African swine fever virus thymidylate kinase gene: sequence and transcriptional mapping

A putative thymidylate kinase gene of African swine fever virus has been identified at the left end of the SalI I' fragment of the virus genome. The gene, designated A240L, has the potential to encode a protein of 240 amino acids with an M(r) of 27,754 and is transcribed early after infection. Primer extension analysis indicates that transcription is initiated a short distance from the first ATG codon of open reading frame A240L. The deduced amino acid sequence of this open reading frame shows significant similarity with the human, yeast and vaccinia virus thymidylate kinases, the degree of identity being 23.7, 25 and 23.5%, respectively. The putative African swine fever virus thymidylate kinase sequence is essentially collinear with the other thymidylate kinase sequences, but contains a carboxy-terminal extension of 37 amino acids rich in glutamic and aspartic acids. The A240L protein conserves the ATP-binding and nucleotide/nucleoside-binding domains characteristic of thymidylate kinases.

Polyprotein processing in African swine fever virus: a novel gene expression strategy for a DNA virus

This report shows that African swine fever virus (ASFV)--a large DNA-containing virus--synthesizes a polyprotein to produce several of its structural proteins. By immunoprecipitation analysis, we have found that ASFV polyprotein is a 220 kDa myristoylated polypeptide (pp220) which, after proteolytic processing, gives rise to four major structural proteins: p150, p37, p34 and p14. Processing of the ASFV polyprotein takes place at the consensus sequence Gly-Gly-X and occurs through an ordered cascade of proteolytic cleavages. So far, polyprotein processing as a mechanism of gene expression had been found only in positive-strand RNA viruses and retroviruses. According to the results presented here, ASFV is the first example of a DNA virus that synthesizes a polyprotein as a strategy of gene expression.

African swine fever virus encodes two genes which share significant homology with the two largest subunits of DNA-dependent RNA polymerases

A random sequencing strategy applied to two large SalI restriction fragments (SB and SD) of the African swine fever virus (ASFV) genome revealed that they might encode proteins similar to the two largest RNA polymerase subunits of eukaryotes, poxviruses and Escherichia coli. After further mapping by dot-blot hybridization, two large open reading frames (ORFs) were completely sequenced. The first ORF (NP1450L) encodes a protein of 1450 amino acids with extensive similarity to the largest subunit of RNA polymerases. The second one (EP1242L) codes for a protein of 1242 amino acids similar to the second largest RNA polymerase subunit. Proteins NP1450L and EP1242L are more similar to the corresponding subunits of eukaryotic RNA polymerase II than to those of vaccinia virus, the prototype poxvirus, which shares many functional characteristics with ASFV. ORFs NP1450L and EP1242L are mainly expressed late in ASFV infection, after the onset of DNA replication.

Characterization of two African swine fever virus 220-kDa proteins: a precursor of the major structural protein p150 and an oligomer of phosphoprotein p32

Two kinds of unrelated African swine fever virus proteins of 220 kDa have been identified by means of two-dimensional gel electrophoresis and immunoprecipitation analysis. One species, named pp220 and identified as the precursor of the major structural protein p150, was found to be a moderately acidic protein (pl near 7) expressed after the replication of the viral DNA. The second species, a cluster of 220-kDa proteins with slightly different isoelectric points (pl ranging from 5 to 6), was found to be a homooligomeric complex formed by an early 32-kDa protein. This component was identified as the viral phosphoprotein p32, the most immunogenic early protein of African swine fever virus. A detailed characterization of its oligomeric structure is reported.

Sequence and characterization of the major early phosphoprotein p32 of African swine fever virus

The gene encoding protein p32, the most abundant and immunogenic protein induced by African swine fever virus at early times of infection, has been mapped in the EcoRI C' fragment of the genome of the Vero cell-adapted virus strain BA71V. Sequencing analysis has shown the existence of an open reading frame, named C'204L, encoding 204 amino acids. The protein is phosphorylated in serine residues located in the 115 N-terminal amino acids and was phosphorylated when expressed in cells infected with a vaccinia virus recombinant. Protein p32 is not glycosylated in spite of the presence of two putative N-glycosylation sites in the deduced amino acid sequence of the polypeptide. Immunofluorescence experiments have shown that the protein is localized in the cytoplasm of infected cells and not in the plasma membrane. In addition, the protein has been found in the soluble fraction and not in microsomes from BA71V-infected Vero cells. Low levels of the protein have been detected in the medium from infected swine macrophages, which probably corresponds to nonspecific release of cytoplasmic proteins. The protein encoded by other virus isolates shows different electrophoretic mobilities, indicating variability of p32.

African swine fever virus encodes a DNA ligase

Sequencing of the EcoRI N' fragment of African swine fever virus (ASFV) DNA revealed an open reading frame encoding a protein similar to ATP-dependent DNA ligases. When the gene encoding this protein was expressed in Escherichia coli, a protein of the expected molecular mass was labeled in bacterial extracts upon incubation with [alpha-32P]ATP. The recombinant protein comigrated in SDS-PAGE with the putative viral DNA ligase detected in extracts of infected cells. We demonstrate that the recombinant protein is a DNA ligase by dissociation of the protein-[32P]AMP adduct with pyrophosphate and nicked DNA. The putatively adenylylated lysine in ASFV is surrounded by two arginine residues, instead of by two hydrophobic amino acids as in the other ATP-dependent DNA ligases. This might explain the high concentration of pyrophosphate necessary to revert the DNA ligase--AMP adduct in ASFV, 10- to 100-fold higher than that required for other DNA ligases. A comparison of the amino acid sequences reported for ATP-dependent DNA ligases disclosed three new amino acid motifs around the adenylylation site of these enzymes. ASFV DNA ligase has little similarity to the other enzymes at the ends of the molecule, but conserves the amino acid motifs of the central region.

Localization of the African swine fever virus attachment protein P12 in the virus particle by immunoelectron microscopy

The African swine fever virus attachment protein p12 was localized in the virion by immunoelectron microscopy. Purified virus particles were incubated, before or after different treatments, with p12-specific monoclonal antibody 24BB7 and labeled with protein A-colloidal gold. Untreated virus particles showed labeling only in lateral protrusions that followed the external virus envelope. Mild treatment of African swine fever virions with the nonionic detergent octyl-glucoside or with ethanol onto the electron microscope grid resulted in a heavier and more homogeneous labeling of the virus particles. In contrast, the release of the external virus proteins by either octyl-glucoside or Nonidet-P40 and beta-mercaptoethanol generated a subviral fraction that was not labeled by 24BB7. Preembedding, labeling, and thin-sectioning experiments confirmed that the antigenic determinant recognized by 24BB7 was localized into the external region of the virus particle but required some disruption to make it more accessible. From these results we conclude that protein p12 is situated in a layer above the virus capsid with, at least, one epitope predominantly not exposed in the virion surface; this epitope may not be related to the virus ligand-cell receptor interaction.

Haemostatic abnormalities in African swine fever a comparison of two virus strains of different virulence (Dominican Republic '78 and Malta '78)

African swine fever (ASF) virus strains cause haemorrhage by producing a variety of defects, which vary in severity from strain to strain. To distinguish the main haemostatic defects leading to haemorrhage, two groups of pigs were infected with moderately virulent (Dominican Republic '78) and less virulent (Malta '78) ASF virus strains. Mortality rate and severity of clinical observations were greater in pigs infected with DR '78 virus compared with pigs infected with Malta '78 virus. The animals became febrile from day 3 to 4 onwards at a time when the viraemia was high (10(7) to 10(8) HAD50/ml). No difference was found during the period observed in their pattern of viraemia or pyrexia. Thrombocytopenia developed in both groups but with different kinetics, suggesting two different mechanisms of sequestration of platelets. When coagulation tests were performed, significant abnormalities were found, including evidence for disseminated intravascular coagulation. These abnormalities were much less pronounced in the group infected with Malta '78. Antithrombin III activity did not change significantly in either group. Decreased plasminogen activity was found in the early phase of disease in DR '78 infected pigs. These results indicate that when haemorrhage does occur in DR '78 infected pigs, it is a consequence of more pronounced degrees of haemostatic impairment probably due to a marked endothelial injury and/or generation of procoagulant activity.

African swine fever virus guanylyltransferase

The gene coding for the guanylyltransferase of African swine fever virus has been identified and sequenced. The gene, designated NP868R, is located within fragments EcoRI N' and D of the virus genome (BA71V strain) and encodes a protein with a predicted molecular mass of 99.9 kDa that shares significant similarity with the large subunit of both vaccinia and Shope fibroma virus capping enzymes, with percentages of identity of 20.6 and 21.8%, respectively. A protein of 95 kDa was induced in Escherichia coli cells transformed with a recombinant plasmid carrying the NP868R gene. The E. coli expressed protein, as well as a protein of the same molecular weight present in African swine fever virus particles, form a covalent complex with GTP that can be reversed by pyrophosphate, two characteristic reactions of guanylyltransferases. An examination of the amino acid sequences of the African swine fever virus, poxvirus, and yeast guanylyltransferases has revealed a common motif around a lysine residue at the amino-terminal part of the proteins [Y(V, A)X2K(T, A)DG] which resembles the adenylylation site of DNA ligases (Tomkinson, A. E., Totty, N. F., Ginsburg, M., and Lindahl, T. (1991). Proc. Natl. Acad. Sci. USA 88, 400-404). This lysine residue could be the guanylylation site in these enzymes.

Structure and expression in E. coli of the gene coding for protein p10 of African swine fever virus

The gene encoding protein p10, a structural protein of African swine fever (ASF) virus, has been mapped, sequenced and expressed in E. coli. Protein p10 was purified from dissociated virus by reverse-phase HPLC, and its NH2-terminal end identified by automated Edman degradation. To map the gene encoding protein p10, a mixture of 20-mer oligonucleotides based upon a part of the amino acid sequence was hybridized to cloned ASF virus restriction fragments. This allowed the localization of the gene in fragment Eco RI K of the ASF virus genome. The nucleotide sequence obtained from this region revealed an open reading frame encoding 78 amino acids, with a high content of Ser and Lys residues. Several of the Ser residues are found in Ser-rich regions, which are also found in some nucleic acid-binding proteins. The gene coding for protein p10 has been inserted in an expression vector which contains the promoter for T7 RNA polymerase. The recombinant plasmid was used to produce the ASF virus protein in E. coli. The bacterially produced p10 protein shows a strong DNA binding activity with similar affinity for both double-stranded and single-stranded DNA.

High-level expression in Escherichia coli of the gene coding for the major structural protein (p72) of African swine fever virus

The gene encoding the major structural protein (p72) of African swine fever virus (ASFV) has been expressed in Escherichia coli using a T7 RNA polymerase system. The use of a recombinant plasmid which contains the entire gene inserted between the T7 promoter and the transcription terminator of the expression vector allowed us to obtain a high expression level of the intact viral protein. This polypeptide, which appears in the insoluble fraction of the bacterial extracts, showed an intense reaction with the antibodies present in the sera of ASFV-infected animals, as demonstrated by Western blot and enzyme-linked immunosorbent assay. The recombinant protein was purified by size-exclusion high-performance liquid chromatography and used to develop a serological test of the disease.

Transcriptional mapping of a late gene coding for the p12 attachment protein of African swine fever virus

The transcriptional characterization of the gene coding for the p12 attachment protein of the African swine fever virus is presented. The results obtained have been used to generate the first detailed transcriptional map of an African swine fever virus late gene. Novel experimental evidence indicating the existence of major differences between the mechanisms controlling the transcription of late genes in African swine fever virus and poxviruses is provided.

Virus-host interactions in African swine fever: the attachment to cellular receptors

Biochemical and morphological techniques have shown that African swine fever virus (ASFV) enters susceptible cells by a mechanism of receptor-mediated endocytosis. The virus binds to a specific, saturable site in the cell and this interaction is required for a productive infection. A structural ASFV protein of 12kDa (p12) has been identified to be involved in the recognition of the cellular receptor, on the basis of the specific binding of the polypeptide to sensitive Vero cells. Protein p12 is externally located in the virus particle, forming disulfide-linked dimers with an apparent molecular mass of 17kDa. The gene has been mapped within the central region of the BA71V strain genome. Sequencing analysis has shown the existence of an open reading frame encoding a polypeptide of 61 amino acids characterized by the presence of a putative transmembrane domain, and a cysteine rich region in the C-terminal part which may be responsible for the dimerization of the protein. Transcripts of the p12 gene were only synthesized during the late phase of the infectious cycle. No posttranslational modifications of the polypeptide, such as glycosylation, phosphorylation or fatty acid acylation, have been found. The comparison of the amino acid sequence of protein p12 from 11 different virus strains has revealed a high degree of conservation of the polypeptide.

An erythroid species-specific antigen of swine detected by a monoclonal antibody

A monoclonal antibody (1AC11) has been produced which recognized the glycophorin of swine red blood cells. 1AC11 was specific for swine membrane erythrocytes. No other swine cells (leukocytes, macrophages, kidney and testis cells) nor red blood cells from all the tested mammalian species (goat, human, sheep, cattle, horse, rabbit, cat and guinea pig) were recognized. There was no blood group activity detected. Immunocytochemical analysis of blood vessel in the swine pituitary tissue showed that besides membrane erythrocytes, cytoplasmic molecules were recognized in some cells. Immunoblot analysis of both membrane and aqueous phase of chloroform/methanol fractions from swine erythrocytes showed that the monoclonal antibody 1AC11 reacts with the major sialoglycoprotein of apparent molecular weight 45,000 daltons.

Transcriptional analysis of multigene family 110 of African swine fever virus

A transcriptional analysis of the 3.2-kb region of the African swine fever virus genome containing the five members of the multigene family 110 is presented. The mRNAs corresponding to the genes studied have short leader sequences with no intervening AUG codons before the translational start site, and their 3' ends map within a conserved sequence motif formed by a stretch of seven or more consecutive thymidylate residues. The possible role of this sequence as a signal for the 3'-end formation of African swine fever virus mRNAs is discussed. While four of the genes studied are actively transcribed from the beginning of the infection until the onset of virus DNA replication, the transcription of one of the members of the multigene family 110, the L270 gene, is silenced at an earlier time. A detailed analysis, including in vitro translation of mRNAs isolated from infected Vero cells, revealed that the L270 gene belongs to a small subset of early genes, designated immediate early, whose transcription is silenced before the onset of virus DNA replication. The transcriptional data obtained enabled us to generate the first detailed transcriptional map of a region of the African swine fever virus genome, thus opening the possibility of studying the cis-acting sequences involved in transcriptional control of the viral genes.

A gene homologous to topoisomerase II in African swine fever virus

A putative topoisomerase II gene of African swine fever virus was mapped using a degenerate oligonucleotide probe derived from a region highly conserved in type II topoisomerases. The gene is located within EcoRI fragments P and H of the African swine fever virus genome. Sequencing of this region has revealed a long open reading frame, designated P1192R, encoding a protein of 1192 amino acids, with a predicted molecular weight of 135,543. Open reading frame P1192R is transcribed late after infection into a 4.6-kb RNA. The deduced amino acid sequence of this open reading frame shares significant similarity with topoisomerase II sequences from different sources, with percentages of identity between 23 and 29%. The evolutionary relationships among the topoisomerase II sequences of ASF virus, eukaryotes and prokaryotes were analyzed and a phylogenetic tree was established. The tree indicates that the ASF virus topoisomerase II gene was present in the virus genome before protozoa, yeasts, and metazoa diverged.

Comparison of the sequence of the gene encoding African swine fever virus attachment protein p12 from field virus isolates and viruses passaged in tissue culture

Comparison of the amino acid sequence of the African swine fever virus attachment protein p12 from different field virus isolates, deduced from the nucleotide sequence of the gene, revealed a high degree of conservation. No mutations were found after adaptation to Vero cells, and a polypeptide with similar characteristics was present in an IBRS2-adapted virus. The sequence of the 5' flanking region was conserved among the isolates, whereas sequences downstream of the gene were highly variable in length and contained direct repeats in tandem that may account for the deletions found in different isolates. Protein p12 was synthesized in swine macrophages infected with all of the viruses tested.

Amino acid sequence and structural properties of protein p12, an African swine fever virus attachment protein

The gene encoding the African swine fever virus protein p12, which is involved in virus attachment to the host cell, has been mapped and sequenced in the genome of the Vero-adapted virus strain BA71V. The determination of the N-terminal amino acid sequence and the hybridization of oligonucleotide probes derived from this sequence to cloned restriction fragments allowed the mapping of the gene in fragment EcoRI-O, located in the central region of the viral genome. The DNA sequence of an EcoRI-XbaI fragment showed an open reading frame which is predicted to encode a polypeptide of 61 amino acids. The expression of this open reading frame in rabbit reticulocyte lysates and in Escherichia coli gave rise to a 12-kDa polypeptide that was immunoprecipitated with a monoclonal antibody specific for protein p12. The hydrophilicity profile indicated the existence of a stretch of 22 hydrophobic residues in the central part that may anchor the protein in the virus envelope. Three forms of the protein with apparent molecular masses of 17, 12, and 10 kDa in sodium dodecyl sulfate-polyacrylamide gel electrophoresis have been observed, depending on the presence of 2-mercaptoethanol and alkylation with 4-vinylpyridine, indicating that disulfide bonds are responsible for the multimerization of the protein. This result was in agreement with the existence of a cysteine-rich domain in the C-terminal region of the predicted amino acid sequence. The protein was synthesized at late times of infection, and no posttranslational modifications such as glycosylation, phosphorylation, or fatty acid acylation were detected.

Role of the host cell nucleus in the replication of African swine fever virus DNA

An examination by autoradiography of African swine fever virus-infected alveolar macrophages pulse labeled with [3H]thymidine showed that, at early times of viral DNA replication, the grains were localized exclusively in the nucleus in 20% of the cells, while in 45% the label was found in the cytoplasm. In the remaining 35%, newly synthesized DNA was detected in both the nucleus and the cytoplasm. At later times, the percentage of cells with grains in the nucleus decreased considerably. Pulse-chase experiments indicated that the DNA synthesized in the nucleus is then transported to the cytoplasm. The presence of virus-specific DNA sequences in the nucleus was confirmed by in situ hybridization of infected macrophages. Similar hybridization experiments with African swine fever virus-infected VERO cells followed by confocal microscopy also indicated the existence of a nuclear stage in the localization of the viral DNA. These results suggest a mechanism for African swine fever virus DNA replication with an initial stage in the nucleus followed by a cytoplasmic phase. Specific nuclear forms associated with the hybridization signal have been observed in African swine fever virus-infected macrophages and VERO cells. The nuclear forms seen in macrophages are consistent with a mechanism for the egress of the viral DNA from the nucleus that involves initial budding at the nuclear membrane.

Genetic manipulation of African swine fever virus: construction of recombinant viruses expressing the beta-galactosidase gene

Homologous recombination is shown to be specifically induced in Vero cells by infection with African swine fever (ASF) virus. The frequency of recombination induced by ASF virus infection between cotransfecting plasmids is comparable to that found after infection with the prototype poxvirus, vaccinia virus. The induction of recombination is accompanied by replication of the plasmid templates in the ASF virus-infected cells. An ASF virus insertion/expression plasmid vector containing the Escherichia coli reporter gene beta-galactosidase (beta-gal) fused to a viral promoter sequence was constructed. Recombination between homologous sequences present in both the plasmid vector and the virus genome led to the generation of recombinant viruses expressing the beta-gal gene. Visual screening of beta-gal+ plaques allowed the isolation and plaque purification of recombinant ASF viruses. The characterization of a beta-gal+ virus isolate showed that the beta-gal gene had been stably inserted into the thymidine kinase locus of the virus genome, thus demonstrating that controlled genetic manipulation of ASF virus can be achieved by homologous recombination in infected cells.

Interferon-gamma production by African swine fever virus-specific lymphocytes

Peripheral blood mononuclear cells (PBMC) from inbred pigs that were immunized with autologous macrophages infected with the African swine fever (ASF) virus BA71V, a nonvirulent virus isolate, proliferated and produced interleukin-2 in response to homologous and heterologous isolates of the ASF virus. They produced, however, interferon (IFN) only when challenged in vitro with homologous or attenuated isolates of the ASF virus, but not with heterologous or virulent isolates. The IFN was pH 2 labile and was neutralized by specific serum to porcine recombinant IFN gamma.

Genes homologous to ubiquitin-conjugating proteins and eukaryotic transcription factor SII in African swine fever virus

The nucleotide sequence of the 6004-bp EcoRI I fragment of African swine fever virus DNA has been determined. Translation of the sequence revealed eight closely spaced open reading frames (ORFs), three of them reading rightward and five leftward. Northern blot hybridization analysis indicated that ORFs I73R and I78R were transcribed early in infection, whereas ORFs I177L, I196L, and I329L were expressed at late times. Transcripts for ORFs I215L, I226R, and I243L were detected at low levels in early RNA and at higher levels in late RNA. The intergenic regions between genes I73R/I329L and I78R/I215L were characterized by the presence of direct repeats in tandem. Direct repetitions were also found within ORF I196L. The protein encoded by ORF I329L contained a putative cleavable signal peptide and an internal transmembrane domain, and that encoded by ORF I177L had an amino-terminal hydrophobic region with the characteristics of a "start-stop" sequence. ORF I243L encoded a protein similar to the eukaryotic elongation factor SII. The protein encoded by ORF I215L was homologous to the family of ubiquitin-conjugating proteins.

African swine fever virus fatty acid acylated proteins

Labeling experiments with [3H]palmitic and [3H]myristic acids of African swine fever virus-infected Vero cells have shown that 11 proteins induced during infection are covalently bound to myristic acid and that palmitic acid was not attached to viral proteins. The time course of synthesis of the myristylated polypeptides and the requirements of viral DNA replication indicated that the myristylated proteins, with the exception of a 13-kDa protein, belong to the late class of viral proteins. The myristic moiety was not released by hydroxylamine treatment, suggesting that the fatty acid is bound to the polypeptide chain through an amide linkage. The purification of [3H]myristic acid-labeled extracellular virus particles demonstrated that the myristylated 28- and 13-kDa proteins incorporated into the virion.

Highly efficient expression of proteins encoded by recombinant vaccinia virus in lymphocytes

Using a recombinant vaccinia virus (VV) that expresses E. coli beta galactosidase (beta-Gal) to infect lymphocytes, we show that enzymometrically or immunologically detectable beta-Gal expression is less pronounced among T cells than among B cells. VV infection caused growth inhibition of B cells, but barely affected T-cell proliferation in vitro. Moreover, the production of infectious viral particles was less pronounced in T lymphocytes. Kinetic studies revealed that after an initial dose-dependent growth inhibition, T cells continued to proliferate without the doubling time being affected by VV infection. Nonetheless, the T cells do express proteins encoded by recombinant VV, such as beta-Gal, or secrete soluble proteins such as interleukin-4, though at a lower efficiency at the per cell level than B lymphocytes. In conclusion, the physiology of T cells appears to be less perturbed by VV than that of B cells, although the virus is capable of directing expression of recombinant genes to T lymphocytes.

A general method to cleave a known DNA sequence at any site

We describe a new method for obtaining DNA fragments starting at a desired point where there is no recognition sequence for any known restriction endonuclease. A single-stranded DNA containing the fragment of interest is annealed to a synthetic oligonucleotide hybridizing at the 5' end of the required fragment. Then, a partially double-stranded DNA is synthesized using the Klenow fragment of DNA polymerase I in the presence of the four deoxynucleoside triphosphates. The remaining single-stranded regions are removed by digestion with a single-strand nuclease, and the resulting 5' blunt-ended fragment is finally released by digestion with a restriction endonuclease at any site downstream its 3' end. The usefulness of the method was exemplified here by insertion of an epidermal growth factor-like African swine fever virus gene immediately downstream of the ribosome binding site of an expression vector.

The sequences of the ribonucleotide reductase genes from African swine fever virus show considerable homology with those of the orthopoxvirus, vaccinia virus

Two African swine fever virus (ASFV) recombinant plasmids containing large inserts of DNA have been sequenced at random, and translations of the DNA sequence have been compared to libraries of vaccinia virus protein sequences. Among other genes identified by their extensive homology with vaccinia virus genes were the large and small subunits of ribonucleotide reductase. A 5.5-kb fragment from the Malawi (LIL20/1) strain of ASFV was identified as containing the genes for both these subunits. The fragment has been sequenced and the two genes have been found to be in a head-to-head orientation. The sequences are compared to other sequenced ribonucleotide reductase genes, and the evolutionary implications discussed.

Repetitive nucleotide sequencing of a dispensable DNA segment in a clonal population of African swine fever virus

Repetitive nucleotide sequencing of a dispensable genomic segment of a clonal population of African swine fever (ASF) virus has been carried out to estimate the mutant frequency to neutral alleles. Since no mutations have been detected in a total of 54026 nucleotides screened, the maximum mutant frequency is 5.5 x 10(-5) substitutions/nucleotide (95% confidence level). The result renders very unlikely the occurrence of hypermutational events during ASF virus DNA replication, at least within the selected DNA fragment.

Isolation and molecular characterization of the swinepox virus thymidine kinase gene

Swinepox virus (SPV), the only member of the Suipoxvirus genus, shows little antigenic relatedness or DNA homology to members of the other poxvirus genera. A SPV thymidine kinase (TK) gene was detected and mapped to the left end of the HindIII G fragment using degenerate oligonucleotide probes. Cloning and sequencing of a 1.8-kb HindIII-BamHI fragment containing the SPV TK gene revealed an open reading frame (ORF) of 181 amino acids yielding a predicted polypeptide of Mr 20.6 kDa with significant homology to both poxvirus and vertebrate thymidine kinases. Comparison with other TK protein sequences showed that the SPV thymidine kinase was closely related to the TK genes of avipoxviruses (52.0%) and vertebrates (57.1-59.7%). The TK gene from African swine fever virus (ASF) showed little homology (30.5%) to the SPV TK gene suggesting that these two viruses are not closely related though they share many biochemical features and infect a single, common mammalian host (swine). The SPV TK gene, like that of other poxviruses, is transcribed early, and when cloned into a TK- strain of vaccinia converted the virus to a TK+ phenotype. BUdRR mutants of SPV contained frameshift, deletion, and missense mutations in the TK ORF.

Escherichia coli thymidine kinase: nucleotide sequence of the gene and relationships to other thymidine kinases

The thymidine kinase (TK)-encoding gene (tdk) of Escherichia coli is located at min 27 of the E. coli genetic map. Sequence analysis of this region revealed an open reading frame of 205 codons. Identification of this region as the E. coli tdk gene was confirmed by its similarity to other TK-encoding genes. The E. coli amino acid (aa) sequence showed significant similarity to the corresponding TK polypeptides of vertebrates and large DNA viruses, but showed no similarity to known herpes virus TK enzymes. Mapping of highly conserved positions among all sequences indicates the importance of these residues for catalytic activity and may facilitate further functional studies. Using a distance matrix method, the evolutionary relationships among the TK aa sequence of poxviruses, eukaryotes and prokaryotes were analyzed and a potential phylogenetic tree was established.

African swine fever virus attachment protein

Treatment of African swine fever virus particles with nonionic detergents released proteins p35, p17, p14, and p12 from the virion. Of these proteins, only p12 bound to virus-sensitive Vero cells but not to virus-resistant L or IBRS2 cells. The binding of p12 was abolished by whole African swine fever virus and not by similar concentrations of subviral particles that lacked the external proteins. A monoclonal antibody (24BB7) specific for p12 precipitated a protein that, when analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis in the absence of 2-mercaptoethanol, showed a molecular mass of 17 kDa (p17*) instead of 12 kDa as found in the presence of 2-mercaptoethanol. The relationship between these two proteins was confirmed by the conversion of p17* to p12 when the former was isolated from polyacrylamide gels in the absence of 2-mercaptoethanol and subsequently treated with the reducing agent. The supernatant obtained after immunoprecipitation with the p12-specific antibody lacked the virus-binding protein.

Fc receptors do not mediate African swine fever virus replication in macrophages

Titration experiments in swine macrophages have shown that African swine fever virus infectivity was not enhanced in the presence of antiviral antibodies. The early viral protein synthesis and the viral DNA replication in swine macrophages infected with virus-antibody complexes were inhibited in the presence of high doses of uv-inactivated virus, which saturated specific virus receptors, but not when Fc receptors were saturated with antibodies. These results indicate that African swine fever virus does not infect swine macrophages through Fc receptors and that the normal entry pathway through virus receptors is not bypassed by the virus-antibody complexes.

Genetic variation and multigene families in African swine fever virus

The genome of a virulent strain (LIS57) of African swine fever virus differs from that of the Vero-cell-adapted strain (BA71V) in several deletions located in the variable regions. The region which contains the most differences is located 8-20 kb from the left end. The DNA sequence of this region was obtained from LIS57 virus DNA and compared with the overlapping sequences of BA71V virus. This comparison revealed that the changes in the variable regions result in differences in the number of genes which belong to the multigene families 360 and 110. Virus isolate LIS57 contains at least 8 genes of the multigene family 360 and 12 genes of the multigene family 110, instead of the 6 and 5 genes, respectively, found in BA71V virus strain. The position of the deletions indicates that new combinations of multigene family members in African swine fever virus DNA may arise by in-frame recombination between homologous genes. These data indicate that the evolution of the multigene families 360 and 110 in African swine fever virus DNA has involved different processes, including gene duplication, divergence of duplicated genes, and gene deletion.

Interaction of African swine fever virus with macrophages

Morphological data obtained by electron microscopy have shown that African swine fever virus adapted to VERO cells enters swine macrophages, its natural host cell, by a mechanism of receptor-mediated endocytosis. Binding studies with 3H-labeled virus and competition experiments with UV-inactivated virus have shown that the virus entry that leads to a productive infection in swine macrophages is mediated by saturable binding sites on the plasma membrane. The virus also penetrated into rabbit macrophages that do not produce infectious virus and initiated the synthesis of some early viral proteins; however, the viral replication cycle was aborted since viral DNA synthesis did not occur. The interaction of ASF virus particles with rabbit macrophages was mediated by nonsaturable binding sites, suggesting that the lack of specific receptors in these cells may be related to the absence of a productive infection. A similar abortive infection was detected in macrophages from other virus-resistant animal species.

Sequence and evolutionary relationships of African swine fever virus thymidine kinase

The thymidine kinase gene of African swine fever virus was mapped in a 1.4-kb EcoRI-PstI fragment located in the left half of the Eco RI K fragment of African swine fever virus DNA by using degenerate oligonucleotide probes derived from regions of the thymidine kinase sequence conserved in several poxviruses, man, mouse, and chicken. The nucleotide sequence of this region revealed an open reading frame of 196 codons, whose translated amino acid sequence showed significant similarity to the thymidine kinases of vaccinia virus, variola virus, monkeypox virus, shope fibroma virus, fowlpox virus, capripox virus, man, mouse, and chicken. The similarity scores obtained after comparison of known thymidine kinase sequences indicated that the African swine fever virus thymidine kinase is more distantly related than the poxvirus thymidine kinases to their cellular homologs. The evolutionary implications of these findings are discussed.

Recovery from autoimmunity of MRL/lpr mice after infection with an interleukin-2/vaccinia recombinant virus

Interleukin-2 (IL-2) is a T-cell derived molecule implicated in the clonal expansion of antigen-activated T cells and in T-cell development. IL-2 is also implicated in autoimmune disease, although its role is still controversial. Murine systemic lupus erythematosus (SLE) is a good model for human SLE as most of the immunological abnormalities in the human disease also seem to be operative in the mouse. Among SLE mice, the MRL/lpr strain develops early in life autoimmune diseases such as immune complex-mediated glomerulonephritis, arthritis and arteritis. Lymphoid abnormalities associated with those diseases in this strain are thymic atrophy and abnormal proliferation of CD3+ CD4- CD8- 'double-negative' T cells, resulting in massive generalized lymph node enlargement. We have therefore now examined the effects of IL-2 on the disease progression in MRL/lpr mice using live vaccinia recombinant viruses expressing the human IL-2 gene. Vaccinated mice showed prolonged survival, decreased autoantibody and rheumatoid factor titres, marked attenuation of kidney interstitial infiltration and intraglomerular proliferation, as well as clearance of synovial mononuclear infiltrates. Inoculation with the IL-2/vaccinia recombinant virus led, in addition, to drastic reduction of the double-negative T-cell population, improved thymic differentiation and restoration of normal values of mature cells in peripheral lymphoid organs.

Analysis of naturally occurring deletion variants of African swine fever virus: multigene family 110 is not essential for infectivity or virulence in pigs

A comparison of uncloned African swine fever virus isolates with their cloned counterparts revealed the presence of genetic variants in three out of seven uncloned field virus populations tested. Five different virus clones were isolated from the uncloned KIR69 virus stock by limit dilution. Structural analysis of the variants showed that they differed by single deletions of 10-16 kilobases in the region located between 6.8 and 27 kilobases from the left DNA terminus. There was no homology between DNA sequences immediately to the left and right of the deletions indicating that the mechanism generating deletion variants was not homologous recombination. The alignment of the restriction maps of the variants with those of other virus isolates indicated that two of the variants lacked the whole multigene family 110. This was confirmed by hybridization of the viral DNA with a degenerate oligonucleotide probe. Each virus variant replicated to high titer and was virulent in domestic pigs. Therefore, the multigene family 110 was not required for viral replication or virulence in domestic pigs. The virus variants were stable upon repeated passage in swine macrophages, indicating that the generation of variants is not a frequent genetic event in vitro.

Multigene families in African swine fever virus: family 360

A group of cross-hybridizing DNA segments contained within the restriction fragments RK', RL, RJ, and RD' of African swine fever virus DNA were mapped and sequenced. Analysis of these sequences revealed the presence of a family of homologous open reading frames in regions close to the DNA ends. The whole family is composed of six open reading frames with an average length of 360 coding triplets (multigene family 360), four of which are located in the left part of the genome and two of which are in the right terminal EcoRI fragment. In close proximity to the right terminal inverted repeat, we found an additional small open reading frame which was homologous to the 5'-terminal portion of the other open reading frames, suggesting that most of that open reading frame has been deleted. These repeated sequences account for the previously described inverted internal repetitions (J.M. Sogo, J.M. Almendral, A. Talavera, and E. Viñuela, Virology 133:271-275, 1984). Most of the genes of multigene family 360 are transcribed in African swine fever virus-infected cells. A comparison of the predicted protein sequences of family 360 indicated that several residues are conserved, suggesting that an overall structure is maintained for every member of the family. The transcription direction of each open reading frame, as well as the evolutionary relationships among the genes, suggests that the family originated by gene duplication and translocation of sequences between the DNA ends.

Multigene families in African swine fever virus: family 110

The genome of African swine fever virus was screened for the existence of repetitive sequences by hybridization between different cloned restriction fragments covering the viral DNA. Several sets of repeated sequences were detected in fragments located close to the DNA ends. One of these groups of repetitions involved fragments located at both ends of the genome. The remaining groups involved fragments that were located exclusively at the left end. The sequence of a 3.2-kilobase segment spanning from 7.5 to 11 kilobases from the left DNA end, which showed a complex pattern of cross-hybridizations, was determined. Two short and three long blocks of direct repeated sequences were found in this DNA region, which accounted for the hybridization results. The repeated sequences formed a family of five homologous genes with an average length of 116 codons (multigene family 110), one of which had a dimeric structure. Transcripts of the five members of the family were detected both in RNA synthesized in vitro by purified African swine fever virions and in RNA isolated at early times after infection. Comparison of the predicted protein sequences revealed a striking conservation of a cysteine-rich domain in the central part of the proteins. In addition, a highly hydrophobic NH2-terminal sequence present in all the proteins suggests that these proteins are processed through the endoplasmic reticulum.

Mapping and sequence of the gene coding for protein p72, the major capsid protein of African swine fever virus

The gene encoding protein p72, the major structural protein of African swine fever virus and one of the most immunogenic proteins in natural infection has been mapped and sequenced. The gene was mapped by using oligonucleotide probes deduced from amino acid sequences of tryptic peptides obtained from purified protein p72. This allowed the location of the gene in fragment EcoRI B of African swine fever virus DNA. The nucleotide sequence obtained from this region revealed an open reading frame encoding 646 amino acids corresponding to a protein with a calculated molecular weight of 73,096 Da. This open reading frame contains the coding information for all the sequenced tryptic peptides from protein p72. A search at the National Biomedical Research Foundation Data Bank did not reveal any significant homology with other described proteins.

Genetic variation of African swine fever virus: variable regions near the ends of the viral DNA

Restriction endonuclease maps of the variable DNA regions of African swine fever virus field isolates from the Iberian peninsula showed that the changes in length are located in the terminal-inverted repetitions and in unique sequences close to the DNA ends. Analysis of nine clones derived from the spleen of an infected pig revealed the existence of frequent length changes within the inverted terminal repetitions. In each clone, changes occurred symmetrically at both terminal-inverted repetitions, suggesting the existence of a terminal-inverted repetition transposition or correction mechanism. Large deletions in unique sequences were detected more frequently in the region located from 8 to 20 kb from the left DNA end. The analysis of this DNA segment from a virulent African swine fever virus isolated in Lisbon (LIS57) showed that this virus strain contains about 8 kb more DNA sequence than the prototype avirulent virus strain (BA71). Hybridization of the additional sequences from LIS57 virus with DNA from different virus field isolates revealed that this DNA region is highly variable in vivo and that it contains several repeated sequences. DNA sequences present around the deletion end points in the variable regions indicate that the deletion process may take place by both homologous and nonhomologous recombination.

The entry of African swine fever virus into Vero cells

The entry of African swine fever virus into Vero cells has been investigated by both biochemical and morphological techniques. A quantitative electron microscopy analysis of the early steps of the infection has shown that African swine fever virus enters Vero cells by a receptor-mediated endocytosis mechanism. The internalization of virus particles is a temperature- and energy-dependent process, since it did not take place at 4 degrees or in the presence of NaF and 2,4-dinitrophenol. To determine the involvement of acidic intracellular vacuoles in the virus entry pathway we have tested the effect of lysosomotropic agents in the infection. Chloroquine, dansylcadaverine, amantadine, methylamine, and ammonium chloride inhibited African swine fever virus production in Vero cells. Dansylcadaverine and chloroquine did not inhibit virus adsorption and internalization; however, in the presence of these drugs, virus particles were retained in cytoplasmic vacuoles and early viral RNA and protein synthesis were not detected, indicating that these compounds inhibit an early step in the infectious cycle, probably the uncoating of the virus particle.

Gly-Gly-X, a novel consensus sequence for the proteolytic processing of viral and cellular proteins

Three African swine fever virus structural proteins of relative molecular weights 150,000, 37,000, and 34,000 (p150, p37, and p34) are derived from precursors with relative molecular weights 220,000, 60,000, and 39,000 (pp220, pp60, and pp39) by proteolytic cleavage after the second Gly residue in the sequence Gly-Gly-Ala/Gly. A search of the National Biomedical Research Foundation Data Bank revealed that several adenovirus proteins, ubiquitin, and an interferon-induced 15-kDa protein are also derived from precursors that are cleaved at the sequence Gly-Gly-X, where X is often an amino acid residue with a hydrophobic side chain. The sequence Gly-Gly-X together with other physical properties of the protein seems to be a recognition sequence for the processing of a variety of viral and cellular proteins.

Saturable binding sites mediate the entry of African swine fever virus into Vero cells

Binding experiments of 3H-labeled African swine fever virus to susceptible VERO cells have shown the presence of saturable binding sites for African swine fever virus on the plasma membrane. The Scatchard analysis of the binding data at equilibrium indicates the existence of about 10(4) cellular receptor sites per cell with a dissociation constant (Kd) of 70 pM. Virus entry into VERO cells is mediated by a saturable component, since tritiated African swine fever virus saturable binding and uptake were competed by the same amounts of unlabeled virus. Similarly, early viral protein synthesis and virus production were inhibited by concentrations of uv-inactivated virus that competed virus attachment to saturable binding sites, suggesting that specific receptors mediate the entry of African swine fever virus particles that initiate a productive infection in VERO cells. African swine fever virus binding to virus-resistant L cells was not mediated by saturable binding sites. As a result of the nonsaturable interaction the virus was not able to enter L cells and neither early viral protein synthesis nor viral DNA synthesis was detected, indicating that the absence of specific receptors for African swine fever virus is a factor that determines the resistance of L cells to the infection.

Variable and constant regions in African swine fever virus DNA

An analysis of the SalI restriction pattern of African swine fever virus DNA showed that the SalI recognition sites did not change after more than 100 virus passages in porcine macrophages. The virus strain BA71V, obtained from the virus isolate BA71 by adaptation to grow in VERO cells, differed from the nonadapted virus in two deletions with a length of 2.5 and 7 kb located close to the DNA ends. A restriction analysis of several virus clones obtained from a naturally infected pig revealed length heterogeneity in both variable regions. A comparison of SalI restriction maps from 23 African swine fever virus field isolates (8 African, 11 European, and 4 American) has shown that the virus genome consists of a central region with a constant length of about 125 kb and two variable regions located close to the DNA ends with a length of 38-47 kb for the left DNA end, and 13-16 kb for the right DNA end. The total length of ASF virus DNA varied between 178 (BA71) and 189 (MOZ64) kb. The 23 African swine fever virus isolates were classified into five groups, according to the arrangement of the SalI sites in the central region. Four groups contained only African isolates, whereas all the European and American isolates belonged to the same group. This distribution of isolates suggests that all non-African virus field isolates have a common origin.

Effect of interferon-alpha, interferon-gamma and tumour necrosis factor on African swine fever virus replication in porcine monocytes and macrophages

Bovine interferon-alpha I1 (IFN-alpha I1) and porcine interferon-gamma (IFN-gamma) inhibited African swine fever virus replication in both porcine monocytes and alveolar macrophages. The most potent antiviral activity was observed with IFN-gamma-treated alveolar macrophages. The production of both a virulent (CC83) and a non-virulent (BA71) isolate of the virus was inhibited. Bovine tumour necrosis factor alpha did not show antiviral activity in either monocytes or alveolar macrophages. Rather, an increase of African swine fever virus production in tumour necrosis factor alpha-treated monocytes was found. An analysis of viral protein synthesis in IFN-alpha I1- and IFN-gamma-treated alveolar macrophages showed an inhibition of synthesis of some viral proteins. The inhibition of late proteins was very pronounced in IFN-gamma-treated cells, and it was probably a consequence of the inhibition of African swine fever virus DNA polymerase activity.

A protein of molar mass 12 kDa incorporates into the membrane of ASF virus-infected cells

An African swine fever virus-induced protein of molar mass 12 kDa (p12) was studied in virus-infected Vero cells using the monoclonal antibody 18B.B11. Protein p12 is incorporated into the membrane of infected cells about 7 h post-infection and is not present in purified African swine fever virus particles. The synthesis of protein p12 is sensitive to cytosine arabinoside.

Mapping and sequence of the gene encoding protein p37, a major structural protein of African swine fever virus

The gene encoding protein p37, one of the major structural proteins of African swine fever (ASF) virus has been mapped and sequenced. Protein p37 was obtained from purified virions and the first 27 amino acids from its NH2-terminal end were identified by automatic Edman degradation. To map the gene encoding protein p37, a mixture of 20-mer deoxyoligonucleotides based upon a part of this amino acid sequence was hybridized to cloned ASF virus restriction fragments. This allowed localization of the gene in fragment KpnI F/HindIII G1 of the African swine fever virus genome. An analysis of the DNA sequence from this region revealed an open reading frame encoding 418 amino acids. In this sequence, the 27 NH2-terminal amino acids determined by sequence analysis of protein p37 are preceded by a stretch of 132 amino acids residues, indicating that protein p37 is synthesized as a polypeptide of higher molecular weight and then post-translationally processed by cleavage of a Gly-Ala bond. This processing event accounts for the antigenic relationship of protein p37 to a virus-induced, nonstructural protein with a relative molecular weight of 60 kD.

Phosphorylation of African swine fever virus proteins in vitro and in vivo

Highly purified African swine fever virus contains a cyclic AMP-independent protein kinase which phosphorylates endogenous virus proteins with a specific activity of about 0.45 pmol/microgram of virus protein. The major substrates for the virion protein kinase in vitro were the structural proteins p10 and p9. Both proteins were phosphorylated preferentially at serine residues. A possible relationship between protein p10 phosphorylation and RNA synthesis in vitro by the virion-associated RNA polymerase is suggested by the finding that N-alpha-tosyl-L-lysyl-chloromethyl ketone inhibited both phosphorylation of p10 and transcription. Two phosphoproteins, with molecular masses of 35 and 17 kDa, were found in African swine fever virus purified from infected Vero cells labeled with [32P]phosphate. A phosphopolypeptide with a molecular mass of about 35 kDa was found in the cytoplasm of infected Vero cells.