Alteration of gammaretroviral vector integration patterns by insertion of histone and leucine zipper into integrase.

Retroviral vectors show long-term gene expression in gene therapy through the integration of transgenes into the human cell genome. Murine leukemia virus (MLV), a well-studied gammaretrovirus, has been often used as a representative retroviral vector. However, frequent integrations of MLV-based vectors into transcriptional start sites (TSSs) could lead to the activation of oncogenes by enhancer effects of the genetic components within the vectors. Therefore, the MLV integration preference for TSSs limits its wider use in clinical applications. To reduce the integration preference of MLV-based vectors, we attempted to perturb the structure of the viral integrase that plays a key role in determining integration sites. For this goal, we inserted histones and leucine zippers, having DNA-binding property, into internal sites of MLV integrase. This integrase engineering yielded multiple mutant vectors that showed significantly different integration patterns compared with that of wild-type vector. Some mutant vectors did not prefer the key regulatory genomic domains of human cells, TSSs. Moreover, a couple of engineered vectors did not integrate into the genomic sites near the TSSs of oncogenes. Overall, this study suggests that structural perturbation of integrase is a simple way to develop safer MLV-based retroviral vectors for use in clinical applications.

Human cytomegalovirus (HCMV) induces human endogenous retrovirus (HERV) transcription.

BACKGROUND: Emerging evidence suggests that human cytomegalovirus (HCMV) is highly prevalent in tumours of different origin. This virus is implied to have oncogenic and oncomodulatory functions, through its ability to control host gene expression. Human endogenous retroviruses (HERV) are also frequently active in tumours of different origin, and are supposed to contribute as cofactors to cancer development. Due to the high prevalence of HCMV in several different tumours, and its ability to control host cell gene expression, we sought to define whether HCMV may affect HERV transcription. FINDINGS: Infection of 3 established cancer cell lines, 2 primary glioblastoma cells, endothelial cells from 3 donors and monocytes from 4 donors with HCMV (strains VR 1814 or TB40/F) induced reverse transcriptase (RT) activity in all cells tested, but the response varied between donors. Both, gammaretrovirus-related class I elements HERV-T, HERV-W, HERV-F and ERV-9, and betaretrovirus-related class II elements HML-2 - 4 and HML-7 - 8, as well as spuma-virus related class III elements of the HERV-L group were up-regulated in response to HCMV infection in GliNS1 cells. Up-regulation of HERV activity was more pronounced in cells harbouring active HCMV infection, but was also induced by UV-inactivated virus. The effect was only slightly affected by ganciclovir treatment and was not controlled by the IE72 or IE86 HCMV genes. CONCLUSIONS: Within this brief report we show that HCMV infection induces HERV transcriptional activity in different cell types.

Clathrin facilitates the morphogenesis of retrovirus particles.

The morphogenesis of retroviral particles is driven by Gag and GagPol proteins that provide the major structural component and enzymatic activities required for particle assembly and maturation. In addition, a number of cellular proteins are found in retrovirus particles; some of these are important for viral replication, but many lack a known functional role. One such protein is clathrin, which is assumed to be passively incorporated into virions due to its abundance at the plasma membrane. We found that clathrin is not only exceptionally abundant in highly purified HIV-1 particles but is recruited with high specificity. In particular, the HIV-1 Pol protein was absolutely required for clathrin incorporation and point mutations in reverse transcriptase or integrase domains of Pol could abolish incorporation. Clathrin was also specifically incorporated into other retrovirus particles, including members of the lentivirus (simian immunodeficiency virus, SIVmac), gammaretrovirus (murine leukemia virus, MLV) and betaretrovirus (Mason-Pfizer monkey virus, M-PMV) genera. However, unlike HIV-1, these other retroviruses recruited clathrin primarily using peptide motifs in their respective Gag proteins that mimicked motifs found in cellular clathrin adaptors. Perturbation of clathrin incorporation into these retroviruses, via mutagenesis of viral proteins, siRNA based clathrin depletion or adaptor protein (AP180) induced clathrin sequestration, had a range of effects on the accuracy of particle morphogenesis. These effects varied according to which retrovirus was examined, and included Gag and/or Pol protein destabilization, inhibition of particle assembly and reduction in virion infectivity. For each retrovirus examined, clathrin incorporation appeared to be important for optimal replication. These data indicate that a number of retroviruses employ clathrin to facilitate the accurate morphogenesis of infectious particles. We propose a model in which clathrin contributes to the spatial organization of Gag and Pol proteins, and thereby regulates proteolytic processing of virion components during particle assembly.

Quantification of reverse transcriptase in ALS and elimination of a novel retroviral candidate.

BACKGROUND: Retroviral involvement in amyotrophic lateral sclerosis (ALS) has been suspected for several years since the recognition that both murine and human retroviruses can cause ALS-like syndromes. Nonquantitative studies have demonstrated the retroviral enzyme reverse transcriptase (RT) in ALS patients' sera, but the amount and source of RT activity are unknown. We therefore developed a quantitative assay to study RT levels in ALS and examined the possibility that the recently discovered human gammaretrovirus XMRV (xenotropic MuLV-related virus) might be the source of the RT activity. METHODS: A quantitative product-enhanced RT assay was used to measure RT activity levels in serum and CSF. XMRV sequences were sought by PCR analysis of DNA and RNA extracted from blood. RESULTS: Fifty percent of ALS patients' sera contained >6 x 10(-8) RT units/mL as opposed to 7% of control sera (p = 0.008). The levels of RT activity in ALS patients were comparable to the levels observed in patients infected with HIV. RT activity was detected in only 1 of 25 CSF samples tested. XMRV sequences were not found in any of 25 nucleic acid extracts obtained from ALS patients' blood. CONCLUSIONS: These findings further support the concept of retroviral involvement in amyotrophic lateral sclerosis (ALS) and demonstrate that serum is more suitable than CSF for assay of reverse transcriptase (RT) activity in this disease. The levels of serum RT activity detected are comparable to those found in HIV infection. XMRV is not detectable in the blood of ALS patients, and the agent responsible for ALS-associated RT activity therefore remains unidentified.

A murine sarcoma virus-associated protein kinase: interaction with actin and microtubular protein.

A low molecular weight (LMW) protein phosphokinase enzyme that binds to actin has been isolated from murine sarcoma virions; this kinase activity is not present in nontransforming murine leukemia viruses. Sephadex G-75 gel filtration and affinity chromatography on actin-Sepharose conjugates allow a significant level of purification of this enzyme. The enzyme associates with microtubular proteins and inhibits the in vitro polymerization of microtubules. This study represents the first isolation of a sarcoma virus-associated protein that possesses the ability to interact directly with two major components of the cytoskeletal system.

Multiple RNase H activities in mammalian type C retravirus lysates.

Lysates of Moloney murine sarcoma-leukemia virus [M-MSV(MLV)], a virus complex grown in the rat cell line 78A-1, were found to contain three RNase H species separable by polycytidylic acid[poly(C)]-agarose chromatography. RNase H activity (RNase H I) associated with RNA-directed DNA polymerase eluted at 0.23 M KCI from poly(C)-agarose. RNase H II, which eluted from poly(C)-agarose at 0.12 M KCI and was not associated with DNA polymerase activity, was shown to be identical to an RNase H species (designated RNase H II) previously isolated from M-MSV(MLV) by a different procedure (G. F. Gerard and D. P. Grandgenett, J. Virol. 15:785-797, 1975). M-MSV(MLV) RNase H II was established to be a random exohybridase that requires free-chain termini in its hybrid substrate for activity. Lysates of Rickard feline leukemia virus also contained RNase H activity not associated with DNA polymerase activity that eluted from poly(C)-agarose at 0.12 M KCl. A third species of enzyme from M-MSV(MLV) lysates, called RNase H III, did not bind to poly(C)-agarose in 0.06 M KCl. RNase H III was purified from lysates of M-MSV(MLV) and M-MLV (grown in mouse cells) by sequential chromatography on poly(C)-agarose, DEAE-cellulose, phosphocellulose, and polyuridylic acid-Sepharose. Purified RNase H III (i) was free of any associated DNA polymerase activity, (ii) had an apparent molecular weight of 30,000 determined by Sephadex G-100 gel filtration, (iii) had an absolute requirement for Mn2+ (1 mM optimum) for the degradation of [3H](A)n.(dT)n, (iv) was inhibited by the presence of any salt in reaction mixtures, and (v) was endoribonucleolytic in its mode of action as indicated by the size distribution of limited degradation products of [3H](A)n.(dT)n. RNase H III was inhibited by antisera prepared against Rauscher MLV and simian sarcoma virus reverse transcriptase, and the quantity of RNase H III and RNase H I present in lysates of M-MLV were reduced and increased proportionately if virus was lysed in the presence of the protease inhibitor phenylmethylsulfonyl fluoride. These results indicate that RNase H III is a proteolytic cleavage product of DNA polymerase-RNase H. Substantial RNase H activity that did not bind to poly(C)-agarose in 0.06 M KCl was also found in lysates of Harvey MSV(MLV), Rauscher MLV, and Rickard feline leukemia virus, but not in lysates of avian myeloblastosis virus.

[Evidence of protein kinase activity in 2 murine oncornaviruses]. / Mise en évidence d'une activité protéine kinase dans 2 oncornavirus murins.

A protein kinase activity has been detected in two strains of murine Oncornaviruses, MSV/MLV and EFV. This activity phosphorylates not only endogenous viral proteins but also exogenous substrates (histones and phosvitin). The stimulation of enzyme activity by detergents along with the increase of specific activity in viruses treated with trypsin during purification suggest that the enzyme is located in the viral particle.

Characterization of an RNA-directed DNA polymerase found in association with murine intracytoplasmic A-particles.

An RNA-directed DNA polymerase was found to be associated with intracytoplasmic A-particles from DBA/2 mouse leukemia cells. The enzyme activity was detected after disrupting the purified particles with 2 M NaCl-20 mM dithiothreitol. The presence of a divalent cation and all four deoxyribonucleoside triphosphates was essential for this enzyme activity. The enzyme had a clear preference for Mg2+ over Mn2+. Cesium sulfate isopycnic gradient centrifugation of the DNA product synthesized in the actinomycin D-containing reaction revealed the presence of DNA-RNA hybrid. Furthermore, the purified DNA product was found to hybridize with RNA isolated from A-particles. These observations strongly indicate that the endogenous A-particle RNA serves as the template for the DNA polymerase.

Assay for type C virus in mouse sera based on particulate reverse transcriptase activity.

Assay of particulate reverse transcriptase activity in the sera from feral mice naturally infected with type C virus provides a sensitive and rapid procedure for the determination of in vivo virus infection. The results compare well with assays for infectious virus and with complement fixation or competitive radio-immunoassays for the p30 internal antigen of the virus.

Differential response of type C and intracisternal type A particle markers in cells treated with iododeoxyuridine and dexamethasone.

Mouse neuroblastoma cells containing intracisternal type A particles were treated with iododeoxyuridine and dexamethasone to induce the release of type C oncornavirus particles. For 5 days after treatment, antigenic markers and DNA polymerase activities specific to particles of each of the two types were assayed in the cells and in pellets obtained by high-speed centrifugation of the culture fluid. There was a marked release of C-particle antigen (p30) and DNA polymerase activity in extracellular particulate form, reaching a maximum on day 3 after treatment and falling thereafter. In contrast, no extracellular A-particle antigen was detected, and A-particle-specific DNA polymerase activity in the medium pellets did not increase from the original very low level. Electron microscopy confirmed the presence of free type C virus particles, but not intracisternal type A particles, in the culture fluid. Although intracellular levels of C-particle antigen rose 20- to 30-fold per milligram of cell protein, intracellular A-particle antigen and DNA polymerase activity did not vary more than two-fold. The relative rate of A-particle synthesis in the treated cells, as judged by incorporation of radioactive amino acids into the major structural protein (P73), was also unchanged over the period of observation. Thus, the induction of type C virus particle formation in cultured neuroblastoma cells had no detectable effect on the quantity, synthesis rate, or location of intracisternal type A particles.

Intracisternal A particles from FLOPC-1 BALB/c myeloma: presence of high-molecular-weight RNA and RNA-dependent DNA polymerase.

Intracisternal A particles from the FLOPC-1 line of BALB/c myeloma have been shown to contain high-molecular-weight RNA (60 to 70S) that is sensitive to RNase, alkali degradation, and heat but resistant to Pronase treatment. The intracisternal A-particle RNA contains tract of poly (A) approximately 180 nucleotides long. As shown in a reconstitution experiment, by antigenic analysis of A-particle preparation and the SC cytopathogenicity assay, the 70S RNA was not due to contamination by type C virus particles. The FLOPC-1 intracisternal A particles also possess an endogenous RNA-dependent DNA polymerase. The enzyme required Mn2+ or Mg2+, dithiothreitol, detergent, and four deoxyribonucleoside triphosphates for maximum activity. Enzymatic activity was maximally stimuated by poly (rC)-oligo (dG)12-18 and less with poly (rG)-oligo (dC)10 or poly (rA)-oligo (dT)12-18 as compare with synthetic DNA/DNA duplex templates such as poly (dA)-oligo (dT)12-18. The enzyme can utilize the A-particle endogenous RNA as template as shown by analysis of the early and late DNA products of the endogenous reaction by CsSO4 isopycnic gradient centrifuation and hybridization of purified 70S or 35S A-particle RNA with the purified complementary DNA product. Approximately 50% of the A-particle complementary DNA also hybridized with oncornavirus RNA.

Relationships between intracisternal type A and extracellular oncornavirus-like particles produced in murine MOPC-460 myeloma cells.

Oncornavirus-like particles of the "A" (both intracisternal and intracytoplasmic) and "B" or "C" (extracellular) types are produced by murine MOPC-460 myeloma cells. This communication describes a comparative study on tracisternal A and extracellular particles. Both types of particles contain an RNA-dependent DNA polymerase activity, traces of 35S and 70 S RNA in addition to larger amounts of degraded RNA, and proteins of approximately 76,000 and 45, 000 daltons. The 76,000-dalton proteins from intracisternal A and extracellular particles have the same cyanogen bromide peptides. Hybridization kinetic analysis indicates that the RNAs in the two particles are identical or very closely related and share partial homology with Moloney leukemia virus RNA. In contrast, the particles appear to have little or no relationship to murine mammary tumor virus as judged by several different criteria. Electron microscope studies indicate that the extracellular particles arise from the budding of core components through the plasma membrane. These results suggest that the intracisternal A and extracellular oncornavirus-like particles produced by MOPC-460 cells are closely related.

Host range studies of FLOPC-1 murine myeloma C particles.

The host range of the C particle produced by FLOPC-1 myeloma cells, FLOPC-1 murine myeloma-associated virus (FL-MuMAV), was assessed in terms of its ability to productively infect and/or induce new viral antigens in a variety of different cell lines. Production of C particle-like structures by cells exposed to FL-MuMAV) was determined by incorporation of [3H]uridine into particles with a density of 1.16 g/ml and/or measurement of RNA-dependent DNA polymerase activity in concentrated culture medium. to FL-MuMAV was capable of infecting NIH/3T3, normal rat kidney (NRK) cell, BALB/c 3T3, and the A31 clone of BALB/3T3 cells but not rabbit cell line, SIRC. Thus, it is an N, B-tropic murine virus as replication in NRK cells has been shown not to delineate a group of murine viruses with a separate host range (M. M. Lieber, C. J. Sherr, and G. J. Todero, 1974). Further neoantigens, reactive with anti-FL-MuMAV serum, were detected on infected cells. Production of the MuMAV-like particle and MuMAV-associated cell antigens in infected NIH/3T3 and NRK cells persisted for three subcultures. The limited production could not be explained by the lack of an RNA-dependent DNA polymerase or high-molecular-weight RNA as the particles possessed both of these properties. The particles produced by infected NIH/3T3 or NRK cells were antigenically and physicochemically similar to FL-MuMAV and not K-MuLV. The MuMAV-like particles produced by infected NIH/3T3 were capable of limited replication in NIH/3T3 and and BALB/3T3 cells, whereas NRK-MuMAV replicated for a limited period in NIH/3T3, NRK, and SIRC cells; i.e., they had a different host range than FL-MuMAV. The particles produced by infected BALB/3T3 and A31 cells had the same host range as FL-MuMAV. In certain situations, isotopically labeled particles with a density of 1.16 g/ml were produced which appeared to lack RNA-dependent DNA polymerase.

Characterization of non-producer human cells induced by Kirsten sarcoma virus.

Non-producer (NP) human cells induced by the Kirsten sarcoma virus were characterized. These morphologically altered NP cells produced neither infectious virus nor complement-fixing antigens of the murine sarcoma-leukemia virus complex. The NP cells did not release RNA-dependent DNA polymerase and type-C virus particles with a density of approximately 1.15 g/ml in sucrose gradients by 3H-uridine labelling. The NP cells produced tumors when transplanted subcutaneously into athymic nude mice. The tumor cells re-established in culture resembled the orginal NP cells, were confirmed as human cells by karyological analysis and were also found to be "non-producer". The sarcoma virus genome in NP cells could be rescued not only by co-cultivation with "helper virus"-releasing cells but also by superinfection with helper type-C viruses. Murine (Rauscher, Ki-MuLV, AT-124 and two other xenotropic viruses), feline, RD-114 and Simian (woolly monkey and baboon) type-C viruses possessed the ability to rescue the sarcoma genome from NP cells but not AKR leukemia virus. In addition, the feline leukemia virus titer obtained by the rescuing technique in NP cells was the same as those obtained in feline embryo and NP cells by CF induction assay.

Murine intracisternal type A particles: a biochemical characterization.

Intracisternal A particle preparations from a murine neuroblastoma cell line (N18) and from a mineral oil-induced murine plasmacytoma (MOPC-104E) contain both an endogenous RNA-dependent DNA polymerase activity and high molecular-weight polyadenylic acid (poly[A])-containing RNA. The DNA polymerase activity is stimulated by oligo(dG)-poly(C) and oligo(dT)-poly(A) and to a lesser extent by oligo(dT)-poly(dA), in agreement with previous reports. The high-molecular-weight RNA is predominantly 35S and contains a poly(A) tract of approximately 220 nucleotides as judged by polyacrylamide gel electrophoresis. Small amounts of 70S RNA are also present. This RNA preparation contains RNA homologous to RNA from type-C particles, as judged by molecular hybridization experiments. However, since this RNA derives only in part from A-particles and in part from other cellular RNA, hybridization of A-particle endogenously synthesized DNA or reverse transcripts of A-particle RNA to purified type C viral 70S RNA may more accurately reflect the relationship of A-particle RNA to RNA from C-particles. None of these DNA transcripts hybridizes significantly to C-particle 70S RNA, although MOPC and N18 DNA transcripts share significant homology. Our interpretation of these results is that murine intracisternal A particles are not closely related genetically to the tested murine type C viruses, although an alternate possibility is that all the A-particle DNA transcripts are copied from only a small part of the genome, which is unrelated to C-particle RNA.

A comparison of the effects of daunomycin and adriamycin on various DNA polymerases.

The effects of the anthracycline antiboties, daunomycin and adriamycin, on the DNA-directed activities of DNA polymerases from murine sarcoma virus, rat liver (high-molecular-weight species), Escherichia coli, and Micrococcus luteus were determined. Under all conditions tested, these compounds had greater inhibitory effect against the viral polymerase than against cellular polymerase. The inhibition of murine sarcoma virus DNA polymerase by daunomycin was competitive with respect to DNA. For viral DNA polymerase it was concluded that the inhibition was predominatly caused by the interaction of duanomycin with the primer-template DNA. Also, an appreciable reversal of the daunomycin-induced inhibition of this polymerase by an increase in Mg-2+ concentration is consistent with the conclusion derived by competition experiments. In contrast, the inhibition of both rat liver and M. luteus DNA polymerases was essentially noncompetitive with DNA. Also, bacterial enzymes wer e less sensitive to inhibition by these drugs than the virion polymerase. The strong and preferential inhibiton of viral DNA polymerase is discussed in relation to a differential sensitivity of normal as compared to tumor cells observed in some cell lines.