Nature of inhibitor of cell-free protein synthesis formed in response to interferon and double-stranded RNA

69-kDa and 100-kDa isoforms of interferon-induced (2'-5')oligoadenylate synthetase exhibit differential catalytic parameters

The (2'-5')oligoadenylate synthetase represents a family of interferon-induced proteins which when activated by double-stranded (ds)RNA polymerizes ATP into 2'-5'-linked oligomers with the general formula pppA(2'p5'A)n, where n > 1, which for convenience are referred to as 2-5A. We studied here the influence of pH, dsRNA concentration and time on oligomeric composition of 2-5A synthesized by purified 69-kDa and 100-kDa isoforms of (2'-5')oligo(adenylate) synthetase. In optimal conditions for activity, the 69-kDa form synthesized higher oligomers of 2-5A molecules whereas the 100 kDa form synthesized preferentially dimeric molecules, which are known not to be functional for the activation of RNase L. This difference does not reflect a differential affinity of the enzymes for the preformed 2-5A dimer, which is found to be a very poor substrate for both enzymes. This latter strongly suggests that the mechanism of elongation is more likely processive. Moreover, we show that both isoforms have efficient nucleotidyl-transferase activity and provide evidence that, in optimized conditions, GTP can be used alone as substrate by these enzymes to generate pppG2'p5'G. Our results clearly demonstrate that the 69-kDa and 100-kDa forms of (2'-5')oligoadenylate synthetase manifest various differential catalytic activities, and favor the hypothesis that these enzymes might have other functions in the cell besides those in the 2-5A system.

Interferon-induced and double-stranded RNA-activated enzymes: a specific protein kinase and 2',5'-oligoadenylate synthetases

Treatment of cells with interferon (IFN) results in the induction of two double-stranded RNA (dsRNA)-activated enzymes: a specific protein kinase and 2'-5' linked oligoadenylate [pppA(2'p5'A)n referred to as 2-5A] synthetases. The protein kinase, when activated by dsRNA, becomes autophosphorylated and catalyzes and phosphorylation of the protein synthesis initiation factor, eIF2. The 2-5A synthetases, when activated by dsRNA, form 2-5A molecules capable of activating a latent endoribonuclease that degrades RNA. By inhibiting initiation of protein synthesis or by degrading of RNA, these enzymes play key roles in two independent pathways that regulate overall protein synthesis.

The binding of the 69- and 100-kD forms of 2',5'-oligoadenylate synthetase to different polynucleotides

Three major forms of 2',5' oligoadenylate (2-5A) synthetase are induced in interferon (IFN)-treated human cells: 40-46, 69, and 100 kD. Here we studied the binding and activation of the 69- and 100-kD forms to single-stranded (ss) or double-stranded (ds) RNAs. The 69- and 100-kD form enzymes purified by immunoaffinity chromatography, were shown to be activated by synthetic dsRNAs poly(I).poly(C) or poly(A).poly(U) whereas ssRNAs poly(I), poly(C), poly(A), poly(U), and poly(G) had no effect. Both enzymes were also partially purified by binding to dsRNA or ssRNA-Sepharose. The synthetases bound to dsRNA Sepharose were partially activated but required the addition of soluble poly(I).poly(C) for maximal activity. The synthetases bound to ssRNA-Sepharose manifested no activity but became activated in the presence of soluble dsRNA, poly(I).poly(C). However, activation of such ssRNA-bound enzymes by dsRNA did not result in their dissociation from the ssRNA-Sepharose. These results indicate the presence of different polynucleotide binding sites on the 69- and 100-kD forms of 2-5A synthetase: a specific dsRNA binding site essential for activation and another polynucleotide binding site or sites which, although not specific, might be important for the optimal conformation of these proteins in cells.

High concentrations of 2-5A, the interferon intracellular mediator, in the blood of children with acute viral infections

We measured the concentration of 2-5A (2',5'-oligoadenylate), an intracellular mediator of the antiviral action of interferon, in the blood of children with acute viral and bacterial infectious diseases. 2-5A concentration was found to be elevated in several children with viral diseases. This elevation seemed transient and was not specific for viral infections. We provide arguments for the use of 2-5A as a marker of the evolution of diagnosed viral diseases.

A nucleotide phosphodiesterase with preference for 2',5'-phosphodiester bonds from Ehrlich ascites carcinoma

We have previously reported that many tumor cell lines express a 5'-nucleotide phosphodiesterase (phosphodiesterase I, EC 3.1.4.1) with properties clearly distinguishable from enzymes of normal tissues (Biochim. Biophys. Acta (1988) 966, 99-106). Such an enzyme with 5'-nucleotide phosphodiesterase activity was purified from Ehrlich ascites carcinoma by measuring the cleavage of thymidine 5'-monophosphate p-nitrophenyl ester (TMP-NP). The enzyme is a soluble protein, has a pH optimum of 7.5, and the molecular mass estimated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis is 67 kDa. The enzyme does not hydrolyze other chromogenic substrates for phosphodiesterases, nor pyrophosphate bond of various nucleotides which are cleaved by 5'-nucleotide phosphodiesterases of normal tissues. But, it hydrolyzes dinucleotides to form 5'-phosphates, and is more active on 2',5'- than on 3',5'-phosphodiester bonds. These results indicate that the TMP-NP splitting enzyme in Ehrlich ascites carcinoma cells is a 2',5'-phosphodiesterase.

Preparation and characterization of polyclonal antibodies specific for the 69 and 100 k-dalton forms of human 2-5A synthetase

Recently, the existence of 40-, 46-, 69- and 100- kDa forms of 2',5'-oligoadenylate (2-5A) synthetase have been established in interferon-treated human cells. Using monoclonal antibodies specific for 69- and 100- kDa forms of 2-5A synthetase, we purified these proteins by immunoaffinity chromatography and raised murine polyclonal antibodies. All immunized mice developed antibodies (anti-69 or anti-100 kDa form) which were characterized by their capacity to immunoprecipitate [35S] cysteine labeled proteins from interferon-treated cells or identify these proteins by electrophoretic transfer immunoblot analysis of extracts from control and interferon-treated cells. The 69 and 100 kDa 2-5A synthetases were induced in different types of human cells, such as Daudi, BJAB, HeLa and differentiated HL-60 cells. These enzymes were not detectable nor induced in MRC5 and undifferentiated HL-60 cells.

Induction of indoleamine 2,3-dioxygenase: a mechanism of the antitumor activity of interferon gamma

The antiproliferative effects of interferon alpha (IFN-alpha) and interferon gamma (IFN-gamma) were found to be cell-dependent. Among the human cell lines examined, IFN-gamma had a greater antiproliferative effect against cell lines that exhibited induction of indoleamine 2,3-dioxygenase, such as the KB oral carcinoma or WiDr colon adenocarcinoma, than against those that lacked the enzyme activity, such as the SW480 colon adenocarcinoma or NCI-H128 small-cell lung carcinoma. Induction of this dioxygenase showed a clear temporal relationship with increased metabolism of L-tryptophan and the depletion of this amino acid in the culture medium. While 70-80% of L-tryptophan remained in the medium of IFN-alpha- or vehicle-treated cells, virtually all of this amino acid was depleted in the medium of the IFN-gamma-treated group following 2-3 days of culture. Supplementing the growth medium with additional L-tryptophan reversed the antiproliferative effect of IFN-gamma against KB cells in a dose- and time-dependent manner. The antiproliferative effects of IFN-alpha and IFN-gamma on SW480 and NCI-H128 cells, which are independent of the dioxygenase activity, and the inability of added L-tryptophan to reverse the effects of IFN-gamma in WiDr cells suggest multiple mechanisms of action of the IFNs. The data show that the antiproliferative effect of IFN-gamma through induction of indoleamine 2,3-dioxygenase, with a consequent L-tryptophan deprivation, is an effective means of regulating cell growth.

Induction of an antiviral state by interferon in the absence of elevated levels of 2,5-oligo(A) synthetase and eIF-2 kinase

A series of clones has been derived from an interferon-resistant murine cell line, Ltk- aprt-, and their antiviral properties have been characterized. In the parental Ltk- aprt- line interferon is unable to establish antiviral properties or to increase the levels of 2,5-oligo(A) synthetase, the 2,5-oligo(A)-activated endonuclease F, 2',5'-phosphodiesterase, or eIF-2 kinase. However, interferon did prevent replication of vesicular stomatitis, Mengo virus, and reovirus in some of the derivative cell lines. The effect of interferon on the levels of the enzymes of the 2,5-oligo(A) and eIF-2 kinase pathways did not correlate directly with the antiviral properties of these cell clones. Greatly increased levels of 2,5-oligo(A) synthetase occurred in one clone without activation of an antiviral state. Another clone exhibited antiviral activity without detectably increased 2,5-oligo(A) synthetase activity. Changes in the levels of endonuclease F and 2',5'-phosphodiesterase were slight in all the clones examined. Neither 2,5-oligo(A) synthetase nor eIF-2 kinase levels were altered by interferon in another clone and yet an antiviral state was established and prevented replication of vesicular stomatitis, Mengo virus, and reovirus. The results show that mechanisms other than the 2,5-oligo(A) and eIF-2 kinase pathways are likely to contribute to the antiviral effects of interferon.

Induction and measurement of 2',5'-oligoadenylate synthetase in Madin-Darby bovine kidney cells and in cattle

2',5'-Oligoadenylate [2',5'-oligo(A)] was separated from 14C-labeled nucleosides produced in the 2',5'-oligo(A) synthetase assay by using 100-microliters columns of Dowex 1. No detectable nucleoside remained on the column after elution with 20 column volumes of water, whereas less than 1% of oligonucleotides were eluted from the column. At least 99% of oligonucleotides were eluted from the column with 1 M NaCl, pH 2. The major product had properties consistent with 2',5'-oligo(A). Exposure to alpha-1 bovine interferon (IFN) caused an increase in cellular 2',5'-oligo(A) synthetase activity which was proportional to the concentration of IFN in the medium up to 10(4) U of IFN per ml and then leveled off at about 15 X control activity. Under the assay conditions used, 2',5'-oligo(A) synthetase activity was directly proportional to the amount of cell extract over a 10-fold range. Cattle inoculated with IBR/BVD/PI-3 modified live virus vaccine showed an increase in 2',5'-oligo(A) synthetase activity in peripheral blood mononuclear leukocytes which persisted for at least 3 days postvaccination. Intramuscular injection of cattle with IFN caused a similar increase in 2',5'-oligo(A) synthetase activity. Changes in 2',5'-oligo(A) synthetase activity should be of value in (i) assessing the response of cattle to experimental viral infections or inoculations with viral vaccines or IFN or (ii) indicating a possible viral etiology in disease.

2',5'-Oligoadenylate synthetase-uninducible alpha/beta-interferon-resistant Friend cells develop an antiviral state when permeabilized with lysolecithin and treated with 2',5'-oligoadenylate oligomers

Variant sublines of Friend erythroleukemia cells (FLC) that do not respond to alpha/beta-interferon (IFN-alpha/beta) by developing an antiviral state but respond partially to IFN-gamma with an induced antiviral state, lack the ability to induce the 2',5'-oligoadenylate (2-5A) synthetase pathway. Exposure of wild-type and variant cells to exogenous 2-5A oligomers made permeable with lysolecithin resulted in 50-70% inhibition of protein synthesis. Further, the replication of vesicular stomatitis virus in IFN-resistant 2-5A synthetase-deficient FLC exposed to 2-5A trimer was inhibited to the same extent as in wild-type cells. Last, a significant cleavage of ribosomal RNA was observed in samples of total RNAs extracted from variant and wild-type permeabilized FLC, but only if they were exposed to 2-5A. These data are compatible with the conclusion that (i) the activation of the 2-5A-dependent endoribonuclease is not impaired in the variant cells, and (ii) the uninducibility of 2-5A synthetase can be bypassed by exogenously introducing its products, which leads to the establishment of a bona fide antiviral state.

2-5A-dependent endoribonuclease activity and effects of A2'p5'A2'p5'A (2-5A core) in mouse pancreatic islets

Glucose-stimulated (pro)insulin biosynthesis is markedly inhibited is mouse pancreatic islets incubated with micromolar concentrations of 2-5A 'core', (A2'p5'A2'p5'A). Total protein synthesis was also reduced, but to a lesser extent, while insulin release and total insulin content of the islets was untouched by 2-5A 'core'. Evidence is given for the presence of a 2-5A-dependent endoribonuclease which mediates these effects.

Injection of mice with antibody to mouse interferon alpha/beta decreases the level of 2'-5' oligoadenylate synthetase in peritoneal macrophages

Injection of conventional or axenic weanling mice with potent sheep or goat antibody to mouse interferon alpha/beta resulted in a decrease in the basal level of 2-5A synthetase in resting peritoneal macrophages and rendered these cells permissive for vesicular stomatitis virus. There was a good inverse correlation between the level of 2-5A synthetase in peritoneal macrophages and the permissivity of these cells for vesicular stomatitis virus. The peritoneal macrophages of 1- and 2-week-old mice had low levels of 2-5A synthetase and were permissive for vesicular stomatitis virus, whereas at 3 weeks (and after) there was a marked increase in the level of 2-5A synthetase in peritoneal macrophages, and these cells were no longer permissive for vesicular stomatitis virus. We suggest that low levels of interferon alpha or beta or both are produced in normal mice, and that this interferon contributes to host defense by inducing and maintaining an antiviral state in some cells.

[Interferons--research, effects and importance]

Interferons are a family of proteins synthesized by the cells of higher organisms as a first reaction to viral infections, preceding the immune response. They are also involved in the regulation of the immune system and therefore included among the lymphokines. Since they had been shown to have antitumor activity in vitro and in experimental animals, they were suggested to be powerful drugs in cancer therapy.

Activation of the ppp(A2'p)nA system in interferon-treated, herpes simplex virus-infected cells and evidence for novel inhibitors of the ppp(A2'p)nA-dependent RNase

High doses (100-1000 reference units/ml) of alpha or beta interferons are required to inhibit the growth of herpes simplex virus types I and II (HSV-I and HSV-II) in human Chang cells. In contrast, much lower doses (10-100 reference units/ml) of interferon inhibit replication of encephalomyocarditis virus (EMCV) in these cells. In the HSV-infected cells these high doses did not prevent the virus-induced shut off of host protein synthesis. The interferons were more effective in reducing the virus yield of HSV-I than of HSV-II. At the above concentrations they inhibited HSV-I protein synthesis but had little apparent effect on that of HSV-II. Similar amounts of (2'-5')oligo(adenylate)s were synthesised in response to HSV-I, HSV-II and EMCV infection of Chang cells after treatment with alpha or beta interferons. No (i.e. less than 1 nM) (2'-5')oligo(adenylate)s were found in control cells or on virus infection alone. Only low levels of ppp(A2'p)nA-specific rRNA cleavage were observed in the interferon-treated HSV-infected cells. In contrast, high levels were found in response to EMCV, despite the fact that ppp(A2'p)nA accumulated to similar levels with each of the three viruses in these cells. High-performance liquid chromatographic analysis of material from interferon-treated Chang cells 18 h after infection with HSV-I or HSV-II, combined with radiobinding, radioimmune and rRNA cleavage assays, confirmed the presence of ppp(A2'p)2A and ppp(A2'p)3A at greater than nanomolar concentration. In addition, apparently equivalent amounts of two other putative (2'-5')oligo(adenylate) derivatives which compete in the radiobinding and radioimmune assays, were present. These compounds were only weak activators of the ppp(A2'p)nA-dependent RNase and under appropriate conditions were capable of inhibiting the activation of this RNase by authentic ppp(A2'p)nA. The presence of these potentially inhibitory compounds provides a possible explanation for the relatively low levels of activation of the ppp(A2'p)nA-dependent RNase in interferon-treated, HSV-infected Chang cells.

The effect of interferon on cells deficient in nucleoside transport or lacking thymidine kinase activity

Mutants of mouse T-lymphoma S49 cells lacking thymidine kinase activity or deficient in nucleoside transport were selected by growth in the presence of 5'-fluorodeoxyuridine and their sensitivity to interferon tested. All five thymidine kinase and both transport deficient mutants were sensitive to the antiproliferative effects of interferon. The replication of encephalomyocarditis virus was also inhibited by interferon, and the intracellular levels of (2'-5')oligo-adenylate synthetase were elevated in all mutants tested. These results suggest that an intact nucleoside transport system or thymidine kinase activity are not essential for the expression of interferon sensitivity in these cells.

Interferon--present and future prospects

The interferons are a group of proteins that have inspired a new era of investigation into biological modification. The interferons are now divided into subgroups characterized by chemical means and correspond to different biological responses which can be observed in terms of the inducer used, and the timing of the response. Identified originally as antiviral agents when homologous cell systems were treated prior to infection, new studies have extended these observations to place the interferons in a central role as a strong force in the regulation of immunologic responses. A marriage of interferonology and cell immunology is enlarging both our understanding of the action of these proteins and our ability to follow the course of an illness and eventually to control its outcome . Genetic engineering has provided a way to process quantities of interferon and provided the molecular sequence of all three classes of IFN including a model of the active site for IFN-alpha. The offshot of the technology developed to study the intracellular processes after interferon treatment have already led to increased sensitivity to detect virally treated diseases. Both the variety of the interferon inducers and the scope of parasites in which it can exert its influence provide a frontier of biological investigation which has at the root of its nature the very secret of life. In addition to cellular phenomena, the positive effects on tumor-bearing organisms and the ill effects on infant animals highlight the potential power of the interferons.

Induction of ppp(A2'p)nA-dependent RNase in murine JLS-V9R cells during growth inhibition

We recently reported that interferon induces the synthesis of ppp(A2'p)nA(n = 2 to greater than or equal to 4) (2-5A)-dependent RNase in the murine cell line JLS-V9R. These cells normally contain very low levels of the nuclease; after interferon treatment, however, they develop levels approaching those found in murine L or Ehrlich ascites tumor cells. Here, we report a similar increase in the nuclease levels in JLS-V9R cells during the transition from the subconfluent actively growing state to the confluent stationary phase. Levels of 2-5A synthetase increased in parallel with the nuclease. The induced levels of both the nuclease and synthetase returned to low basal amounts after trypsinization, dilution, and culturing of the cells at subconfluent densities. The addition of anti-murine interferon (alpha + beta) antibodies to the medium did not affect the induction of the nuclease nor could any interferon be detected in the culture supernatants as determined by the lack of antiviral activity. The increase in the enzymes was not, therefore, due to the spontaneous production of interferon. The induction of the nuclease during confluency preceded an inhibition of [3H]-thymidine incorporation by the cells into DNA. The regulation of the 2-5A-dependent RNase in JLS-V9R cells may, therefore, be related to the control of cell growth.

Conformational characteristics of the trinucleoside diphosphate xyloA2'-5'xyloA2'-5'xyloA. A nuclear magnetic resonance and CD study

In this paper the conformational analysis of the 2'-5' linked xylotrinucleotide xA2'-5'xA2'-5'xA is reported. The title compound is an analogue of A2'-5'A2'-5'A, which compound was shown to display inhibitive effects on protein synthesis. The complete 1H-NMR assignment of the high field spectral region of the xylose trimer is given. Modes of base-base stacking are extracted from coupling constant data at various temperatures. Circular dichroic (CD) spectra confirm the presence of stacked states at low temperature. Xylonucleosides are known to prefer the N-type sugar conformation. However, in the present trimer the S-type conformer is suggested to partake in stacked conformations. Two types of stacking in the two constituent dimer fragments of the trimer are proposed to rationalize the NMR data: xA(1)N-xA(2)S and xA(2)N-xA(3)S.

Conformational analysis of the trinucleoside diphosphate 3'd(A2'-5'A2'-5'A). An NMR and CD study

A 500 MHz and 300 MHz NMR study of the trinucleoside diphosphate 3'd(A2'-5'A2'-5'A) is presented. In addition, circular dichroism is used to study base stacking in the title compound. The complete 1H-NMR spectral assignment of the sugar ring proton signals is given. Information about the sugar ring (N- or S-type conformation) and about the backbone geometry along C4'-C5' and C5'-O5' bonds is obtained from the NMR coupling constants. It is shown that the trimer mainly occurs in the N-N-N stacked state at low temperatures; the presence of a minor amount of N-N-S conformational sequence is indicated.

Interferon-induced differentiation of U937 cells. Comparison with other agents that promote differentiation of human myeloid or monocytelike cell lines

Effects of human fibroblast (beta) or leukocyte (alpha) interferon (IFN) on differentiations of a human histiocytic lymphoma-derived cell line (U937) or promyelocytic leukemia-derived cell line (HL-60) were studied. When cultured with beta-IFN (400-1,000 U/ml), U937 cells showed gross morphologic and microscopic changes consisting of clumping, increased cytoplasmic-to-nuclear ratio, enhanced prominence of cytoplasmic granules, and membrane ruffling. After culture with beta-IFN, the number of U937 cells reactive with B43.4.1 monoclonal antibody, which is specific for human monocytes, natural killer cells, and neutrophils, increased from less than 10% of U937 cells to 47% beta-IFN treatment also enhanced antibody-dependent cellular cytotoxicity against chicken erythrocytes by U937 cells. The same morphologic, phenotypic, and functional changes were also observed when U937 were treated with recombinant or natural alpha-IFN. The effects of alpha-IFN were totally abolished by anti-alpha-IFN serum. In contrast, HL-60, which differentiates toward cells of the monocyte lineage in response to phorbol 12-myristate 13-acetate (based on the above criteria), and toward granulocytes in response to dimethyl sulfoxide, did not differentiate when cultured with alpha- or beta-IFN. No consistent relationship between induction of differentiation and changes in phospholipid methylation were observed.

Interferon affects intracellular calmodulin levels

Interferon lowers calmodulin levels in two cell lines sensitive to its antiproliferative effect. Further, in synchronized cells, interferon strongly inhibits the increase in calmodulin observed when control cells enter the S phase, and concomitantly inhibits DNA synthesis. Calmodulin has been implicated in the control of cell proliferation and an increase in this protein seems to be necessary for the progression of cells into the S phase of the cell cycle. Therefore, the effect of interferon on calmodulin content might constitute part of the molecular mechanism by which interferon inhibits DNA synthesis.

Interferon-induced 56,000 Mr protein and its mRNA in human cells: molecular cloning and partial sequence of the cDNA

Treatment of responsive cells by interferons (IFNs) induces within a few hours a rise in the concentration of several proteins and mRNAs. In order to characterize these IFN-induced mRNA species, we have cloned in E. coli the cDNA made from a 17-18S poly(A)+ RNA of human fibroblastoid cells (SV80) treated with IFN-beta. We describe here a pBR322 recombinant plasmid (C56) which contains a 400 bp cDNA insert corresponding to a 18S mRNA species newly induced by IFN. The C56 mRNA codes for a 56,000 dalton protein easily detectable by hybridization-translation experiments. The sequence of 66 of the carboxy-terminal amino-acids of the protein can be deduced from the cDNA sequence. IFNs-alpha, beta or gamma are able to activate the expression of this gene in human fibroblasts as well as lymphoblastoid cells. The mRNA is not detectable without IFN; it reaches maximum levels (0.1% of the total poly(A)+ RNA) within 4-8 hrs and decreases after 16 hrs.

Interferon-induced synthesis of 2-5A-dependent RNase in mouse JLS-V9R cells

Interferon treatment of mouse JLS-V9R cells resulted in a 10- to 20-fold increase in the levels of the 2-5A (ppp(A2'p)nA)-dependent RNase. The nuclease was monitored in cell extracts by covalent and noncovalent binding of 32P-labeled 2-5A derivatives to the nuclease and by the appearance of 2-5A-mediated ribosomal RNA cleavage products. Untreated JLS-V9R cells contained very low levels of the nuclease (5-10% of that found in control Ehrlich ascites tumor cells). An increase in the nuclease was first detected between 3 and 6 hr after interferon treatment and was completely inhibited by actinomycin D. Control experiments indicated that the increase in the nuclease was not due to an intracellular redistribution of the enzyme. The results indicate that in JLS-V9R cells, interferon controls the activity of the 2-5A system at two levels: by inducing the synthesis of both the 2-5A-dependent RNase and the 2-5A-synthetase.

Induction of an antiviral response by interferon requires thymidine kinase

Mouse fibroblastoid cells (Ltk-) that lack thymidine kinase (tk) activity are unable to respond to murine beta-interferon by establishing antiviral activity or inducing the double-stranded RNA-dependent enzymes, oligo[(2'-5')A] polymerase and Mr 68,000 protein kinase. In contrast, the parental L-929 cell line or clonal derivatives of Ltk- cells into which the herpes virus tk gene was introduced by DNA-mediated gene transfer respond normally to interferon in developing resistance to viral infection and in inducing double-stranded RNA-dependent enzymes. Further evidence for a role of tk in the response to interferon was obtained by isolating revertants of tk+ clones that lost the herpes virus tk gene during growth in BrdUrd-containing medium. In such revertant sublines both tk enzyme activity and viral tk genes were undetectable and treatment with interferon failed to produce an antiviral effect or induce synthesis of the double-stranded RNA-dependent enzymes. Our results indicate that the ability of mouse L cells to respond to beta-interferon is dependent upon the presence of a functional tk gene. We propose that the induction of antiviral responses by interferon stringently requires a metabolite, the level of which is determined by tk activity. The system described may provide a means for elucidating the mechanisms by which responses to interferon are induced.

The ppp(A2'p)nA and protein kinase systems in wild-type and interferon-resistant Daudi cells

Daudi cells, a human lymphoblastoid line, are exceptionally sensitive to the growth inhibitory effects of interferon, 1 unit/ml being sufficient to inhibit cell growth. In addition, interferon treatment of these cells severely inhibits the incorporation of exogenous thymidine into DNA and causes cells to accumulate in the G1(G0) at the expense of the S phase of the cell cycle. The possible involvement of ppp(A2'p)nA(n = 2 to less than or equal to 4) in these effects has been investigated. No (less than 1 nM) ppp(A2'p)nA or (A2'p)nA or alternative products of the ppp(A2'p)nA synthetase [e.g. NAD (2'pA)2] were detected in interferon-treated cells. In addition no evidence was obtained for the occurrence of ppp(A2'p)nA-mediated ribosomal RNA cleavage in these cells even after several days of treatment with relatively high doses of interferon. A line of Daudi cells which is resistant to all three of the above effects of interferon was selected. The wild type and resistant lines were compared with respect to the ppp(A2'p)nA and interferon and double-stranded RNA (dsRNA)-mediated protein kinase systems. The resistant line was not receptor-negative as it responded to interferon by the production of elevated levels of the ppp(A2'p)nA synthetase similar to those observed in extracts from wild-type cells. There was no detectable difference between the lines in the levels of the (2'-5')phosphodiesterase responsible for the degradation of ppp(A2'p)nA. There was, however, about a twofold increase in the ppp(A2'p)nA-dependent endoribonuclease activity in response to interferon with extracts from the wild-type but not the resistant cells. In addition, although the dsRNA-dependent protein kinase activity increased in both types of cell there was a striking reduction in the level of protein phosphorylation in general in response to interferon with material from the wild-type but not the resistant cells.

Conditions required for induction of interferon by rotaviruses and for their sensitivity to its action

Our investigations of interferon induction by rotaviruses showed that only when cells were pretreated with interferon, i.e., primed, could infectious rotaviruses induce significant quantities of interferon. As little as 0.5 U of interferon provided sufficient priming for this induction. UV-irradiated rotaviruses induced significant levels of interferon, and priming only marginally enhanced the yields. Neither heat-inactivated virus nor serum-neutralized virus was able to induce interferon, even when cells were primed. When cells were treated with purified virus double-stranded RNA in the presence of DEAE-dextran to facilitate uptake, interferon was induced, although priming did not enhance yields. These results strongly implicate the viral double-stranded RNA as the effector for interferon induction. The insensitivity of rotaviruses to interferon in vitro was also studied. Results suggested that this lack of sensitivity was not due to any inherent resistance of the virus to the antiviral proteins, but rather to lack of activation of cellular enzymes exhibiting antiviral activity.

Control of the ppp(a2'p)nA system in HeLa cells. Effects of interferon and virus infection

HeLa cells have an unusually high level of ppp(A2'p)nA synthetase (n = 2 to greater than or equal to 4) even in the absence of interferon treatment. In accord with this ppp(A2'p)nA and ppp(A2'p)nA-mediated ribosomal RNA cleavage occur naturally in response to encephalomyocarditis virus infection in control as well as in interferon-treated cells. Despite this, in the absence of interferon treatment, encephalomyocarditis virus grows well in these cells. A possible explanation for this paradox is that the ppp(A2'p)nA dependent RNase is lost or inactivated at later times post-infection in control but not in interferon-treated cells. It appears, therefore, to be the prevention by interferon of the virus-mediated inhibition of the ppp(A2'p)n-dependent nuclease rather than the absolute level or induction of the ppp(A2'p)nA synthetase which is crucial for the activity of the ppp(A2'p)nA system in HeLa cells. These results provide evidence for a further level of control in the ppp(A2'p)nA system and show that limited ppp(A2'p)nA-mediated ribosomal RNA cleavage alone is not sufficient to cause an inhibition of virus growth.

Variation of (2'-5') oligo A synthetase level in lymphocytes and granulocytes of patients with viral infections and leukemia

Using a simplified technique for the determination of oligoisoadenylate synthetase activity, we have compared the cellular level of this interferon (IFN)-induced enzyme in multiple samples of peripheral blood leukocytes. In mononuclear cells (PBMC) of healthy donors the enzyme level was remarkably constant, but in the cells of about 85% of patients with viral infections enzyme activity was significantly elevated. In contrast, the incidence of elevated activity in bacterial infections was low. Synthetase activity could be also detected in granulocytes, although normally its level in these cells was considerably lower than in PBMC. A sharp increase in the enzyme level in granulocytes was found in cells exposed in vitro to IFN, as well as in cells from patients undergoing IFN therapy. Increased synthetase activity was also detected in the granulocytes of patients with viral infections. We have also determined the level of the enzyme in patients with various types of leukemias. In a large proportion of the patients with acute lymphoblastic leukemia (ALL) we found severely decreased enzyme levels (10-20% of control value). The decreased activity could usually be correlated to predominance of blast cells in the peripheral blood.

Inhibition of interferon-mediated induction of indoleamine 2,3-dioxygenase in mouse lung by inhibitors of prostaglandin biosynthesis

Inhibitors of fatty acid cyclooxygenase such as indomethacin (0.1 mM), phenylbutazone (0.3 mM), and aspirin (1 mM) were found to suppress almost completely the interferon-mediated induction of indoleamine 2,3-dioxygenase in mouse lung slices. However, phenacetin, an anti-inflammatory agent devoid of cyclooxygenase inhibitory activity, and sodium salicylate, a weak inhibitor of cyclooxygenase, were much less active under identical conditions. Glucocorticoids including dexamethasone, betamethasone, and cortisone, all of which are inhibitors of phospholipase A2, blocked induction of the dioxygenase by interferon in the nanomolar range, whereas other steroid hormones, such as aldosterone, testosterone, and 17 beta-estradiol, were all but ineffective. These results suggest that the enzymes phospholipase A2 and fatty acid cyclooxygenase, both of which are essential for the biosynthesis of prostaglandins, play an important role in the induction of indoleamine 2,3-dioxygenase by interferon.

A mouse cell line, which is unprotected by interferon against lytic virus infection, lacks ribonuclease F activity

A mouse cell line, NIH 3T3, does not respond to some of the activities of interferon. Even after treatment with high concentrations of interferon the replication of lytic viruses, such as encephalomyocarditis virus (EMCV) and vesicular stomatitis virus (VSV) is not inhibited in these cells. In contrast, interferon treatment of these same cells results in the inhibition of Moloney murine leukemia virus (MMuLV) production. We have analyzed enzymatic pathways which are induced by interferon in these cells. After interferon treatment, the level of the (2'-5')oligoadenylate [(2'-5)An] synthetase activity and the phosphorylation of the 67000-dalton protein (P1) are enhanced in NIH 3T3 cells to approximately the same level as interferon-sensitive mouse L-cells. Moreover, NIH 3T3 and L-cells, contain approximately the same levels of enzymes which inactivate (2'-5')An. Both exogenously added (2'-5')A3 or double-stranded RNA (dsRNA) failed to inhibit protein synthesis in NIH 3T3 extracts even though they were potent inhibitors of L-cell extract-directed protein synthesis. Direct measurements of the (2'-5')An-dependent ribonuclease F (RNase F) failed to detect such activity in NIH 3T3 cells. Our results, therefore, suggest that the presence of RNase F activity is necessary for the interferon-induced antiviral activity against EMCV and against VSV. The induction of protein kinase activity by interferon treatment of NIH 3T3 cells appears to have no direct effect on EMCV and VSV replication.

Interferon enhances the amount of membrane-bound beta2-microglobulin and its release from human Burkitt cells

Human leukocytes interferon (HuIFN-A) increased the amount of beta2-microglobulin on the surface of human Burkitt lymphoma cells (Ramos) and also increased the amount released into the culture medium. The effect was observed 1 h after addition of IFN. These results suggest that the increase in beta2-microglobulin on the cell surface of IFN-treated cells is not due to a decreased shedding of antigen from the cell surface, nor an "unmasking" of surface antigen, but rather to an increased synthesis of antigen.

Effect of interferon on transient shut-off of cellular RNA and protein synthesis induced by Mengo virus infection

Infection of mouse L929 cells with Mengo virus resulted in a rapid shut-off of cellular RNA synthesis followed within the first hours post infection by a gradual decrease in host protein synthesis. Pretreatment of the cells with high doses of interferon, blocking viral multiplication, did not affect the virus-induced shut-off of host macromolecular synthesis. In these interferon-treated cells the 2',5'A-activated nuclease may account for the degradation of viral RNA, soon after its replication. However, the inhibition of host protein synthesis could not be explained by this mechanism. Poly(A)-containing RNA, present in interferon-treated and infected cells, amounted to as much as 70% of that present in interferon-treated, noninfected cells. On the other hand, extracted cytoplasmic RNA was efficiently translated in a reticulocyte lysate, showing that extensive mRNA degradation was not involved in the inhibition of host protein synthesis. In the continued presence of interferon, the virus-induced shut-off was found to be transient. Late in infection, RNA synthesis was found to recover, followed by recovery of protein synthesis and survival of the cells.

Inhibition of vaccinia mRNA methylation by 2',5'-linked oligo(adenylic acid) triphosphate

Extracts of interferon-treated cells synthesize unique 2',5'-linked oligo(adenylic acid) 5'-phosphates in the presence of ATP and double-stranded RNA. 2',5'-linked oligo(adenylic acid) 5'-triphosphate inhibits protein synthesis at nanomolar concentrations by activating RNase. We have observed that oligo(adenylic acid) 5'-monophosphate and 5'-triphosphate are potent inhibitors of vaccinia mRNA methylation in vitro. Both the methylation of mRNA methylation is not due to degradation of the mRNA. Inhibition of the 5'-terminal guanine at the 7 position and the 2'-O-ribose methylation of the penultimate nucleoside are inhibited. Such inhibition of the requisite modification of the 5' terminus of mRNA by 2',5'-linked oligo(adenylic acids) may be a mechanism of interferon action against both DNA and RNA viruses in which mRNAs derived from them are capped.

Lack of systematic correlation between the interferon mediated antiviral state and the levels of 2-5A synthetase and protein kinase in three different types of murine cells

The levels of two dsRNA-dependent enzyme activities, the pppA(2'p5'A)n synthetase (2-5A synthetase) and protein kinase were investigated in control and interferon-treated murine cells: L-929, K/Balb and NIH/3T3. Treatment of these cells with interferon resulted both in the establishment of the antiviral response and the development of anticellular effects. This latter was observed 3 days after treatment with interferon. The levels of 2-5A synthetase and protein kinase in control and interferon-treated cells seemed to vary from one cell type to the other. In L-929 cells, both the 2-5A synthetase and protein kinase were induced by interferon as has been shown previously. On the other hand, treatment of NIH/3T3 cells with interferon resulted in the induction of 2-5A synthetase in the absence of any enhanced protein kinase activity. This lack of protein kinase in interferon-treated NIH/3T3 cells was not due to the presence of high levels of protein phosphatases. A third type of mouse cells, K/Balb cells, contained very high levels of 2-5A synthetase in the absence of any apparent resistance to virus infection. On treatment with interferon the level of 2-5A synthetase in K/Balb cells remained constant while the protein kinase activity was enhanced by several fold. Both control and interferon-treated K/Balb cells showed similar sensitivity to the action of exogenous 2-5A thus suggesting that the 2-5A system (the 2-5A dependent nuclease and the phosphodiesterase which degrades 2-5A) was not altered on treatment with interferon. The significance of these results in relation to the mechanism of action of interferon is discussed.

Interferon-mediated antiviral state in human MRC5 cells in the absence of detectable levels of 2-5A synthetase and protein kinase

Treatment of human HeLa and MRC5 cells with human alpha (leukocyte) and beta (fibroblast) interferon results in the development of an antiviral state against two types of viruses: vesicular stomatitis virus (rhabdovirus) and encephalomyocarditis virus (picornavirus). These cells, however, differ in their ability to synthesize the two double-stranded (ds) RNA-dependent enzymatic activities, pppA(2'p5'A)n synthetase (2-5A synthetase) and protein kinase which have been reported to be induced in several cell lines by interferon. Both the 2-5A synthetase and the protein kinase are enhanced by several fold in HeLa cells on treatment with interferon. In contrast, neither the 2-5A synthetase nor the protein kinase can be detected in MRC5 cell treated or not treated with interferon. The lack of detection of the 2-5A synthetase in MRC5 cells is not associated with the absence of the other components of the 2-5A system (2-5A dependent nuclease and 2'-phosphodiesterase). We have previously shown that MRC5 cells are sensitive to the action of 2-5A and furthermore the inhibitory action of 2-5A on these cells is transient. Mixing experiments between HeLa and MRC5 cell fractions after partial purification on columns of poly(I).poly(C)-Sepharose, showed that the absence of detection of the protein kinase activity in MRC5 cells cannot be attributed to the presence of phosphatases or other inhibitors of phosphorylation in control or interferon-treated MRC5 cell extracts. In addition, we show that the interferon-mediated protein kinase activity in HeLa cell extracts can be precipitated by treatment at pH 5, a procedure which leads to an enhanced level of detectable protein kinase activity in general. Once again, however, MRC5 cell extracts fail to show any interferon-mediated protein kinase activity. These results suggest that either the two enzyme activities are not necessary for the development of the antiviral response induced by interferon or the intracellular events leading to the establishment of the antiviral state vary from one cell system to the other.