A crystal structure of the catalytic core domain of an avian sarcoma and leukemia virus integrase suggests an alternate dimeric assembly.

Integrase (IN) is an important therapeutic target in the search for anti-Human Immunodeficiency Virus (HIV) inhibitors. This enzyme is composed of three domains and is hard to crystallize in its full form. First structural results on IN were obtained on the catalytic core domain (CCD) of the avian Rous and Sarcoma Virus strain Schmidt-Ruppin A (RSV-A) and on the CCD of HIV-1 IN. A ribonuclease-H like motif was revealed as well as a dimeric interface stabilized by two pairs of α-helices (α1/α5, α5/α1). These structural features have been validated in other structures of IN CCDs. We have determined the crystal structure of the Rous-associated virus type-1 (RAV-1) IN CCD to 1.8 Å resolution. RAV-1 IN shows a standard activity for integration and its CCD differs in sequence from that of RSV-A by a single accessible residue in position 182 (substitution A182T). Surprisingly, the CCD of RAV-1 IN associates itself with an unexpected dimeric interface characterized by three pairs of α-helices (α3/α5, α1/α1, α5/α3). A182 is not involved in this novel interface, which results from a rigid body rearrangement of the protein at its α1, α3, α5 surface. A new basic groove that is suitable for single-stranded nucleic acid binding is observed at the surface of the dimer. We have subsequently determined the structure of the mutant A182T of RAV-1 IN CCD and obtained a RSV-A IN CCD-like structure with two pairs of buried α-helices at the interface. Our results suggest that the CCD of avian INs can dimerize in more than one state. Such flexibility can further explain the multifunctionality of retroviral INs, which beside integration of dsDNA are implicated in different steps of the retroviral cycle in presence of viral ssRNA.

A novel self-deleting retroviral vector carrying an additional sequence recognized by the viral integrase (IN).

During retroviral integration, the viral integrase recognizes the attachment (att) sequence (formed by juxtaposition of two LTRs ends) as the substrate of integration. We have developed a self-deleting Avian Leukosis and Sarcoma Viruses (ALSVs)-based retroviral vector carrying an additional copy of the att sequence, between neo and puro genes. We observed that: (i) the resulting NP3Catt vector was produced at neo and puro titers respectively smaller and higher than that of the parental vector devoid of the att sequence; (ii) 61% of NP3Catt proviruses were flanked by LTRs; most of them were deleted of internal sequences, probably during the reverse transcription step; (iii) 31% of clones were deleted of the whole 5' part of their genome and were flanked, in 5', by the additional att sequence and, in 3', by an LTR. Integration of these last proviruses was often imprecise with respect to the viral ends. At total, 77% of proviruses had lost the packaging signal and were not mobilizable by a replication-competent virus and 92% had lost the selectable gene in a single round of replication. Although still to improve, the att vector could be considered as an interesting new safe retroviral vector for gene transfer experiments.

Mutational analyses of the core domain of Avian Leukemia and Sarcoma Viruses integrase: critical residues for concerted integration and multimerization.

During replicative cycle of retroviruses, the reverse-transcribed viral DNA is integrated into the cell DNA by the viral integrase (IN) enzyme. The central core domain of IN contains the catalytic site of the enzyme and is involved in binding viral ends and cell DNA as well as dimerization. We previously performed single amino acid substitutions in the core domain of an Avian Leukemia and Sarcoma Virus (ALSV) IN [Arch. Virol. 147 (2002) 1761]. Here, we modeled the resulting IN mutants and analyzed the ability of these mutants to mediate concerted DNA integration in an in vitro assay, and to form dimers by protein-protein cross-linking and size exclusion chromatography. The N197C mutation resulted in the inability of the mutant to perform concerted integration that was concomitant with a loss of IN dimerization. Surprisingly, mutations Q102G and A106V at the dimer interface resulted in mutants with higher efficiencies than the wild-type IN in performing two-ended concerted integration of viral DNA ends. The G139D and A195V mutants had a trend to perform one-ended DNA integration of viral ends instead of two-ended integration. More drastically, the I88L and L135G mutants preferentially mediated nonconcerted DNA integration although the proteins form dimers. Therefore, these mutations may alter the formation of IN complexes of higher molecular size than a dimer that would be required for concerted integration. This study points to the important role of core domain residues in the concerted integration of viral DNA ends as well as in the oligomerization of the enzyme.

Mutations in the C-terminal domain of ALSV (Avian Leukemia and Sarcoma Viruses) integrase alter the concerted DNA integration process in vitro.

Integrase (IN) is the retroviral enzyme responsible for the integration of the DNA copy of the retroviral genome into the host cell DNA. The C-terminal domain of IN is involved in DNA binding and enzyme multimerization. We previously performed single amino acid substitutions in the C-terminal domain of the avian leukemia and sarcoma viruses (ALSV) IN. Here, we modelled these IN mutants and analysed their ability to mediate concerted DNA integration (in an in vitro assay) as well as to form dimers (by size exclusion chromatography and protein-protein cross-linking). Mutations of residues located at the dimer interface (V239, L240, Y246, V257 and K266) have the greatest effects on the activity of the IN. Among them: (a) the L240A mutation resulted in a decrease of integration efficiency that was concomitant with a decrease of IN dimerization; (b) the V239A, V249A and K266A mutants preferentially mediated non-concerted DNA integration rather than concerted DNA integration although they were found as dimers. Other mutations (V260E and Y246W/DeltaC25) highlight the role of the C-terminal domain in the general folding of the enzyme and, hence, on its activity. This study points to the important role of residues at the IN C-terminal domain in the folding and dimerization of the enzyme as well as in the concerted DNA integration of viral DNA ends.

Analysis of conserved and non-conserved amino acids critical for ALSV (Avian leukemia and sarcoma viruses) integrase functions in vitro.

Retroviral integrase (IN) is the viral enzyme responsible for the integration of viral DNA into host cellular DNA. In vitro, recombinant IN protein is able to catalyze the 3'-processing, strand transfer and disintegration activities. In order to analyze the importance of specific residues of ALSV (Avian leukemia and sarcoma viruses) IN protein, we introduced 31 amino acid substitutions either in residues previously shown by others to be involved in IN oligomerization or in selected conserved and non-conserved residues through the IN sequence. We tested, in vitro, the three catalytic activities of these mutants as well as their capacity to bind DNA. We found that (i) 88% of the substitutions occurring on well-conserved residues have an effect on IN activities (ii) two mutants (S85T in the central catalytic domain and N197C in the C-terminal domain) present a reduced efficiency of DNA binding compared to the wild type protein. Moreover, all mutations made on the dimer interface of C-terminal domain present reduced activities, suggesting an important role of this part of the protein. Finally, for some mutations, we observed differences between the ALSV and HIV (Human immunodeficiency virus) IN corresponding residues.

Substrate sequence selection by retroviral integrase.

Integration of retrovirus DNA is a specific process catalyzed by the integrase protein acting to join the viral substrate DNA (att) sequences of about 10 bases at the ends of the long terminal repeat (LTR) to various sites in the host target cell DNA. Although the interaction is sequence specific, the att sequences of different retroviruses are largely unrelated to one another and usually differ between the two ends of the viral DNA. To define substrate sequence specificity, we designed an "in vitro evolution" scheme to select an optimal substrate sequence by competitive integration in vitro from a large pool of partially randomized substrates. Integrated substrates are enriched by PCR amplification and then regenerated and subjected to subsequent cycles of selection and enrichment. Using this approach, we obtained the optimal substrate sequence of 5'-ACGACAACA-3' for avian sarcoma-leukosis virus (ASLV) and 5'-AACA(A/C)AGCA-3' for human immunodeficiency virus type 1, which differed from those found at both ends of the viral DNA. Clonal analysis of the integration products showed that ASLV integrase can use a wide variety of substrate sequences in vitro, although the consensus sequence was identical to the selected sequence. By a competition assay, the selected nucleotide at position 4 improved the in vitro integration efficiency over that of the wild-type sequence. Viral mutants bearing the optimal sequence replicated at wild-type levels, with the exception of some mutations disrupting the U5 RNA secondary structure important for reverse transcription, which were significantly impaired. Thus, maximizing the efficiency of integration may not be of major importance for efficient retrovirus replication.

Point mutation in avian sarcoma leukaemia virus protease which increases its activity but impairs infectious virus production.

The retrovirus protease (PR), an aspartic PR, is composed of two identical subunits, each containing a conserved tripeptide sequence present at the active site of the enzyme. Asp-Ser-Gly is found in avian sarcoma leukaemia viruses (ASLV) and Asp-Thr-Gly in mammalian oncoretroviruses. We have mutated the conserved sequence at the active site of ASLV PR by converting the Ser and Gly residues to Thr and Ala, respectively. Replacement of Gly with Ala yielded an ASLV PR devoid of proteolytic activity. The Ser to Thr conversion did not alter the substrate specificity of the enzyme. Both wild-type and mutated PRs correctly cleaved viral precursors expressed in bacterial cells, as well as synthetic peptides homologous to ASLV and human immunodeficiency virus type 1 cleavage sites. Bacterially produced ASLV PR with Thr instead of Ser had increased enzymatic activity, as shown by hydrolysis of synthetic peptides. However, this mutation reduced the production of reverse transcriptase-containing particles and infectious virus following transfection of permissive cells with virus DNA.

Proteolytic processing of particle-associated retroviral polyproteins by homologous and heterologous viral proteinases.

Retroviral proteinase(PR)-catalyzed cleavage of the viral Gag and Gag-Pol polyproteins within the nascent virus particle is required for productive viral infection. Kinetic characterization and specificity analyses have been reported for several retroviral PR using oligopeptide substrates. In this study, we performed a comparative analysis of PR from avian, bovine, simian and human retroviruses using polyproteins of human immunodeficiency virus (HIV) type 1 or avian leukosis virus as substrates. Polyproteins were derived from immature virus-like particles purified from culture medium of transfected or recombinant baculovirus-infected cells. Specific cleavage to the correct size intermediate and end products occurred in the presence of detergent and homologous PR. HIV-1 PR cleaved its Gag precursor to completion at a concentration of approximately 25 nM but cleaved the Gag-Pol precursor incompletely even at fourfold higher PR concentration. In contrast to the requirement for high ionic strength for peptide cleavage reported previously, we found that Gag protein cleavage by HIV-1 PR proceeded best at low ionic strength, for both of the protein substrates tested. HIV-2 PR was approximately sixfold less active than HIV-1 PR. PR from avian myeloblastosis-associated virus (MAV) yielded efficient cleavage of the HIV-1 polyprotein only at concentrations above 1 microM. Both enzymes were stimulated by high salt and their cleavage products were identical or very similar to those of HIV-1 PR. A mutant of MAV PR engineered to cleave HIV-1 peptide substrates did not cleave the HIV-1 polyprotein at a concentration of 0.4 microM. The PR of Mason Pfizer monkey virus cleaved this polyprotein very poorly, whereas PR of bovine leukemia virus cleaved it, albeit at different sites.

The v-sea oncogene of avian erythroblastosis retrovirus S13: another member of the protein-tyrosine kinase gene family.

The cloning and sequencing of the oncogene of the avian erythroblastosis virus S13 is described. The oncogene, termed v-sea, was found to be another member of the protein-tyrosine kinase gene family. The oncogene was fused in frame with the retrovirus S13 envelope gene, thus generating a fusion protein with a structure resembling that of a growth factor receptor. Sequence comparisons revealed that the v-sea gene was most closely related to the insulin receptor family of protein-tyrosine kinases, the greatest similarity being with the human MET oncogene.

Mutation of a protein kinase C phosphorylation site in the erbB protein of avian erythroblastosis virus.

Tumor promoter-stimulated phosphorylation of threonine 98 of the erbB protein of avian erythroblastosis virus (AEV) correlates with inhibition of erbB-dependent mitogenesis. To more clearly define the role of phosphorylation of this residue in regulation of the activity of the erbB protein, we have constructed erbB mutations which encode alanine (Ala-98), tyrosine (Tyr-98), or serine (Ser-98) at position 98. The biosynthesis and stability of the three mutant proteins were similar to those of the wild-type erbB protein, and all three retained the ability to transform chicken embryo fibroblasts. Treatment of transformed CEF with 12-tetradecanoylphorbol-13-acetate (TPA) stimulated incorporation of 32Pi into wild-type and mutant erbB proteins and resulted in a slight decrease in the electrophoretic mobilities of all the erbB proteins. Tryptic maps of erbB phosphopeptides showed no endogenous or TPA-stimulated phosphorylation of alanine 98 or tyrosine 98 in cells transformed by the Ala-98 and Tyr-98 mutants. Analysis of tryptic phosphopeptides by high-pressure liquid chromatography revealed that TPA treatment of cells stimulated phosphorylation of other sites of the erbB protein in addition to threonine 98. A high endogenous level of phosphorylation of serine 98 of the Ser-98 mutant protein was found, and TPA treatment of cells did not result in further phosphorylation of this residue. Cells transformed by wild-type and mutant AEV were equally sensitive to TPA-dependent inhibition of growth in soft agar and TPA-dependent inhibition of [3H]thymidine incorporation. TPA treatment inhibited tyrosine phosphorylation to a similar extent in cells transformed by wild-type or Ala-98 AEV. These data indicate that phosphorylation of threonine 98 of the erbB protein is not responsible for TPA-dependent inhibition of growth of AEV-transformed cells or TPA-induced inhibition of erbB-dependent tyrosine phosphorylation. TPA-stimulated phosphorylation of the erbB protein at other sites may mediate these effects. The data also show that subtle changes in a phosphorylation site (i.e., changing threonine to serine) can drastically alter recognition by protein kinases.

Synthetic peptides as substrates and inhibitors of a retroviral protease.

Processing of the gag and pol gene precursor proteins of retroviruses is essential for infectivity and is directed by a viral protease that is itself included in one of these precursors. We demonstrate here that small synthetic peptides can be used as both model substrates and inhibitors to investigate the specificity and molecular parameters of the reaction. The results indicate that a peptide that extends five amino acids but not three amino acids in both directions from a known cleavage site is accurately hydrolyzed by the protease of avian sarcoma-leukosis virus. Substitutions of the amino acids to either side of the peptide bond to be cleaved affect the ability of the peptide (as well as a larger precursor protein) to serve as a substrate. The specificity is more stringent for the amino acid that will become the carboxyl end after cleavage. Some substitutions produced peptides that were not cleaved but could act as inhibitors of cleavage of a susceptible peptide. Thus, small model substrates may be used to explore both the binding and catalytic properties of these important proteases.

Avian sarcoma and leukosis virus pol-endonuclease recognition of the tandem long terminal repeat junction: minimum site required for cleavage is also required for viral growth.

Integration of retroviral DNA is a site-specific reaction involving an endonuclease encoded by the viral pol gene (pol-endo). In vitro the pol-endo from avian sarcoma and leukosis viruses (ASLVs) cleaves both DNA strands near the U5-U3 junction of tandem long terminal repeats (LTR-LTR junction) in single-stranded and replicative form (RF)-I substrates. We have reported previously that the sequences that are required for cleavage of single-stranded substrates by the alpha beta form of the pol-endo differ for the plus and minus strands (G. Duyk, M. Longiaru, D. Cobrinik, R. Kowal, P. deHaseth, A. M. Skalka, and J. Leis, J. Virol. 56:589-599, 1985). This is not the case with RF-I substrates, in which a maximum of 22 base pairs of U5 and 8 base pairs of U3 were required for alpha beta pol-endo cleavage in each strand. Insertion of a palindromic octanucleotide (CATCGATG) at the LTR-LTR junction abolished cleavage in RF-I but not in single-stranded DNA substrates. Deletion of the four nucleotides (TTAA) at the junction prevented cleavage in the plus strand of RF-I DNA, but did not affect cleavage of single-stranded DNA. Furthermore, the alpha beta form of ASLV pol-endo did not recognize heterologous LTR-LTR junction sequences from the reticuloendotheliosis virus or Moloney murine leukemia virus in either substrate form, despite their sequence and structural similarities to the ASLV junction. These results support a role for a sequence-specific interaction between the ASLV pol-endo and the LTR-LTR junction domains that are required for cleavage. By using the infectious Rous sarcoma virus clone pATV8-K, we introduced a set of deletions into the U5 region that would be incorporated into the LTR-LTR junction on viral replication. In the unintegrated provirus, the deletions started 43 base pairs from the LTR-LTR junction and extended various lengths toward the junction. Results of transfection studies with these clones indicated that the U5 sequences that are required for virus production in vivo correspond to those that are required for cleavage of RF-I DNA in vitro.

Circles with two tandem long terminal repeats are specifically cleaved by pol gene-associated endonuclease from avian sarcoma and leukosis viruses: nucleotide sequences required for site-specific cleavage.

The avian retroviral pol gene-encoded DNA endonuclease (pol-endo) has been shown to selectively cleave the viral long terminal repeat sequences (LTRs) in single-stranded DNA substrates in a region known to be joined to host DNA during integration (G. Duyk, J. Leis, M. Longiaru, and A.M. Skalka, Proc. Natl. Acad. Sci. USA 80:6745-6749, 1983). The preferred sites of cleavage were mapped to the unique U5/U3 junctions found only in covalently closed circular DNA molecules containing two tandem LTRs. The cuts occurred three nucleotides 5' to the axis of symmetry of the 12-of-15-base-pair nearly perfect inverted repeat which marks the LTR junction. Experiments with double-stranded supercoiled DNA substrates revealed a similar specificity for nicking. Also, the endonuclease associated with the pol cleavage product, pp32, has the same specificity as the alpha beta form. The limits of sequence required for site-selective cleavage near the U5/U3 junction were established with single-stranded DNA substrates. A domain no larger than 44 base pairs allowed site-selective cleavage in each strand in vitro. Recognition of either strand appeared to be independent of the other, and in each case, the critical sequence was asymmetrically distributed with respect to the U5/U3 junction. The predominant contribution was from the U5 domain; this is consistent with its conservation in the LTR sequences of a number of avian sarcoma and leukosis viruses.

Protein phosphorylation at tyrosine is induced by the v-erbB gene product in vivo and in vitro.

The v-erbB gene product of avian erythroblastosis virus (AEV) has extensive homology with the receptor for epidermal growth factor (EGF). We report here that chicken embryo fibroblasts (CEF) transformed by AEV show enhanced tyrosine phosphorylation of a number of cellular polypeptides, including the 36 kd protein, which is phosphorylated in avian sarcoma virus-transformed fibroblasts, and the 42 kd protein, which is phosphorylated in mitogen-stimulated cells. CEF infected by AEV mutants with deletions in v-erbA showed enhanced tyrosine phosphorylation, whereas CEF infected by mutants with deletions in v-erbB did not. When membranes from AEV-transformed cells were incubated with gamma-32P-ATP, both the v-erbB gene product and the 36 kd cellular protein became phosphorylated at tyrosine. These results indicate that the v-erbB protein induces tyrosine phosphorylation in vivo and in vitro, and suggest that, like the EGF receptor, it possesses tyrosine-specific protein kinase activity.

Antibodies against a synthetic peptide as a probe for the kinase activity of the avian EGF receptor and v-erbB protein.

The transforming protein v-erbB of avian erythroblastosis virus (AEV) displays extensive sequence homology with the presumptive protein-tyrosine kinase domain of the human EGF receptor and with the src protein-tyrosine kinase family of oncogenes. However, no kinase activity has previously been demonstrated for the v-erbB protein. Here antibodies generated against a synthetic peptide from the C terminus of human EGF receptor are shown to immunoprecipitate the EGF receptor from human and avian cells, as well as the v-erbB proteins from AEV-transformed cells that become phosphorylated on tyrosine residues upon the addition of gamma-32P-ATP. The immunoprecipitates are also able to phosphorylate exogenous tyrosine-containing substrates. Hence, it is likely that both avian EGF receptor and v-erbB proteins are protein tyrosine-specific protein kinases. Since the kinase activity of v-erbB protein cannot be regulated by EGF, it is proposed that the tyrosine protein kinase function of v-erbB may be constitutively activated.

Monoclonal antibodies specific for the avian sarcoma virus transforming protein pp60src.

Monoclonal antibodies to avian sarcoma/leukosis virus-coded transforming, structural, and polymerase proteins have been developed. In this first communication, a monoclonal antibody to pp60src will be characterized. The antibody can bind to pp60src in radioimmunoassays, precipitates this antigen in protein kinase tests, and demonstrates in immunofluorescence investigations that pp60src has different intracellular localizations depending on whether RSV-transformed avian or mammalian cells are studied.

Transformation by avian sarcoma viruses leads to phosphorylation of multiple cellular proteins on tyrosine residues.

Phosphoamino acid compositions were determined for 10 size classes of cellular proteins, separated by electrophoresis through one-dimensional sodium dodecyl sulfate-polyacrylamide gels. Phosphotyrosine-containing proteins were observed in uninfected chicken embryo fibroblasts in every size class analyzed, ranging from approximately 20,000 to greater than 200,000 daltons. Transformation of chicken embryo fibroblasts by Rous sarcoma virus or PRC II avian sarcoma virus led to increases in phosphorylation of proteins at tyrosine residues in all of these size classes. A large fraction of the phosphotyrosine-containing protein molecules observed in Rous sarcoma virus-transformed cells was larger than 100,000 daltons with a second broad peak in the 35,000- to 60,000-dalton range. This study suggests that there are a number of substrates of viral or cellular tyrosine-specific protein kinases, which have not yet been identified by other methods.