-bration curves after analysis of the gel photo using themageQuant TL program.

(7iB

5 (2007) 4540–4553 4541

The mice were immunized intramuscularly in both tibialisnterior muscles with mixtures of either two or four differ-nt DNA constructs. For the immunization with two DNAonstructs, the DNA mixtures with various ratios of DNAonstructs were prepared and a total of 50 �g of DNA (25 �gach muscle) per mouse and per immunization was appliedn a total volume of 100 �l. For the immunizations with fourNA constructs, a DNA mixture containing equal amountsf each DNA construct was prepared and a total of 100 �g ofNA (50 �g each muscle) per mouse and per immunizationas applied.In both cases, mice were treated by injection of 100 �l

ardiotoxin (Latoxan, 0.062 �g/�l) on day 0, which was fol-owed by three consecutive DNA immunizations on days 5,9 and 38–42. Blood samples were collected 11–14 days afterhe third immunization by cardiac puncture.

.3. ELISA

Mice sera were screened for the presence of anti-HPVnd -BPV antibodies by ELISA, whereby the correspondingLPs diluted in PBS served as antigens. All sera were tested

n duplicates.VLPs used in ELISA were produced by infection of

igh Five cells (Invitrogen) with recombinant baculovirusnd purified by CsCl density gradient centrifugation [20].he optimal VLP concentration for coating was deter-ined by titration of each VLP sample. The dilution, athich the saturation of the plates with VLPs of particu-

ar PV types was reached, was used in further analysis.fter coating with VLPs (5 �g/ml in PBS) overnight, thelates were blocked with a 3% blocking solution (3%kim milk, PBS/0.3% Tween-20). Mice sera diluted 1:25 in.5% blocking solution were added to the plates for furtherncubation. A goat–anti-mouse–antibody conjugated withorseradish peroxidase, diluted 1:3000 (Dianova, Germany)n 1.5% blocking solution, was used as secondary anti-ody. The measurement at 405 nm was carried out 10–20 minfter the 100 �l staining solution (100 mM NaAc, 44 mMaH2PO4, pH 4.2, 1 mg/ml ABTS, 0.012% H2O2) was

dded. All the incubation steps were carried out for 1 h at7 ◦C and followed by washing three times with PBS/0.3%ween-20.

.4. Pseudovirion production

Production of pseudovirions was performed by cotrans-ection of 293TT cells with a plasmid containing theumanized HPV 16 L1 and L2 genes and an SV40 originpCDNA4-HPV16L1h-L2h/SV40ori) together with a secondlasmid encoding the reporter protein secreted alkaline phos-hatase (SEAP) under the control of the CMV promotor

pCMVSEAP) following a procedure described earlier [21].× 106 293TT cells were seeded on 10 cm culture dishes

n DMEM (Sigma), supplemented with 10% FCS (Gibco,RL), 1% penicillin/streptomycin (Life Technologies) and

4 accine 2

1facwbcfhp5zwomata

2

t(phwdaoiaefccld

2

epcDaCb0raATm

2

aaBr[obTtw4e(tsqBFgb

3

3p

ibemcrDTIvaw0sctlf

1

542 M. Gasparic et al. / V

25 �g/ml hygromycin (Roche) 18 h in advance. The cotrans-ection was performed using METAFECTENETM (Biontex)ccording to the manufacture’s instructions with minorhanges. Fifteen micrograms of both plasmids were mixedith 70 �l METAFECTENETM in 700 �l of medium, incu-ated for 20 min and added to the cells. After 5 h a mediumhange was performed and the cells were incubated at 37 ◦Cor 3–4 days. For pseudovirion extraction, the cells werearvested by trypsination, washed once with PBS and resus-ended in 1 ml modified PBS (containing 1 mM CaCl2 and.6 mM MgCl2) per 5 × 107 cells. After addition of 4 �l Ben-onase (250 U/ml, Sigma) and 50 �l Brij58 (Sigma), cellsere incubated on ice for 5 min followed by the additionf NaCl to a final concentration of 710 mM to induce theaturation of the pseudovirions during a 10 min incubation

t 4 ◦C. The cellular lysate was cleared by centrifuga-ion (10 min, 3500 × g) and the pseudovirions were storedt −70 ◦C.

.5. Neutralization assay

As target cells 293TT cells were seeded at a concentra-ion of 15,000 cells per well on 96-well plates in DMEMSigma), supplemented with 10% FCS (Gibco, BRL), 1%enicillin/streptomycin (Life Technologies) and 125 �g/mlygromycin (Roche). The following day, the pseudovirionsere diluted in DMEM (1:5000) and mixed with the sera atifferent dilutions. After 15 min incubation at room temper-ture, the medium of the 293TT cells was replaced by 200 �lf the pseudovirion solution. The following controls werencluded: untreated cells, cells treated with pseudovirionslone, and cells treated with pseudovirions in the pres-nce of a known neutralizing polyclonal antiserum specificor HPV 11, 16, or BPV 1. Detection of SEAP in cellulture supernatant was performed 5 days later with thehemiluminescent SEAP Reporter Gene Assay (Roche) fol-owing the manufacture’s instructions. All sera were tested inuplicates.

.6. Transfection of 293T cells

Cotransfection experiments of 293T cells with plasmidsncoding for the HPV 16 L1, HPV 11 L1 and BPV 1 L1roteins and with the empty vector were done using the cal-ium phosphate method [22]. Fifteen micrograms of eachNA were mixed with 50 �l of a 3 M CaCl2 solution andtotal volume of 500 �l was completed with water. TheaCl2/DNA solution was gently mixed with 500 �l 2× BBSuffer (280 mM NaCl, 50 Mm BES; 0.75 mM Na2HPO4,.75 mM NaH2PO4, pH 7.5) and incubated for 15 min at

oom temperature. The medium of the cells was removed andfter one washing step with PBS fresh medium was added.fterwards the transfection solution was added drop by drop.he cells were incubated overnight at 35 ◦C, 3% CO2, andedium was changed after 16–18 h.

TiFat

5 (2007) 4540–4553

.7. VLP preparation and purification

High Five insect cells (Invitrogen) were grown at 27 ◦C todensity of 2 × 106 cells/ml as 250 ml suspension culture inspinner flask using Ex-Cell 405 serum-free medium (JRHiosciences). The cells were infected with baculoviruses

ecombinant for HPV 16 L1 protein as previously described20] and collected 3 days after infection. For purificationf VLPs, pellets were resuspended in 20 ml of extractionuffer (5 mM MgCl2, 5 mM CaCl2, 150 mM NaCl, 0.01%riton X100, 20 mM Hepes pH 7.4 and 1 mM PMSF). Pro-

ein extraction was done using a French Press, and cell lysatesere cleared by centrifugation at 11,700 × g for 10 min at◦C. The supernatant was loaded on top of a two-step gradi-nt consisting of 7 ml 40% sucrose on top of 7 ml of 58% CsClw/v) and centrifuged in a SW28 rotor (Beckman Ultracen-rifuge) at 96,500 × g at 10 ◦C for 3 h. The interphase betweenucrose and CsCl was collected, mixed and transferred to auick seal tube (Beckman). Centrifugation was done in aeckman 70 Ti rotor with 184,000 × g at 20 ◦C for 16–18 h.ractions were collected by puncturing the tubes with a 20auge needle and stored at 4 ◦C. Using ELISA and Westernlot analysis, VLP-containing fractions were detected.

. Results

.1. DNA immunization I: combining twoapillomavirus L1 genes

Our objective was to determine whether L1-directedmmunity against multiple papillomaviruses can be inducedy simultaneous immunization with expression constructsncoding for different L1 proteins. First, we decided to deter-ine the minimal amount of DNA of a given L1 expression

onstruct that is required for the induction of an anti-L1esponse in the background of related as well as unrelatedNA. To this end, we immunized six groups of mice (seeable 1A) with seven different ratios (three mice per ratio,–VII, Table 1A) of HPV 16 L1 vector and either the emptyector, vectors encoding the L1 of other papillomaviruses, orvector encoding a non-papillomavirus gene (GFP). Miceere immunized with a total of 50 �g DNA containing 0,.5, 2.5, 5, 10, 25 or 50 �g of HPV 16 expression con-truct, respectively. Mice were pre-treated by injection ofardiotoxin on day 0 and then immunized a total of threeimes on days 5, 19, and 38–42. Eleven to 14 days after theast booster immunization blood was collected and analyzedor the presence of anti-L1 antibodies by VLP-based ELISA.

In the group of mice immunized with mixtures of HPV6 L1 together with the empty expression vector (group 2,able 1A) 14 out of 15 (17 out of 18 including the three mice

mmunized with HPV 16 L1 DNA only; see Table 2 andig. 1) that received HPV 16 L1 DNA developed antibodiess determined by VLP-based ELISA (Fig. 2). More impor-ant, all mice that were immunized with the lowest dose (i.e.

M. Gasparic et al. / Vaccine 25 (2007) 4540–4553 4543

Table 1Immunization protocol for experiments A–D

0pwstLsfwH

e

g3pp(s1w

.5 �g) of HPV 16 L1 DNA were anti-HPV 16 L1 antibody-ositive. In this group the ratio of 16 L1 to unspecific DNAas 1:100 suggesting that at least 100 different L1 expres-

ion construct could be applied simultaneously. As expected,itering selected sera (Table 1A, group 5: HPV 16/HPV 111) indicated that the amount of DNA injected influences thetrength of the antibody response. The mean titer of the serarom the mice immunized with 50 �g of HPV 16 was 1:1600

hereas in the sera of mice immunized with 25 and 10 �g ofPV 16 L1 DNA the titer was 1:200 (data not shown).Comparable results were obtained when the HPV 16 L1

xpression construct was injected either alone (Table 1A,

aupd

roup 1: 9/15 positive), with the vector encoding GFP (group: 10/15 mice receiving HPV 16 DNA were HPV 16 L1ositive) or with HPV 18 (group 6: 15/15 positive). In theresence of the constructs expressing either HPV 18 L1group 6) or GFP (group 3) 0.5 �g of the HPV 16 L1 expres-ion construct were again sufficient to induce an anti-HPV6 response in five out of six mice. When the HPV 16 DNAas administered alone (group 1), this amount did not induce

measurable response (0/3 positive) which indicates that

nrelated DNA might contribute to the immune responseossible due to tracer effects or due to the fact that largeroses of unmethylated CpG motifs contribute to the induc-

4544M

.Gasparic

etal./Vaccine25

(2007)4540–4553

Table 2Results of DNA immunization I–IV

Group no. Group HPV 16positive

PV Xpositivea

Doublepositivea

HPV 11/16/18/BPV 1 positive

16 Neut. 11 Neut. BPVNeut.

16 + BPVNeut.

16 + 11Neut.

11 + BPVNeut.

16 + 11+ BPV

Immunization A1 16/–b 9/15 n.a. n.a n.a. 9/15 0/15 0/15 0/15 0/15 0/15 0/152 16/emptyb 14/15 n.a. n.a. n.a. 14/15 2/15 0/15 0/15 2/15 0/15 0/153 16/GFPb 10/15 n.a. n.a. n.a. 9/15 4/15 0/15 0/15 4/15 0/15 0/154 16/BPV 1b 2/15 14/15 1/15 n.a. 0/15 6/15 12/15 0/15 0/15 4/15 0/155 16/11b 5/15 15/15 5/15 n.a. 4/15 14/15 0/15 0/15 4/15 0/15 0/156 16/18b 15/15 12/15 12/15 n.a. 10/15 0/15 0/15 0/15 0/18 0/18 0/15(1–6) 16 (50 �g) mice 16/17 n.a. n.a. n.a. 16/18 0/18 0/18 0/18 0/3 0/18 0/18(5) 11 (50 �g) mice n.a. 3/3 n.a. n.a. 0/3 3/3 0/3 0/3 0/3 0/3 0/3(6) 18 (50 �g) mice n.a 2/2. n.a. n.a. 0/3 0/3 0/3 0/3 0/3 0/3 0/3(4) BPV (50 �g) mice n.a. 3/3 n.a. n.a. 0/3 1/3 3/3 0/3 0/3 1/3 0/3(1–3) Empty/GFP/–c 0/9 n.a. n.a. n.a. 0/9 0/9 0/9 0/9 0/9 0/9 0/9

Immunization B7 11/16/18/GFP 1/8 n.a. n.a. 8/1/1/0 1/8 8/8 0/8 0/8 1/8 0/8 0/88 11/16/GFP/BPV 1/8 n.a. n.a. 8/1/0/4 0/8 7/8 2/8 0/8 0/8 2/8 0/89 11/GFP/18/BPV 0/8 n.a. n.a. 8/0/1/7 0/8 8/8 5/8 0/8 0/8 5/8 0/810 11/16/18/BPV 1/8 n.a. n.a. 8/1/0/3 1/8 7/8 1/8 1/8 1/8 1/8 1/811 GFP/16/18/BPV 0/8 n.a. n.a. 1/0/0/7 0/8 0/8 4/8 0/8 0/8 0/8 0/8

Immunization C + D12 16 left 11/BPV right 7/8 n.t. n.t. n.a. 7/8 7/8 1/8 1/8 7/8 1/8 1/813 11/16/BPV right 2/8 n.t. n.t. n.a. 0/8 7/8 0/8 0/8 0/8 0/8 0/814 16 week 1 11/BPV week 2 4/4 n.t. n.t. n.a. 4/4 4/4 1/4 1/4 4/4 1/4 1/415 11/BPV week 1 16 week 2 5/8 n.t. n.t. n.a. 4/8 6/8 2/8 2/8 2/8 2/8 1/8

The numbers show mice positive for anti-L1 antibodies and for neutralization of PV pseudovirions. PV X means positive for antibodies against L1 of the PV in the particular immunization group (HPV 11, 18 orBPV 1). n.a, non applicable; n.t., not tested. Although for immunization experiment A each set of mice consisted of seven groups with three animals each, results for animals immunized with HPV 16 L1 DNAonly or with DNA X only are listed separately.

a Only mice immunized with two different L1s.b Excludes the three mice immunized with 50 �g HPV 16 L1 and three mice immunized with DNA X, these are listed separately.c All mice that were not immunized with any of the L1s.

M. Gasparic et al. / Vaccine 25 (2007) 4540–4553 4545

F PV 16 LF betwee

te

esaoaipitH

Dhoawwwf0tI(I1la

3

atiWtticl

3p

igdogwati

ig. 1. Summary of immunization experiments A–D. The percentage of Hisher exact test was used to calculate statistical significance of differences

ion of humoral immune responses by providing an adjuvantffect.

Applying the HPV 16 L1 expression construct in the pres-nce of BPV 1 L1 or HPV 11 L1 (groups 4 and 5) had aignificant impact on the efficacy by which anti-HPV 16 L1ntibodies were induced (Fig. 3). In the presence of HPV 11nly 4 out of 15 mice developed anti-HPV 16 L1 antibodiesnd in the presence of BPV 1 only 2 out of 15 were positiven HPV 16 L1 ELISA. In addition to the smaller number ofositive mice, positive sera were only found in those animalsmmunized with larger doses of the HPV 16 expression vec-or (5 �g in presence of BPV 1 L1; 10 �g in the presence ofPV 11 L1).In none of the 18 mice that had not received HPV 16

NA anti-HPV 16 L1 antibodies were detected. On the otherand, 16 out of the 17 mice immunized with HPV 16 L1nly (50 �g doses) were anti-HPV 16 L1 positive. There wasrobust response against HPV 18 (14/18 mice immunizedith HPV 18 were positive), BPV 1 (17/18 mice immunizedith BPV 1 L1 positive) and HPV 11 (18/18 mice immunizedith HPV 11 positive) which is probably also due to the

act that, compared to the HPV 16 L1 construct (dose range.5–50 �g) on average larger doses (dose range 25–50 �g) ofhese expression constructs were administered (see Table 1).n the group of mice immunized with HPV 16 and HPV 18 L1group 6) 13 out of 15 sera were positive for both antigens.

n the groups immunized with HPV 11 (group 5) or BPV

(group 4) the number of double positive sera was muchower (5/15 and 1/15, respectively), due to suppression of thenti-HPV 16 response (see Fig. 3).

rpmd

1 antibody-positive sera in the different animal groups are indicated. Then various experimental groups. For details see Table 2.

.2. Cross-reactivity

While generally no specific reactions were detectedgainst HPV 18 or BPV 1 VLPs in mice that had not receivedhe L1 genes of these PVs, some of the sera from mice notmmunized with HPV 11 L1 reacted with the HPV 11 VLPs.

e interpret this result by the inhomogeneity and/or par-ial disruption of the HPV 11 VLPs used for ELISA thathereby expose conserved linear epitopes. On the other handt seems likely that DNA immunization not only inducesonformation-specific antibodies but also antibodies againstinear L1 epitopes.

.3. DNA immunization II: combining fourapillomavirus L1 genes

Although the results described above suggested annhibitory effect of HPV 11 and BPV 1 L1 on the immuno-enicity of the HPV 16 L1 expression construct, we wanted toetermine whether it is possible to simultaneously apply threer four different L1 vaccines. To this end, we immunized fiveroups of mice (Table 1B, groups 7–11; 8 mice per group)ith different combinations of HPV 11, 16, 18, BPV 1 L1

nd GFP expression vectors. In group 10 the mice receivedhe L1 constructs in equal amounts (25 �g each construct),n groups 7, 8, 9, and 11 one of the four L1 constructs was

eplaced by the GFP construct. The results confirmed ourrevious finding of pairwise immunizations. All of the 32ice that were immunized with HPV 11 L1 (groups 7–10)

eveloped an anti-HPV 11 response. There was little or no

4546 M. Gasparic et al. / Vaccine 25 (2007) 4540–4553

Fig. 2. Detection of anti-L1 antibodies and neutralizing activity in mice immunized with HPV 16 L1 + empty vector. Sera of mice were immunized with sevendifferent ratios (50/0 to 0/50 �g) of HPV 16 DNA and empty vector, three mice per ratio and were screened by VLP-based ELISA (top) for the presence ofa ttom).v oup 2.51

rttHto1pi(rwT

e1p

3

lpsc

issBw

nti-HPV 16 L1 antibodies and neutralization of HPV 16 pseudovirions (boector, the last group with 50 �g of empty vector. Except one mouse in the gr6 L1 antibody positive.

esponse against HPV 18 L1 (groups 7, 9, 10 and 11) whenhe reactivity was compared to the group of mice not exposedo HPV 18 L1 DNA. Only 3 of the 32 mice immunized withPV 16 L1 developed anti-HPV 16 antibodies, confirming

hat the HPV 16 response can be suppressed by the presencef either HPV 11 or BPV 1 L1 or both. The response to BPVL1 was partially influenced by the presence of other L1

roteins as well. There was a good anti-BPV 1 L1 responsen the two groups lacking either HPV 16 L1 (7/8 positive)group 9) or HPV 11 L1 (7/8 positive) (group 11), but aeduced response in the two groups where HPV 16 and 11ere both present (3/8 and 4/8 positive) (see Figs. 1 and 4 andable 2).

Of the 40 mice immunized with three or four L1xpression constructs only one was triple positive whereas5 mice were antibody-positive against two different L1roteins.

1tac

The first three mice were immunized with 50 �g of HPV 16 L1 expression/47.5 all mice in the six groups exposed to HPV 16 L1 DNA were anti-HPV

.4. Neutralizing responses

Next, we wanted to assess whether DNA immunizationeads to the induction of neutralizing antibodies, whetherrotection against multiple PV types can be achievedimultaneously and whether DNA immunization inducesross-neutralizing antibody responses.

To determine whether L1 DNA immunization leads to thenduction of neutralizing antibodies, we analyzed the mouseera in a pseudovirion based neutralization assay [21,24]. Allera were tested for their ability to neutralize HPV 11, 16 andPV 1 pseudovirions at a 1:50 dilution. In addition the seraere tested for neutralization of HPV 16 pseudovirions at a

:500 dilution. Since an inhibition of pseudovirions by con-rol sera of up to 55% was observed, we arbitrarily selectedcut off of 75% or greater of pseudovirion inhibition to dis-riminate between neutralization-positive and -negative sera.

M. Gasparic et al. / Vaccine 25 (2007) 4540–4553 4547

Fig. 3. Detection of anti-L1 antibodies in mice immunized with HPV 16 L1 + BPV 1 L1. (A) Mice were immunized with seven different ratios (50/0 to 0/50 �g)of HPV 16 L1 and BPV 1 L1 expression constructs. Anti-HPV 16 (top) and anti-BPV 1 L1 (bottom) antibodies were detected by VLP-based ELISA and eachbar represents one serum at a 1:50 dilution. Fourteen out of 15 mice immunized with both BPV 1 L1 and HPV 16 L1 DNA were BPV 1 L1 sero-positive. Inc re imm1 ere det1 y six of

3r

af

Ec

FdE

ontrast, only two of these mice were anti-HPV 16 L1 positive. (B) Mice we1 L1 expression constructs. Anti-HPV 16 and anti-HPV 11 L1 antibodies w1 L1 and HPV 16 L1 DNA were HPV 11 L1 sero-positive. In contrast, onl

.5. Correlation of antibodies and neutralizingesponses

In general, there was a good correlation of neutralizingctivity and reactivity in VLP-based ELISA (Figs. 2 and 5)or all three PV types assayed. Nevertheless, there were

ftnt

ig. 4. Detection of anti-L1 antibodies in mice immunized with HPV 16, 11, 18 anifferent L1 expression constructs. Antibodies against the four L1 proteins (anti-HPLISA, each bar represents one serum at a 1:50 dilution. While all sera were positiv

unized with seven different ratios (50/0 to 0/50 �g) of HPV 16 L1 and HPVected by VLP-based ELISA. All of the 15 mice immunized with both HPVthese mice were also anti-HPV 16 L1 positive.

LISA-positive sera that repeatedly failed to neutralize theorresponding HPV type. This was more frequently found

or HPV 11 positive sera and might be explained by the facthat DNA immunization induces some responses against non-eutralizing epitopes conserved among different PVs and thathese antibodies are detected by ELISA if VLPs become par-

d BPV 1 L1. A group of eight mice was immunized with a mixture of fourV 16, anti-HPV 11, anti-HPV 18, anti-BPV 1) were detected by VLP-basede for anti-HPV 11 antibodies, only one of the sera reacted with HPV 16 L1.

4548 M. Gasparic et al. / Vaccine 25 (2007) 4540–4553

ccine 2

tniosiadc1

tEpwoPrnaa

3

tIwatiwSoweiHBpB

3e

do

bstawD

oitimiaoHwaifmcianwbcHa

3o

t1LsiasB

Fin1w1r

M. Gasparic et al. / Va

ially denatured. On the other hand, we observed that seraon- or weakly reactive in ELISA were strongly neutral-zing. As ELISA-negative/neutralization-positive sera werenly observed in groups of mice that had received the corre-ponding HPV L1 DNA for immunization this observations likely not an artefact. The correlation of neutralizationnd ELISA was weaker when sera were used at higherilutions (1:500) in the neutralization assay which indi-ates that neutralizing titers for a number of sera was below:500.

Eighty-five mice of groups 1–11 were immunized withwo or more PV L1 expression constructs, 80 of these wereLISA-positive for any PV L1 and 42 were double ELISA-ositive (Table 2). However, only 24 sera of the 85 miceere able to neutralize two different PV types, only 1 outf 40 sera from mice immunized with at least three differentVs was able to neutralize all three PV types tested. Thiseflects again the fact that double or triple immunization didot induce high titers of antibodies against all L1 equally wellnd it is therefore difficult to induce multiple neutralizingctivities.

.6. Is there an indication of cross-neutralization?

We also wanted to determine if there was an indica-ion for the induction of PV-cross-neutralizing antibodies.n fact, with no exception, only sera from mice immunizedith HPV 16 and/or BPV 1 L1 expression constructs were

ble to neutralize HPV 16 or BPV 1 pseudovirions, respec-ively. In contrast, 13 sera obtained from the 107 mice notmmunized with HPV 11 L1 (Tables 1 and 2, groups 1–11)ere able to efficiently neutralize HPV 11 pseudovirions.ix of these mice had been treated with HPV 16 L1, sixther mice with HPV 16 L1 and BPV 1 L1 and one mouseith BPV 1 L1 only. Of the 13 sera 9 were positive for

ither HPV 16 neutralization (6 sera) or for BPV 1 neutral-zation (3 sera). The remaining three sera, neutralizing onlyPV 11, were all from mice immunized with HPV 16 andPV 1 L1. Of these, one serum was anti-BPV 1 L1 ELISA-ositive and two were ELISA-negative for both, HPV 16 andPV 1.

.7. DNA immunization III: topical separation of L1xpression constructs in DNA vaccination

The immunization experiments in which two or moreifferent L1 expression constructs were injected simultane-usly as a mixture indicate that there is a strong interference

itns

ig. 5. Immunization against HPV 11, 16 and BPV 1 L1: topical and chronologicalmmunized with HPV 11, 16 and BPV 1 L1 expression constructs and antibody respeutralization assay (HPV 11, 16 and BPV 1). Mice were immunized as indicated in6 + HPV 11 + BPV 1 right; HPV 16 week 1/BPV 1 + HPV 11 week 2; HPV 11 + BPith HPV 16 L1 DNA alone. Separation of the injection of the HPV 16 L1 express6 L1 responses. For the neutralization assay, pseudovirions were neutralized withepresents one serum, diluted 1:50.

5 (2007) 4540–4553 4549

etween different L1 proteins on the induction of a VLP-pecific antibody response. We wanted to determine whetherhis interference is due to immunodominance of one L1 overnother or whether there is interference in particle assemblyhen two or more L1 genes are expressed in one cell uponNA vaccination.To address these questions, we separated the injections

f the expression constructs either topically or chronolog-cally. Two groups of mice were immunized according tohe scheme in Table 1C. In one group, HPV 16 DNA wasnjected into the left side (tibialis anterior muscle) while a

ixture of the HPV 11 and BPV 1 expression construct wasnjected into the right side of the mice. In another group,ll three expression constructs were injected together intone side. In mice where there was a topical separation ofPV 16 injection from the site where HPV 11 and BPV 1ere injected seven out of eight mice developed a strong

nti-HPV 16 response (Fig. 5) all of which were neutraliz-ng. In this group, seven out of eight mice were also positiveor HPV 11 L1 neutralizing antibodies but only one of theice developed anti-BPV 1 L1 neutralizing antibodies. In

ontrast, in the mice into which the three constructs were co-njected, only two out of the eight mice showed a (weak)nti-HPV 16 L1 antibody response but were negative foreutralization. In this group, again seven out of eight miceere positive for anti-HPV 11 L1 neutralizing antibodiesut none of the eight sera was able to neutralize BPV 1onfirming the previous observation that HPV 11 L1 andPV 16 L1 interfere with the induction of anti-BPV 1 L1

ntibodies.

.8. DNA immunization IV: chronologically separationf injection

Next, we alternated injection of HPV 16 L1 DNA withhe injection of HPV 11/BPV 1 L1 DNA. In one group HPV6 L1 was injected on weeks 1, 3, 5 and HPV 11/BPV 11 was injected on weeks 2, 4, and 6. In another group thischeme was reversed. This immunization protocol led to thenduction of anti-HPV 16 L1 antibodies in four out of fournd five out of eight mice, respectively (Fig. 5). Both groupshowed a robust response against HPV 11 L1 but not againstPV 1 L1.

Taken together, these data indicate that it is possible to

nduce immune responses against multiple HPV types, buthere is a strong interference of different L1s during immu-ization which can be overcome by spatial or temporaleparation of the vaccination.

separation to overcome suppression of anti-HPV 16 responses. Mice wereonses were measured by ELISA (HPV 16 L1, top graph) and pseudovirionsthe bottom graph: HPV 16 DNA left/BPV 1 + HPV 11 right; PBS left/HPVV 1 week 1/HPV 16 week 2. As positive control, two mice were immunizedion construct from that of HPV 11 or BPV 1 L1 led to increased anti-HPVa type-specific control serum (antibody) or left untreated (none). Each bar

4 accine 2

3

tiiioioftFw

LtC1tewoits

Fcntatb

550 M. Gasparic et al. / V

.9. Interference in VLP assembly

Topical separation of injection of HPV 16 L1 DNA fromhe site where HPV 11 and BPV 1 DNA was injected led to thenduction of anti-HPV 16 L1 responses which suggests thatmmunodominance is not the mechanism for the observednterference. To provide evidence for interference on the levelf L1 expression or VLP assembly, we transfected 293T cellsn vitro with HPV 16 L1 expression constructs either aloner in presence of BPV 1 or HPV 11 L1. Two days after trans-ection cells were lysed and presence of HPV 16 VLPs in

he crude extracts was measured by antigen-capture ELISA.ig. 6A shows the results of three independent experimentshich were carried out on different days. When HPV 16

1fL

ig. 6. Recovery of HPV 16 VLPs after co-expression of different L1 genes. 293T constructs. (A) Three days after transfection, extracts of the transfected cells wereutralizing antibody recognizing HPV 16 VLPs and capsomers. In case of cotransransfections 45 �g DNA were used. The experiment was repeated three times indepll three experiments. VLP yield was set to 100% for the cells transfected with HPVhose of HPV 16 L1 only slightly reduced when co-expressed with other L1 protelotting of extracts of transfected cells.

5 (2007) 4540–4553

1 DNA was cotransfected together with HPV 11 L1 DNAhe amount of HPV 16 particles was reduced by about 38%.otransfection of HPV 16 L1 with BPV 1 L1 or with HPV1 L1 plus BPV 1 L1 led to a reduction of HPV 16 par-icles in both cases by about 78%. Results of the controlxperiment, in which HPV 16 L1 DNA was cotransfectedith the empty pUF3 vector, indicated that the interferencebserved in above described cotransfections is unlikely due toncreased DNA amount in the transfection experiment or dueo a titration of transcription factors by the expression con-tructs. When we analyzed the total amount of either HPV

6 L1 or BPV 1 L1 protein in cells after simultaneous trans-ection of three different constructs, we observed that BPV 11 levels remained constant, even when the BPV 1 L1 con-

ells were transiently transfected with HPV 11, 16 and BPV 1 L1 expressione analyzed by antigen-capture ELISA using a conformation-specific and

fection of two constructs a total of 30 �g DNA was used, whereas for tripleendently, the error bar shows the standard deviation of ELISA duplicates of16 L1 + empty vector. (B) Total amounts of BPV 1 L1 remain unchanged,

ins as shown by HPV 16 L1-specific (left) or BPV 1 L1-specific Western

ccine 2

sHoidiav

4

amtdBalcaLt

tbbbwDaabwaaioLrnpepvetdrtawpct

wngmciiytaptHfbLL

tascVaonEtdrinfn1inagootea1

mieaGI

M. Gasparic et al. / Va

truct was co-expressed with HPV 16 and HPV 11 L1. ForPV 16 we observed, in comparison to the reduced amountf HPV 16 VLPs, only a slight decrease of HPV 16 L1 proteinn the presence of BPV 1 L1 or HPV 11 L1 (Fig. 6B). Thisata suggests that presence of either HPV 11 or BPV 1 L1nhibits HPV 16 L1 particle formation also in vivo providing

possible explanation for the observed interference in theaccination experiments.

. Discussion

Prophylactic papillomavirus vaccines became recentlyvailable, the major challenge for future candidate papillo-avirus vaccines is to provide cheaper, simpler vaccination

hat covers more different types. The commercial vaccineseveloped by Merck and Co. Inc. and GlaxoSmithKlineiologicals each cover the two high risk HPV types 16nd 18 leaving out at least 13 other oncogenic papil-omaviruses responsible for almost 30% of all cervicalancer cases. In this study, we analyzed whether co-dministration of DNA expression vectors for different PV1 genes is able to induce immunity against multiple PV

ypes.In previous studies, we and others demonstrated that high

iters of anti-L1 antibodies can be induced with genetic, L1-ased vaccines, although, not surprisingly, efficiency lacksehind the usage of the exceptionally immunogenic protein-ased vaccination using L1 VLPs [16–19,26–30]. Here weere able to confirm that very low amounts (i.e. 0.5 �g) ofNA are sufficient to induce neutralizing anti-L1 responses,

lthough larger amounts of DNA led to higher L1-specificntibody titers. This provoked us to investigate whether com-ined administration of different L1 expression constructsould be a simple solution for the generation of vaccines

gainst multiple papillomaviruses. Unfortunately, the shortnswer to this question is no. In most instances, we observednterference between different L1 vaccines in the inductionf antibody responses. In particular HPV 11 and BPV 11 vectors suppressed the anti-HPV 16 and anti-HPV 18

esponses. Likewise, HPV 16 and 11 L1 seem to have aegative effect on the anti-BPV 1 L1 response. This sup-ression can be overcome by separation of immunizationither topically or chronologically. Although this does notrovide a feasible solution in view to potential prophylacticaccines in humans, it suggests that the observed interfer-nce in simultaneous application of L1 genes is not dueo selective preferences of the immune system but ratherue to direct effects on production or presentation of cor-ectly folded L1 antigen. This is also consistent with the facthat the commercial vaccines effectively induce responsesgainst two or four different HPV types [9,31], however it

ould need to be determined whether co-administration ofurified VLPs of different HPV types under less favourableonditions such as limiting amount of proteins, would leado a well-balanced response against all types. Although

thes

5 (2007) 4540–4553 4551

e do not have an explanation for the interference phe-omena, it is thinkable that co-expression of different L1enes within the same transfected cells leads to the accu-ulation of misfolded L1 aggregates which are unable to

ontribute to the humoral response. In a previous studyt has, however, been shown that mixed particles consist-ng of the L1 types of HPV 6 and 16 can be produced ineast, indicating that different L1s can join in the forma-ion of particles and that these particles induce immunitygainst both HPV types [32]. This is consistent with ourreliminary observation obtained by immunoprecipitation,hat upon cotransfection the L1 proteins of BPV 1 L1 andPV 16 L1 are able to co-assemble. The fact that the inter-

erence is asymmetrical, i.e. HPV type-specific, could thene explained by different expression levels of the different1 genes, e.g. an excess of BPV 1 L1 titrates out HPV 161.

We were also able to demonstrate that DNA immuniza-ion induces neutralizing antibodies. This was as expected,s VLP-based ELISA was already generally accepted asurrogate marker for neutralization [33] although this con-lusion is largely drawn from studies in which sera fromLP-immunized individuals were analyzed by both assays

nd these sera were indeed strongly neutralizing. We alsobserved that ELISA reactivity correlated generally well witheutralizing activity. Nevertheless, a number of sera wereLISA-positive but non-neutralizing. Some of these reac-

ions can be explained by the fact that VLPs are partiallyenatured in the assay and thereby show PV type cross-eactivity. It is conceivable that DNA immunization per senduces also a significant amount of antibodies that are non-eutralizing. A small fraction of sera showed clear evidenceor the presence of cross-neutralizing activity, i.e. immu-ization with HPV 16 DNA led to neutralization of HPV1 pseudovirions. Although this observation was found onlyn very few sera, it might indicate the presence of sharedeutralizing epitopes on HPV particles which, however,re under normal circumstances not sufficiently immuno-enic. Nevertheless, it needs to be determined whether thebserved cross-neutralization is due to the specific naturef DNA immunization, which likely leads to the forma-ion of partial assembly intermediates or whether it can bexplained by specific properties of the HPV 11 pseudovirions,s cross-neutralization was exclusively observed for HPV1.

In conclusion, we observe that DNA immunization againstultiple HPV types is yet not feasible as we observed strong

nterference between different L1 proteins in three differ-nt set of experiments. Recently, a number of improvements,mong them in vivo electroporation or co-administration ofM-CSF have led DNA vaccines into human trial [34–36].

n addition, use of viral vectors could significantly enhance

he effectiveness of genetic vaccines [16,27,37–39], it wouldave to be re-evaluated whether interference between differ-nt L1 genes could be overcome by more potent vaccinationtrategies.

4 accine 2

A

nDg

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

552 M. Gasparic et al. / V

cknowledgements

We are grateful to Petra Galmbacher for excellent tech-ical assistance. This study was supported in part by theKFZ-Canceropole Grand-Est cooperation and by a researchrant from the Deutsche Krebshilfe (10-1912-Kl I).

eferences

[1] de Villiers EM, Fauquet C, Broker TR, Bernard HU, zur Hausen H.Classification of papillomaviruses. Virology 2004;324(1):17–27.

[2] Clifford GM, Smith JS, Plummer M, Munoz N, Franceschi S. Humanpapillomavirus types in invasive cervical cancer worldwide: a meta-analysis. Br J Cancer 2003;88(1):63–73.

[3] Munoz N, Bosch FX, de Sanjose S, Herrero R, Castellsague X,Shah KV, et al. Epidemiologic classification of human papillomavirustypes associated with cervical cancer. N Engl J Med 2003;348(6):518–27.

[4] zur Hausen H. Papillomaviruses and cancer: from basic studies to clin-ical application. Nat Rev Cancer 2002;2(5):342–50.

[5] Walboomers JM, Jacobs MV, Manos MM, Bosch FX, Kummer JA,Shah KV, et al. Human papillomavirus is a necessary cause of invasivecervical cancer worldwide. J Pathol 1999;189(1):12–9.

[6] Harper DM, Franco EL, Wheeler CM, Moscicki AB, RomanowskiB, Roteli-Martins CM, et al. Sustained efficacy up to 4.5 years of abivalent L1 virus-like particle vaccine against human papillomavirustypes 16 and 18: follow-up from a randomised control trial. Lancet2006;367(9518):1247–55.

[7] Villa LL, Costa RL, Petta CA, Andrade RP, Ault KA, Giuliano AR,et al. Prophylactic quadrivalent human papillomavirus (types 6, 11,16, and 18) L1 virus-like particle vaccine in young women: a ran-domised double-blind placebo-controlled multicentre phase II efficacytrial. Lancet Oncol 2005;6(5):271–8.

[8] Harper DM, Franco EL, Wheeler C, Ferris DG, Jenkins D, Schuind A, etal. Efficacy of a bivalent L1 virus-like particle vaccine in prevention ofinfection with human papillomavirus types 16 and 18 in young women:a randomised controlled trial. Lancet 2004;364(9447):1757–65.

[9] Villa LL, Ault KA, Giuliano AR, Costa RL, Petta CA, Andrade RP,et al. Immunologic responses following administration of a vaccinetargeting human papillomavirus types 6, 11, 16, and 18. Vaccine2006;24(27–28):5571–83.

10] Roden RB, Hubbert NL, Kirnbauer R, Christensen ND, Lowy DR,Schiller JT. Assessment of the serological relatedness of geni-tal human papillomaviruses by hemagglutination inhibition. J Virol1996;70(5):3298–301.

11] Roden RB, Greenstone HL, Kirnbauer R, Booy FP, Jessie J, Lowy DR,et al. In vitro generation and type-specific neutralization of a humanpapillomavirus type 16 virion pseudotype. J Virol 1996;70(9):5875–83.

12] Roden RB, Hubbert NL, Kirnbauer R, Breitburd F, Lowy DR, SchillerJT. Papillomavirus L1 capsids agglutinate mouse erythrocytes througha proteinaceous receptor. J Virol 1995;69(8):5147–51.

13] White WI, Wilson SD, Bonnez W, Rose RC, Koenig S, Suzich JA. Invitro infection and type-restricted antibody-mediated neutralization ofauthentic human papillomavirus type 16. J Virol 1998;72(2):959–64.

14] Chen XS, Garcea RL, Goldberg I, Casini G, Harrison SC. Structureof small virus-like particles assembled from the L1 protein of humanpapillomavirus 16. Mol Cell 2000;5(3):557–67.

15] Cheung YK, Cheng SC, Sin FW, Xie Y. Plasmid encoding papillo-mavirus type 16 (HPV16) DNA constructed with codon optimization

improved the immunogenicity against HPV infection. Vaccine2004;23(5):629–38.

16] Kuck D, Lau T, Leuchs B, Kern A, Muller M, Gissmann L, et al.Intranasal vaccination with recombinant adeno-associated virus type 5against human papillomavirus type 16 L1. J Virol 2006;80(6):2621–30.

[

5 (2007) 4540–4553

17] Kuck D, Leder C, Kern A, Muller M, Piuko K, Gissmann L, et al.Efficiency of HPV 16 L1/E7 DNA immunization: influence of cel-lular localization and capsid assembly. Vaccine 2006;24(15):2952–65.

18] Leder C, Kleinschmidt JA, Wiethe C, Muller M. Enhancement ofcapsid gene expression: preparing the human papillomavirus type16 major structural gene L1 for DNA vaccination purposes. J Virol2001;75(19):9201–9.

19] Liu F, Wang JB, Zuo YG, Liu YH, Ma DL. Construction and immuno-genicity of human papillomavirus type 6b L1 recombinant plasmid.Chin Med Sci J 2004;19(3):233–6.

20] Muller M, Zhou J, Reed TD, Rittmuller C, Burger A, Gabels-berger J, et al. Chimeric papillomavirus-like particles. Virology1997;234(1):93–111.

21] Pastrana DV, Buck CB, Pang YY, Thompson CD, Castle PE, FitzGer-ald PC. Reactivity of human sera in a sensitive, high-throughputpseudovirus-based papillomavirus neutralization assay for HPV16 andHPV18. Virology 2004;321(2):205–16.

22] Chen CA, Okayama H. Calcium phosphate-mediated gene transfer: ahighly efficient transfection system for stably transforming cells withplasmid DNA. Biotechniques 1988;6(7):632–8.

23] Zolotukhin S, Potter M, Hauswirth WW, Guy J, Muzyczka N. A“humanized” green fluorescent protein cDNA adapted for high-levelexpression in mammalian cells. J Virol 1996;70(7):4646–54.

24] Buck CB, Pastrana DV, Lowy DR, Schiller JT. Efficient intracellularassembly of papillomaviral vectors. J Virol 2004;78(2):751–7.

25] Mossadegh N, Gissmann L, Muller M, Zentgraf H, Alonso A, TomakidiP. Codon optimization of the human papillomavirus 11 (HPV 11)L1 gene leads to increased gene expression and formation of virus-like particles in mammalian epithelial cells. Virology 2004;326(1):57–66.

26] Lin YL, Borenstein LA, Ahmed R, Wettstein FO. Cottontail rabbitpapillomavirus L1 protein-based vaccines: protection is achieved onlywith a full-length, nondenatured product. J Virol 1993;67(7):4154–62.

27] Liu DW, Chang JL, Tsao YP, Huang CW, Kuo SW, Chen SL.Co-vaccination with adeno-associated virus vectors encoding humanpapillomavirus 16 L1 proteins and adenovirus encoding murine GM-CSF can elicit strong and prolonged neutralizing antibody. Int J Cancer2005;113(1):93–100.

28] Matsumoto K, Kawana K, Yoshikawa H, Taketani Y, Yoshiike K,Kanda T. DNA vaccination of mice with plasmid expressing humanpapillomavirus 6 major capsid protein L1 elicits type-specific anti-bodies neutralizing pseudovirions constructed in vitro. J Med Virol2000;60(2):200–4.

29] Rocha-Zavaleta L, Alejandre JE, Garcia-Carranca A. Parenteral andoral immunization with a plasmid DNA expressing the human papil-lomavirus 16-L1 gene induces systemic and mucosal antibodiesand cytotoxic T lymphocyte responses. J Med Virol 2002;66(1):86–95.

30] Sundaram P, Tigelaar RE, Brandsma JL. Intracutaneous vaccinationof rabbits with the cottontail rabbit papillomavirus (CRPV) L1 geneprotects against virus challenge. Vaccine 1997;15(6–7):664–71.

31] Koutsky LA, Harper DM. Current findings from prophylactic HPVvaccine trials. Vaccine 2006;24(Suppl 3):S114–21 [Chapter 13].

32] Buonamassa DT, Greer CE, Capo S, Yen TS, Galeotti CL, Bensi G.Yeast coexpression of human papillomavirus types 6 and 16 capsidproteins. Virology 2002;293(2):335–44.

33] Pagliusi SR, Dillner J, Pawlita M, Quint WG, Wheeler CM, FergusonM. International Standard reagents for harmonization of HPV serol-ogy and DNA assays—an update. Vaccine 2006;24(Suppl 3):S193–200[Chapter 23].

34] Buchan S, Gronevik E, Mathiesen I, King CA, Stevenson FK, Rice J.

Electroporation as a “prime/boost” strategy for naked DNA vaccinationagainst a tumor antigen. J Immunol 2005;174(10):6292–8.

35] Zhao YG, Peng B, Deng H, Chen G, Yang F, Shao M, et al. Anti-HBV immune responses in rhesus macaques elicited by electroporationmediated DNA vaccination. Vaccine 2006;24(7):897–903.

ccine 2

[

[

[

Aujeszky’s disease virus. Vet Immunol Immunopathol 1999;70(3/4):277–87.

M. Gasparic et al. / Va

36] Giannini SL, Hanon E, Moris P, Van Mechelen M, Morel S, DessyF, et al. Enhanced humoral and memory B cellular immunity usingHPV16/18 L1 VLP vaccine formulated with the MPL/aluminiumsalt combination (AS04) compared to aluminium salt only. Vaccine2006;24(33/34):5937–49.

37] Sedegah M, Weiss W, Sacci Jr JB, Charoenvit Y, Hedstrom R, GowdaK, et al. Improving protective immunity induced by DNA-basedimmunization: priming with antigen and GM-CSF-encoding plasmidDNA and boosting with antigen-expressing recombinant poxvirus. JImmunol 2000;164(11):5905–12.

[

5 (2007) 4540–4553 4553

38] Somasundaram C, Takamatsu H, Andreoni C, Audonnet JC, Fischer L,Lefevre F, et al. Enhanced protective response and immuno-adjuvanteffects of porcine GM-CSF on DNA vaccination of pigs against

39] Song R, Liu S, Adams RJ, Leong KW. Enhancing efficacy of HIV gagDNA vaccine by local delivery of GM-CSF in murine and macaquemodels. J Interferon Cytokine Res 2006;26(6):380–9.