Download General enquiries on this form should be made to:

General enquiries on this form should be made to:
Defra, Science Directorate, Management Support and Finance Team,
Telephone No. 020 7238 1612
E-mail:
[email protected]
SID 5



Research Project Final Report
Note
In line with the Freedom of Information
Act 2000, Defra aims to place the results
of its completed research projects in the
public domain wherever possible. The
SID 5 (Research Project Final Report) is
designed to capture the information on
the results and outputs of Defra-funded
research in a format that is easily
publishable through the Defra website. A
SID 5 must be completed for all projects.
1.
Defra Project code
2.
Project title
This form is in Word format and the
boxes may be expanded or reduced, as
appropriate.
3.
ACCESS TO INFORMATION
The information collected on this form will
be stored electronically and may be sent
to any part of Defra, or to individual
researchers or organisations outside
Defra for the purposes of reviewing the
project. Defra may also disclose the
information to any outside organisation
acting as an agent authorised by Defra to
process final research reports on its
behalf. Defra intends to publish this form
on its website, unless there are strong
reasons not to, which fully comply with
exemptions under the Environmental
Information Regulations or the Freedom
of Information Act 2000.
Defra may be required to release
information, including personal data and
commercial information, on request under
the Environmental Information
Regulations or the Freedom of
Information Act 2000. However, Defra will
not permit any unwarranted breach of
confidentiality or act in contravention of
its obligations under the Data Protection
Act 1998. Defra or its appointed agents
may use the name, address or other
details on your form to contact you in
connection with occasional customer
research aimed at improving the
processes through which Defra works
with its contractors.
SID 5 (Rev. 3/06)
Project identification
SE1512
Identification of in vitro and in vivo correlates for
protection of pigs by candidate African swine fever virus
vaccine
Contractor
organisation(s)
Institute for Animal Health
54. Total Defra project costs
(agreed fixed price)
5. Project:
Page 1 of 16
£
885,000
start date ................
01 July 2006
end date .................
31/03/2010
6. It is Defra’s intention to publish this form.
Please confirm your agreement to do so. ................................................................................... YES
NO
(a) When preparing SID 5s contractors should bear in mind that Defra intends that they be made public. They
should be written in a clear and concise manner and represent a full account of the research project
which someone not closely associated with the project can follow.
Defra recognises that in a small minority of cases there may be information, such as intellectual property
or commercially confidential data, used in or generated by the research project, which should not be
disclosed. In these cases, such information should be detailed in a separate annex (not to be published)
so that the SID 5 can be placed in the public domain. Where it is impossible to complete the Final Report
without including references to any sensitive or confidential data, the information should be included and
section (b) completed. NB: only in exceptional circumstances will Defra expect contractors to give a "No"
answer.
In all cases, reasons for withholding information must be fully in line with exemptions under the
Environmental Information Regulations or the Freedom of Information Act 2000.
(b) If you have answered NO, please explain why the Final report should not be released into public domain
Executive Summary
7.
The executive summary must not exceed 2 sides in total of A4 and should be understandable to the
intelligent non-scientist. It should cover the main objectives, methods and findings of the research, together
with any other significant events and options for new work.
African swine fever virus (ASFV) causes a haemorrhagic fever with very high mortality in domestic pigs.
The introduction of ASF to Georgia in 2007 and spread to neighbouring countries, emphasises the
continuing risk that ASF poses to pig farming and food security worldwide. Since a vaccine is not available
for ASF, control of outbreaks relies on rapid diagnosis, implementation of movement restrictions and
slaughter of pigs on affected farms. Development of a vaccine would provide an alternative control method
to mass slaughter and would be particularly important in regions where infected wildlife reservoirs
including soft ticks and wild suids are present. We previously showed that protection induced by
immunization with the non-pathogenic ASFV strain OURT88/3 is dependent on T lymphocytes that
express CD8 on the cell surface. Protection was also shown to correlate with induction of a subset of
these T lymphocytes that express in addition to CD8, high levels of CD4 and also perforin. The results
suggested that induction of this T lymphocyte subset might be used as an indicator of whether pigs are
protected.
This project has further advanced vaccine development by: 1) Developing in vitro assays which can be
used to predict the pathogenesis of ASFV strains. These tests will be used to prescreen candidate
attenuated vaccine strains and thus reduce the requirement for animal experiments. 2) Identifying
protective immune responses. This involved further validation of the correlation in protection with induction
of a CD8+ T lymphocyte response and development of an assay to predict cross-protection between
genetically different ASFV isolates. In addition we started a large scale screen of the ASFV proteins to
identify those which can induce a protective immune response. Identification these would enable
vaccines based on delivery of just those proteins to be developed. Knowledge of the key protective
proteins will also facilitate prediction of cross-protection based on sequence of the genes which code for
the protective proteins. 3) Commercial production of attenuated ASFV vaccines will require a cell line
which can support growth of the vaccine strain. We have tested a number of cell lines to identify a suitable
cell line for vaccine production. The main achievements for each of these 3 objectives are summarised
below.
1)
Development of in vitro assays to aid selection of attenuated ASFV vaccine candidates. Several
assays were tested to identify those which could distinguish between infection with non-pathogenic
compared to pathogenic ASFV strains. All of these assays involved in vitro culture of cells from noninfected pigs. Those assays which distinguished between infection with non-pathogenic and pathogenic
isolates included measurement of host transcriptional responses in macrophages infected with isolates
which varied in pathogenesis. This analysis identified that some messenger RNAs for proteins involved in
activating the host defence’s against virus infection are increased in amount in cells infected with the nonpathogenic compared to pathogenic isolates. Those which showed the greatest differences included
mRNAs for chemokines which are messengers secreted from cells to attract different components of the
SID 5 (Rev. 3/06)
Page 2 of 16
immune system. The other assays tested included measurement of host defence pathway activation in
infected macrophages, effect of ASFV infection on activation and function of naïve immune cell
populations including NK (Natural Killer), NKT, T cells, B cells, dendritic cells (DCs). The results showed
that co-culture of ASFV infected cells with lymphocytes greatly increases the proportion of cells with NK
and NKT cell phenotype. These cells were activated to produce an important cytokine for immune
response activation, IFN . The results showed that infection with non-pathogenic isolates induced more
IFN than infection with pathogenic isolates. Although the trend was for pathogenic isolates to induce
greater B cell stimulation compared to non-pathogenic isolates, this assay was not fully optimised.
2) Identifying protective immune responses. Immunisation of pigs with the non-pathogenic OURT88/3
isolate followed by closely related OURT88/1 isolate is known to protect pigs against lethal challenge with
closely related isolates and this protection depends on CD8+ lymphocytes. In this project we developed an
assay to predict the ability of the OURT88/3 isolate to cross-protect pigs from challenge with more
distantly related isolates. This assay measures stimulation of immune lymphocytes from OURT88/3 and
OURT88/1 immunised pigs with different ASFV isolates and measurement of the numbers of IFN 
producing cells as an indication of cross-recognition. The results showed that OURT immune lymphocytes
were stimulated by other isolates from the same genotype, including Benin 97/1 as well as by isolates
from other genotypes including Uganda 1965 isolate from genotype X but not by genotype VIII isolate
Malawi LIL 20/1. Cross-protection experiments were carried out to validate these predictions and this
confirmed that, as predicted, pigs immunised with OURT88/3 and OURT88/1 isolates were protected
against lethal challenge with Benin 97/1 and Uganda 1965 isolates. Previously it was shown that
OURT88/3 and OURT88/1 immunised pigs were not protected against challenge with genotype VIII isolate
Malawi Lil20/1. These experiments also provided the opportunity to further validate a clinical scoring
system for ASFV and to further confirm the correlation of protection with induction of the
CD4hiperforin+CD8+ T lymphocyte subset. They also provide materials to validate the use of quantitative
PCR to measure viraemia and to optimise protocols for this. Samples were also provided to validate
themultiplex PCR for ASFV and CSFV diagnosis funded by DEFRA. Samples collected are also available
for future validation of new diagnostic assays.
An alternative approach to vaccine development is to identify the ASFV proteins which can induce a
protective immune response and incorporate the genes encoding these or recombinant proteins
expressed from them into appropriate vaccine delivery vehicles. To accomplish this we have started
collaborating with Prof Bert Jacobs and Dr Kathryn Sykes at Arizona State University USA. To screen for
protective antigens we plan to clone all of the genes encoded by the ASFV Georgia 2007/1 isolate for
delivery to pigs as a DNA prime and recombinant vaccinia virus boost . Antibody and T lymphocyte
response to individual ASFV proteins will be measured using the individual proteins expressed in an in
vitro system. The DNA vaccines and rVVs have been delivered to pigs in pools of up to 20. Following this
initial screen for immunogenicity we will select those genes which induce the best immune responses and
test in pools for ability to confer protection against lethal challenge. So far we have optimised the
procedures for delivery of genes by DNA prime and rVV boost and screened 42 ASFV proteins for their
ability to induce antibody and T cell responses.
3) Screening of cell lines which support replication of ASFV field isolates. Attempts made to rationally
modify existing cell lines so that they could support replication of ASFV filed isolates were not successful.
The modifications tested included stable expression of genes which inhibit the Interferon response to
infection and stable expression of a putative cellular receptor for ASFV CD163. In collaboration with
Friedrich Loeffler Institute existing cell lines were tested for their ability to support replication of filed
isolates. This identified a wild boar cell line WSLR which could support replication of some ASFV field
isolates.
Project Report to Defra
8.
As a guide this report should be no longer than 20 sides of A4. This report is to provide Defra with
details of the outputs of the research project for internal purposes; to meet the terms of the contract; and
to allow Defra to publish details of the outputs to meet Environmental Information Regulation or
Freedom of Information obligations. This short report to Defra does not preclude contractors from also
seeking to publish a full, formal scientific report/paper in an appropriate scientific or other
journal/publication. Indeed, Defra actively encourages such publications as part of the contract terms.
The report to Defra should include:
 the scientific objectives as set out in the contract;
 the extent to which the objectives set out in the contract have been met;
 details of methods used and the results obtained, including statistical analysis (if appropriate);
 a discussion of the results and their reliability;
 the main implications of the findings;
SID 5 (Rev. 3/06)
Page 3 of 16


possible future work; and
any action resulting from the research (e.g. IP, Knowledge Transfer).
Background
African swine fever virus (ASFV) causes major economic losses in affected countries and threatens food security
worldwide. The recent spread in 2007 of ASF to Georgia and neighbouring countries including the Russian
Federation increases the risk to other countries. There is no vaccine against ASFV limiting the options for disease
control. Following an outbreak of ASF near St Petersburg in October 2009 FAO called on countries to increase
vigilance and to renew efforts to produce a vaccine. A recently published scientific opinion from EFSA
(http://www.efsa.europa.eu/en/scdocs/doc/1556.pdf) estimated that there is a moderate risk that ASF will remain
endemic in the Trans-Caucasus and Russian Federation and a moderate risk that ASF will be introduced in the EU.
The risk of further spread in the EU is moderate to high depending on the farming sector to which ASF is introduced.
In this project we have extended our previous work in ASFV vaccine development. We have previously shown that
inoculation of pigs with a non-pathogenic ASFV isolate OURT88/3 can protect pigs against challenge with related
isolates and that CD8+ T cells are required for this protection. We are working to construct rationally attenuated
ASFV vaccine strains by sequential deletion of genes involved in virulence and immune evasion. To limit the number
of pig experiments required to test these candidate vaccine strains ideally we need to have assays which can be
used in vitro to predict the pathogenesis of virus isolates and their ability to induce protective immunity. The first
objective in this project was to test a number of in vitro assays to predict pathogenesis. We have succeeded in
identifying in vitro assays which can be used to predict pathogesis of isolates although future work will be needed for
further validation.
The second objective was to identify protective immune response(s) induced by non-pathogenic ASF viruses. The
identification of correlates of protection is essential for effective testing of vaccines. We have further validated the
correlation of protection against lethal ASFV challenge with the stimulation of a particular subset of memory T cells
that are CD4hiCD8+perforin+ T cells. We have also developed an assay that can be used to predict whether
immunity induced by the OURT88/3 isolate can cross-protect against challenge with different ASFV isolates. We
confirmed that OURT88/3 isolate can protect pigs against challenge with African isolates of the same genotype and
a different genotype (genotype X). Development of subunit vaccines is desirable for use in countries in which ASF is
not endemic. In order to identify which ASFV antigens can induce a protective response we have begun a genome
wide screen to identify these.
The practical use of vaccine requires that there is a cell line in which vaccine can be produced. To achieve the third
objective we have tested a number of cell lines for their ability to support ASFV replication and identified one cell line
that can be used for this purpose.
Objectives
The objectives of the project are: 1) To develop in vitro assays that will aid the selection of attenuated ASF vaccine
candidates. 2) To identify protective immune response(s) induced by non-pathogenic ASF viruses. 3) To construct
porcine cell lines expressing inhibitors of interferon and putative receptors for ASFV and test these for their ability to
support high level replication of candidate attenuated African swine fever virus vaccines.
Objective 1) To develop in vitro assays that will aid the selection of attenuated ASF vaccine candidates.
1.i Transcription profiling of macrophages infected in vitro with candidate vaccine strains (milestones 2, 7
and 11).
Transcription of host macrophage genes was compared during the course of infection with either non-pathogenic
or pathogenic ASFV isolates. The aim was to identify differences in the host response to infection which could be
used as indicators for the pathogenesis of different isolates and candidate vaccine strains. A pig oligonucleotide
(70mer) microarray containing over 13000 target genes was used to study alveolar macrophage gene expression
in response to infection with four different pathogenic isolates of ASFV. 626 genes from virus-infected cells were
shown to be significantly changed in levels of transcription compared to mock-infected cells. These differentiallyexpressed genes could be broadly separated into six main groups. The differentially expressed genes were very
similar in macrophages infected with all of these pathogenic isolates.
Transcription levels of 42 key genes involved in activating and regulating the host’s response to infection were
compared over 8 time points using quantitative reverse transcription PCR. The results showed that many cytokine
genes were highly up-regulated following infection. At early times post-infection the non-pathogenic strain
OURT88/3 induced significantly higher levels of mRNA (up to 8 fold) of the gene encoding Tumour Necrosis
Factor alpha (TNF) compared to the pathogenic strain (Fig 1). TNF has a primary role in the regulation of
immune cells, the induction of apoptosis and the inhibition of viral replication. The pathogenic strain of ASFV may
be inhibiting the expression of host TNF hence may be inhibiting the host immune response to viral infection.
SID 5 (Rev. 3/06)
Page 4 of 16
A second set of genes up-regulated upon infection with ASFV includes the chemokines. The products of these
genes can act as chemoattractants for leukocytes, monocytes, neutrophils and other effector cells from the blood
to sites of pathogen infection. One such gene MIP-1 is very highly expressed (>10 fold increase) at 2 hours
post-infection in OURT88/3 infected cells (Fig 2) compared to Malawi (2 fold increase). Thus the highly
pathogenic strain Malawi Lil20/1 may suppress host expression of cytokine and chemokine genes including MIP1 and TNF leading to reduced clearing of virus infection.
1.ii In vitro assays for the effects of ASFV infection on immune cell function. a) Activation of host defence
pathways (milestones 3, 8 and 13)
In vitro infections of porcine macrophages with virus deletion mutants were carried out, to identify possible
correlates of pathogenesis and induction of protective immunity. Production of type I interferons (IFN) was
measured using a bioassay. Type I interferons mediate induction of an antiviral state which limits virus replication.
Infection with wild type ASFV suppressed production of type I IFN. Infection of macrophages with mutants from
which the genes DP71L (short or long forms) were deleted also suppressed production of IFN. Infection with a
virus mutant, from which 6 members of MGF360 and 2 members of MGF530 had been deleted, induced greater
production of type I IFN. Deletion of these MGF genes had previously been shown to reduce virulence of the virus
for pigs showing a possible correlation between virulence and induction of type I IFNs.
1.ii b) Effect of ASFV on NK and NKT cells (Milestones 1 and 10).
Pigs have large numbers of circulating NK cells and previous studies have suggested that there may be a
correlation between increased NK cell activity and protection of vaccinated pigs against lethal ASFV challenge
(DEFRA SE1511 final report, 2006; Leitão et al., 2001). Thus we examined whether the interaction of naïve pig
lymphocytes with ASFV results in activation or inhibition of NK and NKT cell function using IFN- ELISPOT
assays and by comparing the proportion of activated NK and NKT cell subsets by FACS analysis. When naïve
pig lymphocytes were co-cultured with ASFV-infected syngeneic cells, the proportion of both activated NK and
NKT cell phenotypes dramatically increased (Fig 1iib). Further experiments indicated that naïve pig lymphocytes
co-cultured in vitro with ASFV-infected cells produced IFN (examined by ELISPOT assay). Interestingly, tissueculture adapted (attenuated Uganda and Ba71V) and non-virulent virus (OURT88/3) isolates induced IFN
production by naïve pig lymphocytes whereas pathogenic virus isolates generally failed to induce IFN. To
determine which cell populations were required for production of IFN different cell subpopulations were depleted
prior to co-culture with ASFV-infected cells. These cell depletion experiments indicated that NK cells and some T
cells (most likely NKT cells) are responsible for IFNproduction from naïve pig lymphocytes stimulated by ASFV
infected cells whereas  Tcells were not involved in producing IFN The IFNγ ELISPOT assay was also used to
determine the ability of ASFV deletion mutants to stimulate naïve pig lymphocytes for IFNγ production.
Interestingly, BeninCD2v, which contains a deletion of the CD2v gene from the Benin 97/1 isolate induced much
higher levels of IFN from naïve lymphocytes than the parental Benin97/1 virus. However, deletion of the CD2v
gene from the Malawi LIL20/1 isolate (MalawiCD2v) did not increase IFN production in comparison to the
parental Malawi LIL 20/1 isolate. The results indicate that the CD2v protein is involved in suppressing
IFNproduction from naïve lymphocytes co-cultivated with ASFV infected cells. They also indicate that the Malawi
LIL20/1 isolate has additional mechanisms that may perform this function which can compensate for loss of the
CD2v gene. At the moment the mechanism involved is not clear but it is possible that the ability to suppress IFN
production by naïve lymphocytes is correlated with virulence. This needs to be further evaluated in future.
Nevertheless, this system is potentially useful tool to assess ASFV virulence in vitro.
1.ii c) Effect of ASFV on B-cells (Milestones 1, 6 and 10):
One of the major characteristics of the pathogenesis of ASFV infection in domestic pigs is a massive B-cell
apoptosis (Ramiro-Ibáñez et al., 1997; Oura et al., 1998) and hypergammaglobulinaemia in chronically infected
pigs (Pan et al., 1970). This contradictory outcome of B-cell fate in ASFV-infected pigs is due to non-specific
polyclonal B-cell activation. This phenomenon can be reproduced in vitro using purified porcine B-cells
(Takamatsu et al., 1999) and provides the opportunity to determine if there is a correlation between ASFV
virulence and non-specific B-cell proliferation. If so induction of non-specific B-cell proliferation will provide a
valuable tool with which to evaluate the pathogenesis of field isolates and gene deletion mutants of ASFV, in vitro.
In the original assays (Takamatsu et al, 1999), B- cell co-stimulation following co-culture with ASFV infected cells
was performed using either human or murine CD154(CD40L). Thus the first objective of this milestone was to
clone and express porcine CD154 (CD40L), which may increase the sensitivity of the assay. A clone containing
porcine CD40L was obtained from Germany and the CD40L gene was subcloned into pcDNA3.1zeo(+) vector to
enable stably transfected cc and dd haplotype pig kidney cell lines (CCK2 and MAX respectively) expressing
CD154 to be selected by zeocin resistance. Transient transfection showed that MAX cells expressed CD40L both
intracellularly and crucially on the cell surface (Fig 1.iiC-1). However CD40L was not expressed in CCK2 cells
and it was not possible to generate stable cell lines. Although it was not possible to generate a stably transfected
cell line expressing porcine CD40L, transient transfection of MAX cells with the gene expressing porcine CD154
SID 5 (Rev. 3/06)
Page 5 of 16
and a hybridoma producing anti-CD40L monoclonal antibodies (5C8) will be very useful tools for pig immunology
studies in general. The ability of ASFV isolates to stimulate non-specific B cell activation was therefore tested with
the original method using murine CD154. Although in general, virulent virus induced greater B cell stimulation, as
shown for example in Fig 1iiC-2, this original method induced high background (due to mytomycin C treatment of
cells rather than -irradiation to reduce background) and may not be sensitive enough to determine virulence by
this method alone. A number of isolates were tested in the assay, the Georgia 2007/1 isolate, currently circulating
in Russia and the Trans Caucasus region, induced similar B cell stimulation as the pathogenic OURT88/1 isolate
whereas the Benin 97/1 isolate inhibited non-specific B cell proliferation.
1.ii d) Effect of ASFV on DC function (Milestones 1 and 10):
ASFV predominantly infects cells of the monocyte/macrophage lineage (myeloid), including dendritic cells (DCs).
DCs are the major professional antigen presenting cell and initiate immune responses by secreting cytokines and
presenting antigens to various types of lymphocytes. As DCs are important in the initiation of a primary immune
response, a desirable property of any live attenuated ASFV vaccine candidate is that it will not inhibit DC function.
Bone marrow derived DCs were generated by culturing bone marrow cells with recombinant Flt3 ligand and used
to assess the role of ASFV infection on DC function. DCs cultured with Flt3 ligand were shown to express DCLAMP (CD208) as a marker for maturation. Both pathogenic OURT88/1 and non-pathogenic OURT88/3 isolates
of ASFV replicated in these DC cultures as measured byTCID50 and qPCR. It was shown previously that an
attenuated isolate of ASFV induced production of DC maturation signals (TNFα and INFα) in primary
macrophages {Gil, 2008} and apoptotic/necrotic bodies, which are also produced following ASFV infection, are
known to be potent activators of porcine DC cultures {Paillot, 2001}. Therefore virus stocks (OURT88/1,
OURT88/3, Benin 97/1, Ba71, Ba71v, Virulent Uganda, Attenuated Uganda, Malta 78 and Malawi Lil/20) used in
these studies were partially purified by superspeed centrifugation to eliminate cytokines and apoptotic/necrotic
bodies from virus stocks and this purification eliminated the non-specific induction of MHC class II on DCs.
Partially purified ASFV did not significantly affect the surface expression of MHC class II, CD209 or CD80/86 on
DCs. On the other hand, partially purified ASFV did affect the ability of DCs to uptake and process antigen.
However there was a lot of variability between experiments and between the affect of different viruses and there
was no clear evidence of a difference in the effect between virulent ASFV and attenuated ASFV that can induce
immunity, or attenuated ASFV that does not induce immunity. Unfortunately it was not possible to repeat this
partial purification once the SAPO licence was removed from the Main Laboratory at Pirbright therefore it was not
possible to complete all of the experiments on DC function as outlined in the original milestones. ASFV is also
known to induce a number of molecules that may modulate DC function. The effects of TNFα and INFα are well
documented, but those of Prostaglandin E2 (PGE2) are not. Therefore the effect of PGE2 was assessed on DC
phenotype and antigen uptake. So far no effect of PGE2 has been demonstrated on porcine DCs using the
parameters tested either alone or in the presence of TNFα.
From these experiments we conclude that ASFV does not induce major changes in DC phenotype or function.
Thus changes in cultured DC phenotype and function cannot be used to predict whether a given vaccine
candidate will induce immunity.
2. Identification of protective immune response(s) induced by non-pathogenic ASFV.
2.i Recognition of geographically diverse strains of ASFV by CD4hiCD8+perforin + lymphocytes induced
by the non-pathogenic ASF virus OURT88/3 (Milestone 5):
We have demonstrated that the majority of pigs infected with non-pathogenic ASFV OURT88/3 genotype I isolate
do not develop clinical signs of disease and are protected from subsequent challenge with a pathogenic but
closely related ASFV genotype I strain OURT88/1 (Oura et al 2005; DEFRA SE1511 final report 2006).
Lymphocytes obtained from these immunised pigs, prior to challenge, produce IFN following stimulation in vitro
with OURT88/1 isolate and the activated lymphocytes are predominantly a subclass that is CD4hiCD8+perforin+.
In contrast, lymphocytes from pigs that are not completely protected following challenge with OURT88/1 virus,
showed a lower frequency of ASFV specific IFN producing memory T cells and CD4hi CD8+perforin+ T cells.
These results indicate that the ability to prime CD4hiCD8+perforin+ T cells correlates with protection and could
provide an indicator of the ability of vaccination to induce cross-protection between different ASFV isolates. Since
most ASFV isolates kill pigs before induction of antibodies, there are relatively few antisera available to perform
virus neutralisation assays and this method cannot be used to predict cross-protection.
ASFV isolates have been grouped into 22 genotypes by partial sequencing of the B646L gene, which encodes
the major virus capsid protein VP72. In order to determine the potential for cross-protection between different
isolates of ASFV, lymphocytes from pigs immunised with OURT88/3 and boosted with OURT88/1 (both genotype
I isolates) were stimulated in vitro with a number of ASFV isolates from genotype I and other genotypes and their
ability to produce IFN determined by ELISPOT assays. West African isolate Benin 97/1 (genotype I) was able to
stimulate these immune lymphocytes at identical levels to the homologous viruses. However interestingly, not all
genotype I isolates stimulate OURT88/3 and OURT88/1 immune lymphocytes to a similar degree. The genotype I
isolate Lisbon 57 stimulated lymphocytes to a reduced level, whereas one isolate of genotype X (virulent Uganda
SID 5 (Rev. 3/06)
Page 6 of 16
1965) was shown to be highly cross-reactive (Fig 2i-1). Genotype VIII (Malawi Lil/20) showed almost no crossreactivity and in fact OURT88/3-OURT88/1 immune pigs were never protected against Malawi isolates before.
Cross-reactivity was also tested against the following isolates Zambia Lil13/33 (genotype I), Georgia 2007/1
(genotype II), Tengani (genotype V), South Africa :RSA (genotype IV) with varying degrees of cross-reactivity
observed using lymphocytes from different immune pigs (Fig 2i-2). As two isolates, Benin 97/1 (West African
isolate, virulent virus: genotype I), and an East African isolate (Uganda; virulent: genotype X) showed good crossreactivity by this IFN production assay, these two isolates were used in cross-protection experiments. Three
experiments were carried out, one at Pirbright and the other 2 at AFSSA Plougfragan, France. In all 3
experiments cross-protection of pigs immunised with OURT88/3 followed by OURT88/1 isolate to challenge with
a lethal dose of Benin97/1 isolate was tested. Of a total of 14 pigs primed with OURT88/3 and boosted with
OURT88/1 only 2 pigs were not protected from Benin 97/1 challenge (Fig 2.i-2). The two unprotected pigs
showed the lowest frequency of IFN producing lymphocytes and of ASFV specific CD4hiCD8+perforin+ T cells
prior to the challenge. This supports the correlation between protection the induction of ASFV specific IFN
producing CD4hiCD8+perforin+ T cells. One cross-protection experiment was performed with Uganda challenge
and all pigs were protected, although a few pigs showed minor clinical signs. In one group of pigs we
demonstrated that priming and boost with OURT88/3 was equally effective as priming with OURT88/3 followed by
boost with OURT88/1 in inducing a protective response.
These experiments were also used to validate a clinical scoring system for ASFV. In addition use of a quantitative
PCR assay to measure viraemia was compared with haemadsorbtion assay and found to give equivalent results.
The experiments also generated samples from infected animals which can be used in future to validate diagnostic
assays, for example we are planning to use the samples to validate methods using robotic extraction of nucleic
acid from pig blood and the samples have been used to help validate the multiplex PCR assay for ASFV and
CSFV diagnosis funded by DEFRA.
In these experiments we demonstrated that 1) immunisation with OURT88/3 and OURT88/1 can induce crossprotection against virulent virus from the same or different genotype and 2) cross-protection can be predicted by
the ability of ASFV-specific T cells to recognise different ASFV isolates in an IFNγ ELISPOT assay.
2.ii Preliminary identification of ASFV antigens which induce humoral and cellular immune responses in
pigs (Milestones 12 and 14)
Experimental inoculation with attenuated ASFV strains has shown that protection can be achieved with little
replication of inoculums or challenge virus. Previous studies have shown that CD8+ T cells are required for this
protection (Oura et al., 2004) and another study has shown that passive transfer of antibodies from protected pigs
can confer protection against lethal challenge (Onisk et al., 1994). Three ASFV antigens have been identified
which are targets for neutralising antibodies ( Zsak et al., 1993, Gomez-Puertas et al., 1996) but the most
important antigens involved in inducing a cellular immune response have not been defined. Defining the antigens
which are required for induction of a protective immune response is important for the development of subunit
vaccines as well as helping to predict cross-protection between isolates.
The first objective of this milestone was to identify the optimum vaccination protocol for ASFV antigen delivery
using a DNA prime and recombinant vaccinia virus (rVV) boost strategy. To optimise conditions for gene gun
delivery of DNA, plasmids expressing luciferase and green fluorescent protein were delivered to pig skin.
Optimum conditions were identified as 450 pounds per square inch (psi) to the ears of pigs using Helios gene gun
with 5 to 10 ug DNA doses on PEI-charged gold micro/nanoparticles.
The next objective was to optimise the DNA prime/ rVV boost conditions required to induce optimal immune
responsiveness in pigs. For this purpose, ASFV p72, p30 antigens and well characterised model antigens (human
ATT, HIVgp160 and influenza HA) were used. The conditions tested included the number of gene gun
inoculations (between 2 to 10 shots per time) and shot intervals, duration between last gene gun and rVV boost,
rVV application via scarification or needle-free device (Injex 30), number of mixed genes in a single gene gun
shot and delivery of pools of constructs as mixed or individual genes loaded on gene gun bullets. Incorporation of
porcine CpG motif into DNA was also examined. Immune responses to the individual antigens, provided by
collaborators at Arizona State University (ASU) as in vitro translated antigens loaded on beads, were examined at
3 and 4 weeks post rVV boost using IFN ELISPOT and proliferation assays. Serum antibody levels to individual
antigens was measured using the in vitro translated antigens on beads to measure responses by ELISA assay or
using a protein microarray developed at ASU. Optimisation experiments were performed by 3 separate pilot
experiments consisting 1 to 3 pigs per group. The optimum immunisation protocol obtained was double DNA
gene gun prime repeated after a 3 week interval and 3weeks later rVV boost using Injex 30.
Using the optimised procedures for DNA prime and rVV boost, determined as described above, we next
measured the immune response to individual ASFV antigens. We had previously sequenced the entire coding
region of the Georgia 2007/1 isolate of ASFV and identified the open reading frames (ORFs) encoded. Our
collaboators at Arizona State University have cloned all of these ~200 ORFs in plasmid constructs for gene gun
delivery and for generation of in vitro translated antigens. In addition, cloning of the individual ASFV ORFs into
rVVs is in progress with 42 having been produced so far. The first 40 antigens were delivered as 2 pools of 20
SID 5 (Rev. 3/06)
Page 7 of 16
antigens. Three groups of 3 pigs were inoculated by a DNA prime / rVV boost regime. After 4 weeks post boost,
lymphocytes from these pigs were examined by IFN ELISPOT and T cell proliferation assays, and serum was
collected. For these assays individual in vitro translated antigens coupled to beads were used to stimulate
lymphocytes from immunised pigs and the antibody response to individual antigens was also measured by ELISA
and using a protein array developed by ASU. The immunisation protocol so far tested works well inducing
reasonably good T cell responses, although the responses to individual antigens varied between pigs. Selected
results are shown in Fig 2.ii. A further study funded by our follow on DEFRA project is in progress to compare the
response to these antigens in groups of inbred Babraham pigs with that in outbred pigs. Future experiments will
continue to use these DNA prime rVV boost methods to screen the remainder of the ASFV ORFs to induce
cellular and antibody immune responses in pigs. This will allow us to rank all of the ASFV antigens for their ability
to induce cellular and antibody responses in pigs and to select subsets of these antigens to test their ability to
induce a protective immune response to lethal ASFV challenge.
3 Construction of pig cell lines permissive for high level replication of candidate vaccine strains and field
isolates (Milestones 4 and 9).
ASFV isolates obtained from the field do not replicate well in established cell lines and primary porcine
macrophage cultures are usually used to grow these viruses in culture. Field isolates can be adapted to replicate
in established cell lines but this process of adaptation usually results in a reduced ability of the virus to replicate in
primary macrophage cultures and in pigs. Development of a cell line for replication of field isolates of ASFV
without losing the ability to replicate in primary macrophages and in pigs will be necessary for the exploitation of
attenuated ASFV vaccines. These cell lines would also be useful for the isolation of virus for diagnostic purposes.
One possibility is that ASFV infection of an established pig macrophage cell line (IPAM) is inhibited by prior
induction of an antiviral state by type I IFNs. Our studies have shown that production of IFN in this cell line is
constitutively active. We constructed stably transfected IPAM pig macrophage cell lines expressing either the NPro gene from CSFV which inhibits induction of type I IFN or SV5 virus V protein which also inhibits activation of
type I IFN induced pathways. Expression of the transgene was confirmed using specific antibodies. These cell
lines were infected with 6 different ASFV field isolates or deletion mutants and levels of infection monitored by
immunofluorescence using antibodies against an early (p30) or late (p72) virus protein. The results showed no
increase in infection rate of these cell lines compared to the original IPAM cells.
Previous studies have shown a correlation between expression of the cell surface receptor CD163 and ability of
ASFV to replicate in intermediate to late stages of macrophage differentiation. This suggested that the CD163
protein may act as a receptor mediating ASFV binding or entry. We investigated if cell surface expression of the
CD163 protein by cells that are resistant to ASFV infection resulted in ASFV susceptible cells. We generated
PK15 cell lines expressing the full length porcine CD163 gene and compared infection rates of these cell lines
and parental cells with different ASFV isolates. Isolates tested included the tissue-culture adapted Uganda strain
which we showed infected both cells expressing CD163 and parental cells at equal levels. In contrast, little or no
infection of parental PK15 cells or PK15 cells expressing CD163 was observed with virulent field isolates,
including Uganda and Benin 97/1. The results show that expression of CD163 neither increases the proportion of
cells that become infected by a tissue-culture adapted virus nor confer susceptibility of cells to infection with field
isolates.
The ability of ASFV field isolates (including Malawi LIL 20/1 and Benin 97/1) to replicate in a variety of cell lines
was empirically tested using cell lines obtained by IAH and in collaboration with Dr Guenther Keil, Friedrich
Loeffler Institute Germany. More than 40 cell lines were tested and one cell line was identified which supported
replication of ASFV field isolates to an extent. This cell line is a wild boar cell line WSL-R which was kindly
provided by the FLI.
Discussion of results and their implications
The results from this project have advanced work towards development of an ASFV vaccine in several ways. The
results and samples generated have also helped in the development and use of new diagnostic tests. In the work
described in objective 1 we aimed to identify assays that could be used in vitro to predict the pathogenesis of
ASFV isolates. These assays are useful to screen candidate vaccine strains, rationally produced by sequential
deletion of genes involved in immune evasion and virulence, before testing in pigs. We tested a number of
different assays and identified two which could distinguish between between infection of cells with pathogenic and
non-pathogenic isolates. One of these assays was the measurement of differences, detected by quantitative
reverse transcriptase PCR at early times post-infection of macrophages, of mRNA levels for several host genes
including those encoding chemokines and proinfllamtory cytokines. Higher levels of these mRNAs were induced
by infection with non-pathogenic isolates compared to pathogenic isolates. The second assay was co-culture of
SID 5 (Rev. 3/06)
Page 8 of 16
ASFV infected cells with lymphocytes greatly increases the proportion of cells with NK and NKT cell phenotype.
These cells were activated to produce an important cytokine for immune response activation, IFN . The results
showed that infection with non-pathogenic isolates induced less IFN than infection with pathogenic isolates.
Further work will be needed to validate these assays.
A second objective was to identify the mechanisms of protective immunity induced by the non-pathogenic isolate
OURT88/3. We further validated that induction of protection was correlated with induction of a specific memory T
cell subset that is CD4hiCD8+perforin+ . This assay can be used to screen candidate vaccines. We also
developed an assay which can be used to predict cross-protection of immunised pigs to lethal challenge with
different strains. We confirmed the results by demonstrating that pigs immunised with the non-pathogenic isolate
OURT88/3 are protected from lethal challenge with virulent African isolates from the same genotype (Benin
genotype I) and also from a different genotype (Uganda genotype X). This demonstration of cross-protection is
important for the practical implementation of vaccination in countries where more than one genotype of virus is
circulating (eg in Mozambique). The ability to predict cross-protection by an in vitro assay is also critical for
selection of strains or antigens to include in vaccine development and use. These experiments were also used to
validate a clinical scoring system for ASFV and the samples generated were used to validate diagnostic assays.
First they were used to validate methods for qPCR from blood samples and validate that this assay gave very
similar results to the gold standard haemadsorbtion assay. Secondly they were used to help validate the multiplex
CSFV, ASFV PCR assay developed jointly by IAH and VLA.
Little information is available on the ASFV antigens that can induce protective immunity. We have started a
genome wide screen to identify these by delivery of pools of antigens using a DNA prime and recombinant
vaccinia virus boost regime. To date 40 antigens have been screened for their ability to induce antibody and
cellular responses in pigs. In future those which induce the “best” responses will be tested for their ability to
induce protection against lethal challenge in pigs. This information is critical for the development of subunit
vaccines which would probably be the preferred vaccines for use in countries where ASFV is not endemic.
Further work is needed to continue screening of ASFV antigens for immunogenicity and to identify optimal pools
of small groups of antigens that can induce protective immunity in pigs. Ultimately these would have to be
incorporated into a different delivery vector for vaccine use in pigs. The identification of key protective antigens
will also be useful for prediction of cross-protection between isolates. samples generated in this experiment were
used
Our third objective was to identify cell lines that could support replication of ASFV field isolates and candidate
vaccine strains. We showed that stable expression of genes that suppress interferon induction and the putative
cellular receptor for ASFV , CD163, did not make non-susceptible cells susceptible to infection. By empirically
testing more than 40 cell lines one cell line (developed at Friedrich Loeffler Institute) was identified which
supported replication of a number of ASFV isolates. This is practically very useful for ASFV research and very
important for the eventual use of attenuated ASFV vaccines. Further work is needed to optimise growth
conditions for this cell line. This is likely to involve further selection of the cell line.
SID 5 (Rev. 3/06)
Page 9 of 16
Fig1ii-b. ASFV can induce activation of NK and NKT lymphocyte in naïve pig lymphocytes. The top figure
indicates that co-culture of naïve lymphocytes with ASFV-infected cells resulted in an increase of total perforin
positive lymphocyte (either small dense [SDL] or larger granular [LGL]). Lower figure shows proportion of NK,
NKT and CTL phenotypes within perforin positive lymphocytes co-cultured cells without or ASFV infected cells.
SID 5 (Rev. 3/06)
Page 10 of 16
Figure 1ii-C-1: MAX cells transiently expressing porcine CD40L: MAX cells were transfected with
pcDNA3.1zeo(+)-CD40L for 48 hours and then stained with 5C8 (red), TW1 (green) or the DNA dye DAPI (blue).
Panels A and B show intracellular staining for CD40L (red) and ERp57 (green) after permeabilisation with 0.2%
Triton X-100. Panels C and D show surface labelling for CD40L (red) only with no permeabilisation. DAPI dye
shows the position of the cell nucleus (blue).
35000
30000
25000
20000
B + K-47
15000
D k-47
10000
5000
0
BG
SID 5 (Rev. 3/06)
OURT88/3
OURT88/1
Malawi
Page 11 of 16
Figure 1ii-C-2. An example of virulent ASFV-induced non-specific B cell proliferation (shown as 3H-TdR uptake,
cpm). In general, virulent virus induced higher B cell proliferation. However in this assay CD154 (CD40L)
expressed K-47 itself cause high background (BG) (delta K47 is show in brown colour), thus assay sensitivity is
not high.
Fig 2i-1A. ASFV isolate cross-reactivity assessed by IFN ELISPOT using OURT88/3-OURT88/1 immune pig
lymphocytes. Portuguese isolate OURT88/3 and OURT88/1 and West African isolate Benin 97/1 shows very
good cross-reactivity and Central African isolate of Uganda (genotype X) but no cross-reactivity to Malawi isolate.
% cross-reactivity compared to Benin
140
120
100
Benin
80
SA
Madagascar
60
Georgia
40
Tengani
Liv 13/33
20
0
Fig 2i-1B. ASFV isolate cross-reactivity assessed by IFN ELISPOT using OURT88/3-OURT88/1 immune pig
lymphocytes. Cross-reactivity to genotype II (Madagascar and Georgia) was variable between two immune pigs
used.
SID 5 (Rev. 3/06)
Page 12 of 16
Fig.2i-2. OURT immune pigs are cross-protected against challenge with African isolates of ASFV OURT88/3OURT88/1 immune pigs in experiment 1 and experiment 3 were challenged with West African isolate Benin97/1
virus and all immune pigs were protected. OURT88/3-OURT88/1 immune pigs in experiment 2 were challenged
with either Benin 97/1 or Central African isolate Uganda virus in. All pigs were protected against Uganda
challenge. Unlike experiments 1 and 3, two of immune pigs were not protected from Benin 97/1 challenge, but
the remaining three immune pigs were completely protected.
Fig. 2.ii An example of ASFV antigen screening by IFN ELISPOT assay. Pig 9 T cell response to antigen 54 and
128 as determined by IFN production. Both antigens also induced proliferation of lymphocytes from pig 9 (data
not shown).
SID 5 (Rev. 3/06)
Page 13 of 16
qRT-PCR Results
Malawi
2hr
Malawi vs OURT88/3
OURT88/3
2hr
Malawi
4hr
OURT88/3
4hr
TNF
3.5
6.9
5.0
7.1
IL-1
3.4
4.4
3.8
4.1
IL-1b
4.7
5.5
5.1
6.41
GM-CSF
2.5
4.0
3.6
4.0
MIP-1
1.0
4.0
1.9
5.1
RANTES
-1.8
1.6
1.1
2.1
AMCF-II
1.2
3.0
3.2
4.3
Up regulated
Down regulated
Table 1. Differentially expressed cytokine and chemokine genes following infection of macrophages with African
swine fever virus isolate pathogenic isolate Malawi LIL 20/1 and non-pathogenic isolate OURT88/3. RNA was
extracted from infected or mock-infected macrophages at different times post-infection and the levels of
transcripts for different genes measured by quantitative reverse transcriptase PCR. The table shows fold changes
compared to mock-infected macrophages on a log 2 scale for selected chemokine and cytokine genes.
References to published material
9.
This section should be used to record links (hypertext links where possible) or references to other
published material generated by, or relating to this project.
SID 5 (Rev. 3/06)
Page 14 of 16
Papers in Peer-Reviewed Journals
H.-H. Takamatsu, M.S. Denyer, C. Stirling, S. Cox, N. Aggarwal, P. Dash, T.E. Wileman, P.V.
Barnett (2006) Porcine  T cells: Possible roles on the innate and adaptive immune responses
following virus infection. Veterinary Immunology Immunopathology, 112:49-61.
M.S. Denyer, T.E. Wileman, C.M.A. Stirling, B. Zuber, H.-H. Takamatsu (2006) Perforin expression
can define CD8 positive lymphocyte subsets in pigs allowing phenotypic and functional analysis of
Natural Killer, Cytotoxic T, Natural Killer T and MHC un-restricted cytotoxic T-cells. Veterinary
Immunology and Immunopathology, 110:279-292.
D.G. Chapman, Tcherepanov V, C. Upton, L.K. Dixon (2008) Comparison of the sequences of nonpathogenic and pathogenic African swine fever virus isolates. Journal of General Virology 89,
397-408.
Anette Bøtner, Agustin Estrada Peña, Alessandro Mannelli, Barbara Wieland, Carsten Potzsch,
Cristiana Patta, Emanuel Albina, Ferdinando Boinas, Frank Koenen, James Michael Sharp, Linda
Dixon, Mo Salman, Sánchez-Vizcaíno, Sandra Blome, Vittorio Guberti, Sofie Dhollander, Milen
Georgiev, Jordi Tarres und Tilemachos Goumperis (2010) Scientific opinion on African swine fever.
EFSA Journal 8(3):1556 [149 pp.]. doi:10.2903/j.efsa.2010.1556
Costard S, Wieland B, de Glanville W, Jori F , Rowlands R, Vosloo W, Roger F, Pfeiffer DU , Dixon
LK (2009) African swine fever: how can global spread be prevented? Philosophical Transactions
of the Royal Society B-Biological Sciences. 364, 2683-2696
Rowlands, R.J., Michaud, V., Heath, L., Hutchings, G.., Oura, C.., Vosloo, W.., Dwarka, R.,
Onashvili, T. Albina, E. Dixon, L K (2008) African swine fever virus isolate Georgia,2007. Emerging
Infectious Diseases 14, 1870-1874
Book Chapters
Dixon L K., Abrams CC.,,Chapman DG. Zhang F. (2008) African swine fever virus In: Animal
Viruses: Molecular Biology eds Mettenleiter TC, Sobrino F Caster Academic press, pp 475-521
Dixon LK, Chapman DG (2008) African swine fever virus Encyclopedia of Virology Elsevier Ltd.
pp 43-5 Netherton C.L., K. Moffat, E. Brooks, and T. Wileman (2007) A guide to viral inclusions,
membrane rearrangements, factories and viroplasm produced during virus replication. Advances in
Virus Research 70: 101-182
Presentations

K. King & H. Takamatsu 2009 : Protection of European domestic pigs from virulent African isolates
of African swine fever virus by experimental vaccination. The 412th Veterinary Research Club
meeting. 19th June 2009.

Takamatsu H.-H., Marques J., Fishbourne E., Hutet E., Cariolet R., Chapman D., Hutchings G.,
Taylor G., M.-F. Le Pottier & Dixon L.K. (2008) Protection of European domestic pigs from virulent
African isolates of African swine fever virus challenge by experimental vaccination. Third EPIZONE
Theme 5 meeting, El Escorial, Spain, 22-24 October 2008.

Takamatsu H.-H. (2006)How many more?: Porcine CD8+ lymphocyte subset and their functions. 2nd
EVIW, Paris, Sep 4th to 6th. Abstract page 52.
Oral Presentation at Annual Farm Animal Genomics Workshop Cambridge 27th, 28th Sep
Zhang F, Hopwood P, Abrams C, Downing A, Murray F, Talbot R, Archibald A, Lowden S, Dixon
LK. (2006) Macrophage transcriptional responses following in vitro infection with African swine fever
virus.
Chapman DG, Tscherapanov V, Upton C, Dixon LK sequence comparisons of high and low
virulence strains of African swine fever virus European Society for Veterinary Virology Lisbon 24th to
27th Sep 2006
Abrams C, Zhang F, Hopwood P, Archibald A, Talbot R, Lowden S, Dixon LK Microarray analysis of
pig macrophage gene expression following African swine fever virus infection European Society for
Veterinary Virology Lisbon 24th to 27th Sep 2006
LK Dixon, H Takamatsu, DG Chapman, L Goatley, K King, C oura, E Albina, V Michaud, C Martins,
L Heath, W Vosloo, RME Parkhouse. Development of attenuated vaccines for African swine fever
virus. Animal Health Research: Recent Developments and Future Directions Hinxton UK Jan 24 th to
26th 2006
LK Mulumba-Mfumu, B Wieland, B Lombe, J Mande, S Kayimbe, G Hutchings, L Dixon.
SID 5 (Rev.
3/06)
PageAfrica:
15 of serological
16
Epidemiology
of African swine fever in Central
status of pigs in the Democratic
Republic of Congo
SID 5 (Rev. 3/06)
Page 16 of 16