Transcription and transport of virus-specific ribonucleic acids in African green monkey kidney cells abortively infected with type 2 adenovirus

A suppressive effect of Sp1 recruitment to the first leader 5' splice site region on L4-22K-mediated activation of the adenovirus major late promoter

Transcription from the adenovirus major late promoter (MLP) requires binding of late phase-specific factors to the so-called DE element located approximately 100 base pairs downstream of the MLP transcriptional start site. The adenovirus L4-22K protein binds to the DE element and stimulates transcription from the MLP via a DE sequence-dependent mechanism. Here we use a transient expression approach to show that L4-22K binds to an additional site downstream of the MLP start site, the so-called R1 region, which includes the major late first leader 5' splice site. Binding of L4-22K to R1 has a suppressive effect on MLP transcription. L4-22K binds to the distal part of R1 and stimulates the recruitment of Sp1 and other cellular factors to a site overlapping the first leader 5' splice site. Binding of Sp1 to the 5' splice site region had an inhibitory effect on L4-22K-activated MLP transcription.

Adenovirus L4-22K stimulates major late transcription by a mechanism requiring the intragenic late-specific transcription factor-binding site

The adenovirus major late promoter (MLP) generates a primary transcript that undergoes a complex pattern of regulated alternative RNA splicing and polyadenylation events. The late-specific activation of the MLP requires binding of two infected-cell specific transcription factor complexes, DEF-A and DEF-B, to the so-called DE sequence located downstream of the MLP start site. Previous studies have shown that DEF-B is a homodimer of the viral IVa2 protein and suggested that DEF-A is a heterodimer of IVa2 and an unknown protein. Two proteins from the adenoviral L4 unit have been suggested as DEF-A candidates. Here we have examined L4-22K and L4-33K for possible DEF-A activity. We show that L4-22K stimulates transcription from the MLP in a DE sequence dependent manner both in vivo and in vitro, and that L4-22K binds to the DE sequence in vitro. Further, the position of the L4-22K DNA binding site in a promoter does not appear to be critical for function. Thus, tethering L4-22K either to a position upstream or downstream of the MLP start site, or upstream of a minimal E1B promoter, resulted in an activation of transcription. We also show that the viral pIX promoter is a natural target, activated by L4-22K. Collectively, our results are compatible with the hypothesis that L4-22K may be the elusive component of DEF-A that partakes in activation of the MLP.

Suppression of the translation defect phenotype specific for a virus-associated RNA-deficient adenovirus mutant in monkey cells by simian virus 40

Human cells infected with adenovirus type 2 (Ad2) or Ad5 require VAI RNA for efficient translation of viral mRNAs at late times after infection. The Ad5 mutant dl-sub720 synthesized neither virus-associated I (VAI) nor VAII RNAs, and infection of human cells with this mutant resulted in reduced virion polypeptide synthesis. Infection of monkey cells with this mutant also resulted in drastic reduction of polypeptide synthesis compared with wild-type (WT) adenovirus infections. Steady-state levels of hexon-specific mRNA were found to be comparable in WT- and mutant-infected monkey cells. The in vitro translation experiments showed that double-mutant- and WT-infected cells contained comparable levels of translatable hexon mRNA (and other adenovirus late mRNAs), suggesting that the severe inhibition of hexon protein synthesis in the VA mutant involves a translation block. Preinfection of monkey cells with simian virus 40 fully restored the efficient translation of this mRNA in the VA mutant infections to the level observed in WT-infected cultures. These results raise the possibility that simian virus 40 may encode or induce factors that suppress the translation block that occurs during adenovirus infections in the absence of the VA RNAs.

Globin synthesis in hybrid cells constructed by transplantation of dormant avian erythrocyte nuclei into enucleated fibroblasts

The polypeptides synthesized by mature embryonic erythrocytes prepared from the peripheral blood of 14- to 15-day-old chicken embryos were analyzed by two-dimensional gel electrophoresis. Fewer than 200 species of polypeptides were detected; the major polypeptides made at this time were identified as the alpha A-, alpha D-, and beta-globin chains. The dormant erythrocyte nuclei were next reactivated to transcriptional competence by transplantation into enucleated mouse or chicken embryo fibroblasts, with frequencies of cytoplast renucleation of about 50 and 90%, respectively. Since large numbers of hybrid cells could be constructed, a biochemical analysis was possible. Electrophoretic analysis of the [35S]methionine-labeled polypeptides made in the hybrid cell types showed that polypeptides having the mobilities of only two (alpha A and alpha D) of the three major adult globin chains were made as major constituents of the hybrid cells. However, analysis of 14C-amino acid-labeled polypeptides revealed that a beta-like polypeptide that lacked methionine was also synthesized in large amounts. This polypeptide was tentatively identified as the early embryonic globin species rho. Globin synthesis was detected as early as 3 h after nuclear transplantation and as late as 18 h, the last time measured in these experiments. It appeared that globin polypeptides made at very early times were translated at least partially from chicken messenger ribonucleic acid introduced into the hybrid cells during fusion, whereas those made at later times were translated primarily from newly synthesized globin messenger ribonucleic acid. The potential usefulness of this hybrid cell system in analyzing mechanisms regulating globin gene expression is discussed.

Globin synthesis in hybrid cells constructed by transplantation of dormant avian erythrocyte nuclei into enucleated fibroblasts

The polypeptides synthesized by mature embryonic erythrocytes prepared from the peripheral blood of 14- to 15-day-old chicken embryos were analyzed by two-dimensional gel electrophoresis. Fewer than 200 species of polypeptides were detected; the major polypeptides made at this time were identified as the alpha A-, alpha D-, and beta-globin chains. The dormant erythrocyte nuclei were next reactivated to transcriptional competence by transplantation into enucleated mouse or chicken embryo fibroblasts, with frequencies of cytoplast renucleation of about 50 and 90%, respectively. Since large numbers of hybrid cells could be constructed, a biochemical analysis was possible. Electrophoretic analysis of the [35S]methionine-labeled polypeptides made in the hybrid cell types showed that polypeptides having the mobilities of only two (alpha A and alpha D) of the three major adult globin chains were made as major constituents of the hybrid cells. However, analysis of 14C-amino acid-labeled polypeptides revealed that a beta-like polypeptide that lacked methionine was also synthesized in large amounts. This polypeptide was tentatively identified as the early embryonic globin species rho. Globin synthesis was detected as early as 3 h after nuclear transplantation and as late as 18 h, the last time measured in these experiments. It appeared that globin polypeptides made at very early times were translated at least partially from chicken messenger ribonucleic acid introduced into the hybrid cells during fusion, whereas those made at later times were translated primarily from newly synthesized globin messenger ribonucleic acid. The potential usefulness of this hybrid cell system in analyzing mechanisms regulating globin gene expression is discussed.

Adenovirus type 2 expresses fiber in monkey-human hybrids and reconstructed cells

Adenovirus type 2 protein expression was measured by indirect immunofluorescence in monkey-human hybrids and in cells reconstructed from monkey and human cell karyoplasts and cytoplasts. Monkey-human hybrid clones infected with adenovirus type 2 expressed fiber protein, whereas infected monkey cells alone did not. Hybrids constructed after the parental monkey cells were infected with adenovirus type 2 demonstrated that fiber synthesis in these cells could be rescued by fusion to uninfected human cells. Thus, human cells contain a dominant factor that acts in trans and overcomes the inability of monkey cells to synthesize fiber. Cells reconstructed from infected human karyoplasts and monkey cytoplasts expressed fiber, whereas cells reconstructed from infected monkey karyoplasts and human cytoplasts did not. These results are consistent with the hypothesis that the block to adenovirus replication in monkey cells involves a nuclear event that prevents the formation of functional mRNA for some late viral proteins including fiber polypeptide. Furthermore, they suggest that the translational apparatus of monkey cells is competent to translate functional fiber mRNA synthesized in human cells.

Subgenomic viral DNA species synthesized in simian cells by human and simian adenoviruses

DNA synthesized after infection of simian tissue culture cells (BSC-1 or CV-1) with human adenovirus type 2 or 5 or with simian adenovirus 7 was characterized. It was demonstrated that as much as 40% of the virus-specific DNA in nuclei of infected monkey cells consists of subgenomic pieces. No subgenomic viral DNA species were detected in the nuclei of human (HeLa) cells infected with these adenovirus types. Restriction analysis showed that these short viral DNA molecules contain normal amounts of the sequences from the ends of the viral genome, whereas internal regions are underrepresented. The production of subgenomic DNAs is not correlated with semipermissive infection. Although adenovirus types 2 and 5 are restricted in monkey cells, these cells are fully permissive for simian adenovirus 7. HR404, an adenovirus type 5 mutant which is not restricted in monkey cells, produced the same percentage of subgenomic DNAs as did its wild type (restricted) parent, and coinfection of monkey cells with adenovirus type 5 DNAs. The array of predominant size classes among the heterogeneously sized short DNAs is serotype specific. Extensive plaque purification and comparison of wild-type adenovirus type 5 with several viral mutants indicated that the distribution of aberrant sizes of DNA is characteristic of the virus and not a result of random replicative errors and then enrichment of particular species.

Simian virus 40 mutants with deletions at the 3' end of the early region are defective in adenovirus helper function

Coinfection of monkey cells with simian virus 40 (SV40) and adenovirus type 2 (Ad2) increased the Ad2 yield 1,000-fold over that obtained by Ad2 infection alone of monkey cells (A. S. Rabson, G. T. O'Conor, I. K. Berezesky, and F. J. Paul, Proc. Soc. Exp. Biol. Med. 116:187-190, 1964). The ability of viable mutants of SV40 that contain deletions at various sites in the viral DNA to enhance Ad2 growth in monkey cells was examined. Only those mutants with deletions near the 3' end of the early region were deficient in providing this helper function. Mutants dl1265, lacking 39 base pairs at map position 0.18, and dl1263, lacking 33 base pairs at map position 0.20 (H. van Heuverswyn, C. Cole, P. Berg, and W. Fiers, J. Virol. 30:936-941, 1979), were approximately 4 and 30% as effective as wild-type SV40, respectively. The extent of enhancement of Ad2 yield depended on the multiplicity of infection by SV40, but not by Ad2 (at a multiplicity of infection of </=50), as well as on the relative times of infection by Ad2 and SV40. Increasing the SV40 multiplicity of infection or infecting cells with SV40 wild type or mutants prior to Ad2 infection increased the Ad2 yield dramatically. The T antigens of wild-type SV40, dl1263, and dl1265 were examined. We attempt to correlate defects in helper function, alterations in the T antigen structure, and the DNA sequence of the mutants as determined by van Heuverswyn et al. (J. Virol. 30:936-941, 1979).

Stabilization of adenovirus nuclear RNA by intercalating drugs

The effect of the intercalating drugs proflavine, ethidium bromide and daunomycin on the rate of degradation of newly synthesized adenovirus RNA was examined. As shown previously for heterogeneous nuclear RNA (HnRNA) and 45 S precursor ribosomal RNA (pre-rRNA), proflavine immediately stabilizes newly synthesized adenovirus RNA, while ethidium bromide stabilizes the RNA after a 30--60 min lag period. In contrast to its effect on HnRNA and pre-rRNA, daunomycin also stabilizes newly synthesized adenovirus RNA. These results are consistent with the hypothesis that the processing of late adenovirus nuclear RNA involves cleavage of base-paired regions by specific cellular nucleases.

Possible role of the 72,000 dalton DNA-binding protein in regulation of adenovirus type 5 early gene expression

Relative abundances of early virus RNA species in the cytoplasm of cells infected with wild-type adenovirus type 5 (WT Ad5) and a temperature-sensitive "early" mutant, H5ts125 (ts125), were compared by hybridization kinetics using separated strands of HindIII restriction endonuclease fragments of Ad5 DNA. 1-beta-D-Arabinofuranosylcytosine (ara-C) was used to limit transcription to early virus genes in cells infected by WT virus. At 40.5 degrees C, a restrictive temperature for ts125, three to seven times as much virus RNA from all four early regions of the genome accumulated in the cytoplasm of cells infected by the mutant as accumulated in cells infected by WT. At 32 degrees C, no such difference in the relative abundances of cytoplasmic virus RNA was observed. The capacity to synthesize a 72,000-dalton (72K) virus polypeptide, presumably the single-stranded DNA-binding protein that is defective in ts125 at restrictive temperatures, was compared in cells infected at 40.5 degrees C in the presence of ara-C with the mutant or WT Ad5. The rate of 72K polypeptide synthesis, measured by sodium dodecyl sulfate-polyacrylamide gradient gel electrophoresis of [35S]methionine-labeled polypeptides and autoradiography, was greater at 15 h after infection in ts125-infected cells than in cells infected by WT. A time course experiment showed that the rate of synthesis of the 72K polypeptide increased continuously in ts125-infected cells during the first 15 h of infection, relative to the rate in WT-infected cells. These data are consistent with the hypothesis that Ad5 early gene expression is modulated by the product of an early gene, the 72K DNA-binding protein.

Isolation of a variant of human adenovirus serotype 2 that multiplies efficiently on monkey cells

A variant of adenovirus serotype 2 (Ad2) that overcomes the block to multiplication of the wild-type Ad2 on monkey cells is described. The variant was selected, after nitrous acid mutagenesis, by sequential passage on monkey cells. This variant forms plaques with similar efficiency on human and monkey cells cells. The kinetics of its growth and its burst size on human and monkey cells are similar. These growth properties are similar to those found for wild-type Ad2 on monkey cells when the block is overcome by coinfection with simian virus 40.

Genome expression and mRNA maturation at late stages of productive adenovirus type 2 infection

RNA from adenovirus 2-infected KB cells was annealed in liquid with RNA in vast excess to viral heavy (l) and light (r) 32P-labeled DNA strands. Hybridization kinetics were analyzed by computer to estimate the number of viral RNA abundance classes, their relative concentrations, and the fraction of each DNA strand from which they originated. Early whole cell RNA extracted 5 h postinfection annealed rapidly to 10 to 15% of l and r strands and then slowly to final values of 60 and 40% of l and r strands. By 9 h postinfection the expression of late genes was apparent and whole cell RNA annealed to 20 and 75% of l and r strands. Whole cell RNA extracted between 12 and 36 h postinfection annealed to 7 to 15% and 75 to 90% of l and r strands. Late nuclear RNA hybridized to 10 and 90% of l and r strands, and late polyribosomal RNA hybridized to 20 and 75% of l and r strands. Based upon kinetic analyses, we estimate that mRNA synthesized exclusively during late stages arises from about 6 to 8% and 45 to 49% of l and r strands. This assumes that the early class I mRNA (in low concentration late) originates from 8 to 10% and 6 to 10% of l and r strands and that early class II mRNA (in high concentration late) is derived from 2% and 8 to 13% of l and r strands. Mixing experiments indicated that early mRNA is a subset of RNA extracted from polyribosomes late after infection and that late nuclear RNA contains sequences complementary to early l strand class I nRNA. RNA-RNA hybrids were isolated from late mRNA containing sequences from 60% of l and r strands, but it is not known when these were synthesized, and therefore whether complementary RNA transcripts are synthesized late after infection, as they are known to be synthesized early. These results demonstrate that portions of the genome are transcribed into RNA sequences that remain confined to the nucleus and are not exported to polyribosomes as mRNA.

Adenovirus transcription. IV. Synthesis of viral-specific RNA in human cells infected with temperature-sensitive mutants of adenovirus 5

Cytoplasmic RNA sequences produced in HeLa cells infected with the adeno-virus 5 temperature-sensitive mutants ts1, ts2, ts9, ts17, ts18, ts19, ts20, ts22, ts49, ts36, and ts125 were characterized by hybridization to DNA probes generated by strand separation of restriction endonuclease fragments of adenovirus 5 DNA. Two ""early'' mutants defective in DNA synthesis, ts125 and ts36, fail to make wild-type levels of all previously reported classes of late RNA at the nonpermissive temperature. At 40.5 degrees C, both ts125 and ts36 synthesize a wild-type complement of early cytoplasmic RNA 16 h after infection. Under these conditions, no ""late'' cytoplasmic RNA sequences were observed. Similarly, nuclear RNA present in these cells resembled early cytoplasmic RNA rather than late nuclear RNA. All the late adenovirus 5 temperature-sensitive mutants synthesized normal wild-type levels of late cytoplasmic RNA at the nonpermissive temperature, except ts2, which appears to overproduce certain cytoplasmic species.

Viral RNA synthesis and levels of DNA-dependent RNA polymerases during replication of adenovirus 2

The rates of RNA synthesis in cultured human KB cells infected by adenovirus 2 were estimated by measuring the endogenous RNA polymerase activities in isolated nuclei. The fungal toxin alpha-amanitin was used to determine the relative and absolute levels of RNA polymerases I, II, and III in nuclei isolated during the course of infection. Whereas the level of endogenous RNA polymerase I activity in nuclei from infected cells remained constant relative to the level in nuclei from mock-infected cells, the endogenous RNA polymerase II and III activities each increased about 10-fold. These increases in endogenous RNA polymerase activities were accompanied by concomitant increases in the rates of synthesis in isolated nuclei of viral mRNA precursor, which was quantitated by electrophoretic analysis on polyacrylamide gels. The cellular RNA polymerase levels were measured with exogenous templates after solubilization and chromatographic resolution of the enzymes on DEAE-Sephadex, using procedures in which no losses of activity were apparent. In contrast to the endogenous RNA polymerase activities in isolated nuclei, the cellular levels of the solubilized class I, II, and III RNA polymerases remained constant throughout the course of the infection. Furthermore, no differences were detected in the chromatographic properties of the RNA polymerases obtained from infected or control mock-infected cells. These observations suggest that the increases in endogenous RNA polymerase activities in isolated nuclei are not due to variations in the cellular concentrations of the enzymes. Instead, it is likely that the increased endogenous enzyme activities result from either the large amounts of viral DNA template available as a consequence of viral replication of from replication or from functional modifications of the RNA polymerases or from a combination of these effects.

Block to multiplication of adenovirus serotype 2 in monkey cells

The block to adenovirus 2 (Ad2) multiplication in monkey cells can be overcome by coinfection with simian virus 40 (SV40). To identify this block we have compared the synthesis of Ad2 proteins in monkey cells infected with Ad2 alone (unenhanced) or with Ad2 plus SV40 (enhanced). Synthesis of viral proteins in enhanced cells was virtually identical to that found for permissive infection of human cells by Ad2 alone. In contrast, the unenhanced cells were strikingly deficient in the production of the IV (fiber) and 11.5K proteins whereas the synthesis of 100K and IVa2 was normal. Synthesis of a number of other proteins such as II, V, and P-VII was partially reduced. A similar specific reduction in synthesis of these proteins was found when their messages were assayed by cell-free translation. This result suggests that the block to Ad2 protein synthesis is at the RNA level rather than with the translational machinery of monkey cells. Analysis of the complexity and the concentration of Ak2-specific RNAs, using hybridization of restriction endonuclease fragments of the Ad2 genome to increasing concentrations of RNA, shows that although all species of late Ad2 mRNA are present, the concentration of several species is reduced sevenfold or more in unenhanced monkey cells as compared with enhanced cells. These species come from regions of the genome known to encode the deficient proteins. A model for the failure of adenovirus to multiply in monkey cells, based on abnormal processing of specific adenovirus messages, is presented.

Analysis of early adenovirus 2 RNA using Eco R-R1 viral DNA fragments

Adenovirus 2 RNA synthesized early in productive infection was analyzed by RNA-DNA hybridization. Hybridization experiments were performed with adenovirus 2 DNA and wit, the six adenovirus 2 DNA fragments generated 0y digestion with the restriction endonuclease Eco R.R1. Duplex formation between RNA and -32P-labeled viral DNA was assayed by S(1) nuclease digestion. RNA from the cytoplasm annealed 12 percent of the total viral DNA and the following percentage of each of the R.R1 fragments: 6 percent of R1-A, 24 percent of R1-B, 0 percent of R1-F, 40 percent of R1-D, 13 percent of R1-E, and 22 percent of R1-C. The early cytoplasmic RNA is composed of two sequence classes: class I, present in greatly reduced quantities at late times in infection (18 h), and class II, which remains at high concentrations at 18 h. In hybridization-inhibition experiments, hybridization of class II RNA is inhibited by late cytoplasmic RNA, whereas hybridization of class I RNA is not blocked by late cytoplasmic RNA (J. J. Lucas and H. S. Ginsberg, 1971; E.A. Craig and H. J. Raskas, 1971). To determine the location of class I and II sequences on the genome, membrane bound DNA fragments were used in hybridization-inhibition experiments. These studies demonstrated that the early cytoplasmic transcripts of R1-D belong to class II, whereas R1-C transcripts are class I sequences. The cytoplasmic RNAs transcribed from fragments A and B contain both class I and class II sequences. Analysis of cytoplasmic RNA fractionated by size demonstrated that the class I sequences include a 19 S RNA transcribed from R1-B and class II sequences include a 20S RNA derived from R1-D. Nuclear RNA purified from cultures early in infection was annealed with -32P-labeled R1 fragments. With all six fragments the nuclear RNA annealed as much or more of the DNA than did cytoplasmic RNA. Eco R1-F annealed at least 25 percent with early nuclear RNA, whereas no sequences homologous to R1-F were detected in early cytoplasmic RNA. When cultures were labeled from 2 to 6 h after infection, at least 5 percent of the -3H-labeled early nuclear viral RNA annealed to Eco R1-F. Some of these nuclear transcripts from R1-F appear to be covalently linked to sequences transcribed from a contiguous region of the genome (Eco R1-B). 8.4 percent of the RNA selected by hybridization of R1-F reannealed to R1-B, whereas no more than 1.5 percent reannealed to R1 fragments A, D, E, or C.

Post-transcriptional restriction of human adenovirus expression in monkey cells

Infection of the continuous simian cell lines CV-1 and BSC-1 with human adenovirus type 2 (Ad2) is abortive. However, the restriction of Ad2 reproduction in these cells can be overcome by increasing the Ad2 infectious dose or by coinfection with simian virus 40. Vero, another established simian cell line free of detectable endogenous simian virus 40 DNA, is not restricted in its ability to promote Ad2 growth even at low input multiplicities of Ad2 and in the absence of SV40 helper. The amount of structural Ad2 proteins in total cell extracts of enhanced BSC-1 cells is at least two orders of magnitude higher than that of unenhanced cells. In contrast, comparable quantities of Ad2 mRNA specifying these proteins are found in both the enhanced and the unenhanced cell. Both sets of mRNA can be translated in a cell-free system with equal efficiency.

Sequence relationships between adenovirus 2 early RNA and viral RNA size classes synthesized at 18 hours after infection

Synthesis of cytoplasmic viral RNA was studied during infection of cultured human (KB) cells with adenovirus 2. At 6 h, before viral DNA synthesis began 5% of the poly(A)-containing RNA hybridized to viral DNA; by 12 h and at later times more than 80% was virus specified. At 18 h after infection, four major size classes of cytoplasmic viral RNA were identified among the poly(A)-containing molecules. These size classes migrated as 27S, 24S, 19S, and 12 to 15S in polyacrylamide gels. The three larger size classes could also be identified in denaturing formamide gels. Hybridization of the 27S, 24S, and 19S viral RNAs was not inhibited by RNA harvested from cells at early times in infection. Therefore, these three major RNAs must code for late viral proteins. Hybridization of the 12 to 15S RNA was partially inhibited by RNA from cultures harvested at early times, suggesting that in this size class some of the RNA labeled at 18 h codes for early viral proteins.

Synthesis of complementary RNA sequences during productive adenovirus infection

Liquid RNA.DNA hybridization with separated strands of adenovirus type 2 DNA revealed that late nuclear RNA can hybridize to about 85% of the 1-strand and 10-15% of the h-strand, whereas late cytoplasmic RNA hybridizes to 65-70% and 25% of the l- and h-strand, respectively. With separated strands from the six EcoRI fragments of adenovirus type 2 DNA as probes, it was shown that late nuclear RNA hydridizes to 85-90% of the l-strand from all six EcoRI fragments. Since late cytoplasmic RNA hybridizes to 40-50% of the h-strand from both fragments EcoRI-B and EcoRI-C, complementary viral RNA sequences are synthesized during adenovirus infection. Complementarity between nuclear and cytoplasmic RNA could also be demonstrated by showing that late cytoplasmic RNA which had been preincubated with late nuclear RNA hybridized to a smaller fraction of the h-strand of fragment EcoRI-C than without preincubation. Double-stranded RNA which contains sequences that correspond to at least 60% of the viral genome was isolated from infected cells. However, less than 2% of the newly synthesized late RNA became double-stranded after incubation under annealing conditions, which suggests that RNA derived from one of the strands is present at a low concentration. Accordingly, it was shown that nearly all viral cytoplasmic RNA which is synthesized late after infection is derived from the l-strand.

Localization of adenovirus 2 messenger RNA's to segments of the viral genome defined by endonuclease R-R1

Adenovirus 2 mRNAs synthesized in productive infection were assigned to specific regions of the genome by hybridization to unique fragments of viral DNA. Radioactive viral RNA synthesized early or late in infection was first fractionated by polyacrylamide gel electrophoresis. Eluted RNAs were then tested for complementary hybrid formation with each of the six fragments of adenovirus 2 DNA generated by cleavage with endonuclease R.R1. Early RNA species migrating as 13S, 19S, and 20S RNAs were identified as transcription products of fragments A, B, and D, respectively. In addition to the 13S RNA transcribed from A fragment DNA, 13S RNA also hybridized to the D and E fragment DNAs; 11S RNA annealed to both A and B fragments. The RNA that hybridized to fragment C DNA was heterogeneous in size. Viral RNA synthesized late in infection included 27S, 24S, 19S, and 11S size classes, all of which annealed to A fragment DNA. Additional RNA migrating as 24S hybridized to E and C fragment DNA, and 23S RNA annealed to F fragment DNA. The results of the hybridizations of size fractionated RNAs with viral DNA fragments enabled formation of a partial map of viral mRNAs with respect to the adenovirus 2 genome. Some of the viral RNAs may represent transcripts which overlap R1 cleavage sites, because in at least three instances hybridization revealed viral RNAs which have the same electrophoretic mobility and anneal to fragments that are contiguous on the genome.

Evidence for the presence of RNA in the purified virions of vaccinia virus

Vaccinia virus, strain WR, was propagated in HeLa cells, L mouse fibroblats, or primary chicken embryo fibroblasts in the presence of [5- (3)H]uridine. Carefully purified virions were found to contain significant amounts of labeled trichloroacetic acid-precipitable material which was rendered acid soluble when digested with pancreatic RNase or hydrolyzed in alkali. Controlled degradation of virions with Nonidet P-40 and 2-mercaptoethanol demonstrated that 65 to 80% of the [5- (3)H]uridine-labeled molecules resided in the viral core. When the total nucleic acids were extracted from viral cores prepared from virions propagated in HeLa cells, 30 to 50% of the total incorporated [5- (3)H]uridine was found in RNA; in L mouse fibroblasts, 40 to 50%; in primary chicken embryo fibroblasts, 50 to 60%. The RNA molecules do not appear to be covalently linked to the viral DNA genome but sediment in sodium dodecyl sulfate-sucrose gradients as 8 to 10S species relative to ribosomal RNA.