Characterization of RNA from noninfectious virions produced by sarcoma positive-leukemia negative transformed 3T3 cells

Developing novel lentiviral vectors into clinical products

Gene therapy vectors based on murine retroviruses have now been in clinical trials for over 20 years. During that time, a variety of novel vector pseudotypes were developed in an effort to improve gene transfer. Lentiviral vectors are now in clinical trials and a similar evolution of vector technology is anticipated. These modifications present challenges for those producing large-scale clinical materials. This chapter discusses approaches to process development for novel lentiviral vectors, highlight considerations, and methods to be incorporated into the development schema.

Detection of ecotropic replication-competent retroviruses: comparison of s(+)/l(-) and marker rescue assays

Guidelines for testing gene therapy products for ecotropic replication-competent retrovirus (Eco-RCR) have not been delineated as they have for amphotropic viruses. To evaluate biologic assays that can detect these viruses, we compared an S(+)/L(-) assay and a marker rescue assay designed specifically for Eco-RCR detection. Moloney murine leukemia virus (Mo-MuLV) obtained from the American Type Culture Collection was used as the positive control. For marker rescue, NIH 3T3 cells were transduced with a retroviral vector expressing the neomycin phosphotransferase gene (3T3/Neo). Inoculation and passage of test material in 3T3/Neo cells for 3 weeks (amplification) and subsequent testing in the S(+)/L(-) assay or the marker rescue assay increased the level of sensitivity for virus detection greater than 10-fold compared with direct inoculation of D56 S(+)/L(-) cells. When serial dilutions of Mo-MuLV stock were evaluated, six of six cultures had detectable virus by the S(+)/L(-) and marker rescue assays at dilutions of 10(-5) and 10(-6). At the 10(-7) dilution, five of six assays had detectable virus in both assays. The ability to detect virus-infected cells was also evaluated in a modification that substituted cells for supernatant. Fifteen 3T3/Neo cultures inoculated with 10(6) 293 cells containing 100 or 10 Mo-MuLV/3T3 cells were all positive by marker rescue. For dilution with 1 virus-infected cell per 10(6) 293 cells, 10 of 15 cultures were positive. At the 0.1-cell dilution only 2 of 15 cultures were positive. If we hope to detect one infected cell in a test article, the probability of detecting virus if the assay is performed in triplicate is 96.3%. In summary, after 3 weeks of amplification the S(+)/L(-) and marker rescue assays can detect virus with similar sensitivities. We prefer the marker rescue assay because of the more reliable growth features of NIH 3T3 cells compared with the D56 cell line. For laboratories analyzing clinical materials, this report may prove useful in establishing detection assays for Eco-RCR.

Evidence for type-C retrovirus production by Burkitt's lymphoma-derived cell line

Burkitt's lymphoma cell line, P3HR-I, was found to secrete virions with properties of known type-C RNA tumor viruses. The viral particles had a buoyant density of 1.16 g/ml in sucrose gradients and contained a high-molecular-weight RNA and an RNA-instructed DNA polymerase. The viral polymerase was active in an endogenous reaction requiring the presence of the four deoxyriboside triphosphates and manganese ions, and was sensitive to RNase. The DNA product of the endogenous reaction specifically hybridized to P3HR-I viral 60 to 70S RNA. Electron microscopic examination of ultrathin sections of P3HR-I cells revealed immature, mature and budding virions typical of type-C retroviridae. Nucleic acid hybridization assays showed no sequence homoblastosis virus, murine oncornaviruses, simian sarcoma virus or RD114 virus.

Poly(U)-agarose affinity chromatography: specific, sensitivity selectivity, and affinity of binding

These studies were done to determine four basic intrinsic properties of poly(U)-agarose affinity columns. Specificity of binding studies demonstrated that binding to these columns is highly specific with greater than 90% complementary binding and less than or equal to 3% noncomplementary binding. Sensitivity of binding studies indicated that a minimum sequence of 10 adenylates is required for detectable complementary binding. Selectivity of binding studies revealed that nonsequential adenylates in native RNAs and randomly distributed adenylates in synthetic poly(A)-poly(C) co-polymers did not bind to poly(U)-agarose affinity columns. Whereas, affinity of binding studies demonstrated that A=U complementary base pairing is independent of chain-lengths of greater than or equal to 25 adenylates and dependent of chain-lengths of less than 25 adenylates. Thus the data demonstrates that poly(U)-agarose affinity chromatography is scientifically sound and expedient for the detection and isolation of poly(A)-containing cellular and viral RNAs.

Presence of 5'-terminal cap structures in virus-specific RNA from feline leukemia virus-infected cells

The F-422 line of feline thymus tumor cells, chronically infected with the Rickard strain of feline leukemia virus (R-FeLV), was labeled with 32P, and the total cytoplasmic RNA was isolated. The RNA was centrifuged through sucrose gradients, and R-FeLV virus-specific RNA (vRNA) was located by hybridization of portions of the gradient fractions to R-FeLV complementary DNA. vRNA classes with average sedimentation coefficients of approximately 36S, 28S, 23S, and 15S were identified. Each class of RNA was recovered by hybridized with mercurated R-FeLV complementary DNA, and the hybrids were chromatographed on columns of sulfhydryl-Sepharose to separate them from unhybridized cellular RNA. Although insufficient amount of 36S and 28S vRNA were obtained for further analysis, the 23S and 15S VRNA classes were analyzed to determine the nature of their 5' termini. Each of these vRNA classes was found to contain stoichiometric amounts of cap structures per unit length of RNA, consistent with the presence of one cap per molecule. The structure of the 23S vRNA cap was found to be m7G5'ppp5'GmpAp, whereas that of the 15S vRNA cap was m7G5'ppp5'GmpGp.

Characterization of Gazdar murine sarcoma virus by nucleic acid hybridization and analysis of viral expression in cells

Gazdar murine sarcoma virus (Gz-MSV) and Moloney murine sarcoma virus (M-MSV) are closely related. The complete M-MSV-specific nucleic acid sequences constituted a major portion of Gz-MSV-specific sequences. The MSV-specific sequences in both Gz-MSV and M-MSV genomes shared homology with hamster leukemia virus nucleic acid sequences. Both rat cells (S+L+) and hamster (S+L-) cells expressed two viral proteins of 68,000 and 70,000 daltons. These proteins were immunologically related to p60 purified from m1 virions of M-MSV.

Selective decrease in the rate of cleavage of an intracellular precursor to Rauscher leukemia virus p30 by treatment of infected cells with actinomycin D

The cleavage of an intracellular 67,000- to 70,000-dalton precursor, termed Pr4 to Rauscher leukemia virus (RLV) p30 protein proceeded at a slower rate when virus-producing cells were treated with actinomycin D (AMD). Treatment with AMD also caused a slight accumulation of Pr4 in purified early virus particles produced by a cell line which usually produces virions that contain little Pr4. The cleavage of other intracellular viral precursor polypeptides was not affected by treatment with AMD. Treatment of infected cells with cycloheximide, on the other hand, allowed the cleavage of Pr4 to proceed at the usual rate for a short period of time before further cleavage was drastically slowed or prevented. The cleavage of several other viral precursor polypeptides was also inhibited by treatment with cycloheximide. Different lines of evidence suggest that the mechanism of action of AMD is not due to a possible indirect effect on protein synthesis. Thus, the rate of cleavage of Pr4 was not affected by the length of pretreatment with AMD between 1 to 8 h. In addition, the combined effect of AMD and cycloheximide, at their maximal inhibitory concentrations, was greater than the effect of either drug alone, indicating the involvement of two at least partially different mechanisms in the action of AMD and cycloheximide. Furthermore, AMD did not affect the pulse labeling of viral precursor polypeptides. These results suggest that the interaction with viral RNA, whose production is inhibited by AMD, accelerates the cleavage of Pr4 to p30 during virus assembly. A hypothetical model is presented to illustrate th possible advantages of having a step in virus assembly in which genomic RNA interacts with a precursor to capsid proteins before the cleavage of that precursor.

RNA processing and RNA tumor virus origin and evolution

The results of molecular hybridization experiments with high-molecular-weight RNA isolated from RNA tumor viruses and DNA from normal cells suggest that RNA tumor virus genomes originate from cell genes. Some RNA tumor viruses (here called class 1) appear to have been generated in recent times in that their RNA is closely related in nucleotide sequence to certain cell genes (class 1 genes). A second class of RNA tumor viruses (here called class 2) is more distantly related to genomic information of normal cells. Structural properties of the RNA of RNA tumor viruses lead us to propose that the tumor virus RNA is originated when RNA transcripts of class 1 genes are processed by a mechanism we call "paraprocessing." We postulate that RNA paraprocessing is normally used only at particular times during differentiation and is characterized by the cytoplasmic appearance of high-molecular-weight RNA chains containing terminal polyadenylic acid (200 residues). Paraprocessing of class 1 gene transcripts in committed or differentiated cells is considered to be aberrant in transcription that can lead to the generation of an RNA tumor virus genome. If the paraprocessed class 1 gene transcript codes for a reverse transcriptase, replication of the RNA becomes possible. Transfer of the replicating RNA to a new cell can result in genetic change such that the virus genome mutates, differing from the original progenitor genes. We propose that this genetic change causes class 1 viruses to become class 2. These ideas are applied to evidence concerning the biology of infection of RNA tumor viruses and concerning the involvement of RNA tumor viruses in human cancer. Genetic change can also occur during the origination of an RNA tumor virus genome by repeated reverse transcription and recombination (45) or by genetic alteration of particularly changeable cell genes ("hot spots") (43).

RNA of simian sarcoma-associated virus type 1 produced in human tumor cells

Simian sarcoma-associated virus type 1 propagated in human rhabdomyosarcoma cells exhibited characteristics typical of oncornaviruses but seemed to have several aberrant properties. It had a buoyant density of 1.14 g/cm3, had RNA-dependent DNA polymerase activity, seemed to be labile to high salt concentrations, and contained little 50 to 60S RNA but relatively large amounts of human ribosomal RNA. In addition to 50 to 60S RNA, purified virions contained smaller RNA molecules with sedimentation coefficients of 28 to 30S, 18 TO 20S, and 4 to 10S. Unlike the 50 to 60S RNA species, the smaller virion-associated RNAs lacked polyadenylic acid, and the 28 to 30S RNA had an average base composition similar to that of human ribosomal RNA. Upon heat denaturation, the native 50 to 60S RNA genome yielded polyadenylic acid-containing 28 to 30S subunits that degraded in to 18 to 20S molecules upon further heat treatment. The 50 to 60S viral RNA had a guanine plus cytosine content of 56%.

Characterization of DNA polymerase and RNA associated with A-type particles from murine myeloma cells

The RNA-dependent DNA polymerase present in intracisternal A-type particles from mouse myeloma tumor cells has been studied. This polymerase can use either endogenous A particle RNA or an exogenous synthetic polynucleotide [poly (rA)] as a template. The DNA reaction product is small (4S-10S) and over 90% of it hybridizes to A particle RNA, whereas up to 50% of it hybridizes to murine sarcoma-leukemia virus RNAs. The RNA isolated from purified A particles is generally of low molecular weight (5S-15S) but contains small amount of 70S and 35S components. These results suggest that A-type particles may be related to C-type oncornaviruses.

Polyriboadenylate sequences at the 3'-termini of ribonucleic acid obtained from mammalian leukemia and sarcoma viruses

The location of poly(A) sequences in the RNA of mammalian RNA-tumor viruses was determined by enzymatic analyses. The 56-64S viral genomic RNAs, the 20-40S viral subunit RNAs, and the 4-5S poly(A) sequences excised from these viral RNAs were subjected to either hydrolysis with a 3'-OH specific exoribonuclease from Ehrlich ascites tumor cells or phosphorolysis from the 3'-termini with polynucleotide phosphorylase from Micrococcus luteus. Purified adenosine-labeled poly(A) fragments, excised from genomic viral RNAs by RNase A and T(1) digestion, were hydrolyzed with the 3'-OH specific exoribonuclease for various periods of time. Poly(U) filter binding studies of the residual poly(A) indicated that 97% of the poly(A) fragments were hydrolyzed. Adenosine-labeled genomic and subunit viral RNAs and excised poly(A) fragments were phosphorolyzed from their 3'-termini for various periods of time with polynucleotide phosphorylase. The degree of phosphorolysis was monitored by poly(U) filter binding studies, and CCl(3)COOH insolubility and solubility determinations. There was an initial preferential rate of phosphorolysis of the poly(A) sequences of genomic and subunit viral RNAs as compared to the total adenosine-labeled viral RNAs. The data from these two different enzymatic mechanisms of action indicated conclusively that the poly(A) sequences were located at the 3'-termini of genomic and subunit viral RNAs.

Intracellular viral RNA species in mouse cells nonproductively transformed by the murine sarcoma virus

The size and quantity of virus-specific RNA in five non-virus-producing mouse cells transformed by the Moloney isolate of murine sarcoma virus (MSV) was determined. Hybridization of RNA from transformed cells with the [(3)H]DNA product of the RNA-directed DNA polymerase of the murine sarcoma-leukemia virus was used to detect and quantitate virus-specific RNA. The amount of virus-specific RNA in non-virus-producing cells was less than one-sixth of that found in virus-producing cells. A striking correlation was found between the amount of intracellular virus-specific RNA and the degree of agglutination by conconavalin A previously reported for the four non-virus-producing NIH/3T3 cell lines (Salzberg and Green, 1974). A major RNA subunit sedimenting at 26 to 28S was detected in all five MSV-transformed non-virus-producing cells. This could represent the RNA genome of defective MSV.

Deficiency of 60 to 70S RNA in murine leukemia virus particles assembled in cells treated with actinomycin D

Production of particles with the ultrastructural appearance of C-type virions persisted for at least 6 h in actinomycin D-treated cells infected with murine leukemia virus. This phenomenon occurred despite severe inhibition of viral RNA synthesis. Virus particles present in a 6-h harvest sedimented in sucrose gradients with the buoyant density characteristic of RNA tumor viruses (1.16 g/cm(3)) and exhibited high levels of reverse transcriptase activity in response to the exogenous template polyriboadenylic acid.oligo deoxythymidylic acid in the range of untreated controls. However, RNase-sensitive endogenous activity was only (1/5) the level found in controls. This observation correlated with a marked reduction in infectivity. Kinetic studies on the appearance of labeled RNA in banded virions revealed that within the first hour after addition of actinomycin D, particles contained 60 to 70S RNA and two low-molecular-weight RNA species corresponding to 8 and 4S RNA. After approximately 1 h of incubation with actinomycin D, 60 to 70S RNA could not be detected and 4S RNA was the predominant species. These findings suggest that murine leukemia virus particles assembled in the presence of actinomycin D are deficient in 60 to 70S viral RNA.

Detection of naturally occurring antibodies to RNA-dependent DNA polymerase of murine leukemia virus in kidney eluates of AKR mice

Specific antibodies to the RNA-dependent DNA polymerase (reverse transcriptase) of murine type C viruses have been isolated from the renal glomeruli of both leukemic and nonleukemic AKR mice where they presumably had been deposited as immune complexes. The antibodies were shown to have sedimentation coefficients of 26S to 28S and 5S to 7S on sucrose rate zonal centrifugation. Inactivation with monospecific antisera to various mouse immunoglobulins identified antibodies as being in both immunoglobulin (IGM) and IgG classes. In addition, these antibodies only reacted with the reverse transcriptase from murine and feline type C viruses, but not the polymerase from avian myeloblastosis virus (AMV). Our results provide additional evidence for the lack of immunological tolerance and demonstrate the presence of another immune complex system in AKR kidneys.