Mechanism of interferon action. Effect of double-stranded RNA and the 5'-O-monophosphate form of 2',5'-oligoadenylate on the inhibition of reovirus mRNA translation in vitro

The effect of reovirus double-stranded RNA (dsRNA) and 5'-O-monophosphate form of 2',5'-oligoadenylate (pA(2'p5'A)2) on the translation and degradation of reovirus messenger RNA and on protein phosphorylation was examined in extracts prepared from interferon-treated mouse L fibroblasts. The following results were obtained. 1) The enhanced degradation of reovirus [3H]mRNA observed in the presence of either dsRNA or the 5'-O-triphosphate form of 2',5'-oligoadenylate (pppA(2'p5'A)3) was completely blocked by pA(2'p5'A)2. 2) The dsRNA-dependent phosphorylation of protein P1 and the alpha subunit of eukaryotic initiation factor (eIF-2) depended in a similar manner upon the concentration of dsRNA and was optimal at low dsRNA concentrations (0.1 to 1 microgram/ml). However, high concentrations of dsRNA (greater than 100 micrograms/ml) drastically reduced the phosphorylation of both P1 and eIF-2 alpha. Neither P1 nor eIF-2 alpha phosphorylation was affected by either pA(2'p5'A)2 or pppA(2'p5'A)3. 3) The translation of reovirus mRNA in vitro was inhibited by the addition of either low concentrations of dsRNA or pppA(2'p5'A)3. Whereas pA(2'p5'A)2 completely reversed the pppA(2'p5'A)3-mediated inhibition of translation, the inhibition mediated by low concentrations of dsRNA was only partially reversed by pA(2'p5'A)2. Under conditions where the pppA-(2'p5'A)3mediated degradation of reovirus mRNA was blocked, the translation of reovirus mRNA was still inhibited by low but not by high concentrations of dsRNA in a manner that correlated with the activation of P1 and eIF-2 alpha phosphorylation. These results suggest that the pppA(2'p5'A)n-dependent ribonuclease is not required and that protein phosphorylation may indeed be sufficient for the dsRNA-dependent inhibition of reovirus mRNA translation in cell-free systems derived from interferon-treated mouse fibroblasts.     

Synthesis of (2'-5')oligoadenylate and activation of an endoribonuclease in interferon-treated HeLa cells infected with reovirus.

Treatment with interferon protected HeLa cells from infection with reovirus. This virus apparently activated an antiviral mechanism that was detected by the presence of (2'-5')oligoadenylate [(2'-5')An] in intact cells. The (2'-5')An was previously shown to activate an endoribonuclease, RNase L. We measured (2'-5')An by a sensitive competition-binding assay in cells infected at different multiplicities and for different lengths of time. Nanomolar concentrations of (2'-5')An were detected in cells infected at a multiplicity of greater than 5 after 2 h of infection, the time at which the infecting virions were uncoated. The level of (2'-5')An increased up to 6 h postinfection but declined afterward. To establish whether viral mRNAs were cleaved by RNase L, we analyzed the RNA extracted from infected cells by a highly specific hybridization assay on Northern blots. Full-sized reovirus mRNAs were detected in control infected cells, but not in interferon-treated infected cells, at 6 h postinfection. At this time, a nuclease activity could be detected in these cells by demonstration of cleavage of rRNA, degradation of cellular mRNA, and polysome breakdown in the presence of emetine. Since this inhibitor freezes ribosomes, cleavage of mRNA between ribosomes could only be accounted for by an endonuclease, presumably RNase L.     

Impairment of reovirus mRNA 'cap' methylation in interferon-treated mouse L929 cells.

Reovirus mRNAs synthesized in vitro by the virionassociated enzyme have a 5' 'cap 1' structure (m7G(5')ppp(5')GmpCp...). However, about one third to one half of the reovirus mRNAs formed in mouse L929 cells have a 5' 'cap 2' structure (m7G(5')ppp(5')GmpCmp...) and the rest have a 5' 'cap 1' structure. The finding that virus mRNA 'cap' methylation is impaired in extracts of interferon-treated cells prompted us to study the effect of interferon on virus mRNA 'cap' methylation in vivo. Using labeling with [3H]-guanosine and dual labeling with [3H]methionine and [14C]uridine we compared the 5' structures of reovirus mRNAs accumulating between 5 and 11 h after infection in: L929 cells treated with 390 to 2600 U/ml of a partially purified mouse interferon preparation and untreated L929 cells. The treatment resulted in a 70 to 98% decrease in the 24 h virus yield and in a 50 to 55% decrease in the label accumulated in virus mRNAs. The 'capping' of virus mRNAs and the methylation of their 5' terminal and adjacent G residues were not diminished in interferon-treated cells. However, the percent of 'cap 2' termini was 36 to 47% lower in virus mRNAs from interferon-treated cells than in virus mRNAs from control cells. The interferon treatment did not result in the appearance of additional methylated nucleotides in the virus mRNAs.     

Maintenance of protein synthesis in spite of mRNA breakdown in interferon-treated HeLa cells infected with reovirus.

Interferon induces the synthesis of an enzyme which synthesizes 2',5'-oligoadenylate [2',5'-oligo(A)] when activated by double-stranded RNA. The 2',5'-oligo(A) in turn activates an endonuclease (RNase L). Concentrations of 2',5'-oligo(A) sufficient to activate RNase L are formed in interferon-treated HeLa cells infected with reovirus, and a large fraction of cellular mRNA is degraded (T. W. Nilsen, P. A. Maroney, and C. Baglioni, J. Virol. 42:1039-1045, 1982). We report here that in spite of this mRNA degradation, protein synthesis was not significantly inhibited in these cells. When mRNA synthesis was inhibited with 5,6-dichloro-1-beta-D-ribofuranosylbenzimidazole, protein synthesis was markedly decreased, as shown by reduced incorporation of labeled amino acids and a decrease in polyribosomes. This suggested that the turnover of mRNA could be compensated for by increased production of mRNA. The relative concentration of specific mRNAs was measured with cloned cDNA probes. The amount of these mRNAs present in control cells was comparable to that in interferon-treated cells infected with reovirus, whereas it was decreased in the latter cells treated with 5,6-dichloro-1-beta-D-ribofuranosylbenzimidazole.     

Impairment of reovirus mRNA methylation in extracts of interferon-treated Ehrilich ascites tumor cells: further characteristics of the phenomenon.

We reported earlier that the methylation of unmethylated reovirus mRNA (reo mRNAU) by the cellular methylating enzymes is impaired in extracts of uninfected, interferon-treated Ehrilich ascites tumor cells (S30INT). We find now that after the methylation of reo mRNAU has stopped in S30INT, the RNA can be reisolated and further methylated in an extract of control cells (S30C). Thus the impairment of methylation in S30INT cannot be due to cleavage or irreversible inactivation of reo mRNAU. Freshly added reo mRNAU can be methylated in S30INT in which the methylation of previously added reo mRNAU has stopped. This indicates that the impairment is due to the depletion of S-adenosylme thionine (the methyl donor), the accumulation of S-adenosylhomocysteine (an inhibitor of methylation), or the irreversible inactivation of reo mRNAU. Freshly added reo mRNAU can be methylated in S30INT in which the methylation of previously added reo mRNAU has stopped. This indicates that the impairment is not due to the depletion of S-adenosylmethionine (the methyl donor), the accumulation of S-adenoxylhomocysteine (an inhibitor of methylation), or the irreversible inactivation of the methylating enzymes. It may be due, however, to the unavailability of reo mRNAU for methylation. The extent of the impairment of reo mRNAU methylation in S30INT decreases with an increasing concentration of reo mRNAU but is not affected by added poly (U), ribosomal RNA, or encephalomyocarditis virus RNA (an mRNA that is probably not capped or methylated at its 5' end). The methylation of reo mRNAU is also impaired in an extract from cells that have not been treated with interferon but with the interferon inducer poly(I) - poly(C). The inhibitor is apparently a macromolecule that is inactivated during incubation. It decreases the methylation at the 7 position of the 5' terminal guanylate residue. In vitro, the rate of reo mRNA synthesis by reovirus cores in the presence of S30INT is the same as in the presence of S30C. However, the methylation of the de novo synthesized reo mRNA by the core-associated methylating enzyme(s) in vitro is inhibited by S30INT but not by S30C. The relevance of these phenomena to the inhibition of reovirus replication in interferon-treated cells remains to be established.     

Inhibition of protein synthesis in reovirus-infected HeLa cells with elevated levels of interferon-induced protein kinase activity

Protein synthesis was inhibited in one line of interferon-treated HeLa cells (line 2) upon infection with reovirus, but not in different HeLa cells (line 1) treated in the same way. The inhibition resulted in polysome runoff, suggesting that it was due to an impairment of peptide chain initiation. Interferon induces the synthesis of a protein kinase, which is activated in cell-free systems by double-stranded RNA and phosphorylates the alpha subunit of eukaryotic initiation factor 2, thus inhibiting the initiation of protein synthesis. Therefore, we measured the level of this protein kinase in extracts prepared from the two HeLa cell lines. Cells of line 2 showed about 3-4 times more protein kinase activity than cells of line 1. The inhibition of protein synthesis upon infection with reovirus was correlated with an increased phosphorylation of the alpha subunit of eukaryotic initiation factor 2 in interferon-treated cells labeled with 32P. The kinase was presumably activated in intact cells by viral double-stranded RNA, but this activation resulted in inhibition of protein synthesis only in cells with elevated levels of the kinase.     

Specific protein phosphorylation in interferon-treated uninfected and virus-infected mouse L929 cells: enhancement by double-stranded RNA.

The enhanced phosphorylation of specific protein(s) observed in extracts from interferon-treated cells (in the presence of ATP and double-stranded [ds] RNA) was also seen in intact mouse L929 cells upon treatment with dsRNA, polyriboinosinic.polyribocytidylic acid [poly(rI.rC)] or reovirus dsRNA, using 32Pi as radiolabel. Labeling of a 65,000-dalton protein(s) with 32P was greatly increased in interferon-treated cells in the presence of added dsRNA, suggesting that the expression in vivo of the kinase activity involved is regulated by dsRNA. This was used as a test system to investigate whether the activity of interferon-induced enzyme(s) is stimulated following virus infection, possibly owing to the accumulation of dsRNA. No obvious increase in 32P-labeling of 65,000-dalton protein(s) was observed upon infection of interferon-treated cells with mengovirus or vesicular stomatitis virus. A basal level of 32P-labeling of the 65,000-dalton protein(s) was detected in interferon-treated cells in the absence of added dsRNA, indicating a basal level of expression of the kinase activity involved. The possible implications of these results are discussed.     

Interferon treatment of Ehrlich ascites tumor cells: effects on exogenous mRNA translation and tRNA inactivation in the cell extract.

We reported earlier that in cell extracts that were prepared from interferon-treated Ehrlich ascites tumor cells and preincubated and passed through Sephadex G-25 (S60INT), the translation of exogenous mRNA (viral and host) was impaired and the impairment could be overcome to a large extent by adding a crude tRNA preparation from Ehrlich ascites tumor cells but not from Escherichia coli. We find now that the rate of inactivation of some tRNA's (especially those specific for leucine, lysine, and serine) but not those of many others is faster in S30INT than in corresponding extracts from control cells. This increased rate of tRNA inactivation may perhaps account for the need for added RNA to overcome at least partially the impairment of translation in S30INT. The relationship of the increased rate of tRNA inactivation to the antiviral effect of interferon is unclear. So far no significant difference has been detected in the amount of tRNA needed to overcome the impairment of encephalomyocarditis virus RNA translation in S30INT between tRNA from interferon-treated cells and tRNA from control cells. Futhermore, no difference was found in the rate of inactivation in S30INT between leucine-specific tRNA's from interferon-treated and from control cells. tRNA's specific for leucine and lysine were not inactivated (unless very slowly) during incubation under out conditions in an extract from interferon-treated (or from control) cells unless the extract had been passed through Sephadex G-25 or dialyzed. The translation fo exogenous mRNA was, however, impaired in an extract from interferon-treated cells that had not been passed through Sephadex G-25. This impairment was apparently not overcome by added tRNA.     

Indiscriminate degradation of RNAs in interferon-treated, vaccinia virus-infected mouse L cells.

In this report we used Northern blot hybridization analysis to characterize the fate of several species of viral RNA transcribed from internal and terminal regions of vaccinia DNA in interferon-treated, infected mouse L cells grown in suspension. All species of viral RNAs were expressed but were reduced in amount. Larger-sized RNAs were reduced more than smaller-sized RNAs. This reduction appears to be related to the activation of the interferon-mediated double-stranded RNA-dependent 2-5A synthetase-endoribonuclease system, as the rRNA cleavage pattern characteristic of this system was observed early in infection and in cell extracts in response to exogenous 2-5A. Thus, in interferon-treated, vaccinia-infected mouse L cells in suspension, there is indiscriminate degradation of viral and cellular RNAs, and this RNA breakdown might play a role in the interferon-mediated inhibition of protein synthesis.     

Interferon Treatment of Ehrlich Ascites Tumor Cells: Effects on Exogenous mRNA Translation and tRNA Inactivation in the Cell Extract

We reported earlier that in cell extracts that were prepared from interferon-treated Ehrlich ascites tumor cells and preincubated and passed through Sephadex G-25 (S30_INT), the translation of exogenous mRNA (viral and host) was impaired and the impairment could be overcome to a large extent by adding a crude tRNA preparation from Ehrlich ascites tumor cells but not from Escherichia coli. We find now that the rate of inactivation of some tRNA's (especially those specific for leucine, lysine, and serine) but not those of many others is faster in S30_INT than in corresponding extracts from control cells. This increased rate of tRNA inactivation may perhaps account for the need for added RNA to overcome at least partially the impairment of translation in S30_INT. The relationship of the increased rate of tRNA inactivation to the antiviral effect of interferon is unclear. So far no significant difference has been detected in the amount of tRNA needed to overcome the impairment of encephalomyocarditis virus RNA translation in S30_INT between tRNA from interferon-treated cells and tRNA from control cells. Furthermore, no difference was found in the rate of inactivation in S30_INT between leucine-specific tRNA's from interferon-treated and from control cells. tRNA's specific for leucine and lysine were not inactivated (unless very slowly) during incubation under our conditions in an extract from interferon-treated (or from control) cells unless the extract had been passed through Sephadex G-25 or dialyzed. The translation of exogenous mRNA was, however, impaired in an extract from interferon-treated cells that had not been passed through Sephadex G-25. This impairment was apparently not overcome by added tRNA.     

Double-stranded RNA causes synthesis of 2',5'-oligo(A) and degradation of messenger RNA in interferon-treated cells

Interferon-treated HeLa cells were incubated with [3H]uridine to label mRNA and were then exposed to the double-stranded RNA poly(inosinic acid).poly(cytidylic acid) (In.Cn). The incubation with In.Cn greatly enhanced the decay of mRNA. When the cells were incubated in this way in the presence of cycloheximide, which blocks ribosome movement along mRNA, extensive polysome degradation was detected in interferon-treated cells. Products of degradation of mRNA were recovered from monosomes which were presumably formed as a result of endonucleolytic breaks of mRNA. This endonucleolytic activity was correlated with the formation of 2',5'-oligo(A) by an enzyme induced by interferon and activated by double-stranded RNA; the 2',5'-oligo(A) was previously shown to activate an endonuclease in cell extracts. The 2',5'-oligo(A) levels in cells were measured by a competition-binding assay. Details of the procedure used are described, including synthesis of highly radioactive (2'-5')pppA3[32P]cytidine 3',5'-diphosphate, separation of 2',5'-oligo(A) binding from degrading activities, and specificity of the assay.     

Ribosomal RNA cleavage, nuclease activation and 2-5A(ppp(A2''p)nA) in interferon-treated cells.

Ribosomal RNA (rRNA) in intact ribosomes is cleaved into discrete products on incubation of reticulocyte lysates or L-cell extracts with ppp(A2'p)3A. Cleavage of rRNA may, therefore, provide a useful assay for 2-5A (ppp)A2'p)nA; n = 2 to 4) or for the presence of a 2-5A-dependent nuclease. The results with reticulocyte lysates differed from those obtained in the L-cell-free system in that (a) a different RNA cleavage pattern was produced (with added L-cell ribosomes) and (b) cleavage was fully activated by the analogue ppp(A2'p)3A3'pCp. As might be expected from the relatively high levels of 2-5A present in interferon-treated, encephalomyocarditis virus (EMC)-infected L-cells, rRNA extracted from these cells was also cleaved. The cleavage pattern observed overlapped with that obtained on incubation of an L-cell-free system with 2-5A. Thus, not only is 2-5A present, but the 2-5A-dependent nuclease also appears to be active, in interferon-treated, EMC-infected L-cells.     

In vitro translation of polyadenylated RNA isolated from control and interferon-treated human promyelocytic HL-60 leukemic cells

Polyadenylated RNA has been isolated from control and interferon-treated HL-60 cells by centrifugation through cesium chloride and oligo(dT)-cellulose column chromatography. The affinity column-purified RNA is poorly translated in the mRNA-dependent rabbit reticulocyte lysates but is an excellent template for in vitro protein synthesis using the wheat germ cell extracts. The discrepancy in the efficiency of HL-60 mRNA utilization in the two commonly used cell-free protein synthesizing systems is attributable to an inhibitory component present in the polyadenylated RNA. This contaminant is most likely double-stranded RNA based on (i) the ability of 2-aminopurine (3-5 mM) or high concentrations of penicillium chrysogenum double-stranded RNA (10-15 micrograms/ml) to overcome the inhibition exerted by the component, and (ii) the ability of the component to promote the enzymatic conversion of ATP into 2-5A by the highly purified rabbit reticulocyte 2-5A synthetase.     

Mechanism of Interferon Action: Inhibition of Viral Messenger Ribonucleic Acid Translation in L-Cell Extracts

Encephalomyocarditis (EMC) virus ribonucleic acid (RNA) stimulated the incorporation of (14)C-amino acids into polypeptides in cell-free systems using preincubated S10 extracts from L cells. Incorporation was linear for over 2 hr. Analysis of the tryptic peptides derived from the polypeptide products formed in response to EMC RNA showed them to be virus specific. The major product, a polypeptide of 140,000 in molecular weight, migrated on sodium dodecyl sulfate-polyacrylamide gels with one of the virus-specific polypeptides present in EMC-infected cells. A minor component of molecular weight about 230,000 may correspond to the product of complete translation of the EMC virus genome. Little or no effect of interferon or vaccinia virus infection was observed in the preincubated, cell-free system. The EMC RNA-stimulated incorporation of (14)C-amino acids into polypeptides was not inhibited in extracts derived from L cells early in virus infection, from interferon-treated cells, or from cells subjected to both treatments. Interferon treatment did appear to have a slight inhibitory effect on chain elongation in this system. However, treatment of cells with highly purified interferon before virus infection caused a decrease of about 80% in the capacity of non-preincubated cell extracts to translate added EMC RNA. This effect did not extend to the translation of polyuridylic acid and could be reversed by preincubation of the extracts at 37 C for 20 min. The inhibition of translation was manifest at interferon concentrations as low as 5IU/ml, and in this respect closely paralleled the inhibition of virus growth. Inactivation of the antiviral activity of the interferon by heating or digestion with trypsin also abolished the effect on cell-free protein synthesis. The EMC-specific polypeptides formed in reduced amounts in extracts of interferon-treated vaccinia-infected cells were smaller than those formed in extracts of untreated, vaccinia-infected cells. Thus, inhibition of initiation or elongation of polypeptides, or both, can be demonstrated in cell-free systems employing non-preincubated extracts from interferon-treated, virus-infected cells. These results indicate that antiviral activity of interferon is directed against the translation of viral messenger RNA.