FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W1988981229> ?p ?o ?g. }

• W1988981229 endingPage "17428" @default.
• W1988981229 startingPage "17423" @default.
• W1988981229 abstract "Double-stranded RNA-dependent protein kinase (PKR) is suggested to play an important role in both the antiviral and antiproliferative arms of the interferon response. To gain insights into the molecular mechanisms underlying PKR's growth regulatory properties, we examined the biological and biochemical properties of PKR variants containing either a mutation in catalytic domain II (PKR-M1) or a deletion of RNA binding domain I (PKR-M7) in both reticulocyte translation extracts and in vitro kinase assays with purified reagents and compared these results with those using the same mutants stably expressed in vivo. While wild-type PKR (PKR-WT) efficiently inhibited mRNA translation in a reticulocyte extract, the inactive PKR-M1 had no effect. The PKR-M7 mutant was modestly inhibitory in this assay. The PKR-M1 variant was able to reverse the translational inhibitory effects and increased eukaryotic initiation factor (eIF)-2α phosphorylation levels caused by addition of double-stranded RNA to reticulocyte extract, whereas PKR-M7 could not. Both PKR-M1 and PKR-M7 functioned as transdominant inhibitors of PKR-WT in our in vitro kinase assays. While the inhibition by PKR-M1 required a vast excess of mutant to shut down PKR function, PKR-M7 inhibited PKR-WT at approximately stoichiometric levels. To complement these experiments, we compared growth rates and α phosphorylation levels in transformed cell lines overexpressing either PKR-M1 or PKR-M7. Levels of endogenous eIF-2α phosphorylation were significantly more diminished in PKR-M7 overexpressing cells compared with PKR-M1. These paradoxical data will be discussed in terms of the potential molecular mechanisms underlying malignant transformation caused by the PKR variants. Double-stranded RNA-dependent protein kinase (PKR) is suggested to play an important role in both the antiviral and antiproliferative arms of the interferon response. To gain insights into the molecular mechanisms underlying PKR's growth regulatory properties, we examined the biological and biochemical properties of PKR variants containing either a mutation in catalytic domain II (PKR-M1) or a deletion of RNA binding domain I (PKR-M7) in both reticulocyte translation extracts and in vitro kinase assays with purified reagents and compared these results with those using the same mutants stably expressed in vivo. While wild-type PKR (PKR-WT) efficiently inhibited mRNA translation in a reticulocyte extract, the inactive PKR-M1 had no effect. The PKR-M7 mutant was modestly inhibitory in this assay. The PKR-M1 variant was able to reverse the translational inhibitory effects and increased eukaryotic initiation factor (eIF)-2α phosphorylation levels caused by addition of double-stranded RNA to reticulocyte extract, whereas PKR-M7 could not. Both PKR-M1 and PKR-M7 functioned as transdominant inhibitors of PKR-WT in our in vitro kinase assays. While the inhibition by PKR-M1 required a vast excess of mutant to shut down PKR function, PKR-M7 inhibited PKR-WT at approximately stoichiometric levels. To complement these experiments, we compared growth rates and α phosphorylation levels in transformed cell lines overexpressing either PKR-M1 or PKR-M7. Levels of endogenous eIF-2α phosphorylation were significantly more diminished in PKR-M7 overexpressing cells compared with PKR-M1. These paradoxical data will be discussed in terms of the potential molecular mechanisms underlying malignant transformation caused by the PKR variants. The interferon-induced, ds1RNA-dependent protein kinase, 1The abbreviations used are: dsdouble-strandedPKRdsRNA-dependent protein kinaseeIFeukaryotic initiation factor. PKR, is the most studied member of the eIF-2α-specific kinase subfamily(1Meurs E. Chong K.L. Galabru J. Thomas N. Kerr I. Williams B.R.G. Hovanessian A.G. Cell. 1990; 62: 379-390Abstract Full Text PDF PubMed Scopus (822) Google Scholar, 2Samuel C.E. J. Biol. Chem. 1993; 268: 7603-7606Abstract Full Text PDF PubMed Google Scholar). Other members of this family include the reticulocyte lysate heme-sensitive eIF-2α kinase, commonly referred to as HCR or HRI (3Chen J. Pal J. Throop M.S. Gehrke L. Kuo I. Pal J.K. Brodsky M. London I.M. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 7729-7733Crossref PubMed Scopus (168) Google Scholar) and the yeast GCN-2 protein kinase which is involved in the translational regulation of GCN-4(4Dever T.E. Feng L. Wek R.C. Cigan A.M. Donahue T.F. Hinnebusch A.G. Cell. 1992; 68: 585-596Abstract Full Text PDF PubMed Scopus (569) Google Scholar). PKR is a cAMP-independent, serine-threonine kinase, characterized by two distinct kinase activities: first an autophosphorylation, which represents the activation reaction; and second, phosphorylation of eIF-2α(5Galabru J. Hovanessian A.G. J. Biol. Chem. 1987; 262: 15538-15544Abstract Full Text PDF PubMed Google Scholar, 6Hovanessian A.G. J. Interferon Res. 1989; 9: 641-647Crossref PubMed Scopus (163) Google Scholar). This second phosphorylation event can lead to limitations in functional eIF-2 and a resultant inhibition in protein synthesis inhibition(7Hershey J.W.B. Annu. Rev. Biochem. 1991; 60: 717-755Crossref PubMed Scopus (845) Google Scholar, 8Merrick W.C. Microbiol. Rev. 1992; 56: 291-315Crossref PubMed Google Scholar, 9Rhoads R.E. J. Biol. Chem. 1993; 268: 3017-3020Abstract Full Text PDF PubMed Google Scholar). In addition to its translational regulatory role, recent evidence now suggests that PKR may play a role in signal transduction and transcriptional control, possibly through the I-κB/NF-κB pathway(10Kumar A. Haque J. Lacoste J. Hiscott J. Williams B.R.G. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 6288-6292Crossref PubMed Scopus (514) Google Scholar, 11Maran A. Maitra R.K. Kumar A. Dong B. Xiao W. Li G. Williams B.R.G. Torrence P.F. Silverman R.H. Science. 1994; 265: 789-792Crossref PubMed Scopus (211) Google Scholar). double-stranded dsRNA-dependent protein kinase eukaryotic initiation factor. PKR was initially identified as an interferon-inducible enzyme that may become activated during virus infection and thus play an important role in the regulation of protein synthesis in infected cells(12Samuel C.E. Virology. 1991; 183: 1-11Crossref PubMed Scopus (526) Google Scholar, 13Katze M.G. Semin. Virol. 1993; 4: 259-268Crossref Scopus (47) Google Scholar, 14Katze M.G. J. Interferon Res. 1992; 12: 241-248Crossref PubMed Scopus (85) Google Scholar, 15Katze M.G. Trends Microbiol. 1995; 3: 75-78Abstract Full Text PDF PubMed Scopus (160) Google Scholar). More recently PKR, which is constitutively expressed in mammalian cells, has been implicated as playing a role in the control of cell growth and proliferation(16Koromilas A.E. Roy S. Barber G.N. Katze M.G. Sonneberg N. Science. 1992; 257: 1685-1689Crossref PubMed Scopus (496) Google Scholar, 17Lengyel P. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 55893-55895Crossref Scopus (194) Google Scholar, 18Meurs E. Galabru J. Barber G.N. Katze M.G. Hovanessian A.G. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 232-236Crossref PubMed Scopus (417) Google Scholar). The expression of wild-type human PKR cDNA in yeast caused increased phosphorylation of eIF-2α, a reduction in protein synthetic rates, and a decreased proliferation rate(19Chong K.L. Schappert K. Meurs E. Feng F. Donahue T.F. Friesen J.D. Hovanessian A.G. Williams B.R.G. EMBO J. 1992; 11: 1553-1562Crossref PubMed Scopus (290) Google Scholar, 20Dever T.E. Chen J.-J. Barber G.N. Cigan A.M. Feng L. Donahue T.F. London I.M. Katze M.G. Hinnebusch A.G. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 4616-4620Crossref PubMed Scopus (188) Google Scholar). Conversely, the transfection of cDNAs encoding nonfunctional PKR variants into murine cells resulted in malignant transformation, as measured by their ability to produce tumors in nude mice, suggesting a possible tumor suppressor role for PKR(16Koromilas A.E. Roy S. Barber G.N. Katze M.G. Sonneberg N. Science. 1992; 257: 1685-1689Crossref PubMed Scopus (496) Google Scholar, 18Meurs E. Galabru J. Barber G.N. Katze M.G. Hovanessian A.G. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 232-236Crossref PubMed Scopus (417) Google Scholar). The nonfunctional variants of PKR have included a mutant in catalytic domain II, in which the essential lysine at position 296 has been changed to an arginine (18Meurs E. Galabru J. Barber G.N. Katze M.G. Hovanessian A.G. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 232-236Crossref PubMed Scopus (417) Google Scholar), the DII or PKR-M1 variant, as well as a mutant in which a crucial 6-amino acid segment (Δ6) has been deleted(16Koromilas A.E. Roy S. Barber G.N. Katze M.G. Sonneberg N. Science. 1992; 257: 1685-1689Crossref PubMed Scopus (496) Google Scholar). More recently we have determined that introduction of PKR regulatory domain variants, which completely lack the first RNA binding domain (referred to as PKR-M7), into NIH 3T3 cells also induced their malignant transformation (21Barber G.N. Wambach M. Thompson S. Jagus R. Katze M.G. Mol. Cell. Biol. 1995; 15: 3138-3146Crossref PubMed Scopus (139) Google Scholar). The current study was undertaken to begin to define the molecular mechanisms underlying the transforming action of two separate PKR variants, the catalytically dead PKR-M1 (18Meurs E. Galabru J. Barber G.N. Katze M.G. Hovanessian A.G. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 232-236Crossref PubMed Scopus (417) Google Scholar, 22Katze M.G. Wambach M. Wong M.-L. Garfinkel M.S. Meurs E. Chong K.L. Williams B.R.G. Hovanessian A.G. Barber G.N. Mol. Cell. Biol. 1991; 11: 5497-5505Crossref PubMed Scopus (174) Google Scholar) and the RNA binding mutant, PKR-M7, which retains minimal function and is severely deficient in the binding of dsRNA(21Barber G.N. Wambach M. Thompson S. Jagus R. Katze M.G. Mol. Cell. Biol. 1995; 15: 3138-3146Crossref PubMed Scopus (139) Google Scholar). Utilizing purified recombinant PKR variants, we present evidence that PKR-M1 and PKR-M7 have different effects on protein synthetic activity in reticulocyte lysate, although both can function in vitro as dominant negative inhibitors of the wild-type PKR (PKR-WT). However, our in vitro data, taken together with the eIF-2α phosphorylation results acquired from the PKR-M1 and PKR-M7 overexpressing cell lines, suggests that the transdominant transforming action of PKR-M7 is through a translational regulatory pathway, whereas PKR-M1 may trigger transformation through an eIF-2α-independent pathway. Expression of recombinant wild-type PKR was carried out in Escherichia coli as described by Barber et al.(23Barber G.N. Tomita J. Hovanessian A.G. Meurs E. Katze M.G. Biochemistry. 1991; 30: 10356-10361Crossref PubMed Scopus (51) Google Scholar). The PKR-M1 catalytically inactive domain II mutant (Lys → Arg296) was constructed and expressed in insect cells using the baculovirus expression system as described previously(24Barber G.N. Tomita J. Garfinkel M.S. Hovanessian A.G. Meurs E. Katze M.G. Virology. 1992; 191: 670-679Crossref PubMed Scopus (46) Google Scholar). The PKR-M7 regulatory domain variant (24Barber G.N. Tomita J. Garfinkel M.S. Hovanessian A.G. Meurs E. Katze M.G. Virology. 1992; 191: 670-679Crossref PubMed Scopus (46) Google Scholar) was expressed as a histidine-tagged fusion protein. This vector was constructed by introducing a NdeI site at the second ATG methionine at amino acid position 98 with the resultant fragment cloned into the pET15b vector (Novagen)(25Lee T.G. Tang N. Thompson S. Miller J. Katze M.G. Mol. Cell. Biol. 1994; 14: 2331-2342Crossref PubMed Google Scholar). Extracts containing the PKR variants were prepared as earlier described(23Barber G.N. Tomita J. Hovanessian A.G. Meurs E. Katze M.G. Biochemistry. 1991; 30: 10356-10361Crossref PubMed Scopus (51) Google Scholar, 24Barber G.N. Tomita J. Garfinkel M.S. Hovanessian A.G. Meurs E. Katze M.G. Virology. 1992; 191: 670-679Crossref PubMed Scopus (46) Google Scholar). Both the PKR-WT and PKR-M1 proteins were purified utilizing the PKR monoclonal antibody (26Laurent A.G. Krust B. Galabru J. Svab J. Hovanessian A.G. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 4341-4345Crossref PubMed Scopus (108) Google Scholar) as described earlier(5Galabru J. Hovanessian A.G. J. Biol. Chem. 1987; 262: 15538-15544Abstract Full Text PDF PubMed Google Scholar, 22Katze M.G. Wambach M. Wong M.-L. Garfinkel M.S. Meurs E. Chong K.L. Williams B.R.G. Hovanessian A.G. Barber G.N. Mol. Cell. Biol. 1991; 11: 5497-5505Crossref PubMed Scopus (174) Google Scholar). The histidine fusion PKR-M7 variant was purified by Ni(II) column (Novagen) according to manufacturer's protocol(25Lee T.G. Tang N. Thompson S. Miller J. Katze M.G. Mol. Cell. Biol. 1994; 14: 2331-2342Crossref PubMed Google Scholar). Protein concentrations were determined by comparing the purified PKR proteins with bovine serum albumin standards after SDS-polyacrylamide gel electrophoresis and Coomassie Blue staining. PKR autophosphorylation and substrate phosphorylation assays were conducted as described by Katze et al.(22Katze M.G. Wambach M. Wong M.-L. Garfinkel M.S. Meurs E. Chong K.L. Williams B.R.G. Hovanessian A.G. Barber G.N. Mol. Cell. Biol. 1991; 11: 5497-5505Crossref PubMed Scopus (174) Google Scholar). Briefly purified recombinant PKR was incubated in low salt buffer containing 100 mM KCl, 25 mM Tris-HCl, pH 7.2, 10% glycerol, 2 mM MgCl2, 2 mM MnCl2, 1 mM dithiothreitol, 3 mg/ml bovine serum albumin, and 5 εM [γ-32P]ATP, 1000 Ci/mmol). Activator poly(I): poly(C) was added to the mix together with purified rabbit eIF-2 as described(22Katze M.G. Wambach M. Wong M.-L. Garfinkel M.S. Meurs E. Chong K.L. Williams B.R.G. Hovanessian A.G. Barber G.N. Mol. Cell. Biol. 1991; 11: 5497-5505Crossref PubMed Scopus (174) Google Scholar, 27Carroll K. Elroy-Stein O. Moss R. Jagus R. J. Biol. Chem. 1993; 268: 12837-12842Abstract Full Text PDF PubMed Google Scholar). Rabbit reticulocyte lysates were prepared and used as described(28Jagus R. Methods Enzymol. 1987; 152: 267-276Crossref PubMed Scopus (26) Google Scholar). For analysis in PKR overexpressing cell lines, cells at similar densities were rinsed twice in ice-cold phosphate-buffered saline and lysed in 20 mM HEPES, pH 7.2, 2 mM EDTA, 100 mM KCl, .05% SDS, 0.5% Elugent, 10% glycerol, 20 εg/ml chymostatin, 50 nM microcystin, 1 mM dithiothreitol. The 10,000 × g supernatant was clarified with BPA-1000 (Toso-Haas, Philadelphia). Supernatant (100 εg of protein) was first subjected to immunoprecipitation using eIF-2α-specific monoclonal antibody. Immunoprecipitates were then resuspended in the VSIEF sample buffer and fractionated by vertical slab gel electrophoresis (27Carroll K. Elroy-Stein O. Moss R. Jagus R. J. Biol. Chem. 1993; 268: 12837-12842Abstract Full Text PDF PubMed Google Scholar) to separate phosphorylated from nonphosphorylated forms of eIF-2α. For reticulocyte analysis, aliquots of the lysate were taken and similarly treated. Proteins were transferred to Immobilon P and subjected to immunoblotting using monoclonal antibody to eIF-2α(27Carroll K. Elroy-Stein O. Moss R. Jagus R. J. Biol. Chem. 1993; 268: 12837-12842Abstract Full Text PDF PubMed Google Scholar, 29Barber G.N. Thompson S. Lee T.G. Strom T. Jagus R. Darveau A. Katze M.G. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 4278-4282Crossref PubMed Google Scholar). A detailed description of the PKR-M1 and PKR-M7 cell lines, including an analysis of functional activity and physical levels of PKR variants, can be found elsewhere(18Meurs E. Galabru J. Barber G.N. Katze M.G. Hovanessian A.G. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 232-236Crossref PubMed Scopus (417) Google Scholar, 21Barber G.N. Wambach M. Thompson S. Jagus R. Katze M.G. Mol. Cell. Biol. 1995; 15: 3138-3146Crossref PubMed Scopus (139) Google Scholar, 30Meurs E. Watanabe Y. Barber G.N. Katze M.G. Chong K.L. Williams B.R.G. Hovanessian A.G. J. Virol. 1992; 66: 5805-5814Crossref PubMed Google Scholar). To determine growth rates, cells were plated at 5 × 104/60-mm dish and cell density determined at absorbance at 600 nm and recorded every 24 h. Prior to trypsinization, cell monolayers were extensively washed to remove dead cells. Cell medium (Dulbecco's modified Eagle's medium, 200 εg/ml G418, plus 10% fetal calf serum) was replaced daily. We previously reported that cell lines overexpressing the PKR-M1 and PKR-M7 variants were malignantly transformed and tumorigenic in nude mice(18Meurs E. Galabru J. Barber G.N. Katze M.G. Hovanessian A.G. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 232-236Crossref PubMed Scopus (417) Google Scholar, 21Barber G.N. Wambach M. Thompson S. Jagus R. Katze M.G. Mol. Cell. Biol. 1995; 15: 3138-3146Crossref PubMed Scopus (139) Google Scholar). The current study was initiated to delineate molecular mechanisms of transformation by these PKR variants. To accomplish this goal, highly purified PKR proteins were prepared and their activities analyzed both in reticulocyte lysate and in vitro kinase assays. The recombinant PKR-WT and PKR-M1 kinases were prepared from E. coli(23Barber G.N. Tomita J. Hovanessian A.G. Meurs E. Katze M.G. Biochemistry. 1991; 30: 10356-10361Crossref PubMed Scopus (51) Google Scholar) and baculovirus-infected insect cells(24Barber G.N. Tomita J. Garfinkel M.S. Hovanessian A.G. Meurs E. Katze M.G. Virology. 1992; 191: 670-679Crossref PubMed Scopus (46) Google Scholar), respectively, and immunopurified using PKR-specific monoclonal antibody bound to CnBr-activated Sepharose(26Laurent A.G. Krust B. Galabru J. Svab J. Hovanessian A.G. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 4341-4345Crossref PubMed Scopus (108) Google Scholar). Since this monoclonal antibody fails to react with PKR variants lacking RNA binding domain I(22Katze M.G. Wambach M. Wong M.-L. Garfinkel M.S. Meurs E. Chong K.L. Williams B.R.G. Hovanessian A.G. Barber G.N. Mol. Cell. Biol. 1991; 11: 5497-5505Crossref PubMed Scopus (174) Google Scholar), we expressed PKR-M7 as a histidine fusion protein which can be readily purified by passage over a nickel column. The recombinant wild-type and mutant proteins were first analyzed for their effects on protein synthesis regulation in reticulocyte extracts (Fig. 1A). While PKR-WT dramatically reduced protein synthetic rates, the catalytically inactive PKR-M1 had no effect on mRNA translation in the reticulocyte extracts. The decrease caused by PKR-WT occurred in the absence of added dsRNA, thus suggesting that activation is caused by either endogenous RNAs or other polyanions present in the lysate. Moreover the protein kinase is already partially active when purified from E. coli(23Barber G.N. Tomita J. Hovanessian A.G. Meurs E. Katze M.G. Biochemistry. 1991; 30: 10356-10361Crossref PubMed Scopus (51) Google Scholar). In contrast to PKR-M1, the RNA binding domain variant, PKR-M7, did reduce translation rates in these extracts at high concentrations, although minimally compared with PKR-WT. We then correlated the variants' effects on translation with an analysis of endogenous eIF-2α phosphorylation in these extracts (Fig. 1B). Several conclusions can be made concerning these experiments. (i) In the absence of added PKR, levels of α phosphorylation in the lysates are minimal (Fig. 1B, lane 3). (ii) Addition of 10 εg/ml PKR-WT caused essentially a 100% conversion to phosphorylated eIF-2α in the reticulocyte extracts (lane 8). (iii) Addition of PKR-M1 had little effect on endogenous α phosphorylation levels and even appeared to reduce levels at the lower concentrations tested. (iv) Due to its minimal functional activity, PKR-M7 did increase α phosphorylation, although modestly compared with PKR-WT. It should be emphasized that we measure the steady state levels of α phosphorylation and not the rate of phosphorylation by PKR in these in vitro assays. We proceeded to utilize the reticulocyte system to further examine the differing biological properties of PKR-M1 and PKR-M7. It is well established that addition of dsRNA to reticulocyte lysates severely compromises mRNA translation rates presumably due to activation of the endogenous rabbit PKR(31Farrell P.J. Balkow K. Hunt T. Jackson R.J. Trachsel H. Cell. 1977; 11: 187-200Abstract Full Text PDF PubMed Scopus (446) Google Scholar). We took advantage of this system to test whether PKR-M1 or PKR-M7 could interfere with activation of the endogenous kinase by the addition of dsRNA. In the presence of 150 ng/ml poly(I): poly(C), protein synthesis became inhibited after a lag period, reducing the incorporation of [14C]valine by 35-40% after a 30-min incubation (Fig. 2A). The addition of increasing levels of PKR-M1 restored protein synthesis nearly to control levels. Approximately 7.5 εg/ml of the PKR-M1 variant, equivalent to to more than 10 pmol of PKR-M1/100 εl, was needed to completely restore protein synthetic activity, a level considerably higher than the estimated endogenous level of PKR of less than 0.1 pmol/100 εl of reticulocyte lysate.2 2M. Gray and R. Jagus, unpublished observations. Since further addition of dsRNA could reverse this restoration (data not shown), it is likely that PKR-M1 is functioning to inhibit the endogenous PKR by sequestering the activator. In contrast to the recovery of mRNA translation caused by PKR-M1, the regulatory domain variant PKR-M7 failed to reverse the inhibitory effects of dsRNA (Fig. 3A). Consistent with its inability to bind to and therefore sequester dsRNA, the highest concentrations of PKR-M7 had modest effects on protein synthetic rates. Examination of endogenous eIF-2α phosphorylation levels confirmed the protein synthesis data. Whereas the catalytically inactive PKR-M1 variant quantitatively reduced α phosphorylation levels in the presence of dsRNA (Fig. 2B), PKR-M7 had no effects on these phosphorylation levels (Fig. 3B).Figure 3:Effects of PKR-M7 on translational inhibition by dsRNA addition and hemin deprivation in the reticulocyte translation system. A, cell-free extracts of rabbit reticulocytes were incubated as indicated in the absence or presence of 150 ng/ml dsRNA poly(I): poly(C) and in the presence of 1 εM hemin. Separate incubations were carried out in the absence of hemin (-h). Indicated amounts of pure PKR-M7 were added to reactions and protein synthetic activity measured as described above. B, eIF-2α phosphorylation levels were measured separately in the presence of dsRNA (+dsRNA) and the absence of hemin (-h) after incubation with increasing amounts of PKR-M7 as described in the legend to Fig. 1.View Large Image Figure ViewerDownload Hi-res image Download (PPT) The reticulocyte translation system also contains a heme-sensitive eIF-2α kinase that can be activated by incubation in the absence of added hemin(3Chen J. Pal J. Throop M.S. Gehrke L. Kuo I. Pal J.K. Brodsky M. London I.M. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 7729-7733Crossref PubMed Scopus (168) Google Scholar). This can be used to test whether the different PKR variants can act as dominant negative inhibitors by binding to eIF-2α and preventing its phosphorylation. If this occurred one would expect an enhancement of protein synthesis rates and resultant decrease in α phosphorylation levels in hemin-deprived lysates. However, neither PKR-M1 (Fig. 2A) nor PKR-M7 (Fig. 3A) were able to prevent the severe decreases in protein synthetic rates caused by the omission of hemin. Similarly neither variant prevented the increases in eIF-2α phosphorylation levels caused by the absence of hemin (Fig. 2B and 3B). These data strongly suggest that any transdominant effects of the PKR-M1 or PKR-M7 variants observed in vivo or in vitro is unlikely due to sequestration of the PKR substrate. Thus far we have examined the biological properties of the variants in reticulocyte extracts but have not yet directly demonstrated that PKR-M7 or PKR-M1 can act as transdominant inhibitors in vitro. To accomplish this and examine the stoichiometry of such a reaction, we analyzed the enzymatic activity of PKR-WT which was incubated with either the catalytic or regulatory domain variants. In the absence of any variant, the PKR-WT efficiently phosphorylated exogenously added eIF-2α in the presence of poly(I): poly(C) (Fig. 4A, lane 1). The addition of increasing amounts of PKR-M1 variant reduced the phosphorylation of eIF-2α (Fig. 4A, lanes 2-6). However, levels of PKR-M1 up to 10 times higher than the levels of wild type were required to significantly reduce α phosphorylation. We could not observe a decrease in PKR-WT autophosphorylation due to the concomitant increased phosphorylation of the 68-kDa polypeptide which likely results from trans-phosphorylation of PKR-M1 by PKR-WT (since both PKR proteins comigrate) as also observed by others(32Thomis D.C. Samuel C.E. J. Virol. 1993; 67: 7695-7700Crossref PubMed Google Scholar). PKR-M1 alone is catalytically dead and unable to autophosphorylate itself(22Katze M.G. Wambach M. Wong M.-L. Garfinkel M.S. Meurs E. Chong K.L. Williams B.R.G. Hovanessian A.G. Barber G.N. Mol. Cell. Biol. 1991; 11: 5497-5505Crossref PubMed Scopus (174) Google Scholar). Importantly, the ability of PKR-M1 to prevent phosphorylation of eIF-2α by PKR-WT could be reversed (α phosphorylation increasing 3-5-fold) by increasing the concentration of dsRNA again, suggesting that the variant functions by sequestering activator (Fig. 4B, lanes 2-4). We next examined the activity of recombinant PKR-M7 in a similar assay. When assayed alone, the PKR-M7 variant possessed minimal activity compared with PKR-WT (Fig. 5A). Although autophosphorylation of the histidine fusion PKR protein is detectable at the highest dsRNA concentrations, this mutant cannot appreciably phosphorylate the eIF-2α substrate (Fig. 5A, lanes 7 and 8). PKR-M7 is, however, an effective inhibitor of wild-type kinase function (Fig. 5B). Moreover, roughly equal amounts of PKR-M7 compared with PKR-WT can significantly reduce both the autophosphorylation and eIF-2α phosphorylating ability of the wild-type kinase. Furthermore, excess dsRNA was not able to reverse this inhibition (data not shown). This is in marked contrast to the previous experiments in which vast excesses of PKR-M1 were required for a similar effect that was reversed by increasing activator concentration. It is relevant to note that PKR-WT and PKR-M7 coincidentally comigrate due to the presence of the histidine tag despite the latter's smaller size. In contrast to the PKR-M1 variant, however, little trans-phosphorylation of PKR-M7 by PKR-WT was observed for unknown reasons. The data presented thus far demonstrated that both PKR-M1 and PKR-M7 can function as transdominant inhibitors of the wild-type kinase in vitro, resulting in reduced α phosphorylation. An important question that needed to be answered is whether these variants functioned in a similar way in vivo, inside a mammalian cell. We therefore analyzed both the growth rates and endogenous eIF-2α phosphorylation levels in PKR-M1 expressing NIH 3T3 cells and compared these cells with those expressing the PKR-M7 regulatory domain variant. A detailed description of the PKR-M1 and PKR-M7 cell lines, including analyses of PKR variant levels, can be found elsewhere(18Meurs E. Galabru J. Barber G.N. Katze M.G. Hovanessian A.G. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 232-236Crossref PubMed Scopus (417) Google Scholar, 21Barber G.N. Wambach M. Thompson S. Jagus R. Katze M.G. Mol. Cell. Biol. 1995; 15: 3138-3146Crossref PubMed Scopus (139) Google Scholar, 30Meurs E. Watanabe Y. Barber G.N. Katze M.G. Chong K.L. Williams B.R.G. Hovanessian A.G. J. Virol. 1992; 66: 5805-5814Crossref PubMed Google Scholar). Although we previously tested the tumorigenicity of these PKR-M1 cell lines(18Meurs E. Galabru J. Barber G.N. Katze M.G. Hovanessian A.G. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 232-236Crossref PubMed Scopus (417) Google Scholar), we never compared PKR-M1 and PKR-M7 growth rates nor did we examine α phosphorylation levels in the absence of virus infection or interferon treatment(30Meurs E. Watanabe Y. Barber G.N. Katze M.G. Chong K.L. Williams B.R.G. Hovanessian A.G. J. Virol. 1992; 66: 5805-5814Crossref PubMed Google Scholar). Our present analysis revealed that the growth rate of the PKR-M1 overexpressing cells did not differ from the control cells expressing the neomycin-resistant gene alone as revealed by the slope of the curve (Fig. 6A). The PKR-M1 cells did, however, grow to somewhat higher densities which quickly leveled off after approximately a week in culture. In contrast, the PKR-M7 cell lines grew both at a faster rate and to higher densities with growth rates not levelling off. We then compared endogenous eIF-2α phosphorylation levels in the three different cell lines at both day 5 and day 7 after plating (Fig. 6B). At both days, phosphorylation levels were more drastically reduced in the PKR-M7 cell line compared with the PKR-M1 cell lines. Indeed phosphorylated eIF-2α was barely detectable only at day 7 in the PKR-M7 cell extracts. Laser densitometry quantitation at day 7 revealed that eIF-2α phosphorylation levels in PKR-M1 cells were approximately 2.5-fold lower than in control cells expressing the neomycin-resistant gene alone, whereas levels in PKR-M7 cells were reduced approximately 50-fold. We have presented evidence that two PKR variants possess distinct biological activities in reticulocyte extract, although both can act in vitro as dominant negative inhibitors of PKR. These data lead us to speculate that the inhibition may be occurring through different mechanisms. First, it is unlikely that either mutant is working by sequestration of the eIF-2 substrate, since neither prevented phosphorylation of eIF-2α by the heme regulated kinase. The catalytically inactive PKR-M1 mutant is likely inhibiting PKR-WT by sequestering the dsRNA activator, since (i) PKR-M1, at large excesses, can reverse the damaging effects of translation in reticulocyte extract caused by dsRNA addition; (ii) large amounts of PKR-M1 are required for transdominant inhibition of PKR-WT in our in vitro kinase assays; and (iii) this inhibition of the wild-type kinase can be reversed by further addition of excess dsRNA. In contrast, the regulatory domain mutant, PKR-M7, can inhibit PKR-WT at approximately equal concentrations, cannot reverse dsRNA effects in reticulocyte lysate, nor can its action in vitro be reversed by addition of more activator. We conclude, therefore, that PKR-M7 is probably inhibiting kinase activity through a direct interaction, forming inactive heterodimers with the wild-type protein kinase, although we concede we have no concrete data at this time to support this model. Others have presented evidence that PKR may need to dimerize to become fully functional(33Langland J.O. Pettiford S. Jiang B. Jacobs B.L. J. Virol. 1994; 68: 3821-3829Crossref PubMed Google Scholar, 34Romano P.R. Green S.R. Barber G.N. Mathews M.B. Hinnebusch A.G. Mol. Cell. Biol. 1995; 15: 365-378Crossref PubMed Google Scholar). It was important to address the biological relevance of these in vitro observations. As earlier mentioned, overexpression of either PKR-M1 or PKR-M7 in NIH 3T3 cells induced their malignant transformation and allowed these cells to cause tumors in nude mice (18Meurs E. Galabru J. Barber G.N. Katze M.G. Hovanessian A.G. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 232-236Crossref PubMed Scopus (417) Google Scholar, 21Barber G.N. Wambach M. Thompson S. Jagus R. Katze M.G. Mol. Cell. Biol. 1995; 15: 3138-3146Crossref PubMed Scopus (139) Google Scholar). However, important differences do exist between the PKR-M1 and PKR-M7 cell lines, and this may relate to the variants' differing modes of action in vitro. Somewhat unexpectedly we found that the PKR-M7 cell lines grew faster and to higher densities than the PKR-M1 overexpressing cell lines. More importantly, levels of eIF-2α phosphorylation were more dramatically reduced in the NIH 3T3 cell lines overexpressing the regulatory domain variant, PKR-M7, compared with PKR-M1 expressing cell lines. This raises the possibility that the two variants may be triggering transformation through different pathways. The in vivo and in vitro data describing PKR-M7 activity are consistent and strongly suggest that PKR-M7 functions through the inactivation of PKR activity by mechanisms that do not involve the sequestration of activator. This then results in reduced eIF-2α phosphorylation levels and accelerated growth rates in cell lines expressing the variant. This situation is very similar to cell lines overexpressing the Δ6 PKR mutant (16Koromilas A.E. Roy S. Barber G.N. Katze M.G. Sonneberg N. Science. 1992; 257: 1685-1689Crossref PubMed Scopus (496) Google Scholar) and the PKR cellular inhibitor P58(35Barber G.N. Thompson S. Lee T.-G. Strom T. Jagus R. Darveau A. Katze M.G. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 4278-4282Crossref PubMed Scopus (121) Google Scholar). In both these cases, decreases in endogenous PKR function lead to dramatic reductions in eIF-2α phosphorylation levels, accelerated growth rates, and subsequent ability of these cells to cause tumors in nude mice. The scenario appears to be different for PKR-M1. Despite the variant's in vitro properties, PKR-M1 does not cause a dramatic reduction of eIF-2 phosphorylation in overexpressing cell lines nor does it appreciably up-regulate growth rates of these cells in culture (Fig. 6). Furthermore, our earlier study showed that the presence of PKR-M1 failed to prevent excessive phosphorylation of eIF-2α in interferon-treated cells infected by encephalomyocarditis virus(30Meurs E. Watanabe Y. Barber G.N. Katze M.G. Chong K.L. Williams B.R.G. Hovanessian A.G. J. Virol. 1992; 66: 5805-5814Crossref PubMed Google Scholar). Taken together, these results suggest one of two possibilities regarding the variant's ability to induce malignant transformation: (i) PKR-M1 may still work by down-regulating PKR function by sequestering dsRNA activator. However, since eIF-2α phosphorylation levels are not dramatically diminished in vivo, PKR-M1 also may be inhibiting other, nontranslational activities of PKR, e.g. those functions involved with signal transduction and transcription(10Kumar A. Haque J. Lacoste J. Hiscott J. Williams B.R.G. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 6288-6292Crossref PubMed Scopus (514) Google Scholar, 11Maran A. Maitra R.K. Kumar A. Dong B. Xiao W. Li G. Williams B.R.G. Torrence P.F. Silverman R.H. Science. 1994; 265: 789-792Crossref PubMed Scopus (211) Google Scholar). Consistent with this hypothesis is the recent report demonstrating that a nearly identical catalytically inactive PKR molecule can alter the activation state of NF-κB(10Kumar A. Haque J. Lacoste J. Hiscott J. Williams B.R.G. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 6288-6292Crossref PubMed Scopus (514) Google Scholar). (ii) An alternative explanation is that PKR-M1 transforms cells via mechanisms completely independent of any known PKR regulatory pathway. It may be that PKR-M1 overexpression is disrupting a PKR-independent pathway by binding to and/or sequestering unknown proteins which would then result in malignant transformation. This would then suggest that the variant does not function as a bona fide transdominant inhibitor of PKR in overexpressing cells and would be in agreement with a recent report showing that the domain II mutant cannot function as a dominant negative inhibitor in yeast(34Romano P.R. Green S.R. Barber G.N. Mathews M.B. Hinnebusch A.G. Mol. Cell. Biol. 1995; 15: 365-378Crossref PubMed Google Scholar). However it would be in contradiction to our earlier findings suggesting PKR-M1 can function dominant negatively in vivo using transient transfection assays(36Barber G.N. Wambach M. Wong M.-L. Dever T.E. Hinnebusch A.G. Katze M.G. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 4621-4625Crossref PubMed Scopus (88) Google Scholar). In any case, it seems unlikely that the in vitro inhibition of PKR and eIF-2α phosphorylation (this report) and effects on protein synthetic rates in reticulocyte extracts caused by PKR-M1 (this report and (37Sharp T.V. Xiao Q. Jeffrey I. Gewert D.R. Clemens M.J. Eur. J. Biochem. 1993; 214: 945-948Crossref PubMed Scopus (33) Google Scholar)) accurately reflect the only biological properties of the PKR-M1 variant. We are grateful to Olga Savinova for technical assistance, the Henshaw laboratory for eIF-2α antibody, Dr. Leonard Jefferson for purified eIF-2, and Marjorie Domenowske for help with the figures. We are grateful to Greg Schaefer for excellent technical assistance." @default.
• W1988981229 created "2016-06-24" @default.
• W1988981229 creator A5011095778 @default.
• W1988981229 creator A5036980616 @default.
• W1988981229 creator A5050436106 @default.
• W1988981229 creator A5066798307 @default.
• W1988981229 creator A5068452671 @default.
• W1988981229 date "1995-07-01" @default.
• W1988981229 modified "2023-09-27" @default.
• W1988981229 title "Molecular Mechanisms Responsible for Malignant Transformation by Regulatory and Catalytic Domain Variants of the Interferon-induced Enzyme RNA-dependent Protein Kinase" @default.
• W1988981229 cites W1514587909 @default.
• W1988981229 cites W153108858 @default.
• W1988981229 cites W1549105044 @default.
• W1988981229 cites W1582095946 @default.
• W1988981229 cites W1587044218 @default.
• W1988981229 cites W1604315363 @default.
• W1988981229 cites W1759060244 @default.
• W1988981229 cites W179070074 @default.
• W1988981229 cites W1968727721 @default.
• W1988981229 cites W1968987342 @default.
• W1988981229 cites W1970812281 @default.
• W1988981229 cites W1974129290 @default.
• W1988981229 cites W1987694369 @default.
• W1988981229 cites W1987875115 @default.
• W1988981229 cites W2012810401 @default.
• W1988981229 cites W2015686762 @default.
• W1988981229 cites W2023530592 @default.
• W1988981229 cites W2027565750 @default.
• W1988981229 cites W2034736009 @default.
• W1988981229 cites W2036853011 @default.
• W1988981229 cites W2040353939 @default.
• W1988981229 cites W2040500676 @default.
• W1988981229 cites W2048465699 @default.
• W1988981229 cites W2049444392 @default.
• W1988981229 cites W2062417484 @default.
• W1988981229 cites W2074882944 @default.
• W1988981229 cites W2080028459 @default.
• W1988981229 cites W2080133175 @default.
• W1988981229 cites W2089033636 @default.
• W1988981229 cites W2118635464 @default.
• W1988981229 cites W2130684922 @default.
• W1988981229 cites W2156065951 @default.
• W1988981229 cites W2165055107 @default.
• W1988981229 cites W2166661434 @default.
• W1988981229 doi "https://doi.org/10.1074/jbc.270.29.17423" @default.
• W1988981229 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/7615547" @default.
• W1988981229 hasPublicationYear "1995" @default.
• W1988981229 type Work @default.
• W1988981229 sameAs 1988981229 @default.
• W1988981229 citedByCount "77" @default.
• W1988981229 countsByYear W19889812292013 @default.
• W1988981229 countsByYear W19889812292014 @default.
• W1988981229 countsByYear W19889812292015 @default.
• W1988981229 countsByYear W19889812292016 @default.
• W1988981229 countsByYear W19889812292017 @default.
• W1988981229 countsByYear W19889812292018 @default.
• W1988981229 countsByYear W19889812292019 @default.
• W1988981229 crossrefType "journal-article" @default.
• W1988981229 hasAuthorship W1988981229A5011095778 @default.
• W1988981229 hasAuthorship W1988981229A5036980616 @default.
• W1988981229 hasAuthorship W1988981229A5050436106 @default.
• W1988981229 hasAuthorship W1988981229A5066798307 @default.
• W1988981229 hasAuthorship W1988981229A5068452671 @default.
• W1988981229 hasBestOaLocation W19889812291 @default.
• W1988981229 hasConcept C104317684 @default.
• W1988981229 hasConcept C161238802 @default.
• W1988981229 hasConcept C181199279 @default.
• W1988981229 hasConcept C185592680 @default.
• W1988981229 hasConcept C2776178377 @default.
• W1988981229 hasConcept C502942594 @default.
• W1988981229 hasConcept C54355233 @default.
• W1988981229 hasConcept C55493867 @default.
• W1988981229 hasConcept C67705224 @default.
• W1988981229 hasConcept C86803240 @default.
• W1988981229 hasConcept C90934575 @default.
• W1988981229 hasConcept C95444343 @default.
• W1988981229 hasConcept C97029542 @default.
• W1988981229 hasConceptScore W1988981229C104317684 @default.
• W1988981229 hasConceptScore W1988981229C161238802 @default.
• W1988981229 hasConceptScore W1988981229C181199279 @default.
• W1988981229 hasConceptScore W1988981229C185592680 @default.
• W1988981229 hasConceptScore W1988981229C2776178377 @default.
• W1988981229 hasConceptScore W1988981229C502942594 @default.
• W1988981229 hasConceptScore W1988981229C54355233 @default.
• W1988981229 hasConceptScore W1988981229C55493867 @default.
• W1988981229 hasConceptScore W1988981229C67705224 @default.
• W1988981229 hasConceptScore W1988981229C86803240 @default.
• W1988981229 hasConceptScore W1988981229C90934575 @default.
• W1988981229 hasConceptScore W1988981229C95444343 @default.
• W1988981229 hasConceptScore W1988981229C97029542 @default.
• W1988981229 hasIssue "29" @default.
• W1988981229 hasLocation W19889812291 @default.
• W1988981229 hasOpenAccess W1988981229 @default.
• W1988981229 hasPrimaryLocation W19889812291 @default.
• W1988981229 hasRelatedWork W1973179284 @default.
• W1988981229 hasRelatedWork W1983702520 @default.
• W1988981229 hasRelatedWork W2008585820 @default.
• W1988981229 hasRelatedWork W2012657116 @default.

• next