Search for protective Epstein Barr virus (EBV) epitopes in the rabbit model.

Search for protective Epstein Barr virus (EBV) epitopes in the rabbit model.

*Rajčáni, J, *Szenthe, K, **Ďurmanová, V, ***Ásványi, B, *Sollner, J, *Lantis, Zs, Stipkovits, L, ****Szathmary, S

*RT-Europe, Virology laboratory, Mosonmagyaróvár, Hungary, **Department of Immunology, Medical Faculty, Com. Univ., Bratislava, Slovakia

***Food Microbiology, University of Western Hungary and ****Galenbio, Mosonmagyaróvár, Hungary

Contents

• Introducing the rabbit model of Epstein-Barr virus (EBV) infection (Rajčáni et al., Intervirology 57, 5, 239-310, 2014).

• Recapitulation of the principles for selection of protective EBV proteins (glycoproteins) and of their immunogenic epitopes (Rajčáni et al., in Recent patents on Anti-infective Drug Discovery 9, 1, 1-15, 2014; Söllner J. et al.: Immunome Res. 4, Jan 7; p.1. 2008).

• A rationally designed epitope vaccine (Szathmary S. et al., Galenbio, CA).

• Presenting an attempt for in vivo efficacy of EBV vaccines: statistical analysis of infection markers instead of the traditional lethality or morbidity based protection tests.

The rabbit model of EBV infection: pitfalls and solutions

• The need of a high infectious dose (108-109 DNA copy per animal) for eliciting symptoms and/or signs of infection (no practical infectious dose titration available).

• Long term observation periods are needed until post-infection signs occur.

• Controversy concerning to the hematologic changes seen in rabbits: do they resemble to infectious mononucleosis (IM) in man?

• Determination of methods for reliable assessment of EBV replication in rabbit tissues: a possible way for grading prevention following immunization.

• Establishment of latency in EBV infected rabbits.

The ELISA test in EBV infected rabbits

• For each plate, one column (8 wells) was used for the control reagents provided by manufacturer (anti-human IgG kit).

• Instead of single serum dilution testing (as recommended), we determined the final dilution, at which the ISR value (immune status ratio) became negative (lower than 1.1) .

• The interpretation of the ELISA results using the capsid antigen (p23/p18) was confusing, when pre-infection sera (interval 0) showed false positive binding at basic dilution (1:20, occasionally even at 1:40).

• In contrast, the EA-D/p54 ELISA test showed more precise results, which correlated with the virus dose administered (more frequent and higher ISR values following the higher virus dose).

• We came to conclusion that EA-D antibodies might reflect the replication of inoculated virus (later confirmed by immunoblot).

• In contrast, VCA ELISA was considered reflecting the antibody response to the EBV antigens administered.

VCA ELISA

0

100

200

300

400

500

600

700

K6 K9 K11 K13 K16 K18 K21 K23 K2 K3

Rabbit number

seru

m d

ilutio

n a

t po

sitiv

e I

SR

interval 0*

day 8

day 28

month 3

*interval 0 = the serum was obtained 7 days before virus inculation

VCA (recombinant p23/p18) coated immunoplates (provided by Trinity Biotech kit). The serum samples were tested at days 8, 28 and 98 p.i.

Above: in 10 rabbits (inoculated with 3-4x108 EBV DNA) there was no false positive binding of pre-infection sera to immunoplates.

Below: in 10 rabbits (infected 1x109 EBV DNA copies) the pre-infection sera showed false positive binding to VCA-coated immunoplates. In 4 of them the anti-VCA response could not be interpreted.

Conclusion: The capsid antibody response seemed delayed even after the higher virus dose; thus, these results seem confusing.

VCA ELISA

0

50

100

150

200

K4 K7 K8 K10 K12 K14 K15 N3 N5 N6

Rabbit number

seru

m d

iluti

on s

how

ing

po

siti

ve I

SR

interval 0*

day 8

day 28

month 3

Antibodies to the capsid antigen in EBV infected rabbits

The early antibody response in EBV infected rabbits EAD ELISA

0

50

100

150

200

K4 K7 K8 K10 K12 K14 K15 N3 N5 N6 K1

rabbit number

Seum

dilu

tion

at

posi

tive

ISR interval 0*

day 8

day 28

month 3

*interval 0 : 7 days before virus infection

Antibodies against EA-D (Trinity Biotech kit): the immunoplates were coated with the recombinant p54, a cofactor of DNA synthesis; the antibodies were determined at days 8, 28, and 98 p.i.

False positive binding was not found: neither in group A (at above, with exception of a single serum, no. K4) nor in group B (below).

The EA-D antibody response was less prominent in group A (above), being more intensive and frequent in group B (below).

The slightly higher inoculation dose (1x109

DNA copies) elicited a more extensive antibody response (as compared to inoculation of 3-4x108 EBV DNA copies)

0

50

100

150

200

K6 K9 K11K13K16K18K21K23 K2 K3

rabbit number

Seru

m d

iluti

on a

t posit

ive

ISR

interval 0

day 8

day 28

month 3

A

B

Conclusion at single glance: the EA-D antibody response better corresponded to the given EBV dose.

The IgG response* as detected by immunoblot (antibodies to structural and non-structural proteins including Zta/BZLF1)

Day 28 Day 98 (autopsy)

capsid early

More antibodies of higher avidity were seen on day 98 p.i. (exception K1/2). No anti-EBNA1 was found.

p54 (BMRF1, DNA polymerase associated), p138 (BALF2, ssDNA binding protein).

* In group A rabbits infected with the lower virus dose

BZLF1 capsid

Early proteins

Positive rates of IgG antibodies against five EBV proteins as seen in immunoblot strips

IB positive rate

0%10%

20%30%40%50%

60%70%

p18 p23 BZLF1 p138 p54

antigens detected

per

cen

t day 28day 98

total

IB positive rate

0%10%20%30%40%50%60%70%80%

p18 p23 BZLF1 p138 p54

antigens detected

per

cen

t

group A

group B

total

Above the time dependent response: At least 65 % of EBV infected rabbits revealed antibodies to the BZLF1/Zta transactivation polypeptide at day 98, while on day 28 this response was less frequent by nearly 20%.

Below the dose dependent response: the frequency of antibodies to each protein was higher following the larger inoculation dose (black columns are the rabbits of group B).

Conclusion: The positive rates of IgG antibodies detected by IB increased by time as well as following the virus dose.

rabbit Capsid proteins Early proteins ISR posit at dilutionnumber EBNA 1 p18 p23 BZLF1 p138 p54 VCA EA-D0244,K4/3 faint faint 40 <20

0258,K7/3 faint faint 40 <20

0220,K8/3 positive positive 160 200257,K10/3 faint 40 <200229, K11/3 faint faint positive 20 200254,K12/3 positive positive positive 160 400216, K13/3 positive 20 <200215,K14/3 positive faint 40 <200264,K15/3 positive 40 200225,K18/3 faint faint 160 <200280, K21/3 positive positive 160 <200209, K23/3 positive positive 40 400149, N3/3 faint faint <20 <200120,N5/3 faint 160 <20

37/K1 faint faint 40 4038/K2 posit posit posit posit 40 16039/K3 faint faint 160 40total 8 14 3 10

Comparison of the IgG response in EBV infected rabbits (as detected by immunoblot and ELISA)*

*by over 3 months, i.e. on day 98 p.i., only 17 out of 20 infected animals were still available) The IB bands and ELISA results correlated in 15 out 17 animals (poor correlation was seen in sera nos. N3/3 and N5/3). A faint reaction to transactivation protein BZLF1/Zta was detected in 2 samples with negative EA-D ELISA (in the absence of p54 band the VCA/p23 antibodies were present). EBNA1 ANTIBODIES WERE MISSING.

Conclusions from serological examinations

• At 3 months p.i., the correlation between IB bands and ELISA results was excellent in 13 out of 17 animals and satisfactory in 2 additional rabbits.

• When comparing the p54 bands with the EA-D ELISA reults, the positive rate of the latter was slightly lower than that of the former.

• At later post-infection interval, the IB reactive antibodies occurred at higher frequency and showed more intensive bands.

• The presence of antibodies to Zta/BZLF1 (transactivator protein) correlating with the p54 antigen band (DNA polymerase cofactor) might indicate the replication of EBV in permissive rabbit tissues.

• No EBNA1 antibody was found by IB. This suggests that latency I (as found in man) has not developed, at least not during the 98 day observation period.

• We conclude that the EBV DNA might have been eliminated (cleared) from white blood cells in the majority of EA-D and IB/p54 positive rabbits (shown up later).

The rare detection of EBV DNA in peripheral white blood cells (WBC)*

1 = K10/2; 2 = K121/2; 3 = K14/2; 4 = N3/2; 5 = N5/2; 6 = K10/2; 7 = K12/2; 8 = K14/2; 9 = N3/2; 10 = N5/2; 11 = F2/K1E; 12 = F2/K2/E; 13 = F2/K1M (pozitive); 14 = F2/K2M; 15 = F2/K1E; 16 = F2/K2/E; 17 = F2/K1M (pozitive); 18 = F2/K2M; 19 = B95-8 DNA extract 1:100; 20 = B95-8 DNA extract 1:1000; 21 = LCL721 DNA extract 1:100. 22 – water control and marker ladder (100 bp to 1200 bp).

Repeated run with several controls

LMP1 primers:LMP1 forward (169 724-169 741)* sequence 5´-GGG CAA GCT GTG GGA ATG-3´ (18bp)LMP1 reversed (169 911-169 892)* sequence 5´-CTC ACC TGA ACC CCC CTA -3´ (18bp)*As described for the B95-8 derived DNA sequence (Baer et al., 1984)

Posit contr

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 M

Survey of LMP1 antigen (by IF) and DNA detection (by classical PCR) in WBC

Combination of positive IF and PCR results

00,5

11,5

22,5

33,5

K10 K12 K13 K14 K18 K21 N3 K1 K2 K3

rabbit number

rela

tive in

cid

en

ce

day 8

day 28

day 98

day 8 day 28 day 98

Number WBC/LMP1WBC/DNA WBC/LMP1WBC/DNA WBC/LMP1WBC/DNA Spleen/DNA

K10 negat negat negat negat negat posit posit

K12 negat negat negat negat posit posit negat

K13 posit posit negat negat negat negat negat

K14 posit* negat posit* negat negat negat negat

K18 negat negat negat negat negat negat posit

K21 negat negat posit posit ND ND negat

N5 negat negat negat negat posit posit negat

K1 negat negat negat negat posit* negat negat

K2 negat negat neg negat posit posit posit

K3 negat negat positive positive negat negat negat

Total 10 2 1 3 3 4 4 3*Notice: no DNA correlation

The EBV DNA presence and LMP1 antigen in WBC agreed at day 8 (once K13), at day 28 twice (K3, K21) and 3 times at day 98 (K12, N5 and K2). The DNA in WBC and spleen correlated at day 98 (K10); WBC and spleen DNA as well as the LMP1 antigen in WBC agreed once at day 98 in K2.

In one case (K18) the DNA was found only; in two cases (K1, K14) single LMP1 positivity was seen which might be regarded non-specific (PCR was negative).

Lower frequency of DNA presence in comparison to serology in EBV infected rabbits

Comparison of serology, DNA and LMP1 detection

0%10%20%30%40%50%60%70%80%90%

100%

EA-D IB EBV DNA LMP1

methods

posi

tive

rate

(per

cen

t)

day 8

day 28

day 98

The positive rate of EA-D antibodies as detected by ELISA compared to that found by IB was lower on day 98; on day 28 both methods showed equal frequency (about 70%).

The EBV DNA as detected by PCR in WBCs was considerably less frequent (it correlated well with the frequency of LMP1 antigen expressed in WBCs).

We conclude that individual tests for assessing EBV infection in rabbits are the following:

IB > EA-D ELISA > EBV DNA (PCR) > LMP1 (IF). The above shown 4 tests were considered

reliable for the efficacy of selected EBV peptides.

Atypical monocytes and high lymphocyte levels in blood smears of EBV infected rabbits

The proportion of infected rabbits (per cent calculated from 20 animals), which developed high percentage of lymphocytes (>60%) and/or that of atypical mononuclear cells (>10%) is shown at individual post-infection intervals. At days 8 and 28 a considerable proportion of animals (close to 30%) revealed the presence of atypical cells. The frequency of animals developing lymphocytosis correlated with the EA-D antibody rate especially on day 28.

These findings were not regarded to represent the IM-like syndrome.

day 8 day 28 day 98

WBC/DNA 9.5 % 9.5 % 17.6 %

WBC/LMP1 9.5 % 9.5 % 20%

WBC/Atyp 35.7 % 29.4% 23.5 %

WBC/Ly 57.1 % 66.6 % 35.3 %

EA-D ELISA 75% 70% 72%

WBC = white blood cells The WBCs separated by Ficoll gradient show lymphocytosis

EBV immunogenic proteins and their epitopes*

Protein Gene Function Epitope(s) gp350/gp220 BLLF1 Binds to CR2/CD21 receptor 6 gp85/gH BXLF2 Adsorption to B cell chemokine receptor 3 Zta BZLF1 Transactivator protein 2 Rta BRLF1 Transactivator protein 7 Regulator protein BMLF1 DNA polymerase processivity factor 3 DNA poly factor BMRF1 DNA polymerase cofactor 2 Bcl-2/vBcl-2 BHRF1/BALF1 B cell (leukemia 2) growth factor 2 EBNA1 BKHF1 Maintenance of the viral episome 3 EBNA2 BYRF1 EBV transcription regulator and activator 4 EBNA3/EBNA3a BERF1/BLRF3 EBV transcription regulator, TCP chaperon 12 EBNA4/EBNA3b BERF2a/b unclear 9 EBNA5/EBNA-LP BamH-W EBNA2 transcription cofactor, leader protein 1 EBNA6/EBNA3c BEFR3/4 Transcription regulation (repression) 10 LMP1 BNLF1 Latent membrane protein, signaling activator 12 LMP2a/b TR/BNLF1b/c Latent membrane protein, signaling activator 17 Total 15 93

*as reviewed by Rajčáni et al., in Recent patents on Anti-infective Drug Discovery (2014)

The EBV epitopes selected for this study

Epitope number Epitope target Protein/gene considered selected* T cell B cell gp350/220/BLLF1 14 4 + gp42/BZLF2 5 1 + gp110/BALF4 4 1 + dUTPase/BLLF3 8 1 + Rta/BRLF1 2 1 + BMRF2 1 1 + vIL-10/BCRF1 2 1 + p138/BALF2 3 1 + LMP1/BNLF1 5 2 + EBNA1/BKRF1 3 1 + EBNA2/BYRF1 2 1 + EBNA3A 3 1 + Total 52 16 5 7

The selection of EBV antigens suitable for formulation of protective vaccine was based on:

1) a rational strategy combining literature data and application of biological principles

2) considering the EBV receptors and pathogenicity mechanisms,

3) computer based reverse vaccinology and systemic biology (by Söllner J. et al.: Analysis and prediction of protective continuous B-cell epitopes on pathogen proteins. Immunome Res. 2008

Jan 7;4, p.1.*number of epitopes coming from each of 12 proteins (the aa. positions epitope can not be shown, sorry)

Manufacturing of the microbead formulations (at GalenBio, California*)

• Proprietary coupling chemistries to attach precise amounts of a carefully selected combination of epitopes and PRR (pathogen recognition receptor) agonists to the microparticles.

• The coupled ligands are stable for years at room temperature.

• Thoroughly characterized, totally controlled manufacturing system – scalable virtually to any volume.

• Over 20 year experience in manufacturing microparticles (particles larger than 100 µm) at commercial scale (1000+ L lots) under ISO/cGMP-compliant conditions for chromatographic separations of FDA/EMEA licensed biologics.

*patented by Szathmary S., et al.

The rationally designed epitope vaccine components

AntigenAntigen

AdjuvantAdjuvant

CarrierCarrier

Vaccine

Peptide epitopes*

Microparticles

PRR agonists & immune modulators

*Always 3 out of 16 selected epitopes (from 12 EBV proteins) were coupled to the carrier in 19x3 (57) combinations. Typically, one B-cell and two T-cell stimulating epitopes (for one helper and one cytotoxic lymphocyte) were mixed. At least one peptide derived from non-structural protein was present.

PRR = pathogen recognition receptor (non-specific)

The synthetic multivalent peptide vaccine Pathogen-mimicking (Stealth) MicroparticlesTM- PMM produced by Galenbio

PRR agonist (LPS)

PRR agonist (R848)

PRR agonist (IC)

T-helper cell epitope 2

Cytotoxic T-cell epitope (1)

B-cell epitope (3)

Cytotoxic T-cell epitope (1)

B-cell epitope (3)

T-helper cell epitope (2)

PRR = pathogen recognition receptor (non-specific)

The PMM have been designed to interact with various receptors (Toll-like, B and T cells). LPS =lipopolysaccharide; R848 = resimiquod; IC = poly I/poly C

Testing the PMM* formulations in vitro

• Directing PMM to target cells – APCs (dendritic cells and macrophages) and theafter monitoring their movements.

• Beads coated with PRR agonists (i.e. PMM) and selected epitopes were applied together, since even short peptides can interfere with some immune stimulation pathways.

• Efficacy of the carrier microbeads (PMM) in vitro was made by following its uptake into CD80/86 cells and consequently measuring interleukin induction in DCs.

* Pathogen mimicking microparticles

Targetting of PMM into DCs and macrophages (APCs detected according to the CD80/86 marker)

Example of FACS analysis of PMM uptake by dendritic cells (DCs) and macrophages.

The particulate beads were

coated with the PRR agonists as well as with the

selected epitopes to apply both components together. The

PMM uptake into cells expressing

the CD80/86 marker was

demonstrated by cell sorter.

CD80

CD86

CD80+/CD86-

CD80-/CD86+

CD80-/CD86-

CD80+/CD86+

Cytokines induced in vitro by the 19 different PMM formulations coated with the EBV epitopes

Production in DCs Interleukin

Natural production Immature cells Mature cells

Function(s)

IL-1 DCs, macrophages

Yes Yes T-cell differentiation, IL-2 receptor expression

IL-2 Th-cells Yes Yes Proliferation of T/CD3 lymphocytes as well of B lymphocytes

IL-6 Macrophages, Th-cells

No Yes B-cell proliferation and differentiation, IgG production

IL-8 macrophages Yes Yes Capillary adhesion and diapedesis, promotion of inflammation

IL-10 Th-cells No Yes Anti-inflammatory lymphokine IL-12 DCs,

macrophages No Yes NK cell activation,

IFNinduction, promotion of CTL response

TNF Macrophages, T-cells

No Yes B cell proliferation, Cytotoxic T cell differentiation, apoptosis indection

IFN T-cells, NK cells No Yes MHC I expression, macrophage activation

Five epitope mixes exerted the most intensive effect in mature DCs.

Well protected immunized rabbits after their challenge developed minimal antibody response (category I)

group 1

020406080

100120140160

day 0 day 8 day 28 day 108

4702 4998 4798 group 16

020406080

100120140160

day 0 day 8 day 28 day 108

823 850 849

Category I consisted either of the immunization groups showing a minimal EA-D antibody response (on day 8 post infection only, groups 1 and 16) or of groups revealing no EA-D response at all (number 6, 12, 13, 15, 17 and 18 not shown). One out of 3 rabbits, immunization group 14 (depicted in the left), revealed a potent EA-D response by day 108 post-challenge (therefore, this immunization group was shifted to category II).

Conclusion: out of 19 immunization groups, 7 ones (1, 6, 12, 13, 15, 16, 17 and 18) showed a minimal or no serological response when examined by EA-D ELISA.

The EA-D response in slightly protected groups (category II) was following challenge signifantly more frequent

The immunized and challenged animals of category II (groups number 3, 4, 5, 7, 8, 9, 10 and 19) encompassed rabbits, which revealed a moderate EA-D response (2 or 3 serum samples positive). Group 14 from the previous slide was included into this category.

Conclusion: category II consisted of 10 immunization groups (3, 4, 5, 7, 8, 9, 10, 14 and 19) showing a minimal protection only.

Not protected immunized rabbits (category III) behaved after challenge as infected controls

Immunized and challenged animals (in groups 2 and 11) showed an EA-D response similar to mock-immunized controls. The EA-D response in the controls was the highest by day 108. In contrast, in immunization group 2 the stronger response occurred on day 28. Nevertheless, both immunization groups showed positive results that did not differ from the control group.

Conclusion: 2 immunization groups (2 and 11) showed an antibody response similar to controls.

Positive rates at antibody testing by immunoblotIB results in category I rabbits

0102030405060708090

100

p18 p23 BZLF1 p138 p54

Antibodies to antigens

posi

tive

rate

s (p

er c

ent) day 28

day 108

IB results in category II rabbits

0,0

10,0

20,0

30,0

40,0

50,0

60,0

p18 p23 BZLF1 p138 p54

antibodies to antigens

po

sit

ive

ra

te

day 28

day 108

IB results in control and category III rabbits

0,0

20,0

40,0

60,0

80,0

100,0

120,0

p18 p23 BZLF1 p138 p54

antibodies to antigens

po

sit

ive

ra

te (

pe

r c

en

t)

contr day 28

contr day 108

category III day 28

category III day 108

As detected by IB, the category I immunization groups showed a minimal or negative antibody response, which differed from that seen in rabbits of categories II and III (as well as from non-immunized controls).

Summary of serological methods (EA-D ELISA and IB positive rates)

EA-D ELISA Method 1 IB Method 2 Serology (both procedures)

Posit 1 Total 1 Posit 2 Total 2 Group  Posit both Total both per cent Score of protection

Category I

1 8 0 6 1 1 14 7.14 Good

0 6 0 4 6 0 10 0 Full

0 8 1 6 12 1 14 7.14 Good

0 9 13(*) Slight (shifted to II)

0 9 0 5 15 0 14 0 Full

1 8 16(*) Slight (shifted to II)

0 9 17(*) Slight (shifter to II)

0 9 1 5 18 1 14 7.14 Good

Category II

3 8 2 5 3 5 13 38.46 None (shifted to III)

3 8 3 4 4 6 12 50 None (shifted to III)

3 6 3 3 5 6 9 75 None (shifted to III)

2 9 6 6 7 8 15 51.33 None (shifted to III)

2 9 5 6 8 7 15 46.66 None (shifted to III)

2 9 1 6 9 3 15 20 slight

2 8 1 4 10 3 12 25 slight

3 6 13 3 15 20 Slight, *transferred from I

1 8 2 6 14 3 14 21.42 Slight

2 6 16 3 14 21.42 Slight, *trasnsferred from I

3 5 17 3 14 14.28 Slight, *transferred from I

3 9 1 6 19 4 15 26.66 slight

Category III

6 9 4 4 2 10 13 76.92 none

5 6 2 4 11 7 10 70 none

7 9 5 6 Control 1 12 15 80 none (included control)

32 58 31 36 Control 2 64 94 68.08 none (separate controls)

Survey of serological results in immunized and challenged rabbits

1. The immunized rabbits of category I (which have been considered for protected against EBV challenge) finally encopassed 5 immunization proups, since 3 groups (number 13, 16 and 17) were shifted into category II due to relatively frequent IB results.

2. The category II of immunized rabbits fell into 2 subcategories: 2A. Slightly protected rabbits (7 groups: 9, 10, 13, 14,16, 17 and

19). 2B. Not protected rabbits (5 groups: 3, 4, 5, 7 and 8) shifted to

category III4. Category III: originally consisted of 2 groups (clearly unprotected

rabbits not differing from the infected non-immunized controls) but later on 5 groups (mentioned under paragraph 2B) were transferred here, so that together 7 immunization groups could not be considered as protected.

EBV DNA and LMP1 antigen detections in rabbits of categories II and III

F1/WBC F1/WBC F2/WBC F2/WBC Aut/WBC Aut/WBC Aut/Spl Group rabbit

LMP1 DNA LMP1 DNA LMP1 DNA DNA   number 

Category II (9 animals out of 26, 35.6 %)

neg negat negat positive* negat negat negat 7 4718

neg negat negat negat positive negat negat   4797

negat negat positive positive negat negat negat 8 4723

neg negat negat positive negat negat negat 9 4999

negat negat positive positive negat negat negat   4997

neg negat negat negat negat negat negat 10 320

neg negat positive positive ND ND ND   995

posit negat positive positive posit posit negat 14 818

neg negat negat negat ND ND ND   819

Category III (3 animals out of 6, 50 %)

neg negat negat negat negat negat positive 2 4710

negat negat posit negat ND ND ND 11 550

neg negat positive positive negat posit positive   530

Conclusion: no EBV DNA was found in category I (protected) rabbits

Summary of positive rates after including all 4 examination techniques (statistical calculations)

EA-D Method 1 DNA Method 2 IB Method 3 LMP1/IF Method 4 Summary    GroupP T P T P T P T positive Total Per cent

Group A                       50 8 0 8 0(1)* 6 0 8 0 30 0,00 10 6 0 6 0 4 0 5 0 21 0,00 60 8 0 10 1 6 0(1)* 8(8)* 1 32 3,13 120 9 0 12 2 6 0 9 3 36 8,33 130 9 0 6 0(1)* 5 0(1)* 8(8)* 0 28 0,00 151 8 0 10 1 5 0 8 2 31 6,45 160 9 0 8 1 5 0 8 1 30 3,33 18

Group B                       5 + 2 = 73 8 0 7 2 5 0 8 5 28 17,85 33 8 0 10 3 4 0 8 6 30 20,00 4 2 9 1 12 5 6 1 9 9 36 25,00 82 9 2 12 1 6 1 9 6 36 16,66 92 8 1 8 1 4 1 7 5 27 18,52 101 8 2 10 2 6 2 8 7 32 21,87 140 9 2 12 2 5 1 9 6 35 17,14 173 9 0 12 1 6 0 9 4 36 11,11 19

Group C                     4 + 3 = 7 

6 9 1 8 4 4 1 7 12 28 42,85 22 6 0 6 3 3 0 5 5 20 25,00 52 9 2 12 6 6 1 9 11 36 30,55 75 6 3 8 2 4 2 6 12 22 54,55 117 9 3 12 5 5 2 8 17 34 50,00 Control 1

32 58 10 75 31 36 9 60 82 231 35,50 Control 2

Summary

• The novel EBV peptide vaccine presented here is a mix of microparticles, to which the immunostimulation molecules have been bound along with the selected EBV peptides.

• The 19 combinations of selected 16 epitopes were tested in vitro as well in vivo (using the rabbit model elaborated in our laboratory).

• When the peptide vaccine was tested in dendritic cells, 5 of selected epitope combinations induced the most prominent cytokine formation.

• In vivo, the protective effect at least 5 mixes (containing 7 selected epitopes out of 16) exerted a significant protective effect.

• In addition, a very moderate (or slight effect) was noticed after immunization with 7 mixes, while no effect was found after immunization with the rest of 7 mixes.

• We conclude that the rabbit model might be used for EBV vaccine efficacy in vivo provided that the frequency of reliable signs of EBV challenge are compared by statistical methods.