Download African Swine fever virus immune evasion genes

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
SE1510
African swine fever immune evasion genes
Contractor
organisation(s)
Institute for Animal Health Compton Lab
Compton
Nr Newbury
Berks
RG20 7NN
54. Total Defra project costs
(agreed fixed price)
5. Project:
Page 1 of 14
£
273,675
start date ................
01 July 2003
end date .................
30 June 2006
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 devastating disease of domestic pigs with very high
mortality rates The disease is endemic in many African countries and sporadic outbreaks occur
in other countries causing major economic losses. Outside Africa ASF is currently endemic in
Sardinia. In Africa in many areas wildlife hosts (warthogs, bushpigs and soft ticks of
Ornithodoros species) are persistently infected providing a source of infection which is unlikely to
be eradicated. The lack of an effective vaccine limits the options for controlling disease and is a
particular problem in areas where the tick vector is present. The development of an effective
vaccine would not only aid pig production in Africa but by controlling ASF in Africa the risk of
accidental introduction elsewhere in the world would be reduced.
ASFV is a large DNA virus which encodes many proteins that are not essential for virus
replication in cells but play an important role in helping the virus survive in its hosts. These
include proteins which interfere with and manipulate the host defence systems to aid virus
survival. In the last ten years major advances have been made in understanding ASFV-encoded
proteins involved in evading host defences and in causing virus virulence. This knowledge has
helped us to develop a rational strategy for producing an attenuated vaccine.
This project has advanced our knowledge of virus proteins involved in immune evasion or
virulence using two different approaches. First, we have identified two new functions for a virus
protein, l14L, which had previously been shown to be required for virus virulence. One of the
new functions is the activation of the host enzyme protein phosphatase 1 (PP1). PP1 regulates a
broad range of host cell activities including transcription and translation of mRNAs. In Herpes
simplex virus infection a similar virus protein to l14L targets PP1 to prevent shut-off of host
protein synthesis that is induced following infection to limit virus replication. Secondly, we have
discovered that the l14L protein inhibits induction of type I interferons by preventing transcription
of the interferon gene. Type I interferons are secreted from infected cells and induce an anti-viral
state in bystander cells that limits virus replication. Possibly these two functions of l14L are
linked such that activation of PP1 is involved in preventing interferon gene transcription. The
wide range of cell processes regulated by PP1 makes it likely that l14L may also target PP1 to
regulate other pathways advantageous for virus replication. Further research is needed to
identify these.
SID 5 (Rev. 3/06)
Page 2 of 14
The other approach we used to identify virus proteins important for virulence and immune
evasion was to sequence the complete genome of a non-pathogenic ASFV isolate OUR T88/3.
This sequence was compared with that of 8 other virulent isolates and one non-virulent isolate
which has been adapted to grow in tissue-culture cells. The analysis showed that most genes
were present in the genomes of all of the different isolates. One feature of the ASFV genome is
that 5 families of related genes, called multigene families, are encoded. Most of the variation
between the different genomes is as a result of gain or loss of members of these multigene
families. One deletion near the left end of the genome of the OUR T88/3 isolate has removed a
sequence of around 8 kbp encoding 6 copies of MGF 360 and 2 copies of MGF 530. These
genes have previously been shown to have roles in virulence and in suppressing the host’s
production of interferon. In addition the OUR T88/3 isolate has interruptions in genes encoding
the CD2v and lectin-like protein. These proteins have roles in impairing the function of T
lymphocytes and in preventing cell death as well as in binding of virus particles and infected cells
to red blood cells. The latter function helps virus dissemination in infected pigs. These gene
deletions and interruptions help to explain why the OUR T88/3 isolate is non-pathogenic. The
information also helps to explain why infection with this isolate induces an effective immune
response to protect pigs challenged with related virulent viruses.
The information we have gained from this analysis and from previous work has allowed us to
design a strategy to construct candidate attenuated ASFV strains by sequential deletion of
genes involved in virulence and immune evasion.
ASFV replicates in macrophages, which are important immune cells involved in activating and
orchestrating the immune response to infection. Understanding how ASFV infection manipulates
the function of macrophages is crucial to gain an understanding of mechanisms of virus
pathogenesis and immune evasion. We used porcine microarrays, containing either 2880 or
14000 genes, to gain a global overview of changes in the genes transcribed into mRNAs in
macrophages at different times following infection with ASFV isolates which vary in virulence.
The results showed that, following infection with a virulent isolate, most of the differentially
expressed genes increased in expression level at early times post-infection and returned to
similar levels to those in uninfected cells at late times post-infection. Many of these were genes
which we expect to be activated as part of the host’s defence response to infection and included
proinflammatory cytokines and chemokines as well as other cell surface and secreted proteins.
ASFV encodes proteins which inhibit host pathways that are needed to activate host defence
genes. Our results therefore support a model in which the macrophage responds to infection by
activating genes involved in the host’s defences. As virus proteins which interfere with this
response are produced early in infection these may help to switch off these host defence genes
latter during infection. Increased expression of some host defence genes was observed in
macrophages infected with the non-pathogenic isolate suggesting that this virus is less effective
at switching off these genes. Possibly this may help to explain why this isolate induces a good
protective immune response and is not virulent. By comparing the pattern of mRNAs expressed
in macrophages infected with different virulent isolates we could distinguish patterns of gene
expression that were typical of the different isolates.
The implications from these results are that we may in future be able to use global transcription
profiling of mRNAs from macrophages infected in vitro with different virus isolates as one
method to predict virulence of different virus isolates and the effectiveness of the host’s defence
response. This will be useful to prioritise which candidate attenuated vaccine strains to test in
pigs. In addition we have identified many new target genes encoding proteins which may have
important roles in virus pathogenesis and immune evasion.
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.
SID 5 (Rev. 3/06)
Page 3 of 14
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;
 possible future work; and
 any action resulting from the research (e.g. IP, Knowledge Transfer).
Objectives
1. To use a porcine cDNA microarray; a) to characterise changes in host macrophage mRNA
expression profiles following infection with ASFV isolates of varying pathogenicity,
b) To determine the effects of individual ASFV immune evasion genes on host macrophage gene
expression.
2 To investigate the mechanism by which the ASFV CD2v protein inhibts lymphocyte proliferation.
3 To identify novel ASFV encoded proteins involved in immune evasion.
4. Determination of complete nucleotide sequence of the low virulence ASFV isolate OURT88/3
Results obtained in relation to the objectives.
Introduction
African swine fever (ASF) is endemic in Sardinia and many African countries south of the
Sahara and has recently spread to previously uninfected countries including Madagascar.
Disease control is difficult since wildlife reservoirs and tick vectors are present in many
countries. Pigs, which recover from infection with less virulent isolates, can remain persistently
infected for long periods providing a reservoir for infection of uninfected herds. Virus isolates
vary in pathogenicity complicating diagnosis. Disease can be spread by movement of infected
pigs and pig meat products and ASF thus constitutes a continuous threat to UK agriculture.
There is no vaccine against ASF so control relies on rapid diagnosis and implementation of
quarantine and slaughter policies.
ASFV replicates mainly in macrophages. By modulating macrophage function the virus can
profoundly affect the host response to infection. Unravelling how the virus manipulates
macrophage function is critical both to understanding the mechanism of virus pathogenesis
and to understanding virus immune evasion mechanisms. ASFV is a large DNA virus and
encodes many proteins that are not essential for the virus to replicate in cells but have
important roles in interactions with the host. These include a number of genes which help the
virus to evade host defence systems (Dixon et al., 2004, 2005). Non-pathogenic ASFV isolates
have been obtained from a chronically infected pig or from ticks in Portugal. Pigs inoculated
with non-pathogenic isolates are protected against challenge with related virulent isolates.
These non-pathogenic viruses therefore provide good models to understand the molecular
basis for their reduced pathogenesis as well as mechanisms of protective immunity.
In this project one of our aims was to improve understanding of how macrophages respond to
infection in vitro with different virulent and non-virulent ASFV isolates by analysing global
macrophage transcription profiles following infection using porcine microarrays. This would
provide information relevant to understanding the molecular basis of virus pathogenesis and
immune evasion and may provide an in vitro method to assay for virus pathogenesis and
stimulation of host defence systems. In addition, we aimed to define how individual ASFV
encoded-proteins modulate host gene transcription to help the virus evade host defences. We
also aimed to identify novel ASFV encoded proteins involved in virulence and immune evasion
by direct study of individual proteins and by determining the complete sequence of a nonpathogenic isolate and comparing this to the sequence of virulent isolates. The information
gained is being used to design a strategy for the rational attenuation of ASFV to produce
candidate vaccines by sequential deletion of genes involved in immune evasion and virulence.
The results obtained in relation to objectives are described below.
SID 5 (Rev. 3/06)
Page 4 of 14
1 To use a porcine cDNA microarray; a) To characterise changes in host macrophage
mRNA expression profiles following infection with ASFV isolates of varying
pathogenicity.
We used two different porcine microarrays to study global changes in transcription of host
genes following infection with ASFV isolates which vary in virulence. First, we used a porcine
cDNA microarray with 2880 genes targeted to those likely to be activated as part of the host’s
response to infection. This array was constructed in a different project in collaboration with
ARK-Genomics Roslin Institute and Edinburgh University. Details of the array are available at
www.ark-genomics.org. This was used to compare changes in host gene expression following
infection of porcine macrophages with the high virulence Malawi LIL 20/1 isolate and the low
virulence isolate OUR T88/3 at 4 hr and 16 hr post-infection. These time points represent one
early time point when virus has entered cells and early virus genes are expressed and one late
time point when production of progeny virions is in progress but infected cells have not yet
entered apoptosis (Zhang et al., in press).
During the course of the project a new Pig OligoArray set from Operon was purchased. This
OligoArray consists of 13,297 70 mer oligonucleotides. In addition the OligoArray contained a
further 360 specific oligos which we designed from genes present on our cDNA microarray but
not contained on the Operon OligoArray The OligoArray set represents porcine cDNAs and
ESTs designed from The Institute of Genome Resarch (TIGR) Tentative Consensus cDNA
sequences. Annotation analysis has specified that the OligoArray contains 8541 unique pig
annotated gene NCBI gene accession numbers. Of these, 6244 oligonucleotides have been
assigned gene ontology terms. Using the OligoArray Set transcriptome, analysis conducted by
Zhao et al 2005 revealed that 11,328 oligos encoded transcripts which were expressed in at
least one pig tissue and that statistical analysis revealed that 1810 genes showed differential
expression among different tissues.
We used this OligoArray to compare macrophage transcription profiles following infection with
the high virulence isolate Malawi LIL 20/1 to those from mock-infected macrophages. In
addition, we compared transcription profiles from macrophages infected with the Malawi LIL
20/1 isolate with those infected with two additional virulent isolates (OUR T88/1, Liv 13/33) and
the moderately virulent isolate Dominican Republic (DR). These comparisons were also
carried out at 4 hr and 16 hr post-infection. Viruses were grown in pig bone marrow cells and
purified to remove cytokines and other factors present in culture media. Porcine macrophage
cultures were infected at a high multiplicity and RNA prepared at 4 hours and at 16 hours postinfection as well as from mock-infected cells.
For the first set of experiments, using the 2880 cDNA array, RNA was prepared from cultures
from 8 separate pigs and pooled in groups of two. RNAs were labelled with either Cy3 or Cy5
fluorescent dyes and pairwise hybridisations to cDNA microarray slides performed. RNA from
infected cells at each time point was compared with that from mock-infected cells. RNA from
macrophages infected with each isolate at both 4 hours and 16 hours post-infection were also
compared directly. Raw data on the relative hybridisation of Cy5 and Cy3 labelled RNA to
each cDNA (spot intensity and background measurements) were extracted by using
Quantarray and normalized in Genespring per spot and per chip (Lowess), then in-slide
averaging was taken of the duplicate targets on each array. One-way ANOVA algorithm was
used to identify differentially regulated targets with a T-test P<0.05.
One hundred and twenty five (125) targets were found significantly altered at either or both 4
hr and 16 hr post-infection compared with mock infection. These targets were assigned into
three groups according to their temporal expression profiles. 86 targets showed increased
expression levels at 4 hr post-infection, but returned to similar expression levels as those in
mock-infected cells at 16 hr post-infection. These encoded a number of different classes of
genes which would be expected to be activated as part of the host response to infection such
SID 5 (Rev. 3/06)
Page 5 of 14
as several proinflammatory cytokines (TNF-α, IL-1β, IFN-β) and chemokines (eg MIP-1α,
MIP1β), other surface and secreted proteins (eg syndecan 2 and galectin 3), proteins involved
in cell signaling and trafficking pathways. The return to the base level of transcription
observed in mock-infected cells of this temporally-regulated class of genes is consistent with
the action of ASFV encoded proteins such as A238L, which inhibit activation of host
transcription factors involved in immunomodulatory gene expression (Powell et al., 1996,
Miskin et al., 1996, Revilla et al., 1998, Granja et al., 2004, 2006). 34 genes showed increased
expression levels at 16 hr post-infection compared to 4 hr post-infection and mock-infected
cells. One host gene showed increased expression levels at both 4 and 16 hr post-infection
compared to mock-infected cells. These latter two classes of genes might include some which
are advantageous for virus replication. The microarray results were validated for 12 selected
genes by quantitative real-time PCR. Levels of protein expression and secretion were
measured for two pro-inflammatory cytokines, IL-1β and TNF-α during a time course of
infection with either the virulent Malawi LIL 20/1 isolate or the OURT88/3 non-pathogenic
isolate. The results revealed differences between these two ASFV isolates in the amounts of
these cytokines secreted from infected cells most notably infection with the virulent Malawi
isolate induced secretion of much larger amounts of IL-1β (Zhang et al., in press). Interestingly
following infection with the non-pathogenic isolate OUR T88/3 transcription of some
immunomodulatory genes was increased compared to the virulent isolate. These include IFN-γ
and some subunits of the immunoproteasome which are induced by IFN-γ.
The second microarray experiment using the pig OligoArray involved the direct comparison of
macrophages infected with strain Malawi LIL20/1 with each of the other three viruses,
OURT88/1, Dominican Republic and Livingstone 13/33. In addition, a separate comparison
was made between Malawi LiL20/1 and mock-infected cells. Macrophages from three separate
pigs were used. Total RNA was extracted from infected cells at 4 hours and 16 hours postinfection. mRNA was amplified and labelled using an Ambion amino allyl direct labelling kit.
The quantity and integrity of the labelled RNA was determined using an Agilent Bioanalyser
and Nanodrop spectrophotometer. RNA was labelled with either Cy3 or Cy5 and hybridised
pairwise to microarray slides in a Genetix hybridisation chamber. Dye-swap hybidisations were
also carried out, which represented two technical replicates. In total 48 separate hybridisations
involving 27 different RNAs were carried out. The data collected represents the results of 1.5
million spot intensity measurements. BlueFuse software was used initially to analyse the raw
intensity individual spot data. Following conversion, TIGR-MIDAS software was used for
subsequent data processing involving the following manipulations :- Global Lowess, SD
Regularisation, Dye-Swap and 2 Fact ANOVA. This analysis identified the same temporal
classes of genes in infected cells compared to mock-infected cells as in the previous
experiment. In total 277 differentially regulated genes were identified. Some differences were
observed in host gene regulation between the different virus isolates. A group of 10 genes
were down regulated at both 4 hrs and 16 hours following infection with the OUR T88/1 isolate
compared to the other isolates. A group of 25 genes were up-regulated at 16 hours postinfection with the OUR T88/1 isolate but unaffected or down-regulated following infection with
the other isolates. Thus the data indicate that although many differentially expressed genes
were common between cells infected with the different isolates, macrophages respond to
infection with different isolates by differentially regulating expression of certain genes. Further
work is required to understand the molecular basis of these differences and to establish if this
can be used to predict the pathogenesis and effectiveness of the immune response against
different isolates.
1 b) To determine the effects of individual ASFV immune evasion genes on host
macrophage gene expression.
Two approaches were used to study the effect of individual ASFV immune evasion proteins on
host gene expression. One compared mRNA expression profiles from cells infected with either
wild type ASFV or a mutant ASFV from which individual genes have been deleted. Using this
SID 5 (Rev. 3/06)
Page 6 of 14
approach the effect of individual genes may be masked if the virus encodes other proteins with
similar functions which can compensate for the deleted gene. In the second approach
individual ASFV genes were expressed from a defective virus vector in transiently transfected
cells. The effects of the expressed protein on host mRNA profiles can then be compared.
mRNA profiles were compared from cells infected with wild type ASFV and with a deletion
mutant lacking the A238L gene using both RTPCR and the porcine cDNA microarray. The
results showed little difference in transcription profile in cells infected with either virus. Since
A238L is predicted to have a large effect on host immunomodulatory gene expression, our
data suggests that ASFV may encode other proteins with similar functions to A238L which can
compensate for the loss of A238L. The A238L gene was cloned under control of a eukaryotic
promoter and its expression studied following introduction into cells by transfection or by
delivery in a recombinant baculovirus. Up to 90% of cells transiently expressed the A238L
protein at high levels when the gene was delivered using the recombinant baculovirus. Only
the A238L gene under control of the eukaryotic promoter is expressed since baculovirus
promoters are not recognized by eukaryotic RNA polymerase. Stocks of recombinant
baculovirus have been prepared and the experimental design optimised using a pig
macrophage cell line (Weingartl et al., 2002) but this experiment has not yet been completed.
This was due to delays encountered since we established conditions for spotting and
hybridising the pig OligoArray set. This involved optimising the choice of slides for spotting, the
spotting buffer and the hybridisation conditions.
2. To investigate the mechanism by which the ASFV CD2v protein inhibts lymphocyte
proliferation
Porcine peripheral blood leucocyte cultures were infected with either ASFV expressing the
intact CD2v protein or ASFV expressing a CD2v protein with an intact cytoplasmic domain but
disrupted cytoplasmic domain. The ability of the non-infected lymphocytes within the culture to
proliferate in response to mitogens was compared. The results showed that, as has been
previously described, ASFV infection inhibits lymphoproliferative ability. Previous results, using
a virus deletion mutant lacking the CD2v gene (Borca et al., 1998), suggested the CD2v
protein was required for this. In our experiments both the virus expressing full length CD2v and
CD2v with a disrupted cytoplasmic domain inhibited lymphocyte proliferation. Thus the
extracellular domain of CD2v alone may be sufficient for this inhibition.
In order to test if the extracellular domain of the CD2v protein was sufficient to inhibit the ability
of lymphocyets to proliferate in response to mitogens we attempted to express this domain as
a fusion with human IGg1 hinge, CH2 and CH3 domains. In parallel we expressed the
extracellular domain of the porcine CD58 protein as a control. Unfortunately we obtained only
very low levels of the expressed CD2v extracellular domain fusion only the control protein was
expressed at high levels. The reason for this was not clear but meant that we were unable to
use this approach to determine the effect of the CD2v extracellular domain.
3. Identification of novel ASFV encoded proteins involved in immune evasion.
l14L protein.
The ASFV l14L protein (also designated DP71L, NL, 23NL) is of great interest because it can
act as a virus virulence factor in ASFV infected pigs (Zsak et al., 1996). We have
demonstrated two novel functions for the l14L protein which are likely to be relevant for
mechanisms of virus virulence and immune evasion. These functions are first, activation of the
cellular protein phosphatase 1 (PP1) which is likely to modulate signalling pathways regulated
by this phosphatase in infected macrophages. Secondly (in collaboration with Emma Poole,
and Antonio Alcami,[Cambridge University] and Steve Goodbourn [London University]) we
showed that the l14L protein inhibits transcriptional activation of the interferon-β gene promoter
and is therefore likely to interfere with type I interferon induction in ASFV infected
macrophages and activation of the antiviral state in infected and uninfected cells.
SID 5 (Rev. 3/06)
Page 7 of 14
The l14L protein shares a conserved C-terminal domain with the Herpes simplex virus
encoded neurovirulence factor ICP34.5 and the host GADD34 protein (Goatley et al., 1999).
The HSV ICP34.5 protein has several different functions, the best described of these is
preventing shut-off of host protein synthesis that occurs following virus infection. The ICP34.5
protein acts as a regulatory subunit of protein phosphatase 1 (PP1) and targets PP1 to
dephosphorylate eiF 2-α and prevent shut-off of protein synthesis that is part of the host’s
defence against virus infection (He et al., 1997, He et al., 1998). In addition ICP34.5 negative
HSV is defective in replicating in non-dividing cells. Evidence suggests that ICP34.5 binds to
proliferating cell nuclear antigen (PCNA) to sites of HSV replication and this may be
particularly important for replication in non-dividing cells which have low levels of PCNA.
Finally ICP34.5 has a role in virus egress although the mechanism by which this function is
mediated is not understood (Brown et al., 1994a, Brown et al., 1994b, Harland et al., 2003).
ICP34.5 contains signals which target it to different cellular compartments and this determines
its function. The growth arrest and DNA damage-inducible protein (GADD34) mediates growth
arrest and apoptosis in response to DNA damage, negative growth signals and protein
misfolding. GADD34 binds to PP1 and can reverse stress-induced translation arrest through
dephosphorylation of eukaryotic initiation factor-2α (Brush et al., 2003). In addition the
GADD34 protein has been shown to interact with the hSNF5/INI1 component of the
hSW1/SNF chromatin remodelling complex (Wu et al., 2002) and have a role in transcription
regulation.
In previous work (Goatley et al., 1999) we showed that the l14L protein is present in the
genomes of all ASFV isolates as either a short form encoding a 71 amino acid protein or a
long form encoding 184 amino acids. The presence of these different forms of the gene is
correlated with the geographical region from which viruses were isolated. Both forms of the
gene encode the C-terminal domain that is similar to the HSV ICP34.5 protein and the
GADD34/myd116 DNA damage response protein. These conserved domains contain motifs
required for binding to PP1 catalytic subunit as well as a nuclear localisation signal (Cheng et
al., 2002, Goatley et al., 1999). In addition both the long form of the DP71L protein and the
HSV ICP34.5 protein contain N-terminal basic domains that act as nucloelar targeting regions.
The l14L protein does not contain a nuclear export signal nor a repeated amino acid sequence
that is present in the HSV ICP34.5 protein and act to anchor the protein in the cytoplasm
(Cheng et al., 2002, Goatley et al., 1999). This could explain why the l14L protein forms are
localised mainly either in the nucleus or nucleolus (Goatley et al., 1999) and not in the
cytoplasm.
The presence of the PP1 docking motif in both forms of the DP71L protein and its similarity to
the HSV ICP34.5 and GADD34 proteins suggested that it may bind to and modulate PP1
activity in infected cells. PP1 is involved in regulating a broad spectrum of cellular activities. Its
activities are regulated by association with regulatory subunits which can either target the
enzyme to dephosphorylate particular substrates or inhibit its activity. To investigate the effect
of l14L expression on PP1 activity we compared activity of PP1 in cells infected with wild-type
ASFV or a deletion mutant from which the l14L gene was deleted.
ASFV infection activaties PP1 and this is dependent on expression of the l14L protein.
Vero cells were harvested at different time points (4 and 18 h.) after infection with the BA71V
isolate or a deletion mutant, Δl14L, lacking the l14L gene or following mock infection: Cell
extracts were prepared in cell lysis buffer (10 mM Tris-HCl, pH 7.6, 140 mM KCl, 4 mM MgCl2,
1 mM DTT, 1 mM EDTA, 0.1% Triton X-100, 0.5% Nonidet P-40, 2,5 mM sodium
pyrophosphate; 1 mM -Glycerophosphate, 1mM sodium orthovanadate and a protease
inhibitors) after washing cells with PBS supplemented with 1mM sodium orthovanadate.
PP1 and PP2A phosphatase activities were determined (Cohen et al. 1988) using purified 32Plabeled phosphorylase a as the substrate. To prepare this substrate, phosphorylase b (10
mg/ml) was phosphorylated by incubation with phosphorylase kinase (0.2 mg/ml) (Quevedo et
SID 5 (Rev. 3/06)
Page 8 of 14
al 2003). Phosphatase activity was measured as the amount of 32P released from the
substrate. Phosphorylase a is a substrate for both PP1 and PP2A. To distinguish between
these two phosphatases activity, assays were performed in the presence of either I2 or
fostriecin, specific inhibitors of PP1 and PP2A respectively. The doses of PP1 and PP2
inhibitors used were predetermined by a set of dose-response experiments in which the effect
of each inhibitor was independently measured.
The results showed that infection of Vero cells with ASFV BA71V isolate resulted in an
increase in total PP1 activity by 1.6 fold compared to levels in mock-infected cells by 4 hours
post-infection and by 2.8 fold by 16 hours post-infection. In contrast, in cells infected with the
Δl14Lrecombinant ASFV lacking the l14L gene, total PP1 activity remained at the same level
as in mock-infected cells at both time points. This result shows that infection with ASFV
activates PP1 and that the l14L gene is required for this activation. To determine if the effect
of the l14L protein is specific for PP1 or if other serine threonine protein phosphatases are
affected, we measured levels of the other major phosphatase PP2A. This analysis showed that
infection of cells with either the wild type BA71V or the Δl14Lrecombinant ASFV lacking the
l14L gene did not increase levels of PP2A activity above that observed in mock-infected cells.
Thus we conclude that activation of PP1 is specific.
Expression of the l14L protein inhibits transcription from the interferon-beta gene
promoter.
Our hypothesis is that the l14L acts as a regulatory subunit of PP1 activating the enzyme by
displacing inhibitory subunits and targeting the enzyme to dephopshorylate specific substrates.
The nuclear location of the short form or the l14L protein and nucleolar localisation of the long
form of the protein suggest the substrates may be localised in these compartments. PP1 has
been shown to regulate both transcription and processing of mRNAs and we therefore
examined a possible role for l14L in transcription regulation. These experiments were carried
out in collaboration with Drs Emma Poole and Antonio Alcami, University of Cambridge, Prof
Steve Goodbourn, London University.
Plasmids expressing either the long form or the short form of the l14L protein were cotransfected into cells with a reporter plasmid containing the luciferase gene downstream from
an interferon-beta gene promoter sequence (from Steve Goodbourn). Following treatment of
cells with double-stranded RNA to induce interferon, activation of transcription from the
interferon-beta promoter was determined by measuring levels of luciferase activity. This
anlaysis showed that expression of the short form of l14L protein inhibited transcriptional
activation from the interferon-beta promoter.
We examined whether the failure of the long form of l14L protein to inhibit transcription may be
because of its nucleolar localisation. We transfected a plasmid expressing a mutant form of the
l14L long form, which lacks the nucleolar targeting signal, into cells and measured
transcriptional activation of the interferon beta promoter. This showed that when targeted to
the nucleus rather than nucleolus the long form of the l14L protein also inhibited transcriptional
activation of the interferon beta promoter.
To investigate if the l14L protein may also inhibit interferon activated pathways we cotransfected a plasmid expressing the l14L protein into cells with a promoter containing an
interferon sensitive response element upstream of a luciferase reporter gene. We stimulated
cells with interferon-beta and measured luciferase activity in cell extracts to monitor promoter
activation. The results showed that the l14L protein did not significantly inhibit transcription
from the interferon-activated promoter.
SID 5 (Rev. 3/06)
Page 9 of 14
The results show that the short form of the l14L protein can inhibit type I interferon induction
although not pathways activated by type I interferon. The protein is therefore likely to be
important in helping ASFV to evade host defences.
Further work is required to determine if activation of PP1 is required for l14L to inhibit
transcription from the interferon beta promoter. Further work is also required to determine
additional functions for the long and short forms of the l14L protein and to investigate if these
functions are linked to the role of the protein in virus virulence.
4. Determination of complete nucleotide sequence of the low virulence ASFV isolate
OURT88/3 Most ASFV isolates cause a severe haemorrhagic fever with high mortality in
domestic pigs. However some isolates (NHP68 and OUR T88/3) have been described which
are non-pathogenic and cause a persistent infection with sporadic low viraemia (Leitao et al.,
2001, Boinas et al., 2004). Pigs infected with these isolates are protected against challenge
with closely related virulent isolates and these virus isolates therefore provide a good model
for mechanisms of protective immunity. To help understand the molecular basis for the
difference in pathogenesis between these low virulence isolates and the high virulence isolates
we determined the complete genome sequence of the OUR T88/3 isolate and compared this
to complete genomes of 8 virulent isolates and one tissue-culture adapted isolate.
OUR T88/3 virus was purified from supernatants of infected porcine macrophage cultures and
DNA extracted from the purified virus. The genome was amplified using the Repli G method
(Qiagen) and sequenced using primers designed from the BA71V complete genome sequence
and partial sequence of the E70 isolate present in the EMBL database. The genome sequence
was assembled into a single contig of 171,719 nucleotides excluding the inverted terminal
repeats for which sequence was not obtained. The estimated error rate was 2 nucleotides over
this complete sequence. The sequence was analysed using software available from Dr Chris
Upton, University of Victoria, Canada on the website www.virology.ca. The complete sequence
encoded 165 predicted open reading frames (ORFs). Of these 109 were conserved in all 10
complete genome sequences available. These included 39 which encoded proteins of known
function either in transcription or replication of the virus genome, virus structural proteins or
roles in evasion of host defences. A further 42 ORFs encoded proteins which either contained
recognised motifs or had similarity to known proteins. The remaining 28 conserved ORFs
encoded proteins with no similarity to other proteins.
In addition to these conserved ORFs all of the virus genomes encoded 5 different multigene
families (MGF) which varied in number when genomes of different isolates were conserved.
The MGF 360 was the largest family and contained between 11 and 19 copies in different
genomes. MGF 505/530 varied between 8 and 10 copies per genome, MGF 110 between 5
and 13, MGF 300 between 3 and 4 and MGF 100 between 2 and 3. The OUR T88/3 isolate
encoded the smaller number of copes of MGF360 (11) and of MGF 505/530 (8) compared to
the virulent isolates. In contrast the OUR T88/3 isolate encoded more copies of MGF 110
compared to some of the virulent isolates.
Pairwise alignment of the complete genomes revealed that a region approximately 6.5 kbp
from the left genome end which encoded copies of MGF110 and MGF 100 had undergone
both additions and deletions in genomes of different isolates. About 20 kbp from the left
genome end a sequence of approximately 8 kbp had been deleted from the OUR T88/3
genome. This contained 6 copies of MGF 360 and 2 copies of MGF 530. The same genes
were also deleted from the genome of the non-pathogenic NH P68 isolate and from the tissue
culture adapted BA71V isolate apart from one copy of MGF 530. Previously, deletion of these
genes from the genome of a virulent isolate had been shown to reduce virulence of the virus
for domestic pigs and to result in increased production of interferon alpha from virus-infected
SID 5 (Rev. 3/06)
Page 10 of 14
macrophages (Afonso et al., 2004). Thus we predict that deletion of these genes from the
genome of the OUR T88/3 isolate is a reason for the reduced virulence of this isolate.
Other changes in the genome of the OUR T88/3 isolate, compared to virulent isolates, include
frameshift mutations in the genes encoding the CD2 like protein (EP402R) and lectin-like
protein (EP153R). These mutations would mean that functional proteins are not expressed and
this would explain the failure of cells infected with this isolate to cause haemadsorbtion of red
blood cells around infected cells. The CD2v protein has been implicated in virus dissemination
in infected pigs and in impairment of lymphocyte function (Borca et al., 1998). Therefore failure
to express the protein may result in attenuation and a more effective immune response against
the virus.
The OUR T88/3 and all of the other virulent isolates contain an additional 4.5 kbp encoding 5
ORFs about 5 kbp from the right end of the genome. The deleted ORFs include one which
encodes an extra copy of the p22 structural protein which is also encoded at the left end of the
genome. Thus in the BA71V tissue-culture adapted isolate genome only one copy of this gene
is encoded whereas in the genomes of all field isolates 2 or 3 copies are encoded. Possible
the deletion of these genes may help to explain why the BA71V isolate is unable to replicate
efficiently in porcine macrophage cultures.
The information from this sequence analysis has provided us with the information to explore
further the molecular basis for the attenuation of the OUR T88/3 isolate and to rationally
construct candidate vaccine strains based on currently circulating virulent isolates by
sequential deletion of genes involved in immune evasion and virulence.
Implications from the findings and possible future work.
Identification of novel ASFV encoded proteins involved in evasion of host defences and
or virulence. The results described have identified novel ASFV genes involved in evading
host defences and virulence. The l14L protein had previously been identified as a virulence
factor for domestic pigs but its mechanism of action was unknown. Here we have identified two
functions which are very likely to be important for evading host defences and in virus virulence.
These two functions, activation of the cellular phosphatase PP1 and inhibition of transcription
from the interferon-beta promoter may be linked. In addition, because PP1 regulates a broad
spectrum of cellular activities, the l14L protein may have additional effects on host cell
function. Further work will be required to investigate both of these questions. Our data shows
that l14L inhibits induction of type I interferons and will therefore inhibit induction of the antiviral
state in bystander as well as infected cells. This would help virus replication in an infected
host.
We sequenced the complete genome of a non-pathogenic isolate OUR T88/3 and identified
one large sequence deletion of about 8 kbp which encodes genes involved in virulence and
immune evasion. We also showed that the genes encoding the CD2v and lectin-like proteins
are interrupted. Together these findings explain at least in part why this isolate is nonpathogenic and induces a protective immune response.
In future we will use this information to rationally construct candidate attenuated ASFV vaccine
strains by sequential deletion of genes involved in virulence and immune evasion from a
virulent W. African isolate. Our strategy is to delete the genes encoding the CD2v and lectinlike genes in one step followed by the 8 genes from MGF 360 and 530 missing from the OUR
T88/3 isolate close to the left end of the genome. These will also be deleted in one step.
Subsequently we will delete additional genes including the l14L gene and the A238L genes.
Testing of these candidate vaccine strains will identify which is the best combination of genes
to delete to produce an effective vaccine. This strategy can then be applied to antigenically
diverse isolates circulating in other parts of Africa. Development of a vaccine will help to
control disease in Africa where many countries suffer continuing losses in commercial pig
farms as well as to poor farmers in rural and peri-urban areas. The threat of ASF also
SID 5 (Rev. 3/06)
Page 11 of 14
effectively limits pig production in Africa depriving people of a relatively cheap source of high
quality protein. Controlling disease in Africa will also reduce the risk of accidental introduction
elsewhere in the world.
Changes in macrophage transcription responses following infection with ASFV isolates
of different virulence
We also used porcine microarrays to study macrophage transcription responses following
infection with ASFV isolates of different virulence. This has given us clues about the
mechanisms of virus pathogensis and immune evasion. Previous work had shown that ASFV
encode several proteins which interfere with signalling pathways in infected macrophages and
may thus inhibit transcriptional activation of host immune response genes. Following infection
with a high virulence isolate we found that most of the 125 differentially regulated genes
identified fell into a class which were increased in transcription level compared to mockinfected cells at early times post-infection but returned to the base level at late times. These
included many which were predicted to be activated as part of the host response to infection
including proinflammatory cytokines and chemokines as well as other cell surface and
secreted proteins (Zhang et al., in press). This is consistent with the action of virus-encoded
proteins which inhibit specific host transcription factors when they are expressed in infected
cells and would be expected to reduce transcription of host immunomodulatory genes.
Interestingly we identified some differences in macrophage transcription responses following
infection with a non-pathogenic isolate. In particular we observed increased transcription of
some genes involved in activation of the immune response (eg IFN-γ) and in antigen
presentation to T cells. This suggests that the host’s immune response may respond more
effectively following infection with the non-pathogenic isolate. Possibly this may be because
the non-pathogenic isolate lacks genes encoding proteins which interfere with this response.
In future we plan to compare global transcription profiles following infection of macrophages in
vitro with candidate vaccine strains as one of a series of in vitro screens. These in vitro assays
aim to predict the likely pathogenesis of the candidate vaccines as well as their ability to
induce a protective immune response. By developing and using these assays we can limit the
number of pig experiments carried in future which will have both cost and ethical benefits.
SID 5 (Rev. 3/06)
Page 12 of 14
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 13 of 14
1.
2.
3.
4.
5.
6.
7.
8.
9.
10
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24
25.
26.
Afonso, C. L., M. E. Piccone, K. M. Zaffuto, J. Neilan, G. F. Kutish, Z. Lu, C. A. Balinsky, T. R.
Gibb, T. J. Bean, L. Zsak, and D. L. Rock. 2004. Journal of Virology 78:1858-1864.
Boinas, F. S., G. H. Hutchings, L. K. Dixon, and P. J. Wilkinson. 2004. Journal of General
Virology 85:2177-2187.
Borca, M. V., C. Carrillo, L. Zsak, W. W. Laegreid, G. F. Kutish, J. G. Neilan, T. G. Burrage,
and D. L. Rock. 1998.. Journal of Virology 72:2881-2889.
Brown, S. M., J. Harland, A. R. Maclean, J. Podlech, and J. B. Clements. 1994. Journal of
General Virology 75:2367-2377.
Brown, S. M., A. R. Maclean, J. D. Aitken, and J. Harland. 1994. I. Journal of General Virology
75:3679-3686.
Brush, M. H., D. C. Weiser, and S. Shenolikar. 2003.. Molecular and Cellular Biology 23:12921303.
Cheng, G. F., M. E. Brett, and B. He. 2002. Journal of Virology 76:9434-9445.
Cohen, P., S. Alemany, B. A. Hemmings, T. J. Resink, P. Stralfors, and H. Y. L. Tung. 1988.
Methods in Enzymology 159:390-480.
Dixon, L. K., C. C. Abrams, G. Bowick, L. C. Goatley, P. C. Kay-Jackson, D. Chapman, E.
Liverani, R. Nix, R. Silk, and F. Q. Zhang. 2004. Veterinary Immunology and Immunopathology
100:117-134.
Dixon, L. K., Escribano, J.M., Martins, C., Rock, D.L., Salas, M.L. and Wilkinson, P.J. (2005).
Asfarviridae. 2005. In: Virus Taxonomy, VIIIth Report of the ICTV (C.M. Fauquet, M.A. Mayo, J.
Maniloff, U. Desselberger, and L.A. Ball, eds), 135-143. Elsevier/Academic Press, London.
Goatley, L. C., M. B. Marron, S. C. Jacobs, J. M. Hammond, J. E. Miskin, C. C. Abrams, G. L.
Smith, and L. K. Dixon. 1999. Journal of General Virology 80:525-535.
Granja, A. G., M. L. Nogal, C. Hurtado, C. del Aguila, A. L. Carrascosa, M. L. Salas, M.
Fresno, and Y. Revilla. 2006. Journal of Immunology 176:451-462.
Granja, A. G., M. L. Nogal, C. Hurtado, V. Vila, A. L. Carrascosa, M. L. Salas, M. Fresno, and
Y. Revilla. 2004. Journal of Biological Chemistry 279:53736-53746.
Harland, J., P. Dunn, E. Cameron, J. Conner, and S. M. Brown. 2003. Journal of Neurovirology
9:477-488.
He, B., M. Gross, and B. Roizman. 1998. Journal of Biological Chemistry 273:20737-20743.
He, B., M. Gross, and B. Roizman. 1997. Proceedings of the National Academy of Sciences of
the United States of America 94:843-848.
Leitao, A., C. Cartaxeiro, R. Coelho, B. Cruz, R. M. E. Parkhouse, F. C. Portugal, J. D.
Vigario, and C. L. V. Martins. 2001.. Journal of General Virology 82:513-523.
Miskin, J. E., C. C. Abrams, L. C. Goatley, and L. K. Dixon. 1998. Science 281:562-565.
Powell, P. P., L. K. Dixon, and R. M. E. Parkhouse. 1996. Journal of Virology 70:8527-8533.
Quevedo, C., M. Salinas, and A. Alcazar. 2003. Journal of Biological Chemistry 278:1657916586.
Revilla, Y., M. Callejo, J. M. Rodriguez, E. Culebras, M. L. Nogal, M. L. Salas, E. Vinuela, and
M. Fresno. 1998. Journal of Biological Chemistry 273:5405-5411.
Weingartl, H. M., M. Sabara, J. Pasick, E. van Moorlehem, and L. Babiuk. 2002. Journal of
Virological Methods 104:203-216.
Wu, D. Y., D. C. Tkachuck, R. S. Roberson, and W. H. Schubach. 2002. Journal of Biological
Chemistry 277:27706-27715.
Zhang F., Hopwood P., Abrams C.C., Downing A., Murray F., Talbot R., Archibald A., Lowden
S., Dixon L.K. Journal of Virology (in press)
Zhao, S. H., J. Recknor, J. K. Lunney, D. Nettleton, D. Kuhar, S. Orley, and C. K. Tuggle.
2005. Genomics 86:618-625.
Zsak, L., Z. Lu, G. F. Kutish, J. G. Neilan, and D. L. Rock. 1996. Journal of Virology 70:88658871.
SID 5 (Rev. 3/06)
Page 14 of 14