Inositol Hexakisphosphate Kinase 2 Mediates Growth Suppressive and Apoptotic Effects of Interferon-β in Ovarian Carcinoma Cells

Interferons (IFNs) regulate the expression of genes that mediate their antiviral, antitumor, and immunomodulatory actions. We have previously shown that IFN-β suppresses growth of human ovarian carcinoma xenografts in vivo and induces apoptosis of ovarian carcinoma cells in vitro. To investigate mechanisms of IFN-β-induced apoptosis we employed an antisense technical knockout approach to identify gene products that mediate cell death and have isolated several regulators of interferon-induced death (RIDs). In this investigation, we have characterized one of the RIDs, RID-2. Sequence analysis revealed that RID-2 was identical to human inositol hexakisphosphate kinase 2 (IP6K2). IP6K2 is post-transcriptionally induced by IFN-β in ovarian carcinoma cells. A mutant IP6K2 with substitutions in the putative inositol phosphate binding domain abrogates IFN-β-induced apoptosis. These studies identify a novel function for IP6K2 in cell growth regulation and apoptosis. Interferons (IFNs) regulate the expression of genes that mediate their antiviral, antitumor, and immunomodulatory actions. We have previously shown that IFN-β suppresses growth of human ovarian carcinoma xenografts in vivo and induces apoptosis of ovarian carcinoma cells in vitro. To investigate mechanisms of IFN-β-induced apoptosis we employed an antisense technical knockout approach to identify gene products that mediate cell death and have isolated several regulators of interferon-induced death (RIDs). In this investigation, we have characterized one of the RIDs, RID-2. Sequence analysis revealed that RID-2 was identical to human inositol hexakisphosphate kinase 2 (IP6K2). IP6K2 is post-transcriptionally induced by IFN-β in ovarian carcinoma cells. A mutant IP6K2 with substitutions in the putative inositol phosphate binding domain abrogates IFN-β-induced apoptosis. These studies identify a novel function for IP6K2 in cell growth regulation and apoptosis. interferon(s) regulator(s) of interferon-induced death inositol phosphate binding domain inositol hexakisphosphate kinase diphosphoinositol pentakisphosphate poly(ADP-ribose) polymerase, a caspase 3 substrate 4,6-diamidino-2-phenylindole terminal dUTP nick-end labeling IFN gene regulatory factor(s) inositol hexaphosphate polymerase chain reaction substitution mutant antisense mutant fluorescein isothiocyanate propidium iodide polyethyleneimine cellulose thin-layer chromatography area under the curve full-length pyrophosphate The interferon (IFN)1family of cytokines stimulate antiviral, antitumor, antiproliferative, and immunoregulatory activities (1Gresser I. Bourali C. Levy J.P. Fontaine-Brouty-Boye D. Thomas M.T. Proc. Natl. Acad. Sci. U. S. A. 1969; 63: 51-57Crossref PubMed Scopus (156) Google Scholar, 2Evinger M. Rubinstein M. Pestka S. Arch. Biochem. Biophys. 1981; 210: 319-329Crossref PubMed Scopus (74) Google Scholar, 3Kalvakolanu D.V. Histol. Histopathol. 2000; 15: 523-537PubMed Google Scholar). Upon binding to receptors, IFNs activate a signaling cascade wherein Janus tyrosine kinases induce tyrosine phosphorylation of signal-transducing activators of transcription proteins (4Stark G.R. Kerr I.M. Williams B.R. Silverman R.H. Schreiber R.D. Ann. Rev. Biochem. 1998; 67: 227-264Crossref PubMed Scopus (3343) Google Scholar, 5Darnell J.E.J. Science. 1997; 277: 1630-1635Crossref PubMed Scopus (3330) Google Scholar). These transcription factors induce expression of hundreds of genes that possess a wide range of activities (6Der S.D. Zhou A. Williams B.R. Silverman R.H. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 15623-15628Crossref PubMed Scopus (1509) Google Scholar). Although a great deal is known about antiviral actions of IFNs, mechanisms responsible for their antitumor actions are unclear.In vivo, they up-regulate expression of tumor-specific antigens, natural killer, and T cell function (7Ortaldo J.R. Mason A. Rehberg E. Moschera J. Kelder B. Pestka S. Herberman R.B. J. Biol. Chem. 1983; 258: 15011-15015Abstract Full Text PDF PubMed Google Scholar, 8Greiner J.W. Hand P.H. Noguchi P. Fisher P.B. Pestka S. Schlom J. Cancer Res. 1984; 44: 3208-3214PubMed Google Scholar, 9Greiner J.W. Guadagni F. Noguchi P. Pestka S. Colcher D. Fisher P.B. Schlom J. Science. 1987; 235: 895-898Crossref PubMed Scopus (197) Google Scholar). IFNs also activate growth suppressive proteins such as pRb (10Kumar R. Atlas I. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 6599-6603Crossref PubMed Scopus (92) Google Scholar, 11Resnitzky D. Tiefenbrun N. Berissi H. Kimchi A. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 402-406Crossref PubMed Scopus (116) Google Scholar), down-regulate c-Myc (12Raveh T. Hovanessian A.G. Meurs E.F. Sonenberg N. Kimchi A. J. Biol. Chem. 1996; 271: 25479-25484Abstract Full Text Full Text PDF PubMed Scopus (59) Google Scholar), and suppress activity of transcription factor E2F (13Melamed D. Tiefenbrun N. Yarden A. Kimchi A. Mol. Cell. Biol. 1993; 13: 5255-5265Crossref PubMed Scopus (67) Google Scholar). Protein kinase R and ribonuclease L, which inhibit viral growth in IFN-treated cells (14Samuel C.E. Kuhen K.L. George C.X. Ortega L.G. Rende-Fournier R. Tanaka H. Int. J. Hematol. 1997; 65: 227-237Crossref PubMed Google Scholar, 15Silverman R.H. D'Alessio G. Riordan J.F. Ribonucleases: Structure and Functions. Academic Press, New York1997: 515-551Crossref Google Scholar), also play a role in growth suppression (16Barber G.N. Wamback M. Thompson S. Jagus R. Katze M.G. Mol. Cell. Biol. 1995; 15: 3138-3146Crossref PubMed Scopus (139) Google Scholar, 17Hassel B.A. Zhou A. Sotomayor C. Maran A. Silverman R.H. EMBO J. 1993; 12: 3297-3304Crossref PubMed Scopus (244) Google Scholar). The family of transcription factors known as IFN gene regulatory factors (IRFs) also mediates the effects of IFNs (18Harada H. Kitagawa M. Tanaka N. Yamamoto H. Harada K. Ishihara M. Taniguchi T. Science. 1993; 259: 971-974Crossref PubMed Scopus (427) Google Scholar). Two members of this family, IRF-1 and interferon consensus sequence-binding protein (IRF-8) are up-regulated by IFN-γ. Deletion of these genes results in myelodysplasias or chronic myelogenous leukemia-like disease (19Willman C.L. Sever C.E. Pallavicini M.G. Harada H. Tanaka N. Slovak M.L. Yamamoto H. Harada K. Meeker T.C. List A.F. Taniguchi T. Science. 1993; 259: 968-971Crossref PubMed Scopus (379) Google Scholar, 20Holtschke T. Lohler J. Kanno Y. Fehr T. Giese N. Rosenbauer F. Lou J. Knobeloch K.P. Gabriele L. Waring J.F. Bachmann M.F. Zinkernagel R.M. Morse III, H.C. Ozato K. Horak I. Cell. 1996; 87: 307-317Abstract Full Text Full Text PDF PubMed Scopus (536) Google Scholar). Because IRFs are transcription factors, and their biological activity depends on genes they induce; characterization of downstream gene products should allow identification of critical regulators of growth suppression. We have previously shown that IFN-β suppresses the growth of ovarian tumor xenografts in nude mice (21Lindner D.J. Borden E.C. J. Interferon Cytokine Res. 1997; 17: 681-693Crossref PubMed Scopus (50) Google Scholar) and that IFN-β induces apoptosis in these cells. To identify death genes we used an antisense technical knockout approach (22Deiss L.P. Kimchi A. Science. 1991; 252: 117-120Crossref PubMed Scopus (172) Google Scholar). In this approach, death regulatory genes are identified by their ability, when expressed in antisense orientation, to confer resistance to death inducers. Using this technique we have identified several genes, regulators of interferon-induced death (RIDs), that enhance IFN-β-activated death. In this study we have characterized one of these genes, RID-2, and identified it as human inositol hexakisphosphate kinase 2 (IP6K2) (23Saiardi A. Erdjument-Bromage H. Snowman A.M. Tempst P. Snyder S.H. Curr. Biol. 1999; 9: 1323-1326Abstract Full Text Full Text PDF PubMed Scopus (300) Google Scholar). IP6K2 catalyzes the synthesis of diphosphoinositol pentakisphosphate (PP-IP5) using inositol hexaphosphate (IP6) as a substrate in the presence of ATP. Hence, PP-IP5 may cause growth suppression. We show that cellular IP6K2 levels are post-transcriptionally enhanced by IFN-β. Overexpression of IP6K2 enhances both growth suppressive and apoptotic activities of IFN-β. A dominant negative inositol phosphate binding domain (IPBD) mutant is highly resistant to both the antiproliferative and apoptotic functions of IFN-β. Thus, our studies ascribe a novel function for IP6K2 in cell growth control via apoptosis. Human IFN-β (Serono), specific activity 2.7 × 108 units/mg; IFN-α2b (Schering Plough), specific activity 3 × 108 units/mg; IFN-γ (Roche Molecular Biochemicals), specific activity 2 × 107units/mg; 3Hinositol and 3HIP6 (PerkinElmer Life Sciences); anti-PARP antibody (Biomol); anti-caspase 3 antibody (Pharmingen); anti-myc antibody (Oncogene Research Products); and horseradish peroxidase goat antirabbit IgG (Pierce) were used in these studies. Cells were treated with IFNs during growth in RPMI 1640 and 5% fetal bovine serum. Growth was monitored using a calorimetric assay (24Skehan P. Storeng R. Scudiero D. Monks A. McMahon J. Vistica D. Warren J.T. Bokesch H. Kenney S. Boyd M.R. J. Natl. Cancer Inst. 1990; 82: 1107-1112Crossref PubMed Scopus (8663) Google Scholar). Each treatment group contained eight replicates. Cells were fixed and stained with sulforhodamine B after 7 days. Bound dye was eluted from cells, and absorbance (A exp) was measured at 570 nm. One plate was fixed 8 h after plating to determine the absorbance representing starting cell number (A ini). Absorbance with this plate and that obtained with untreated cells at the end of the growth period (A fin) were taken as 0 and 100% growth, respectively. Thus, percent control growth = 100% × (A exp −A ini)/(A fin −A ini) expressed as a percent of untreated controls; a decrease in cell number (death) falls on the negative scale. To determine cell cycle distribution, cells were stained with propidium iodide and analyzed by flow cytometry (Beckton Dickinson) using MultiPass software. Total RNA from NIH-OVCAR-3 cells, treated with IFN-β (500 units/ml) for 0, 1, 2, 4, 8, 16, 24, 48, and 72 h, was prepared using RNAZol B (Tel-Test). RNAs were pooled, and total poly(A)+ RNA was isolated (polyAttract; Promega). cDNA libraries were constructed with a commercially available kit (Stratagene). 10 μg of mRNA was used for preparing the cDNA library, using an oligo(dT) primer and a dNTP mixture containing 5-methyl dCTP. After second-strand synthesis,Pfu thermal DNA polymerase was used to create blunt-ended cDNA. cDNAs were ligated to a bifunctional linker, 5′-GCTTGGATCCAAGC-3′. Ligated to the 3′ and 5′ ends of the cDNA, this linker generates HindIII and BamHI sites, respectively (25Meissner P.S. Sisk W.P. Berman M.L. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 4171-4175Crossref PubMed Scopus (135) Google Scholar). The library was digested with HindIII andBamHI, purified on a Sepharose 6B column, and ligated into an episomal vector pTKO1 (from Adi Kimchi, Weizmann Institute, Rehovot, Israel), which carried markers for selection in eukaryotic and bacterial cells (22Deiss L.P. Kimchi A. Science. 1991; 252: 117-120Crossref PubMed Scopus (172) Google Scholar). When cloned into pTKO1 the cDNA is expressed in antisense orientation. The library was transformed into Escherichia coli DH10B, and plasmid DNA was extracted and purified on CsCl gradients. Electroporation of the library (40 μg) into NIH-OVCAR-3 cells (107) was followed by selection with hygromycin B (200 μg/ml) and IFN-β (2000 units/ml) for 4 weeks. All pTKO1-transfected cells (selected similarly) died after 14 days. After 4 weeks selection-surviving colonies were pooled and expanded, and Hirt DNA extracts were prepared. DNA was digested with DpnI and electroporated into E. coli DH10B. Resultant colonies were screened by PCR using pTKO1-specific primers to detect inserts. Episomes (20 μg) were tested for protection against IFN-β-induced death by electroporation into NIH-OVCAR-3 cells. Selection with IFN-β and hygromycin was initiated after 24 h. RID-2 cDNA was digested from pTKO1 and cloned into the pCXN2myc mammalian expression vector (26Kinoshita S. Suzuki H. Ito K. Kume K. Shimizu T. Sugiyama Y. Pharm. Res. 1998; 15: 1851-1856Crossref PubMed Scopus (26) Google Scholar) in which the chicken actin expression of the A substitution mutant of the putative was using with full-length RID-2 as the highly consensus with at 7 of using primers and 5′ PCR products were digested with and transformed into E. and were antisense mutant was by RID-2 into pCXN2myc in antisense orientation. were by were electroporated into NIH-OVCAR-3 cells, and were with After 3 weeks of were pooled for studies. of was monitored by Total RNA (20 μg) was on to and with the PCR of RID-2 cDNA. Total cell protein (20 μg) was on and to were with in our against full-length expressed RID-2. After were with IgG antibody to horseradish peroxidase and using cells were using a commercially available kit and with and propidium iodide Cells were analyzed by flow DNA was using the kit Cells were with using terminal stained with antibody followed by The of cells was by flow activity G. Snyder S.H. Proc. Natl. Acad. Sci. U. S. A. 1996; PubMed Scopus Google Scholar) was in cell extracts of cells by on in and to of Using kinase were in of 10 of using 3HIP6 (PerkinElmer Life as were at for were on and with of and followed by with and by were using polyethyleneimine cellulose thin-layer chromatography The reaction was and in and were into and the cellulose was from the with of with of 3 of and of was by this and PP-IP5 with an of and and with was prepared by of of IP6K2 and 3HIP6 in a kinase reaction as followed by chromatography on a × A and B A and area under the curve was used to determine total of We have previously shown that human IFN-β induces of NIH-OVCAR-3 human ovarian carcinoma in nude mice (21Lindner D.J. Borden E.C. J. Interferon Cytokine Res. 1997; 17: 681-693Crossref PubMed Scopus (50) Google Scholar). To that IFN-β was in we treated tumor cell with After cell growth was measured using a assay on binding of the sulforhodamine B (24Skehan P. Storeng R. Scudiero D. Monks A. McMahon J. Vistica D. Warren J.T. Bokesch H. Kenney S. Boyd M.R. J. Natl. Cancer Inst. 1990; 82: 1107-1112Crossref PubMed Scopus (8663) Google Scholar). of IFN-β growth in NIH-OVCAR-3 cells in with at and death at IFN-β was at growth in at the IFN-β was IFN-γ on cell growth at IFN-β was of the IFNs at of death. of NIH-OVCAR-3 cells treated with IFN-β revealed a with by cells a of To identify apoptotic cells binding were IFN-β was at in the IFN-γ cause binding IFN-β at (200 units/ml) apoptosis after 3 of of units/ml) induced apoptosis. apoptosis induced by IFN-γ (2000 units/ml) was from untreated cells. that as as h after treatment with IFN-β 2 The of by assay the of IFN-β induced levels of and IFN-γ cause an of IFNs on of apoptosis in NIH-OVCAR-3 of cells by flow of in a of cells by flow of To determine cell cycle cells were treated with IFN-β for stained with and analyzed by flow IFN-β cell cycle The of cells in and was after IFN-β or treatment (200 Thus, death of cell cycle or IFN-β death in NIH-OVCAR-3 cells at (200 in and carcinoma IFN-β induced apoptosis with IFN-γ (2000 units/ml) inhibit growth or induce apoptosis in cell tested and To that were by for caspase 3 and was After 8 h of IFN-β of caspase 3 was followed by at h Hence, at one caspase and downstream substrate for cell were IFN-β Because NIH-OVCAR-3 cells Y. H. Cancer Res. 1992; Google Scholar, M. K. K. K. T. Y. P. K. 1995; PubMed Scopus Google Scholar), apoptosis to via We that may the death to the apoptotic in of we to identify gene products using an antisense knockout approach (22Deiss L.P. Kimchi A. Science. 1991; 252: 117-120Crossref PubMed Scopus (172) Google Scholar). In this a gene isolated by antisense Cells are with an antisense cDNA library from tumor cells. RNA expression of the death cell that the antisense mRNA in the presence of death are into ovarian carcinoma cells to the antisense mRNA and isolated in the of To genes we prepared antisense cDNA libraries cloned in the episomal gene expression of antisense RNAs in this library was electroporated into NIH-OVCAR-3 cells and for resistance to hygromycin B and human IFN-β. 4 weeks of selection were pooled, and Hirt extracts were prepared. DNA was digested with DpnI and electroporated into E. coli DH10B. were in the Each was into NIH-OVCAR-3 cells and for cell protection against IFN-β-induced death. After of resistance to IFN-β-induced death. We We RID-2 for of the RID-2 antisense protection against IFN-β-induced death colonies were in cells with pTKO1 vector The RID-2 was on both The for a Sequence analysis revealed that this cDNA was identical to human number (23Saiardi A. Erdjument-Bromage H. Snowman A.M. Tempst P. Snyder S.H. Curr. Biol. 1999; 9: 1323-1326Abstract Full Text Full Text PDF PubMed Scopus (300) Google Scholar). A putative is and of this The consensus is The highly are shown in is for the activity of inositol kinase S. K. T. C. T. Biochem. J. 1997; PubMed Scopus Google Scholar). the are the presence of a number of and the of in the RID-2 to as To study the of IFN-β on we mRNA levels were IFN-β induce IP6K2 mRNA in NIH-OVCAR-3 cells after with cDNA the presence of of RNA in IP6K2 mRNA was induced in or ovarian carcinoma cells To IFN-β enhanced expression of IP6K2 analysis was on cell IFN-β induced IP6K2 protein in a was 4 h 4 and 24 h, IP6K2 protein levels of this revealed that of IP6K2 protein levels at h of IFN-β treatment and Because the protection from death it was to that this was of expression of antisense IP6K2 mRNA and of IP6K2 analysis of cells with vector or antisense IP6K2 was In vector cells one representing IP6K2 mRNA was cells also mRNA in mRNA that in vector cells, to antisense IP6K2 mRNA from the We expression in of analysis that cells contained levels of IP6K2 protein with vector cells We the in IP6K2 protein levels IFN-β with an in In kinase 3HIP6 as substrate were to activity in cell of untreated and cells, using of total cell In the PP-IP5 was by using and 3HIP6 as activity of was after 4 h of IFN-β a at 8 h and to decrease from by 24 h activity was with IP6K2 protein as by When for activity was by IFN-β cells an activity of 8 2 IFN-α2b or IFN-γ enhance activity at to cells antisense IP6K2 mRNA at the activity with cells, by levels of the PP-IP5 IP6K2 a putative we the of this domain by an In this 7 of in the were at and of this mutant in the pCXN2myc vector of the protein with vector was into NIH-OVCAR-3 cells, and cell were IP6K2 was cloned into the of were and they growth of with vector of the mutant was by of protein were in both cell proteins were in cells Growth were in NIH-OVCAR-3 cells that expressed In of were in growth to the cell enhanced suppression of growth in to IFN-β The mutant was resistant to antiproliferative effects of IFN-β. cells were by IFN-β with vector cells, which were of IFN-β that vector cells units/ml) cells cells that expressed IP6K2 mRNA in antisense were resistant to growth at the tested the total suppression of activity in cells their growth was by IFN-β that factors may mediate the antiproliferative effects of IFN-β in these cells. of apoptosis by IFN-β was measured using Cells IP6K2 the of apoptosis treatment with IFN-β (200 was in cells with cells or cells of IP6K2 induce apoptosis in the of IFN-β cells to death by IFN-β. cells IFN-β-induced apoptosis when with vector cells, a of an is critical for of apoptosis. we of IFN-β-induced apoptosis by the mutant was of a decrease in activity of In kinase were to detect these Total into the PP-IP5 was expressed as Cells that expressed IP6K2 the activity with cells, cells an of activity S. K. T. C. T. Biochem. J. 1997; PubMed Scopus Google Scholar). Thus, the as a dominant negative cells the of activity G. Snyder S.H. Proc. Natl. Acad. Sci. U. S. A. 1996; PubMed Scopus Google Scholar), which was activity of vector cells Thus, of activity with of growth suppression and of apoptosis by IFN-β. These the role of IP6K2 as a of growth and apoptosis in to IFN-β IFN-β was the of cell death of the IFN-γ cause death at These that and IFNs to actions. The that IFN-β induced death of cell cycle that gene products regulate growth and apoptosis. Because NIH-OVCAR-3 cells Y. H. Cancer Res. 1992; Google Scholar, M. K. K. K. T. Y. P. K. 1995; PubMed Scopus Google Scholar), it that IFN-β-induced apoptosis via are by a of death tumor and it is the apoptosis in to IFN-β. To this we employed an antisense technical knockout (22Deiss L.P. Kimchi A. Science. 1991; 252: 117-120Crossref PubMed Scopus (172) Google Scholar) and identified the The library used was using mRNA isolated from untreated cells, as as treated with IFN-β. Thus, genes expressed in of apoptosis were in the of Janus tyrosine kinases or signal-transducing activators of transcription for the expression of antisense is by an (22Deiss L.P. Kimchi A. Science. 1991; 252: 117-120Crossref PubMed Scopus (172) Google Scholar). cDNAs to these signaling have in our studies The RID-2 cDNA characterized in this study is identical to Because IFN-β enhances protein levels mRNA it that a regulatory IP6K2 The of antisense IP6K2 was of a of IP6K2 protein and expression of also resistance to IFN-β-induced apoptosis. PP-IP5 L. T. U. G. A. J. Biol. Chem. 1993; Full Text PDF PubMed Google Scholar), and A. Snyder S.H. J. Biol. Chem. 2000; Full Text Full Text PDF PubMed Scopus Google Scholar). The specific of and their are was previously to an have that functions as a substrate in mammalian cells, and (23Saiardi A. Erdjument-Bromage H. Snowman A.M. Tempst P. Snyder S.H. Curr. Biol. 1999; 9: 1323-1326Abstract Full Text Full Text PDF PubMed Scopus (300) Google Scholar, L. T. U. G. A. J. Biol. Chem. 1993; Full Text PDF PubMed Google Scholar, J.W. J. Biol. Chem. 1993; Full Text PDF PubMed Google Scholar) to a family of is to which in mammalian cells is by a kinase to A. Snyder S.H. 1998; PubMed Scopus Google Scholar). The that in cells is and that functions as a substrate in synthesis of of of N. A. J. Biol. Chem. 1995; Full Text Full Text PDF PubMed Scopus Google Scholar) that these play an regulatory is the substrate for and also to A. Snyder S.H. J. Biol. Chem. 2000; Full Text Full Text PDF PubMed Scopus Google Scholar). Because PP-IP5 and it is the may also to growth and apoptosis. The inositol to play a role in the of inositol responsible for and protein kinase and cell growth regulation have The 3 inositol in apoptosis during S. A. Snyder S.H. J. 2000; PubMed Google Scholar). In these of 3 inositol expression by antisense apoptosis. mediates of Overexpression of this growth of P. R. S.H. Proc. Natl. Acad. Sci. U. S. A. 2000; PubMed Scopus Google Scholar). studies inositol as regulators of IFN-β-induced apoptosis. and IP6K2 have cloned (23Saiardi A. Erdjument-Bromage H. Snowman A.M. Tempst P. Snyder S.H. Curr. Biol. 1999; 9: 1323-1326Abstract Full Text Full Text PDF PubMed Scopus (300) Google Scholar). Although and from may also play a role in growth of the IP6K2 dominant negative mutant of growth and apoptosis induced by IFN-β. cells and cells both an of cells were resistant cells to growth that may in the cells. to of both and IP6K2 function the IFN-β resistance of these cells.