Interferon messenger RNA content of human fibroblasts during induction, shutoff, and superinduction of interferon production

Tuning the immune response of dendritic cells to surface-assembled poly(I:C) on microspheres through synergistic interactions between phagocytic and TLR3 signaling

The artificial dsRNA polyriboinosinic acid-polyribocytidylic acid, poly(I:C), is a potent adjuvant candidate for vaccination, as it strongly drives cell-mediated immunity. However, because of its effects on non-immune bystander cells, poly(I:C) administration may bear danger for the development of autoimmune diseases. Thus poly(I:C) should be applied in the lowest dose possible. We investigated microspheres carrying surface-assembled poly(I:C) as a two-in-one adjuvant formulation to stimulate maturation of monocyte-derived dendritic cells (MoDCs). Negatively charged polystyrene microspheres were equipped with a poly(ethylene glycol) corona through electrostatically driven surface assembly of a library of polycationic poly(l-lysine)-graft-poly(ethylene glycol) copolymers, PLL-g-PEG. Stable surface assembly of poly(I:C) was achieved by incubation of polymer-coated microspheres in an aqueous poly(I:C) solution. Surface-assembled poly(I:C) exhibited a strongly enhanced efficacy to stimulate maturation of MoDCs by up to two orders of magnitude, as compared to free poly(I:C). Multiple phagocytosis events were the key factor to enhance the efficacy. The cytokine secretion pattern of MoDCs after exposure to surface-assembled poly(I:C) differed from that of free poly(I:C), while their ability to stimulate T cell proliferation was similar. Overall, phagocytic signaling plays an important role in defining the resulting immune response to such two-in-one adjuvant formulations.

Superinduction of interleukin-6 mRNA in lung epithelial H292 cells depends on transiently increased C/EBP activity and durable increased mRNA stability

Restriction of eukaryotic protein synthesis affects the regulation of some transiently expressed gene transcripts resulting in their superinduction. We determined the transcriptional and post-transcriptional processes implicated in IL-6 mRNA superinduction in a human lung-derived epithelial cell line H292, and their kinetics in the absence and presence of an exogenous stimulus, tumor necrosis factor-alpha (TNF-alpha). Cycloheximide (CHI) at 10 microg/ml, which inhibited protein synthesis for 80%, caused a 80-fold induction of IL-6 mRNA level which was due predominantly to a stabilization of IL-6 mRNA (20-fold) early on. Employing transient transfection protocols we noted a small positive effect of CHI on transcription, mediated by the proximal and the distal C/EBP sites of the IL-6 promoter and paralleled by an increased C/EBP DNA-binding activity, similar to that found for exposure to TNF-alpha alone. TNF-alpha and CHI synergized on IL-6 mRNA expression (200-fold increase) which was due to an increased transcription, corresponding to a further increased C/EBP DNA-binding activity. However, the effect of CHI on IL-6 gene transcription was transient, in support of the need for ongoing protein synthesis for C/EBP activity. These findings indicate that IL-6 mRNA superinduction, at least in H292 cells, is regulated predominantly by modulating the repressive system that ensures a rapid degradation of IL-6 mRNA.

Inducible processing of interferon regulatory factor-2

PRDI-BFc and PRDI-BFi are proteins that bind specifically to a regulatory element required for virus induction of the human beta interferon (IFN-beta). PRDI-BFc is a constitutive binding activity, while the PRDI-BFi binding activity is observed only after cells are treated with inducers such as virus or poly(I).poly(C) plus cycloheximide or in some cells by cycloheximide alone. In this paper we report that PRDI-BFc is interferon regulatory factor-2 (IRF-2), a known transcriptional repressor. In addition, we find that PRDI-BFi is a truncated form of IRF-2, lacking approximately 185 C-terminal amino acids. Thus, PRDI-BFi appears to be generated by inducible proteolysis. Although the affinity of PRDI-BFc/IRF-2 for the IFN-beta promoter does not appear to be affected by the removal of C-terminal amino acids, the ability of PRDI-BFi to function as a repressor in cotransfection experiments is significantly less than that of intact IRF-2. Studies have shown that IRF-2 can block the activity of the transcriptional activator IRF-1, which also binds specifically to the IFN-beta gene promoter. Thus, the inducible proteolysis of IRF-2 may be involved in the regulation of the IFN-beta gene or of other genes in which the ratio of IRF-1 to IRF-2 can affect the level of transcription.

Inducible processing of interferon regulatory factor-2

PRDI-BFc and PRDI-BFi are proteins that bind specifically to a regulatory element required for virus induction of the human beta interferon (IFN-beta). PRDI-BFc is a constitutive binding activity, while the PRDI-BFi binding activity is observed only after cells are treated with inducers such as virus or poly(I).poly(C) plus cycloheximide or in some cells by cycloheximide alone. In this paper we report that PRDI-BFc is interferon regulatory factor-2 (IRF-2), a known transcriptional repressor. In addition, we find that PRDI-BFi is a truncated form of IRF-2, lacking approximately 185 C-terminal amino acids. Thus, PRDI-BFi appears to be generated by inducible proteolysis. Although the affinity of PRDI-BFc/IRF-2 for the IFN-beta promoter does not appear to be affected by the removal of C-terminal amino acids, the ability of PRDI-BFi to function as a repressor in cotransfection experiments is significantly less than that of intact IRF-2. Studies have shown that IRF-2 can block the activity of the transcriptional activator IRF-1, which also binds specifically to the IFN-beta gene promoter. Thus, the inducible proteolysis of IRF-2 may be involved in the regulation of the IFN-beta gene or of other genes in which the ratio of IRF-1 to IRF-2 can affect the level of transcription.

Identification and characterization of a novel repressor of beta-interferon gene expression

We have identified and characterized a novel repressor of human beta-interferon (beta-IFN) gene expression. This protein, designated PRDI-BF1, binds specifically to the PRDI element of the beta-IFN gene promoter and is distinct from previously reported proteins that bind to this sequence. PRDI-BF1 is an 88-kD protein containing five zinc-finger motifs. Cotransfection experiments in cultured mammalian cells revealed that PRDI-BF1 is a potent repressor of PRDI-dependent transcription. PRDI-BF1 blocks virus induction of the intact beta-IFN gene promoter and of synthetic promoters containing multiple PRDI sites. PRDI-BF1 can also block the SV40 enhancer when PRDI sites are located between the enhancer and the promoter. This repression is highly dependent on the location of the PRDI sites, however, indicating that PRDI-BF1 cannot act at a distance. On the basis of the properties of PRDI-BF1 and the observation that PRDI-BF1 mRNA accumulation is virus inducible, we propose that PRDI-BF1 may act as a postinduction repressor of the beta-IFN gene by displacing positive regulatory proteins from the PRDI site of the promoter.

Role of calmodulin/protein kinase C in interferon production by poly(rI).poly(rC) in primed human cell cultures

Human cells treated with trifluoperazine (TFP) or K252a prior to and/or during priming with human interferon-alpha (HuIFN-alpha) produce significantly more IFN on induction with poly(rI).poly(rC) than primed cells in the absence of the drug. Results suggest that calmodulin/protein kinase activity may be involved in priming activity of HuIFN-alpha.

Postinduction turnoff of beta-interferon gene expression

Viral induction of the human beta-interferon (IFN-beta) gene leads to a transient accumulation of high levels of IFN-beta mRNA. Previous studies have shown that the increase in IFN-beta mRNA levels after induction is due to an increase in the rate of IFN-beta gene transcription. In this paper, we show that the rapid postinduction decrease in the level of IFN-beta mRNA is due to a combination of transcriptional repression and rapid turnover of the mRNA. This transcriptional repression can be blocked with cycloheximide, suggesting that the synthesis of a virus-inducible repressor is necessary for the postinduction turnoff of the IFN-beta gene. Analysis of the sequence requirements for IFN-beta mRNA instability revealed two regions capable of destabilizing a heterologous mRNA. One destabilizer is an AU-rich sequence in the 3' untranslated region, and the other is located 5' to the translation stop codon.

Postinduction turnoff of beta-interferon gene expression

Viral induction of the human beta-interferon (IFN-beta) gene leads to a transient accumulation of high levels of IFN-beta mRNA. Previous studies have shown that the increase in IFN-beta mRNA levels after induction is due to an increase in the rate of IFN-beta gene transcription. In this paper, we show that the rapid postinduction decrease in the level of IFN-beta mRNA is due to a combination of transcriptional repression and rapid turnover of the mRNA. This transcriptional repression can be blocked with cycloheximide, suggesting that the synthesis of a virus-inducible repressor is necessary for the postinduction turnoff of the IFN-beta gene. Analysis of the sequence requirements for IFN-beta mRNA instability revealed two regions capable of destabilizing a heterologous mRNA. One destabilizer is an AU-rich sequence in the 3' untranslated region, and the other is located 5' to the translation stop codon.

Activators of protein kinase C enhance accumulation of interferon-beta mRNA in murine cell lines

Murine C127 fibroblasts carrying an expression vector for a human interferon gene (HuIFN-beta, under the control of a constitutive promoter) can be induced to produce murine (Mu) IFN by double-stranded (ds) RNA or virus infection. Fibroblasts treated with the protein kinase C activators 1-oleyl-2-acetylglycerol (OAG) or phorbol-12-myristate-13-acetate (PMA) secrete greater amounts of MuIFN than untreated cells, but the same amount of HuIFN-beta. Accordingly, the level of MuIFN-beta mRNA increases in the presence of protein kinase C activators whereas that of HuIFN-beta mRNA is unchanged. In time course experiments after induction with dsRNA, accumulation of MuIFN-beta mRNA is observed within 30 min in the presence of OAG, when this mRNA cannot be detected in control cells. The protein kinase C activators increase accumulation of MuIFN-beta mRNA, even in the presence of the inhibitor of protein synthesis cycloheximide. A similar increase in MuIFN-beta mRNA is observed in C243 fibroblasts treated with phorbol-12-myristate-13-acetate, but not in parental C127 cells. These findings suggest that protein kinase C does not promote synthesis of regulatory factors controlling transcription of IFN mRNA, but that it may be directly or indirectly involved in activation of such factors in some murine cell lines.

Autocrine induction of major histocompatibility complex class I antigen expression results from induction of beta interferon in oncogene-transformed BALB/c-3T3 cells

By varying growth conditions, we identified a novel mechanism of autocrine regulation of major histocompatibility complex (MHC) class I gene expression by induction of beta interferon gene expression in transformed BALB/c-3T3 cells. Low-serum conditions enhanced MHC class I antigen expression in v-rasKi- and v-mos-transformed BALB/c-3T3 cells but not in untransformed BALB/c-3T3 cells. Transformed and untransformed cells grown under standard serum conditions (10% bovine calf serum) expressed similar cell surface levels of MHC class I antigens. However, low-serum conditions (0.5% bovine calf serum) induced four- to ninefold increases in cell surface levels of MHC class I antigens in both v-rasKi- and v-mos-transformed cells but not in untransformed cells. These increases in MHC class I gene expression were seen at both the mRNA and cell surface protein levels and involved not only the heavy-chain component of the class I antigens but also beta 2 microglobulin. Beta 1 interferon mRNA and beta interferon-inducible 2',5'-oligoadenylate synthetase mRNA were induced by growth under low-serum conditions in transformed BALB/c-3T3 cells, and antibodies to beta interferon blocked the induction of MHC class I antigen expression by serum deprivation in these cells. These results demonstrate that growth under low-serum conditions leads to induction of beta interferon expression in oncogene-transformed cells which then directly mediates autocrine enhancement of MHC class I gene expression.

Autocrine induction of major histocompatibility complex class I antigen expression results from induction of beta interferon in oncogene-transformed BALB/c-3T3 cells

By varying growth conditions, we identified a novel mechanism of autocrine regulation of major histocompatibility complex (MHC) class I gene expression by induction of beta interferon gene expression in transformed BALB/c-3T3 cells. Low-serum conditions enhanced MHC class I antigen expression in v-rasKi- and v-mos-transformed BALB/c-3T3 cells but not in untransformed BALB/c-3T3 cells. Transformed and untransformed cells grown under standard serum conditions (10% bovine calf serum) expressed similar cell surface levels of MHC class I antigens. However, low-serum conditions (0.5% bovine calf serum) induced four- to ninefold increases in cell surface levels of MHC class I antigens in both v-rasKi- and v-mos-transformed cells but not in untransformed cells. These increases in MHC class I gene expression were seen at both the mRNA and cell surface protein levels and involved not only the heavy-chain component of the class I antigens but also beta 2 microglobulin. Beta 1 interferon mRNA and beta interferon-inducible 2',5'-oligoadenylate synthetase mRNA were induced by growth under low-serum conditions in transformed BALB/c-3T3 cells, and antibodies to beta interferon blocked the induction of MHC class I antigen expression by serum deprivation in these cells. These results demonstrate that growth under low-serum conditions leads to induction of beta interferon expression in oncogene-transformed cells which then directly mediates autocrine enhancement of MHC class I gene expression.

Identification of an inducible factor that binds to a positive regulatory element of the human beta-interferon gene

Human beta-interferon gene expression is induced by virus or poly(I).poly(C). This induction is due at least in part to an increase in the rate of transcription and does not require protein synthesis. A 40-base-pair DNA sequence within the beta-interferon promoter, termed the interferon gene regulatory element (IRE), is an inducible enhancer in mouse fibroblasts, and both positive and negative regulatory DNA sequences have been identified within this element. In this paper we identify three factors that bind specifically to two positive regulatory domains within the IRE. Two of these factors are present in nuclear extracts prepared from uninduced and induced cells; one is present only in extracts from induced cells. The functional significance of these binding activities was demonstrated by showing that point mutations within the IRE that decrease human beta-interferon gene transcription in vivo prevent binding in vitro. We propose that induction of the beta-interferon gene involves the modification of a protein to a form that binds specifically to a positive regulatory sequence within the IRE.

Activation of the human beta-interferon gene requires an interferon-inducible factor

beta-Interferon (beta-IFN) gene expression can be induced by poly(I)-poly(C) or virus, but there is considerable variation in the extent of induction between different cell lines. We characterized two poorly inducible human cell lines, HeLa and 143 thymidine kinase negative (143 tk-), to define cellular factors involved in the activation of the beta-IFN gene. We show that the deficiency in beta-IFN induction in these cells can be complemented by fusion to highly inducible mouse cells. We conclude that the human cells are deficient in a trans-acting factor required for B-IFN gene activation. The level of induction of the beta-IFN gene in HeLa and 143 tk- cells can also be increased by priming with IFN before induction. If IFN priming is carried out in the presence of cycloheximide, a approximately 200-fold increase in induction is observed. We conclude that activation of the beta-IFN gene requires an IFN-inducible factor that is only expressed at low levels in unprimed HeLa and 143 tk- cells.

Activation of the human beta-interferon gene requires an interferon-inducible factor

beta-Interferon (beta-IFN) gene expression can be induced by poly(I)-poly(C) or virus, but there is considerable variation in the extent of induction between different cell lines. We characterized two poorly inducible human cell lines, HeLa and 143 thymidine kinase negative (143 tk-), to define cellular factors involved in the activation of the beta-IFN gene. We show that the deficiency in beta-IFN induction in these cells can be complemented by fusion to highly inducible mouse cells. We conclude that the human cells are deficient in a trans-acting factor required for B-IFN gene activation. The level of induction of the beta-IFN gene in HeLa and 143 tk- cells can also be increased by priming with IFN before induction. If IFN priming is carried out in the presence of cycloheximide, a approximately 200-fold increase in induction is observed. We conclude that activation of the beta-IFN gene requires an IFN-inducible factor that is only expressed at low levels in unprimed HeLa and 143 tk- cells.

Control of biologically active interleukin 2 messenger RNA formation in induced human lymphocytes

The regulation of human interleukin 2 (IL-2) mRNA production in induced normal lymphocytes was studied by following the expression of isolated mRNA in microinjected oocytes of Xenopus laevis. Mitogenic stimulation results in the appearance of greatly increased levels of IL-2 mRNA activity. This process requires de novo transcription. Induction is followed promptly by a shutoff of active IL-2 mRNA formation. This shutoff requires the synthesis of a protein repressor and can be prevented by cycloheximide, an inhibitor of translation. The presence of cycloheximide leads to extensive superinduction of IL-2, concomitant with an increase in active IL-2 mRNA formation up to 30-fold over normal levels. The repressor appears to be short-lived, as the addition of cycloheximide after shutoff leads to an immediate resumption of active IL-2 mRNA formation. The shutoff mechanism is restored rapidly upon removal of cycloheximide. The repressed state is readily reversed also by reinduction of the cells, even soon after shutoff has occurred, without a refractory period. The accumulated active IL-2 mRNA decays with a half-life of about 20 hr. The net result is the generation of a relatively short wave of IL-2 mRNA activity, demonstrating the tight control of IL-2 gene expression.

The 5'-flanking region of a human IFN-alpha gene mediates viral induction of transcription

Usually only cells exposed to virus, double-stranded RNA or other inducers synthesize interferon (IFN). Interferon mRNA appears 1-2 h after induction, peaks at 1.5-20 h and decays with a half life of about 30 min. So far, it has not been determined whether induction of interferon is due to transient stabilization of a rapidly turning-over mRNA or to activation of transcription. To clarify this issue we transformed mouse L cells with a hybrid gene in which the 5'-flanking region of the human IFN-alpha 1 gene was followed by the rabbit beta-globin transcription unit. Correctly initiated beta-globin RNA appeared only after viral induction, with the kinetics described for interferon mRNA. Cells transformed with the converse construction, or with the complete rabbit beta-globin gene, constitutively produced correctly initiated transcripts; viral infection decreased the level of transcripts. We conclude that induction acts by activating transcription rather than by reducing turnover, and that the regulatory elements are contained in the 5'-flanking region of the interferon gene.

Multiple human beta interferon genes

Analysis of human beta interferon (IFN) mRNA preparations obtained from poly(I) . poly (C)-induced human diploid fibroblasts (FS-4) and from several similarly induced human-mouse somatic cell hybrids by electrophoresis through agarose-CH3HgOH tube gels led to the detection of at least five translationally active human IFN-beta mRNA species. The results obtained are consistent with the existence of IFN-beta genes on different human chromosomes. Marked cell-dependent variability in the expression of these IFN mRNA species was observed.

The structure of a thirty-six kilobase region of the human chromosome including the fibroblast interferon gene IFN-beta

The isolation of a human genomic cosmid hybrid containing the interferon beta gene has recently been reported (Gross et al., 1981). This hybrid was mapped using single and double digests and cross-hybridisation with the sub-cloned EcoRI and BgIII fragments. Purified fragments and subclones were used as hybridisation probes against chromosomal "Southern" blots to show that at least half of the region has been cloned without alteration. This cannot at present be confirmed for the rest of the region due to the presence of highly repetitive DNA on these cloned fragments. Sequencing data on the 5'-end of the fibroblast IFN-beta gene shows a high density of direct and inverted repeats. The IFN-beta mRNA coding region contains no intron, although the possibility of other transcription starts is not ruled out. The cloned region shows no similarities to known genomic clones containing IFN-alpha genes.

Partial characterization of gamma (immune) interferon mRNA extracted from human lymphocytes

gamma (immune) interferon (IFN-gamma) was induced in cultures of fresh human lymphocytes by combined treatment with a phorbol ester (12-O-tetradecanoylphorbol 13-acetate, TPA) and the T cell mitogen phytohemagglutinin (PHA). Compared to the IFN-gamma yields obtained with PHA induction alone, the inclusion of TPA caused a significant enhancement of IFN-gamma production. Poly(A)-containing mRNA was isolated from mononuclear cells induced with TPA and PHA. Injected into Xenopus laevis oocytes, this mRNA preparation gave rise to IFN activity with characteristic properties of human IFN-gamma. Sucrose density gradient centrifugation analysis showed that IFN-gamma mRNA sedimented at 15 S, suggesting that it contains a total of about 1400 nucleotides.

Translational activity and functional stability of human fibroblast beta 1 and beta 2 interferon mRNAs lacking 3'-terminal RNA sequences

Polyadenylylated mRNA was purified from poly(I).poly(C)- and cycloheximide-superinduced human fibroblast (FS-4) cultures. The mRNA was subjected to electrophoresis through an agarose/CH3HgOH gel, and human fibroblast beta 1 and beta 2 interferon mRNAs were isolated. Each mRNA preparation was phosphorolyzed at 0 degrees C for 20 min by using a molar excess of polynucleotide phosphorylase to produce RNAs lacking poly(A) and then incubated at 37 degrees C for varying lengths of time to allow the phosphorylase to further digest the deadenylylated RNA from the 3' end in a processive and synchronous manner. Removal of the poly(A) (less than or equal to 100 residues) and approximately 100 adjacent residues from human fibroblast beta 1 interferon mRNA (native length, 900 residues, including a 3'-noncoding region of 203 residues) did not alter the translational activity or the functional stability of this mRNA in Xenopus oocytes, whereas deletion of the poly(A) and approximately 200 adjacent residues decreased its translational efficiency. On the other hand, removal of the poly(A) (approximately 200 residues) and approximately 200 adjacent residues from human fibroblast beta 2 interferon mRNA (native length, 1300 residues) did not alter the translational activity or the functional stability of this molecule in oocytes. Thus, neither the poly(A) nor large segments of the 3'-noncoding region (which includes the hexanucleotide A-A-U-A-A-A sequence, at least in the case of beta 1 mRNA) are required for the maintenance of the functional stability of human beta 1 and beta 2 interferon mRNAs in Xenopus oocytes.

Heterogeneity of interferon mRNA species from Sendai virus-induced human lymphoblastoid (Namalva) cells and Newcastle disease virus-induced murine fibroblastoid (L) cells

Cytoplasmic polyadenylated RNA preparations obtained from Sendai-induced human lymphoblastoid (Namalva) cells and from Newcastle disease virus (NDV)-induced murine (L) cells were denatured in 10-12.5 mM CH3HgOH and then electrophoresed in 2% agarose tube gels containing 10 mM CH3HgOH, the RNA eluted from gel slices and translationally active interferon mRNA species located using the Xenopus oocyte assay. The interferons synthesized were characterized as alpha or beta types based on neutralization tests using specific antisera against human or murine interferon-alpha and interferon-beta. At least two species of mRNA for human interferon-alpha and two for human interferon-beta were detected in RNA from Sendai-induced Namalva cells. These are (approximate mRNA length in parentheses) alpha (1.3 kb), alpha (1.9 kb), beta (1.1 kb) and beta (1.9 kb). Two populations of murine interferon mRNA of lengths approximately 1.4 kb and 3 kb were detected in mRNA preparations from NDV-induced L cells by electrophoresis. However, since the translation products of each of these two populations of mRNA consist of both murine interferon-alpha and murine interferon-beta it is likely that both the 1.4 kb and 3 kb populations contain at least one species each of murine interferon-alpha and murine interferon-beta mRNA.

Heterogeneity of poly(I) x poly(C)-induced human fibroblast interferon mRNA species

Three classes of human interferons (IFNs) have been defined on the basis of their immunological properties: the 'Le' or 'alpha' IFN, mainly derived from leukocyte or lymphoblastoid cells; the 'F' or 'beta' IFN, mainly derived from fibroblast cultures; and the 'T', 'immune' or 'gamma' IFN, mainly derived from mitogen- or antigen-stimulated lymphoid cells. Whereas several individual species of Le IFN have been purified to homogeneity, it is generally considered that F IFN represents a single protein. Thus current efforts to clone human fibroblast IFN mRNA sequences are based on the observation that F IFN mRNA sediments in sucrose gradients as a single RNA species of size corresponding to 12-14 S (refs 7-10). We show here, using gel electrohporesis of mRNA, that two populations of translationally active human fibroblast IFN mRNA molecules exist--an abundant '14 S' species and a scarce '11 S' species. Microinjection of either species of mRNA into Xenopus oocytes leads to the synthesis of biologically active F-type human IFN. These data agree with and complement recent RNA hybridization studies of Weissenbach et al.

The Xenopus oocyte as a surrogate secretory system. The specificity of protein export

Combining messenger RNA from one kind of secretory cell with the cytoplasm of another such cell can reveal the nature and specificity of protein export mechanisms. We show that messenger RNAs from secretory cells of chickens, rats, mice, frogs, guinea-pigs, locusts and barley plants, when injected into Xenopus oocytes, direct the synthesis and export of proteins. Chicken ovalbumin, Xenopus albumin, mouse thyroid-stimulating hormone, locust vitellin and guinea-pig milk proteins were identified using specific antibodies, whilst chicken lysozyme and ovomucoid, rat albumin, Xenopus vitellogenin and rat seminal vesicle basic proteins were identified provisionally from their molecular weights. Certain endogenous proteins are sequestered and secreted although most oocyte proteins are not exported. Similarly the major polyoma viral protein and the simian virus 40 and polyoma tumour antigens are retained within the oocyte. Radioactive proteins exported by oocytes programmed with chicken oviduct or Xenopus liver RNA are not re-exported in detectable amounts when injected into fresh oocytes, nor is there secretion of chicken oviduct or guinea-pig mammary gland primary translation products prepared using wheat germ extracts. Thus the export of secretory proteins from oocytes cannot be explained by leakage and may require a cotranslational event. The secretory system of the oocyte is neither cell-type nor species-specific yet is highly selective. We suggest that the oocyte can be used as a general surrogate system for the study of gene expression, from transcription through translation to the final subcellular or extracellular destination of the processed protein.

Le interferon mRNA from human fibroblasts

Human F and Le interferon can be clearly distinguished on the basis of different antigenic properties and host range. After inoculation with Newcastle disease virus (NDV), GM-258 fibroblasts produced Le as well as F interferon; in contrast, only F interferon was detectable after stimulation with poly(I) . poly(C). Polyadenylylated mRNA isolated from fibroblasts induced with poly(I) . poly(C) or NDV was injected into Xenopus laevis oocytes and the interferon activities thus produced were analyzed. Only F interferon production was demonstrable in oocytes injected with mRNA from cells induced with poly(I) . poly(C), whereas both F and Le interferons were made in oocytes injected with mRNA from NDV-induced cultures. The time course of accumulation of F and Le interferon mRNAs in NDV-induced cells corresponded to the kinetics of F and Le interferon synthesis in intact cells. The ratio of F and Le interferons made in oocytes was similar to that observed in intact GM-258 cells. F and Le interferon mRNA activities isolated from GM-258 cells could not be separated by sucrose density gradient centrifugation. However, the profile of F mRNA activity was more heterogeneous and its peak sedimented somewhat more slowly than that of Le interferon mRNA. These results suggest that the varying ratios of F and Le interferon synthesis in different cells after different modes of stimulation are determined at the level of mRNA. The induction mechanisms of F and Le interferon mRNA synthesis appear to be closely related but not identical.

Synthesis of new proteins associated with the induction of interferon in human fibroblast cells

The relative amounts of translatable cellular mRNAs and newly synthesized cellular proteins were examined in poly(I) x poly(C)-induced human fibroblast cells early during induction. At this time interferon and interferon mRNA synthesis are maximal and cells have not acquired their antiviral thesis are maximal and cells have not acquired their antiviral state. Translation of the mRNA from poly(I) x poly(C)-induced cells in a wheat germ cell-free system led to the synthesis of a [35S]methionine-labeled 22,000-dalton protein that is precipitated by antiserum to highly purified human fibroblast interferon. The synthesis of this protein was detected only with the mRNA preparations that, when translated in Xenopus oocytes, coded for the synthesis of biologically active human interferon. Two-dimensional gel analysis of the [35S]methionine-labeled polypeptides translated from the total mRNA of the induced and uninduced cells revealed the presence of 23 new proteins that were translated from mRNAs of the induced cells but not from the mRNAs of the controls. These polypeptides ranged from 15,000 to 70,000 daltons. Thirteen of these proteins were detected in induced cells labeled with [35S]methionine. It is concluded that, in human fibroblasts, poly(I) x poly(C) induces, in addition to interferon, the synthesis of a variety of "interferon-associated" proteins.

Regulation of rat-liver tyrosine-aminotransferase mRNA by hydrocortisone and by N6,O2'-dibutyryladenosine 3',5'-phosphate

The intraperitoneal injection of either hydrocortisone of N6,O2'-dibutyryladenosine 3',5'-phosphate (Bt2cAMP) results in a specific increase in functional tyrosine aminotransferase mRNA (mRNATAT) activity in rat liver that is proportional to the degree of enzyme induction. Both require continuous RNA synthesis. There are several differences in the response to these inducers: (a) the magnitude of the increase is greater following hydrocortisone injection than after Bt2cAMP; (b) the peak response is seen within 1 h following the injection of Bt2cAMP as compared to the 5 h required for the maximal response following hydrocortisone injection; (c) finally, although both responses are rapid, the lag period which precedes the accumulation of functional tyrosine aminotransferase mRNA activity following the injection of hydrocortisone is at least 20 min whereas following Bt2cAMP it is 5-10 min. The administration of actinomycin D to rats 5 h after they were treated with hydrocortisone causes an additional twofold increase in tyrosine aminotransferase enzymatic activity, a phenomenon known as superinduction, but does not prevent the normal decrease in its mRNA seen at this time. This dissociation of enzyme and mRNA activities indicates that superinduction of tyrosine aminotransferase is not due to a selective stabilization of the mRNA which codes for this protein.

Regulation of the stability of poly(I)xpoly(C)-induced human fibroblast interferon mRNA: selective inactivation of interferon mRNA and lack of involvement of 2',5'-oligo(A) synthetase activation during the shutoff of interferon production

The inactivation of interferon mRNA during the shutoff phase of interferon production in poly(I)xpoly(C)-induced human fibroblast cultures is selective. We have determined that the shutoff of interferon production, which takes place from 3 to 8 hr after the beginning of induction, is not associated with an appreciable declined in the rate of bulk cellular protein synthesis or of cellular protein secretion. While the amount of translatable interferon mRNA declined markedly during the shutoff phase, the level of translatable bulk cellular mRNA and the stability of [3H]uridine-labeled mRNA were unaffected. Superinduction with actinomycin D selectively stabilized interferon mRNA with no apparent effect on the stability of bulk cellular mRNA. Furthermore, an activation of the 2',5'-oligo(A) synthetase/endonuclease system does not appear to be involved in the shutoff phenomenon. Uninduced FS-4 cells contained a low basal level of 2'5'-oligo(A) synthetase activity, which was unchanged in poly(I)xpoly(C)-induced cells during the shutoff phase. Treatment of FS-4 cells with interferon for 16-18 hr prior to induction increased the enzyme activity by approximately 200-fold. However, this did not inhibit interferon production after induction with poly(I)xpoly(C) alone or after superinduction with cycloheximide or actinomycin D or both. Furthermore, the rates of decay of interferon production were comparable in cells with either a basal or an increased level of 2',5'-oligo(A) synthetase. Thus a 200-fold increase in 2',5'-oligo(A) synthetase level did not affect either the stability of interferon mRNA or the efficacy of interferon superinduction by metabolic inhibitors.

Identification of the translation products of human fibroblast interferon mRNA in reticulocyte lysates

Messenger RNA was purified from human foreskin fibroblasts FS11, a high interferon-producer line, after induction with synthetic double-stranded RNA. The mRNA was translated in a cell-free protein-synthesis system from rabbit reticulocytes. The translation products, containing biologically active human interferon, were immunoprecipitated by a serum from rabbits immunized against partially purified interferon. Analysis of the immunoprecipitate by polyacrylamide gel electrophoresis in dodecylsulfate shows that the product of human fibroblast interferon mRNA is a 23000-Mr polypeptide. Methods are described for the synthesis and rapid identification of this polypeptide, which should be useful for structural analysis of interferon and isolation of its mRNA.

Superinduction of alpha 2u globulin by actinomycin D: evidence for drug-mediated increase in alpha 2u mRNA

Actinomycin D, an inhibitor of DNA-dependent RNA synthesis, increased the hepatic concentration of alpha 2u globulin, an androgen-inducible protein in the rat. Spayed female rats with a marginally induced state of alpha 2u synthesis showed an approximately 5-fold increase in hepatic alpha 2u globulin within 3-6 hr after treatment with actinomycin D. Initial treatment of these animals with 5 alpha-dihydrotestosterone, followed by actinomycin D, resulted within 2-3 hr in a more than 2-fold increase in hepatic alpha 2u globulin compared to animals treated with the androgen alone. In spite of inhibition of hepatic synthesis of poly(A)-containing RNA to less than 25% of control, superinduction with actinomycin D resulted in a parallel increase in the translatable mRNA for alpha 2u globulin. These results showing increase in both alpha 2u globulin and its translatable mRNA after superinduction with actinomycin D support the concept of post-transcriptional repression of alpha2u synthesis.

Does 3'-terminal poly(A) stabilize human fibroblast interferon mRNA in oocytes of Xenopus laevis?

Polynucleotide phosphorylase (polyribonucleotide:orthophosphate nucleotidyltransferase, EC 2.7.7.8) purified from Escherichia coli was used enzymatically to deadenylate polyadenylated human fibroblast interferon mRNA preparations obtained from human diploid fibroblasts (FS-4 strain) induced by poly(I)-poly(C) (20 microgram/ml) in the presence of cycloheximide (50 microgram/ml, 4 hr). Both the polyadenylated and the deadenylated interferon mRNA preparations were translated into biologically active human interferon when injected into oocytes of Xenopus laevis. In the oocytes the functional stability of deadenylated interferon mRNA was indistinguishable from that of polyadenylated interferon mRNA.

Cycloheximide causes increased accumulation of translatable mRNA for tyrosine aminotransferase and tryptophan oxygenase in livers of cortisol-treated rats

Messenger RNA activities for two cortisol-inducible enzymes, tyrosine aminotransferase and tryptophan oxygenase, have been determined by translation in a wheat germ system. The effects of cycloheximide on the two mRNA activities have been evaluated. Cortisol leads to an increase of the translatable mRNAs for tyrosine aminotransferase and tryptophan oxygenase with a maximum at approximately 6 h. Cycloheximide was administered 4 h after treatment with cortisol; 2 h later, the activities of tyrosine aminotransferase and tryptophan oxygenase mRNA had increased five-fold and two-fold, respectively, compared to the activities reached with cortisol alone. Thereafter the amount of the two translatable mRNAs declined, though 14 h after cortisol administration the mRNA activities were still several fold higher than in control animals. Application of alpha-amanitin together with cycloheximide did not prevent an increased accumulation of specific translatable mRNAs. The increase in tyrosine aminotransferase and tryptophan oxygenase activity by cortisol was immediately blocked by cycloheximide. Whereas tryptophan oxygenase activity rapidly declined after cycloheximide application, tyrosine aminotransferase activity remained at the same level. Approximately 4 h thereafter, both enzyme activities increased again.

Induction and decay of human fibroblast interferon mRNA

Polyadenylylated interferon mRNA, obtained from induced human fibroblasts, was quantitatively assayed by synthesis of biologically active human interferon in Xenopus laevis oocytes. The assay for interferon mRNA was used to distinguish between various hypotheses relating to interferon induction and biosynthesis. The data demonstrate that on induction with poly(I-poly(C) human fibroblasts accumulate interferon mRNA for 1-1.5 hr, after which time the mRNA is rapidly degraded with a half-life (t 1/2) of 18 min. Treatment of cells with cycloheximide prolongs the period of accumulation to 3 hr and decreases the rate of mRNA inactivation (t 1/2 = 49 min). Treatment with actinomycin D decreases the rate of inactivation still further (t 1/2 = 68 min). A comparison of cellular interferon synthesis with the relative amounts of interferon m RNA after simple induction or inductionin the presence of the inhibitors (superinduction) indicated a general correlation. Thus, on induction, the genes for interferon are activated to produce a transcript for a short time. The superinducing treatments prolong the period of accumulation and decrease the rate of degradation of this transcript.