Presented by Linda K. Dixon at the African Swine Fever Diagnostics, Surveillance, Epidemiology and Control Workshop, Nairobi, Kenya, 20-21 July 2011

1. Current Approaches for African Swine
Fever Virus Vaccine Development
Linda K. Dixon1
1 Institute for Animal Health, Pirbright Laboratory, UK

2. IAH Resources for ASFV Research
•High containment (BSL4) laboratory, large
animal facilities and insectary
•OIE Reference Lab for ASFV
•Large collection of ASFV strains and
reagents
•Interdisciplinary research programmes
•3 lines of inbred pigs, colonies of
Ornithodoros ticks
New CL4 Laboratory Complex
being built at IAH Pirbright

3. African swine fever virus
• Large double-stranded DNA virus, genome length 170-190 kbp
• Only member of virus family the Asfarviridae
• Replicates in the cytoplasm – similar strategy to Poxviruses
• Virus particle contains RNA polymerase and other enzymes
needed to start replication cycle – virus DNA is not infectious
• Encodes about 151-167 genes including enzymes required for
replication and transcription of the virus genome
• Many genes ( ~1/3) are not essential for virus replication in
cells but play an important role in virus survival and
transmission
• Replicates mainly in macrophages in vivo
• No vaccine
Institute for Animal Health

4. Nucleo-cytoplasmic large DNA virus superfamily

5. ASFV structure
•ASFV virions have a complex
multilayer structure
•More than 50 proteins
are present
•Extracellular and intracellular
mature virions are both
infectious
a) schematic showing layers
in extracellular virions
b) extracellular virions budding
c) and d) intracellular virus factories
showing immature (IM) and mature
(M) virions
c) chemical fixation, d) high pressure
freezing
Pippa Hawes IAH 200 nm

6. Virus Particle
P72
cD2V
p22
p54
Proteins on surface of
extracellular and
intracellular virus particle
targets for antibody
mediated protection
B438L

7. Virus Genome
160-175 genes Many not essential for replication
182 kbp
Benin 97/1 complete genome Replication MGF100 MGF360
Structural P22 MGF110
unknown evasion MGF505/530

8. Genes involved in immune
evasion/virulence
Inhibitors of host signalling pathways that block
transcription of host immunomodulatory genes
A238L, broad inhibition of host gene transcription.
- Inhibitors of IFN
-Inhibitor of Toll-like receptors TLR 3 and 4
 Adhesion proteins
CD2v, causes binding of infected cells
and virus particles to red blood cells,
impairs lymphocyte proliferation
C-type lectin -resembles NK cell inhibitory
receptors
 Apoptosis inhibitors – IAP and Bcl2 homologues
Comparison of sequences of non-pathogenic
and pathogenic strains

9. Red Blood Cells bound to ASFV Extracellular virus particles
infected macrophages bound to Red Blood Cells
RBC
V
V
V
RBC
ASFV infected macrophage
“Hides” virus particles and infected cells
Courtesy Sharon Brookes

10. Pathogenesis
• Highly virulent isolates ~100% death of pigs within 5 to 12
days. – High viraemia (> 10 8 ) Apoptosis of lymphocytes
Damage to endothelial cells lining blood vesicles, disseminated intravascular
coagulation, haemorrhage
• Moderately virulent isolates cause death of 30 to 50 % of pigs.
- Disease similar to highly virulent isolates but survivors tend
to have lower viraemia (10 4-6). Virus persists in recovered pigs
• Low virulence isolates. Very few deaths. - Occasional low
viraemia 10 2-3 and fever. Virus in tissues. Persistent infection
in pigs.
• Pigs which recover from infection are protected against
challenge with lethal dose of related virulent viruses
• Low virulence isolates provide good model for understanding
protection

11. Non-pathogenic OurT88/3 has deletions and insertions compared to
highly pathogenic Benin97/1 isolate
Left end
MGF110
MGF 530 3FR, NR
MGF110 Benin 97/1 MGF 360 3CL, DL, EL MGF 360 3HL, IL, LL

12. Summary: Genome comparisons Benin 97/1 (highly
pathogenic) compared to OUR T88/3 (non-pathogenic)
• Gene deletions at left end of OURT88/3 genome include
members of MGF360 (6 copies) and MGF530 (2 copies)
implicated in virulence, cell tropism and IFN induction
• CD2v and C-type lectin genes interrupted in OURT88/3. CD2v
implicated in impairing lymphocyte activation
• MGF 300 (1 copy) and MGF 110 (2 copies) in Benin not OUR
T88/3
• MGF 110 (4 copies) and 4 other ORFs in OUR T88/3 not Benin.
• Conserved ORFs encode proteins with 98 to 100% identity.
• Two ORFs encode proteins with variable numbers of tandem
repeats.

13. ASFV Multigene families
• 5 Multigene Families (MGFs)
– A set of genes derived by duplication of an ancestral gene
followed by independent mutational events resulting in a
series of independent genes
• Constitute ~17% - 25% of the coding capacity
• Lack similarity to other known genes, functions unknown
• Vary in gene number between ASFV isolates:
– MGF 100: 2-3 genes per genome
– MGF 110: 5-13 genes per genome
– MGF 300: 3-4 genes per genome
– MGF 360: 11-19 genes per genome
– MGF 530: 8-10 genes per genome

14. Deletion of MGF360 and MGF530 reduces virus growth in
macrophages and virulence in pigs
macrophage virulence in tick IFN
replication pigs replication induction
+ + + -
+ NT + NT
+ NT + NT
+ NT NT
102-103
+ NT +
102-103 102-103
Note these MGF 360 and 530 genes
are also deleted from non-pathogenic
isolate (Chapman et al., 2008)
Zsak et al., 2001, Neilan et al., 2002, Afonso et al,
Burrage et al.,2004

15. Prospects for vaccine development
• Survivors of ASF can resist challenge by related
virulent viruses (eg De Tray 1957, Malmquist 1963,
Handy and Dardiri 1983)
- therefore prospects for ASFV vaccine development
are good

16. Obstacles to ASFV vaccine development
• Inactivated ASF virions do not induce protection
• Serially passaged ASFV vaccine strain used in Portugal
and Spain in 1960s caused post-vaccination reactions in
128,684 of 550,000 vaccinated
-Loss in confidence and need for extensive tests of
vaccine emphasised
• Complexity of virus (~160-175 genes encoded. Virus
particles contain > 50 proteins in several concentric
layers)
• Neutralising antibodies are not effective
• Genetic complexity. Many virus genotypes (22) have
been defined by sequence of the gene encoding the
major capsid protein.

17. However -
• Highest ASFV diversity is in natural hosts (warthogs
and O. moubata ticks) in E and S Africa. Spread of
genotypes to domestic pigs is limited and in some
endemic areas a single genotype is circulating
• In addition cross-protection can be induced between
genotypes (King et al., 2011)
• ASFV is a large DNA virus with more accurate
replication than RNA viruses. This results in a
relatively stable genome.

18. Pigs can be protected:
• Survivors of ASF can resist challenge by related virulent
viruses (eg De Tray 1957, Malmquist 1963, handy and Dardiri 1983)
.
Pigs are protected when:
• Inoculated with viruses attenuated by passage in tissue
culture, eg E75CV (Ruiz Gonzalvo et al., 1986, Gomez-Puertas et al., 1998)
• Inoculated with natural low virulence isolates, eg NHP68,
OurT88/3 ( Leitao et al., 2001, Boinas et al., 2004, Denyer et al., 2006)
- Low sporadic or no viraemia detected, protection close to
100%.
• Inoculated with recombinant virus with single genes deleted
(Lewis et al., 2000, Neilan et al., 2004).
- Viraemia 10 3-6 over ~20 days. High percentage protection

19. Understanding mechanisms of protection
• Identification of correlates of protection for vaccine
development
• Identification of protective immune mechanisms
directs strategies for vaccine development

20. Mechanisms of protection induced by
attenuated viruses: A role for CD8+ T cells
• CD8+ T cells are necessary. Depletion of CD8+ T cells
abrogates protection induced by OURT88/3 (Oura et
al., 2004)
• Protection correlates with frequency of ASF specific
IFN-gamma producing memory T cells
• Ability of different virus isolates to stimulate
lymphocytes from OURT88/3 immune pigs correlates
with cross-protection (King, et al., Vaccine 2011)
• Key virus antigens involved in inducing immunity
mediated by T cells not defined.

21. Mechanisms of protection induced by
attenuated viruses: The role of antibodies
• Pigs can be protected by passive transfer of antibodies from
immune pigs (Onisk et al., 1994). Higher viraemia observed
than in pigs protected by attenuated virus
• Mechanism by which antibodies protect:
- pre virus entry (neutralisation), targets identified p54
(E183L), p30 (CP204L), p72 (B646L)
- post virus entry (infection inhibition), mechanism and targets
not known
• Inhibition of infection in vitro by immune serum correlates
with cross-protection observed in vivo against different
isolates

22. Experimental vaccination with attenuated ASFV strain OURT88/3
OURT88/1 i.m. 10 4
virulent genotype I
OURT88/3 i.m. 10 4 Benin 97/1 i.m. 10 4
or OURT88/3
non-virulent genotype I virulent genotype I
Blood
sampling –
Temperature and
serum and
clinical scores
whole blood
0 7 14 21 28 36 41 49 d56
Termination
of experiment
Introduction of
non-immune
pigs

23. ASFV viraemia and clinical score in vaccinated compared to control pigs
2.5 x 107
Copy number per ml
OURT88/3 OURT88/1 Benin 97/1
VR89
2.0 x 107 VR90
ASFV detected only VR92
1.5 x 107 from non-immune VR97
pigs VR98
VR99
1.0 x 107 VS00
5.0 x 106
Days post
OURT88/3 0
inoculation
20
18 VR89
VR90
Clinical score
Benin 97/1
16 VR92
14 VR97
12 Non-immune pigs VR98
VR99
10 VS00
8
6
4
Days post 2
Benin 97/1 0
1 0 1 2 3 4 5 6 7 8 9 10
challenge

24. IFN-γ ELISPOT 80000
Proliferation assay
1000 8
A VR89 D VR89
800 ● 60000
6 ●
● ●
600
● ●
40000
4
400
●
20000
2 ●
200 ●
ASFV specific IFN-γ production frequency per
0 ▲ ▲ ▲ ▲ ▲ ▲ ▲ 0 0
W-0 W-1 W-2 W-3 W-4 W-5 W-6
●
W-8
0 1 2 3 4 5 6 8 0 1 2 3 4 5 6 8
104)
3H-TdR uptake (∆ cpm X
120000
1000 12
B VR90 ● 100000
10 E VR90 ●
800
● ● ●
● 8
●
80000
600 ●
●
60000
6
400 40000
4
200 20000
2 ●
0 ●
●
▲ ▲ ▲ ▲ ▲ ▲ ▲ 0 0
W-0 W-1 W-2
●
W-3 W-4 W-5 W-6 W-8
0 1 2 3 4 5 6 8 0 1 2 3 4 5 6 8
106 PBMC
1000 100000
10
800
C VR92
● 80000
8
F VR92 ●
600
● 6
60000
● ●
400 4
40000
●
200
● ● 20000
2
●
▲ ● ●
0 ●
0
▲
1
▲
2
▲
3 4
▲
5
▲
6
▲
8
0 0
W-0
0
W-1
1
●
2
W-2 W-3
3
W-4
4 5
W-5 W-6
6
W-8
8
Week post first vaccination
Frequency of IFNγ producing cells increases after 1st immunisation and is boosted after 2nd

25. Anti- ASFV p72 antibody responses Exp 2
14.00 A
12.00
Anti-ASFV antibody
10.00
1803
8.00
1826
titre
6.00
1834
4.00
1845
2.00
0.00
0 10 20 30 40 50 60
14.00 B
Anti-ASFV antibody titre
12.00
1809
10.00
1811
8.00
1829
6.00
1837
4.00
2.00 1844
0.00 1822
0 10 20 30 40 50 60
Days post 1st immunisation
Anti-p72 response rises to day 20 and is boosted by 2nd immunisation. Infection
inhibition assays showed low inhibition of infection in vitro (up to 10 2 )

26. Recognition of diverse strains of ASFV by lymphocytes
from OURT immune pigs correlates with protection
Cross-reactivity of OURT88/3 immune pigs PBMC to
Cross reactivity to OURT88/3
other ASFV isolates : IFN-g ELISPOT Assay
140
120
OURT88/3
OURT88/3 Type I
% Cross-reactivity
Benin 97
Benin -5 Type 1
% cross-reactivity
100
Lisbon 57
Lisbon Type 1
80
Malawi
malawi Type VIII
60
Malta 78
malta Type I
40
Uganda
uganda Type X
20
OURT88/1
OURT 1-6 Type I
0
Pig number VR89 VR90 VR92
Good correlation between IFN-γ cross-reactivity and cross-protection
• OURT88/3, OURT88/1 immune pigs protected against
virulent African isolates ASFV Benin 97 and Uganda
challenge. No cross-protection to Malawi, only partial
protection to Lisbon 57

27. Comparison of complete genomes of Georgia
2007/1 isolate with other ASFV isolates
Kenya 69
Malawi 88
Georgia 2007/1
Mkuzi 79
OURT88/3
BA71V W. Africa
Benin 97/1 Europe
E70
Tengani 62
Warmbaths
E. and S.
Pr 96/4 Africa
0.004
Warthog
Comparison of the concatenated sequences of 125 conserved genes (~40,000 amino acids)
shows the Georgia 2007 isolate is in the same clade as those from Europe and W. Africa
but more distantly related -Chapman et al., Emerging Infectious Diseases 2011

28. Survival of pigs challenged with ASFV isolates form genotype I and X
Challenge of immune
100
Immune - Benin
80
pigs with different ASFV
Percent survival
60
isolates: % survival 40
Exp 1
Exp 1 IAH, UK 20 Benin
Exp 2 ANSES, France –SPF 0
0 5 10
Exp 3 ANSES, France 100
Days post challenge
Exps 1 and 3 100% 80
Immune - Uganda
immunised pigs survived
Percent survival
Immune - Benin
60
challenge with genotype 1 40
Benin
Exp 2
Benin 97/1 20
Exp 2 60%
Uganda
0
0 5 10
immunised pigs survived Days post challenge
challenge with genotype 1 100
OURT88/3 - OURT88/1 - Benin
Benin 97/1 and 100%
OURT88/3 x 2 - Benin
80
Percent survival
genotype X Uganda 60
Some adverse effects of
40
Exp 3
20
immunisation in experiments 0
Benin
in France
0 5 10 15 20
Days post challenge

29. Challenges for attenuated vaccines
• Safety concerns about release of replicating virus
vaccine
• High containment required for production
• Optimised cell culture required for growth of vaccine
strain
• Current strains may not be sufficiently attenuated
• Additional genes involved in virulence deleted from
attenuated strains

30. Virus Genome
160-175 genes Many not essential for replication
182 kbp
Benin 97/1 complete genome Replication MGF100 MGF360
Structural P22 MGF110
unknown evasion MGF505/530

31. Effect of ASFV gene deletions
Gene Function Effect on Effect on Conserved in
virulence replication in isolates
culture
dUTPase, Nucleotide Reduced Reduced Yes
Thymidine metabolism replication in
Kinase macrophages
9GL Virion Reduced Reduced Yes
morphogenesis
MGF Unknown Reduced Reduced No
360/530 IFN induction?
CD2V Binding red Delayed No effect No
blood cells, dissemination
lymphocyte no reduction
function in mortality
DP71L PP1 regulator Can reduce No effect Present as
(short form) long or short
form

32. Effect of ASFV gene deletions
Gene Function Effect on Effect on Conserved in
virulence replication isolates
A238L Inhibitor of None None Yes
host
transcription
C-Type lectin Inhibition of None None No
MHC class I
presentation
IAP Apoptosis None None Yes
inhibition
UK Unknown Reduced None Yes

33. Subunit vaccines
• Partial protection achieved with recombinant
proteins expressed in baculovirus:
- a mixture of proteins p30 and p54 (Gomez-Puertas et al.,
1996) – NB Neilan et al., 2004 reported no protection
- CD2-like protein (or haemmaglutinin) (Ruiz-Gonzalvo et al.,
1999)
• Delay in onset of disease signs and viraemia, some
pigs recover from infection and clear virus

34. Challenges for subunit vaccines
• Identification of additional protective antigens
especially dominant antigens recognised by CD8+T
cells
• Identification of vaccine delivery systems for pigs to
induce cell-mediated and antibody responses
eg host restricted virus vector such as swinepox or
avipox

35. Rapid vaccine development platform
• Collaboration Kathy Sykes, Bert Jacobs, Biodesign
Institute, Arizona State University, IAH Pirbright UK
• Genome wide screen of ORFs encoded by Georgia
2007 ASFV isolate to rank proteins for induction of
cell mediated and antibody responses in pigs
• Genes delivered in pools of 20- 40 to pigs by
DNA/prime recombinant vaccinia virus boost
• Antibody and cell mediated immune responses to
individual antigens measured using individual in vitro
translated proteins
• Test smaller pools “best” antigens for ability to
protect pigs against lethal ASFV challenge

36. Strategy for ASFV Library Construction

37. Genome wide screen for protective ASFV antigens
Collaboration IAH- Biodesign Institute, Arizona State University
Immunize pigs with
expression libraries in
pools by DNA prime
Assay sera, PBMC, RNA for immune responses
recombinant vaccinia
virus boost
Sort and Rank all ORFs
Ab Iso- Th1 Th2 Cyto-
type kine
ORF10 ORF5 ORF69 ORF50 ORF100
Test in Pig Select antigens to ORF113 ORF811 ORF98 ORF63 ORF39
Challenge- test .
.
.
.
.
.
.
.
.
.
Protection . . . . .
Assays . . . . .
. . . . .
. . . . .
. . . . .

38. Proteome-scale protein
production and purification
T7 RBS ATG TRX ORF His Term
In vitro synthesis of proteins
Linear DNAs for in vitro
transcription/translation
Magnetic beads for capture and
purification proteins

39. Challenge/protection experiments
1. Pool top antigens from each bin and immunize pigs with
these pools of antigens by DNA prime and recombinant
vaccinia virus boost.
Immunize Survival readout
Challenge
?
2. Pool top 5-10 antigens from positive bins, and immunize
pigs.
3. Re-test and validate vaccine candidates

40. 3H-TdR uptake, cpm
0
100
200
300
400
500
600
700
ASFV113
ASFV163
individual antigens
ASFV127
ASFV105
Proliferation Assays:
ASFV145
from immunised pigs with
ASFV154
Stimulation of lymphocytes
ASFV083
ASFV006
Pool of 12
ASFV194 3H-TdR uptake, cpm
ASFV132
0
100
150
200
250
300
350
400
450
50
ASFV205
ASFV170
ASFV002
PHA
medium ASFV004
ASFV006
ASFV011
ASFV012
ASFV037
ASFV052
ASFV053
ASFV054
ASFV068
ASFV070
3H-TdR uptake, cpm ASFV07…
ASFV07…
0
100
200
300
400
500
600
Pool of 22
ASFV083t
ASFV105
ASFV111
ASFV122
ASFV128
2 antigens
ASFV167
ASFV179
ASFV113
ASFV127
PHA
medium
Immune responses in pigs immunised with pools of antigens:
Antigen pool complexity does not reduce T cell response level

41. Antigen pool complexity does not reduce
antibody response
Group 1 (pool of 22) vs. VP30 Group 3 (pool of 12) vs. VP30
1.6 127 pre 2
127 pre
1.4 127 post 1.8
1.6 127 post
1.2 1.4
1 1.2
0.8 1
0.8
0.6
0.6
0.4 0.4
0.2 0.2
0 0
394 395 410 419 420
393 396 398 404 405 406
Pig # Pig #
Group 2 (pool of 22) vs. VP30 Group 4 (pool of 2) vs. VP30
1.6
127 pre 2
127 pre
1.4 1.8
127 post 1.6 127 post
1.2
1.4
1 1.2
0.8 1
0.8
0.6
0.6
0.4
0.4
0.2 0.2
0 0
397 407 409 411 412 424 403 417 418 421 423
Pig # Pig #
ELISA assays

42. Summary of Progress: genome wide
antigen screen
• DNA vaccine and protein expression libraries
complete
• rVV library 47 complete
• Immunome screening in pigs – conditions
optimised and 47 antigens tested by DNA
prime rVV boost
• T cell and antibody assays used to rank ORFs
for immune responses
• Challenge experiments in progress

43. Future Priorities Vaccines
• Attenuated vaccines: Rational strategy for
attenuation
• Better knowledge of cross-protection between
genotypes- antigens involved in cross-
potection
• Optimised cell culture
• Subunit vaccines: Identification of protective
antigens especially those which induce CD8+ T
cell responses
• Incorporation and testing in host-restricted
gene delivery systems

44. Future work vaccines
• Subunit vaccines – complete screen for
protective antigens
• Test in pools in challenge experiments
• Select best antigens and clone in host-
restricted vaccine delivery vector

45. Acknowledgements
IAH UK Biodesign Institute Arizona
• Linda Dixon
• Dave Chapman State University
• Lynnette Goatley Center for Infectious Diseases
• Fuquan Zhang • Bert Jacobs
• Charles Abrams • James Jankovich
• Emma Fishbourne • Greg Golden
• Pam Lithgow Center for Innovations in Medicine
• Derah Arav • Kathy Sykes
• Mark Robida
• Geraldine Taylor
ANSES Ploufragan, France
• Haru Takamatsu
• Marie-Frederique le Potier
• Katherine King
• Chris Netherton • Evelyne Hutot
• Josie Golding • Roland Carriolet
• Pippa Hawes Univ. Victoria, Canada
• Chris Upton www.virology.ca
• Don King
• Chris Oura
• Carrie Batten
• Geoff Hutchings

46. Genotypes of ASFV isolates
Penrith et al., 2007
Data from partial sequence of gene
encoding p72 capsid protein

47. Antibody response following DNA prime
rVV boost compared to infection
Uninfected and ASFV-infected pigs vs. VP30 Pre/Post Immunization vs.
1.2 VP30
2.5
1 pre VP30
2
post VP30
0.8
1.5
0.6
1 0.4
0.5 0.2
0 0
1:100 1:500 1:2500
cont 01 cont 04 60 76 105 184
Uninfected and ASFV-infected pigs vs. VP72 Pre/Post Immunization vs. VP72
0.6 1.2
0.5 1
pre VP72
0.4 0.8 post VP72
0.3 0.6
0.2 0.4
0.1 0.2
0 0
cont 01 cont 04 60 76 105 184 1:100 1:500 1:2500