Cells Formation of Stat2:Stat6 Complexes in B Activates Stat6 and Leads to the Α

IFN-α Activates Stat6 and Leads to the Formation of Stat2:Stat6 Complexes in B Cells

This information is current as Sanjay Gupta, Man Jiang and Alessandra B. Pernis of October 2, 2021. J Immunol 1999; 163:3834-3841; ; http://www.jimmunol.org/content/163/7/3834

References This article cites 55 articles, 31 of which you can access for free at: Downloaded from http://www.jimmunol.org/content/163/7/3834.full#ref-list-1

Why The JI? Submit online.

http://www.jimmunol.org/ • Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: by guest on October 2, 2021 http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 1999 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. IFN-␣ Activates Stat6 and Leads to the Formation of Stat2: Stat6 Complexes in B Cells1

Sanjay Gupta, Man Jiang, and Alessandra B. Pernis2

IFN-␣ consists of a family of highly homologous proteins, which exert pleiotropic effects on a wide variety of cell types. The biologic activities of IFN-␣ are mediated by its binding to a multicomponent receptor complex resulting in the activation of the Janus kinase-STAT signaling pathway. In most cell types, activation of Stat1 and Stat2 by IFN-␣ leads to the formation of either STAT homo-/heterodimers or of the IFN-stimulated gene factor 3 complex composed of Stat1, Stat2, and p48, a non-STAT protein. These distinct transcriptional complexes then target two different sets of cis-elements, ␥-activated sites and IFN-stimulated response elements. Here, we report that IFN-␣ can activate complexes containing Stat6, which, until now, has been primarily associated with signaling by two cytokines with biologic overlap, IL-4 and IL-13. Induction of Stat6 complexes by IFN-␣ appears to be cell type speciﬁc, given that tyrosine phosphorylation of Stat6 in response to IFN-␣ is predominantly detected in B cells. Activation of Stat6 by IFN-␣ in B cells is accompanied by the formation of novel Stat2:Stat6 complexes, including an IFN-stimulated gene Downloaded from factor 3-like complex containing Stat2, Stat6, and p48. B cell lines resistant to the antiproliferative effects of IFN-␣ display a decrease in the IFN-␣-mediated activation of Stat6. Activation of Stat6 as well as of Stat2:Stat6 complexes by IFN-␣ in B cells may allow modulation of target genes in a cell type-speciﬁc manner. The Journal of Immunology, 1999, 163: 3834–3841.

nterferon-␣ consists of a family of highly homologous pro- Stat1, and of Stat2 in most cell types (9–11), although activation

teins that exert pleiotropic actions, including immunomodu- of Stat3, Stat4, and Stat5 has also been reported (12–14). http://www.jimmunol.org/ I latory, antiviral, and antiproliferative effects (1–3). These bi- STAT homo-/heterodimers activated in response to IFN-␣ can ologic activities are mediated by the ability of the IFN-␣ species to drive the expression of a subset of IFN-␣-inducible genes, e.g., the bind to a multicomponent receptor complex. Two distinct human gene encoding the transcription factor IFN regulatory factor 1 receptor chains have been cloned so far: the IFN␣R1 and the (IRF-1), by binding to specific promoter elements termed ␥-acti- IFN␣R2 chains (4). However, additional receptor components may vated sites (GAS).3 Unlike other cytokines, IFN-␣ activation of exist, given that transfection of both human receptor chains in STATs also leads to the formation of an additional transcriptional murine cells does not fully reconstitute biologic responses to all complex termed IFN-stimulated gene factor 3 (ISGF3), which is human IFN-␣ subspecies (5). Cell type-specific differences in re- composed of a Stat1:Stat2 heterodimer and a non-STAT protein, ceptor composition may also occur, as suggested by the finding that p48. The ISGF3 complex can then induce the expression of a dis- by guest on October 2, 2021 IFN-␣ interacts differently with lymphoid vs epithelial cells (6). tinct subset of IFN-responsive genes by binding to a different sub- The early intracellular events triggered by IFN-␣ have recently set of cis-acting elements, termed IFN-stimulated response ele- been elucidated and have served as a paradigm for cytokine sig- ments (ISREs) (15). naling (7, 8). Binding of IFN-␣ to its receptor leads to the activa- IFN-␣ has been widely used as an antiviral as well as an anti- tion, via two receptor-associated tyrosine kinases, Tyk2 and Jak1, tumor agent because of its growth-inhibitory effects (16–18). of latent cytoplasmic factors belonging to the STAT (signal trans- Up-regulation of IRF-1 is believed to mediate some of the anti- ducers and activators of transcription) family of proteins. Tyrosine proliferative effects of the IFNs (19). The effectiveness of the phosphorylation of the STATs allows them to homodimerize or growth-inhibitory actions of IFNs is, however, limited by the heterodimerize and to translocate into the nucleus, where they emergence of IFN resistance. In B cells, IFN resistance character- modulate gene transcription. So far, seven distinct members of the izes EBV transformation of B cells as well as certain subclones of STAT family have been identified in eukaryotes. Some STATs Daudi, a Burkitt’s lymphoma cell line, which is normally exquis- display a restricted pattern of activation. For instance, Stat6 com- itely sensitive to growth inhibition by IFN-␣. IFN resistance in plexes are induced primarily in response to two cytokines with EBV-transformed B cells has been attributed to the presence of the overlapping biologic functions, IL-4 and IL-13. In contrast, other EBV nuclear Ag 2 (EBNA2) (20). However, a distinct mechanism STATs can be activated in response to a variety of cytokines (9). is believed to be responsible for the emergence of IFN resistance IFN-␣ has been shown to lead to the tyrosine phosphorylation of in Daudi variants, because the EBV genome present in Daudi cells is defective and carries a deletion of the entire coding region of EBNA2 (21). In both of these systems, IFN resistance leads to the Department of Medicine, Columbia University, New York, NY 10032 selective extinction of the antiproliferative pathways triggered by Received for publication September 22, 1998. Accepted for publication July 23, 1999. IFN-␣, whereas the induction of other IFN-␣ responses is main- The costs of publication of this article were defrayed in part by the payment of page tained at normal levels (20, 22). Studies of the molecular mecha- charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. nisms underlying IFN resistance in both Daudi as well as EBV- 1 This work was supported by a Grant-in-Aid from the American Heart Association, transformed B cells have failed to demonstrate changes in either the Silverberg Award, and the Saydman Trust Fund for Research in Septicemia in honor of Dr. Harold C. Neu at Columbia University. 2 Address correspondence and reprint requests to Dr. Alessandra Pernis, Department 3 Abbreviations used in this paper: GAS, ␥-activated site; IRF, IFN regulatory factor; of Medicine, Columbia University, 630 West 168th Street, New York, NY 10032. ISRE, IFN-stimulated response element; ISGF3, IFN-stimulated gene factor 3; E-mail address: [email protected] EBNA2, EBV nuclear Ag 2; WCE, whole-cell extract.

Copyright © 1999 by The American Association of Immunologists 0022-1767/99/$02.00 The Journal of Immunology 3835 the affinity or number of IFN-␣ receptors. A variety of defects in Antibodies STAT activation has been reported in Daudi resistant clones, in- Rabbit polyclonal antisera against Stat1, Stat2, and p48 were a generous cluding lack of Stat3 activation (12), premature loss of Stat1 phos- gift of Dr. Chris Schindler. Rabbit polyclonal Abs against human Stat3, phorylation (23), and an inability to activate an ISRE binding com- Stat5, and Stat6 were purchased from Santa Cruz Biotechnology (Santa plex (24). A similar defect in ISGF3 activation has not been Cruz, CA). The antiserum against IFN-␣RI was a generous gift of Dr. detected in EBV-transformed B cells (21), while the ability of Oscar Colamonici (University of Tennessee). these cells to activate other STAT complexes has not been thor- Immunoprecipitations oughly investigated. In this study, we have found that exposure of B cells to IFN-␣ Extracts were immunoprecipitated with anti-STAT antisera as described leads to activation of Stat6-containing complexes, which are in- previously (31). The immunoprecipitates were fractionated by 7% SDS- PAGE before immunoblotting with either an anti-phosphotyrosine Ab distinguishable from those induced by IL-4 (25–28). Stat6 activa- (4G10, Upstate Biotechnology, Lake Placid, NY) or anti-STAT antisera. tion in response to IFN-␣ is not observed in non-B cell lines. The Bands were detected by enhanced chemiluminescence (Amersham, Arling- ability of IFN-␣ to activate Stat6 in addition to previously char- ton Heights, IL). acterized STATs allows for the formation of novel Stat2:Stat6 complexes in B cells, including an ISGF3-like complex containing Northern analysis ␣ Stat2, Stat6, and p48. IFN- -mediated activation of Stat6 is mark- Total RNA was extracted by utilizing the ULTRA-SPEC II kit (Bioteck edly reduced in both EBV-transformed B cells and D-R cells. A Laboratories, Houston, TX). Northern blot analysis was performed with 10 decrease in the IFN-␣ inducibility of Stat6 binding to the IRF-1 ␮g of total RNA according to standard protocols. The blot was probed with GAS is associated with a diminished transactivation of the IRF-1 either a human IRF-1 cDNA (a generous gift of Dr. Richard Pine, Public ␣ ␣ Health Research Institute, NY) or a GAPDH cDNA radiolabeled by Phar- Downloaded from gene in response to IFN- . The ability of IFN- to activate distinct macia DNA labeling bead (ϪdCTP) kit. Stat6-containing transcriptional complexes in B cells may allow it to modulate the expression of target genes in a cell type-specific manner. Results IFN-␣ activates B cell-specific IRF-1 GAS binding complexes Materials and Methods IFN-␣ exerts pleiotropic effects on the immune system (1–3). To Cell lines investigate whether the diverse biologic activities exerted by http://www.jimmunol.org/ ␣ The human cell lines Ramos (a kind gift of Dr. Seth Lederman, Columbia IFN- on distinct cell types are reflected in a differential usage of University, New York, NY), BL-2, and BJAB (a generous gift of Dr. Ric- signaling molecules, we exposed a panel of human cell lines to cardo Dalla Favera, Columbia University) are EBV-negative Burkitt’s lym- IFN-␣2a. Cells were then harvested, and WCE were obtained and phomas. Namalwa (a generous gift of Dr. Riccardo Dalla Favera, Columbia assayed by EMSA utilizing as a probe the GAS element of the University) is an EBV-positive Burkitt’s lymphoma cell line. Subclones of IRF-1 promoter. This GAS element is able to mediate binding of the EBV-positive Burkitt’s lymphoma cell line Daudi, which are either IFN sensitive (D-S) or IFN resistant (D-R), were kind gifts of Dr. Sharon Evans a variety of distinct STAT complexes (32). As shown in Fig. 1A, (Roswell Park Cancer Institute, Buffalo, NY) and Dr. A. Hovanessian (In- IFN-␣ is able to activate multiple IRF-1 GAS binding complexes. stitut Pasteur, Paris, France) (29, 30). WIL-2-729HF2 (American Type The mobility and pattern of the faster-migrating complexes was by guest on October 2, 2021 Culture Collection (ATCC), Manassas, VA) and JY cells (a generous gift consistent with the previously described IFN-␣-activated Stat1: of Dr. Riccardo Dalla Favera, Columbia University) are EBV-transformed lymphoblastoid B cell lines. THP-1 (a kind gift of Dr. Kathryne Calame, Stat3 homo-/heterodimers (12). These complexes were detected in Columbia University) is a monocytic cell line. Jurkat (a kind gift of Dr. all of the cell lines tested. Exposure of human B cell lines to Kathryne Calame) is a human T cell leukemic line, and H9 (ATCC) is IFN-␣, however, led to the induction of an additional slow-mobil- derived from a human T cell lymphoma. All cells were grown in IMDM ity complex (complex X). Both EBV-positive (Namalwa and supplemented with 10% FCS (Atlanta Biologicals, Norcross, GA). WI38 Daudi) and -negative (Ramos, BL-2, and BJAB) Burkitt’s lym- VA (a human embryonic lung fibroblast) and WISH (a human epithelial- like amnion tissue-derived cell line) were a kind gift of Dr. Chris Schindler phoma cell lines were found to activate this slow-mobility com- (Columbia University) and were grown in DMEM supplemented with plex, although its intensity varied among the individual B cell lines 10% FCS. (Fig. 1A, and data not shown). This complex was, however, almost Cell culture undetectable or even absent in two EBV-transformed B cell lines (WIL-2 and JY) (Figs. 1A and 6A and data not shown). Kinetic ϫ 6 Cells (20–50 10 ) were incubated at 37°C for varying periods of times experiments revealed that the induction of this B cell-specific com- in a final volume of 10 ml. Cells were stimulated utilizing the following ␣ cytokine concentrations: human IL-4 (100 U/ml; a generous gift of Dr. Paul plex in response to IFN- was short lived (Fig. 1B), because it had Rothman, Columbia University), human IFN-␣2a (1000 U/ml; a generous disappeared after2hofculture with IFN-␣. These kinetics differed gift of Dr. Chris Schindler), and human IFN-␥ (10 ng/ml; PeproTech, sharply from those displayed by the Stat1-containing complexes or Rocky Hill, NJ). by the previously characterized ISGF3 complex binding to an DNA Binding Assays and Cell Extracts ISRE element (Fig. 1B and data not shown) (11, 33). Kinetic experiments in a non-B cell line (THP-1) failed to reveal the presence The preparation and use of DNA oligonucleotide probes for mobility shift assays have been described previously (31). The probes used in these stud- of complex X at any of the time points examined, suggesting that ies were as follows: IRF-1 GAS, 5Ј-gatcGATTTCCCGAAAT-3Ј; CD23b differences in the kinetics of activation are not responsible for lack GAS, 5Ј-gatcGGGTGAATTTCTAAGAAAGGGAC-3Ј; and ISG-15 of induction of complex X in non-B cells (Fig. 1B). ISRE, 5Ј-gatcCTCGGGAAAGGGAAACCGAAACTGAAGCC-3Ј. Dou- Induction of the slow-mobility IRF-1 GAS binding complex by ble-stranded oligonucleotides used as cold competitors were prepared from IFN-␣ was also detected in extracts from fresh peripheral blood single-stranded oligonucleotides (Life Technologies) with the following sequences: ␤CAS-GAS, 5Ј-gatcGACTTCTTGGAATT3-Ј;I⑀-GAS (Ϫ119 mononuclear cells obtained from a patient with chronic lympho- to Ϫ104), 5Ј-gatcAACTTCCCAAGAACAG-3Ј; Ly6E-GAS, 5Ј-gat- cytic leukemia (which contain a large number of circulating ma- cATATTCCTGTAAGT-3Ј (31). STAT antisera were added (final dilution, ture B cells) (data not shown). These results suggest that activation 1:20) for 30–60 min at 4°C, before a standard 20-min (25°C) incubation of of complex X is not due to long-term culture of the Burkitt’s lym- extracts with shift probe. Oligonucleotide competitions were performed by adding a 100-fold excess of cold oligonucleotides for 15 min (25°C) before phoma cell lines. a standard 20-min (25°C) incubation of extracts with shift probe. These data thus indicate that IFN-␣ can induce cell type-specific Whole-cell extracts (WCE) were prepared as described previously (31). IRF-1 GAS binding complexes. 3836 IFN-␣ ACTIVATES Stat6 IN B CELLS Downloaded from

FIGURE 1. A, IFN-␣ induces distinct IRF-1 GAS binding complexes in

human B cell vs non-B cell lines. Extracts were prepared from a panel of http://www.jimmunol.org/ B cell lines (Ramos, Namalwa, and WIL-2) or non-B cell lines (Jurkat, H9, and THP-1) with or without a 20-min stimulation at 37°C with IFN-␣2a (1000 U/ml). The WCE were examined by EMSA using an IRF-1 GAS probe. B, Kinetics of IFN-␣ activation of the B cell-specific IRF-1 GAS binding complex. Extracts were prepared from a human B cell line, Ramos, FIGURE 2. A, GAS oligonucleotide competition pattern exhibited by or a human monocytic cell line, THP-1, before or after stimulation with the IFN-␣-induced IRF-1 GAS binding complexes. Ramos cells were stim- IFN-␣2a (1000 U/ml) for 20 min, 45 min, or 2 h. The WCE were examined ulated with human IFN-␣2a (1000 U/ml) for 20 min. Extracts were then by EMSA using an IRF-1 GAS probe. prepared and examined as described in Fig. 1. Oligonucleotide competition assays were performed either in the absence or in the presence of a 100- fold molar excess of cold GAS oligonucleotides added to the shift reaction by guest on October 2, 2021 The B cell-specific DNA binding complex activated by IFN-␣ as indicated. Competitors included I⑀-GAS, Ly6E-GAS, ␤CAS-GAS, an contains Stat6 irrelevant NF-␬B binding site, and the IRF-1 GAS. B, The CD23b GAS selectively binds the slow-mobility complex activated by IFN-␣ in Ramos. To determine the identity of the DNA binding activity induced by ␣ ␣ Cells either were unstimulated or were stimulated with human IFN- 2a IFN- in B cells, we subjected these complexes to oligonucleotide (1000 U/ml) or human IL-4 (100 U/ml) for 20 min. Extracts were prepared competitions with a panel of GAS elements known to discriminate as described above and examined by EMSA utilizing either a CD23b GAS among distinct STAT complexes (Fig. 2A). The pattern of com- probe (left panel) or an IRF-1 GAS probe (right panel). petition displayed by the slow-mobility complex found in B cells (complex X) was identical with the one previously described for Stat6 (31) and contrasted to the one exhibited by the Stat1:Stat3- complex activated by IFN-␣ was found to migrate with a mobility containing complexes. Specifically, competition was observed identical with that of the IL-4-inducible complex and to contain with the I⑀ GAS (which binds Stat6 but not Stat1) but only min- Stat6 or a protein antigenically related to it. As previously reported imally with the LY6E GAS (which binds Stat1 but not Stat6). The in other cell types (12), the faster mobility complexes contained ␤-CAS GAS competed all complexes in a manner identical with Stat1 homodimers, Stat3 homodimers, and Stat1:Stat3 het- that of the IRF-1 GAS, whereas an irrelevant oligonucleotide con- erodimers. Neither the Stat5 nor the Stat2 antiserum significantly taining an NF-␬B binding site failed to compete any of the affected any of the IRF-1 GAS binding complexes induced by complexes. IFN-␣ in Ramos cells. Interestingly, the inability of non-B cells to The ability of the slow-mobility complex activated in response activate Stat6 in response to IFN-␣ was not due to a global defect to IFN-␣ to target Stat6-selective GAS elements was further con- in Stat6 activation, because, except for Jurkat cells, these cells firmed by simultaneously testing IFN-␣-stimulated extracts from a displayed normal induction of Stat6 complexes in response to IL-4 B cell line, Ramos, with two distinct probes, the IRF-1 GAS and (Fig. 3B and data not shown). These results thus demonstrate that the CD23b GAS (Fig. 2B) (34). In contrast to the IRF-1 GAS, the IFN-␣ is able to activate a B cell-specific DNA binding complex CD23b GAS only detected the slow-mobility complex activated by that is indistinguishable from the IL-4-inducible Stat6 complex by IFN-␣. The mobility of this complex was identical with that of the gel-shift mobility, as well as by oligonucleotide and antiserum IL-4-inducible Stat6 complex, suggesting that complex X might competition patterns. contain Stat6. ␣ To identify further the components of complex X, we subjected Activation of Stat2:Stat6 complexes in B cells by IFN- the IFN-␣-inducible complexes found in Ramos to Ab competi- To confirm that Stat6 was indeed activated in response to IFN-␣ in tions with a panel of STAT antisera (Fig. 3A). Consistent with the B cells, we then performed immunoprecipitations with a Stat6 an- results obtained above, the B cell-specific IRF-1 GAS binding tiserum on extracts from a panel of cell lines cultured with IFN-␣ The Journal of Immunology 3837 Downloaded from

FIGURE 4. Stat6 is tyrosine phosphorylated in response to IFN-␣ in B cell but not in non-B cell lines. Ramos, Daudi, WISH, and WI38 VA cells either were unstimulated or were stimulated with IFN-␣2a (1000 U/ml) for 20 min. Extracts were then immunoprecipitated (IP) with a Stat6 antiserum, resolved by 7% SDS-PAGE, and analyzed by Western blotting http://www.jimmunol.org/ (WB) using an antiphosphotyrosine Ab (top panel). The blot was later ␣ FIGURE 3. A, The IRF-1 GAS binding complexes induced by IFN- in stripped and reprobed with a Stat6 antiserum to ensure equal loading of B cells contain Stat6. Ramos cells either were unstimulated or were stim- immunoprecipitates (middle panel). After restripping, the blot was re- ␣ ulated with human IFN- 2a (1000 U/ml), human IL-4 (100 U/ml), or hu- probed with a Stat2 antiserum to confirm the presence of Stat2 in the Stat6 ␥ man IFN- (10 ng/ml) for 20 min. Extracts were then prepared and exam- immunoprecipitates from IFN-␣-treated extracts (bottom panel). Extracts ined as described in Fig. 1. Ab interference mobility shift assays were from IL-4-stimulated Ramos cells were included as a control for Stat6 conducted by addition of antisera against Stat1, Stat2, Stat3, Stat5, or Stat6 activation. or preimmune serum (control). All antisera were added at a final dilution of 1:20 for 30 min at 4°C before incubation with the IRF-1 GAS probe for by guest on October 2, 2021 20 min at 25°C. B, Non-B cell lines activate Stat6 in response to IL-4 but not to IFN-␣; Ramos, H9, and THP-1 cells either were unstimulated or STATs in addition to Stat6 (Fig. 4) or, as previously described, to were stimulated with human IFN-␣2a (1000 U/ml) or human IL-4 (100 Stat1 (10). We thus subjected Ramos extracts cultured with or U/ml) for 20 min. Extracts were prepared and examined by EMSA utilizing without IFN-␣ to immunoprecipitations with a panel of antisera an IRF-1 GAS probe as described in Fig. 1. against distinct STATs. As shown in Fig. 5A, tyrosine phosphorylation of all of the STATs assayed was observed in response to (Fig. 4). Consistent with our previous observations, tyrosine phos- IFN-␣. Interestingly, in this experiment, we observed that cultur- phorylation of a 100-kDa protein was observed in response to both ing Ramos cells with IFN-␣ led to the activation of Stat5, which, IFN-␣ and IL-4 in Ramos cells (Fig. 4, top panel). Despite com- possibly because of the strong induction of the Stat1:Stat3-con- parable levels of inactive Stat6 in the immunoprecipitates from taining complexes, had not been detected by EMSA. Multiple ty- non-B cell lines (WI38 VA and WISH) (Fig. 4, middle panel), rosine-phosphorylated proteins were detected only in the Stat2 and IFN-␣ treatment led to the tyrosine phosphorylation of Stat6 only Stat6 immunoprecipitates (Fig. 5A, top panel). Stripping and re- in B cell lines (Ramos and Daudi) (Fig. 4, top panel). The inability probing of this blot with antisera against Stat2 revealed that Stat2 of IFN-␣ to activate Stat6 in non-B cells was not due to unrespon- coimmunoprecipitated only with Stat1 or Stat6 but not with Stat3 siveness of these cells to IFN-␣, because IFN-␣ induction of or Stat5 (Fig. 5A, middle panel). As previously described, com- ISGF3 was detected in both WISH and WI38 VA cells. plexing of Stat2 with Stat1 occurred in the absence of ligand (37), These experiments had revealed that an additional tyrosine while formation of the Stat2:Stat6 heterodimer was only detected phosphorylated ϳ110-kDa protein was coimmunoprecipitated on exposure of cells to IFN-␣. Reprobing with Stat6 antiserum with activated Stat6 in IFN-␣-treated extracts (Fig. 4, top panel). demonstrated complexing of Stat6 only with Stat2 but not with Because Stat2 is known to become phosphorylated in response to other STATs (Fig. 5A, bottom panel). Reprobing of this Western IFN-␣ and to migrate on SDS-PAGE gel to a ϳ110-kDa position filter with a Stat1 antiserum confirmed the presence of Stat1 in the (35, 36), we entertained the notion that Stat2 might complex with Stat2 but not in the Stat6 immunoprecipitation (data not shown). Stat6 in B cells. Reprobing of this blot with a Stat2 antiserum Moreover, reprobing with either Stat3 or Stat5 antisera ensured the indeed confirmed that Stat2 can be detected in Stat6 immunopre- adequacy of these immunoprecipitations and established the lack cipitates from IFN-␣-treated B cells (Fig. 4, bottom panel). Low of interaction of these proteins with either Stat2 or Stat6 (data not levels of Stat2 also coimmunoprecipitated with Stat6 in IL-4- shown). treated extracts (Fig. 4, bottom panel), although activation of Stat2 We next proceeded to determine whether the interaction of Stat2 was not detected in response to IL-4 (Fig. 4, top panel). and Stat6 in IFN-␣-stimulated B cells could lead to the formation The ability of IFN-␣ to activate a variety of STATs in B cells of a multimeric complex with p48. To address this possibility, we (Fig. 3A) raised the possibility that Stat2 might interact with other conducted EMSA experiments with the ISG15 ISRE probe (Fig. 3838 IFN-␣ ACTIVATES Stat6 IN B CELLS

5B). These assays revealed that stimulation of Ramos cells with IFN-␣ leads to the induction of two distinct complexes. The faster- mobility IFN-␣-inducible complex was activated by IFN-␣ in both Ramos and WI38 VA and corresponds to the previously described ISGF3 complex. In contrast, the slower-mobility IFN-␣-inducible complex was activated only in Ramos and was found, by super- shifting experiments, to contain Stat2, Stat6, and p48 (Fig. 5B). Thus, stimulation of B cells with IFN-␣2a leads to the interaction of Stat2 and Stat6 and to the formation of ISGF3-like complexes containing Stat2, Stat6, and p48. To assess whether, in B cells, Stat6 can associate with the IFN-␣ receptor, we then conducted immunoprecipitation experiments with an antiserum directed against the IFN␣RI chain (38). These studies revealed that Stat6 interacts with the IFN␣RI chain in the absence of IFN-␣ stimulation (Fig. 5C, bottom panel). Exposure of Ramos cells to IFN-␣ led to a decrease in the association of Stat6 with IFN␣RI as well as to the concomitant tyrosine phosphorylation of the IFN␣RI chain (Fig. 5C, top panel). Interestingly, preassociation of Stat2 with the IFN␣RII chain has been previously reported and is believed to be a critical step in Stat2 activation (39). Downloaded from

B cell lines resistant to IFN-␣ display markedly reduced activation of Stat6 in response to IFN-␣ Our survey of different B cell lines displayed heterogeneity in the intensity of the IFN-␣-inducible Stat6 complex. In particular, we noticed that EBV-transformed B cell lines (i.e., JY or WIL-2), http://www.jimmunol.org/ despite a robust activation of Stat6 in response to IL-4, displayed minimal Stat6 induction on exposure to IFN-␣ (Figs. 1A and 6A and data not shown). EBV-transformed B cell lines have been shown to be resistant to the antiproliferative effects of IFN-␣ on B cells (20). We thus became interested in determining whether IFN-␣ resistance in B cells was accompanied by a decrease in Stat6 activation. To investigate this issue, we utilized two distinct

subclones of the B cell line Daudi. This cell line displays exquisite by guest on October 2, 2021 sensitivity toward growth inhibition by IFN-␣. However, subclones that are resistant to these effects have been derived (29, 40–42). We thus utilized an IRF-1 GAS probe to assess the IFN-␣ inducibility of STAT complexes in IFN-sensitive (D-S) and -resistant (D-R) cell lines. As shown in Fig. 6A, the most striking difference between the two Daudi subclones consisted of a markedly reduced activation of Stat6 complexes in the IFN-resistant cells. Interestingly, both the D-S and D-R subclones were unable to activate Stat6 in response to IL-4 (data not shown). Because the EMSA experiments performed with the ISG-15 ISRE probe had previously identiﬁed a Stat2:Stat6:p48 complex (Fig. 5B), we then proceeded to investigate whether assembly of such a complex was also impaired in IFN-resistant cells. As shown in Fig. 6B, formation of a Stat2:Stat6:p48 complex on IFN-␣ stimulation could not be detected in either D-R subclones or EBV- FIGURE 5. A, Stat2 complexes with Stat6 in an IFN-␣-stimulated B cell line. Ramos cells were either unstimulated or were stimulated with transformed B cells. Consistent with previous reports, ISGF3 com- IFN-␣2a (1000 U/ml). Extracts were then immunoprecipitated (IP) with a plex formation was defective in D-R subclones but occurred panel of anti-STAT antiserum, resolved by 7% SDS-PAGE, and analyzed normally in EBV-transformed B cells (21, 24). To conﬁrm that the by Western blotting (WB) using an antiphosphotyrosine Ab (top panel). lack of Stat6 DNA binding activity in IFN-resistant cells was due The blot was later stripped and reprobed with a Stat2 (middle panel)ora to a lack of appropriate activation of Stat6 in response to IFN-␣, Stat6 (bottom panel) antiserum. B, IFN-␣ activates a Stat2:Stat6:p48 com- we then subjected extracts from D-S as well as D-R subclones to plex in B cell but not in non-B cell lines. Extracts were prepared from a immunoprecipitation with a Stat6 antiserum. As shown in Fig. 6C, human B cell line, Ramos, or a non-B cell line, WI38 VA, before or after stimulation with IFN-␣ led to tyrosine phosphorylation of Stat6 ␣ stimulation with IFN- 2a (1000 U/ml) for 20 min. The WCE were exam- only in D-S but not in D-R cells. ined by EMSA using an ISG15 ISRE probe. Ab interference mobility shift assays were conducted by addition of antisera against Stat2, Stat6, p48, or preimmune serum (control) as described in Fig. 3A. C, Stat6 preassociates with the IFN␣RI chain. Ramos cells either were unstimulated or were stimulated with IFN-␣2a (1000 U/ml) for 20 min. Extracts were then im- tiphosphotyrosine Ab (top panel). The blot was later stripped and reprobed munoprecipitated (IP) with either an IFN␣RI or a Stat6 antiserum, resolved with an IFN␣RI antiserum (middle panel). After restripping, the blot was by 7% SDS-PAGE, and analyzed by Western blotting (WB) using an an- reprobed with a Stat6 antiserum (bottom panel). The Journal of Immunology 3839

Because the IRF-1 GAS has previously been shown to be the critical element regulating IFN-␣ inducibility of the IRF-1 gene (43), we then tested whether the diminished ability of D-R subclones to activate Stat6 binding to the IRF-1 GAS is reﬂected in a decreased induction of the IRF-1 gene in response to IFN-␣.We therefore assayed by Northern analysis total RNA obtained from D-S and D-R cells that had either been unstimulated or stimulated with IFN-␣ (Fig. 6D). Probing the Northern blot with a radiolabeled IRF-1 cDNA probe demonstrated that, when compared with the D-S cells, the D-R subclone displayed a 30% reduction (by densitometry) in the IFN-␣-mediated induction of IRF-1. Thus, in B cells, a decrease in the Stat6 activation in response to IFN-␣ correlates with diminished IFN-␣ inducibility of the IRF-1 gene and with the acquisition of IFN resistance.

Discussion It has long being recognized that IFN-␣ exerts pleiotropic actions on a variety of target cells (1–3). The molecular mechanisms lead-

ing to such diverse effects, however, have not been fully eluci- Downloaded from dated. We have now demonstrated that, in B cells, IFN-␣ leads to the activation of a Stat6-containing complex that is indistinguishable from that activated by IL-4 (25–28). The ability of IFN-␣ to activate Stat6 is intriguing, because, unlike other STATs, Stat6 has been primarily linked to the signaling pathways triggered by cy-

tokines that exhibit overlapping biologic activities, i.e., IL-4 and http://www.jimmunol.org/ IL-13 (9). In contrast to the IL-4-mediated Stat6 activation, however, appearance of the Stat6 complex in response to IFN-␣ is short lived (Fig. 1B and data not shown). This restricted pattern of Stat6 activation may thus allow IFN-␣ to exert B cell-speciﬁc effects without fully mimicking the IL-4-mediated biologic responses. The ability of IFN-␣ to activate Stat6 and possibly to target some IL-4-responsive genes may also play a role in the inappropriate induction of CD23b exhibited by B cells of CLL

patients in response to IFN-␣ (44). by guest on October 2, 2021 IFN-␣-mediated Stat6 activation is accompanied by the formation of novel Stat2:Stat6 complexes. In contrast to Stat1:Stat2 heterodimers that, consistent with previous studies, were detected as preexisting latent complexes (37), association of Stat2 and Stat6 appears to be dependent on the presence of ligand. Assembly of the Stat2:Stat6 dimeric complex may thus require a distinct set of interactions from those leading to Stat1:Stat2 complex formation. Also, in contrast to the Stat1:Stat2 heterodimer, the Stat2:Stat6 heterodimer does not appear to target the IRF-1 GAS, because Ab interference assays with a Stat2 antiserum did not affect the Stat6 complex activated by IFN-␣. It remains to be determined whether the Stat2:Stat6 heterodimer can target a distinct set of GAS-like elements.

were stimulated with human IFN-␣2a (1000 U/ml) for 20 min. Extracts were then prepared and examined utilizing an ISG15 ISRE probe as described in Fig. 5B. C, Stat6 is not tyrosine phosphorylated in response to IFN-␣ in a D-R subclone. Cells from Ramos, D-S, and D-R either were unstimulated or were stimulated with IFN-␣2a (1000 U/ml) for 20 min. Extracts were then immunoprecipitated (IP) with a Stat6 antiserum, resolved by 7% SDS-PAGE, and analyzed by Western blotting (WB) using FIGURE 6. A, IFN-␣ activation of Stat6 is markedly diminished in IFN- an antiphosphotyrosine Ab (upper panel). The blot was later stripped and resistant human B cell lines. Cells from a Daudi subclone sensitive to reprobed with a Stat6 antiserum (lower panel). D, Induction of IRF-1 in IFN-␣ (D-S), a Daudi subclone resistant to IFN-␣ (D-R), and an EBV- response to IFN-␣ stimulation is diminished in DR cells. Cells either were transformed B cell line (JY) were stimulated with human IFN-␣2a (1000 unstimulated or were stimulated with human IFN-␣2a (1000 U/ml) for 1 h. U/ml) for 1 h. Extracts were then prepared and examined utilizing an IRF-1 Total RNA was then extracted, and 10 ␮g of RNA was assayed by North- GAS probe as described in Fig. 1. B, IFN-␣ activation of Stat2:Stat6:p48 ern blotting as per standard protocols. The blot was probed with 32P-la- complexes is absent in IFN-resistant human B cell lines. Cells from D-S beled IRF-1 cDNA (upper panel), stripped, and then reprobed with 32P- subclone, D-R subclone, JY cell line, and a non-B cell line, WI38 VA, labeled GAPDH cDNA (lower panel). 3840 IFN-␣ ACTIVATES Stat6 IN B CELLS

Our studies have also revealed that interaction of Stat6 with Acknowledgments Stat2 can lead to the formation of an ISGF3-like complex com- We thank J. Siu for his critical reading of this manuscript. posed of Stat2, Stat6, and p48. Because previous studies utilizing a yeast two-hybrid system failed to detect a direct interaction between p48 and Stat6 (45), presence of Stat2 and/or specific post- References translational modifications may be required for the formation of 1. Pestka, S. 1997. The human interferon-␣ species and hybrid proteins. Semin. this ternary complex. Consistent with studies from other groups, Oncol. 24(Suppl. 9):4. ␣ 2. Pfeffer, L. 1997. Biologic activities of natural and synthetic type I interferons. we have found that IFN- induces ISGF3 normally in EBV-trans- Semin. Oncol. 24(Suppl. 9):63. formed B cells (20) but not in D-R cells (24). However, both cell 3. Herberman, R. 1997. Effect of ␣-interferons on immune function. Semin. Oncol. types are unable to activate the Stat2:Stat6:p48 complex on stim- 24(Suppl. 9):78. ␣ 4. Pestka, S. 1997. The interferon receptors. Semin. Oncol. 24(Suppl. 9):18. ulation with IFN- . Because transactivation of the ISG15 gene in 5. Ghislain, J., G. Sussman, S. Goelz, L. Ling, and E. Fish. 1995. Configuration of response to IFN-␣ has previously been shown to be defective in the interferon-␣/␤ receptor complex determines the context of the biological re- both cell types (20, 24), these findings suggest that, in B cells, sponse. J. Biol. Chem. 270:21785. 6. Overall, M. L., P. Chambers, and P. J. Hertzog. 1992. Different interactions of activation of the Stat2:Stat6:p48 complex may be required for the interferon-␣ subtypes at the surface of epithelial and lymphoid cells. J. Interferon appropriate IFN-␣ induction of this gene. Res. 1992;12:281. IL-4-mediated Stat6 phosphorylation has been linked to activa- 7. Schindler, C., and J. Darnell. 1995. Transcriptional responses to polypeptide li- gands: the JAK-STAT pathway. Annu. Rev. Biochem. 64:621. tion of Jak1 and Jak3 (46, 47) or of Jak1 alone (48–53). Prelim- 8. Ihle, J., B. Witthuhn, F. Quelle, K. Yamamoto, and O. Silvennoinen. 1995. Sig- inary experiments have failed to detect IFN-␣-inducible activation naling through the hematopoietic cytokine receptors. Annu. Rev. Immunol. 13: of Jak3 in B cells (data not shown). Thus, tyrosine phosphorylation 369. ␣ 9. O’Shea, J. J. 1997. Jaks, STATs, cytokine signal transduction, and immunoregu- Downloaded from of Stat6 in response to IFN- does not appear to be mediated by a lation: are we there yet? Immunity 7:1. different set of Janus kinases, and IFN-␣ activation of Jak1 is 10. Schindler, C., K. Shuai, V. Prezioso, and J. E. Darnell. 1992. Interferon-depen- likely to be sufficient for Stat6 phosphorylation. The mechanism dent tyrosine phosphorylation of a latent cytoplasmic transcription factor. Science ␣ 257:809. underlying the selective ability of IFN- to activate Stat6 in B cell 11. Schindler, C., X.-Y. Fu, T. Improta, R. Aebersold, and J. E. Darnell. 1992. Pro- lines is at present unclear. However, we favor the hypothesis that teins of transcription factor ISGF-3: one gene encodes the 91- and 84-kDa ␣ Stat6 activation may be due either to different posttranslational ISGF-3 proteins that are activated by interferon . Proc. Natl. Acad. Sci. USA 89:7836. http://www.jimmunol.org/ modifications of the IFN␣RI receptor chain in B cells or to the 12. Yang, C., W. Shi, L. Basu, A. Murti, S. Constantinescu, L. Blatt, E. Croze, recruitment of additional, possibly B cell-specific, components to J. Mullersman, and L. Pfeffer. 1996. Direct association of STAT3 with the the IFN-␣ receptor complex. IFNAR-1 chain of the human type I interferon receptor. J. Biol. Chem. 271:8057. 13. Meinke, A., F. Barahmand-Pour, S. Wohrl, D. Stoiber, and T. Decker. 1996. The phenomenon of IFN resistance has long been recognized as Activation of different Stat5 isoforms contributes to cell-type-restricted signaling carrying profound implications for the diverse clinical uses of in response to interferons. Mol. Cell. Biol. 16:6937. ␣ 14. Cho, S., C. Bacon, C. Sudarshan, R. Rees, D. Finbloom, R. Pine, and J. O’Shea. IFN- . Recent studies have suggested that multiple mechanisms 1996. Activation of Stat4 by IL-12 and IFN-␣: evidence for the involvement of may be involved in the acquisition of IFN resistance. For example, ligand-induced tyrosine and serine phosphorylation. J. Immunol. 157:4781. in T cell lines, changes in the ability of the IFN-␣ receptor to 15. Williams, B. R., and S. J. Haque. 1997. Interacting pathways of interferon signaling. Semin. Oncol. 24(Suppl. 9):70. complex with CD45 (54) as well as Stat1-dependent mechanisms 16. Lindner, D., D. Kalvakolanu, and E. Borden. 1997. Increasing effectiveness of by guest on October 2, 2021 have been described (55). IFN resistance in Daudi cells instead has interferon-␣ for malignancies. Semin. Oncol. 24(Suppl. 9):99. been associated with either lack of Stat3 phosphorylation (12) or 17. Dorr, R. 1993. Interferon-␣ in malignant and viral diseases. Drugs 45:177. 18. McLaughlin, P. 1996. The role of interferon in the therapy of malignant lym- premature loss of Stat1 phosphorylation (23). Our finding that phoma. Biomed. Pharmacother. 50:140. Stat6 activation by IFN-␣ was markedly diminished in both D-R as 19. Sato, T., C. Selleri, N. Young, and J. Maciejewski. 1997. Inhibition of interferon well as EBNA2-containing B cell lines suggests that changes in regulatory factor-1 expression results in predominance of cell growth stimulatory effects of interferon-␥ due to phosphorylation of Stat1 and Stat3. Blood 90:4749. Stat6 phosphorylation are also associated with the acquisition of 20. Aman, P., and A. von Gabain. 1990. An Epstein-Barr virus immortalization as- IFN resistance. The decreased/absent Stat6 binding to the IRF-1 sociated gene segment interferes specifically with the IFN-induced anti- GAS in IFN-resistant cells was associated with a diminished in- proliferative response in human B-lymphoid cell lines. EMBO J. 9:147. 21. Kanda, K., T. Decker, P. Aman, M. Wahlstrom, A. von Gabain, and B. Kallin. duction of IRF-1, a gene involved in the regulation of cellular 1992. The EBNA2-related resistance toward ␣ interferon (IFN-␣) in Burkitt’s proliferation. The residual IRF-1 induction detected in D-R cells is lymphoma cells effects induction of IFN-induced genes but not the activation of likely to be mediated by the other STAT complexes, which bind to transcription factor ISGF-3. Mol. Cell. Biol. 12:4930. 22. McMahon, M., G. Stark, and I. Kerr. 1986. Interferon-induced gene expression in the IRF-1 GAS (Fig. 3A), because activation of these complexes wild-type and interferon-resistant human lymphoblastoid (Daudi) cells. J. Virol. can still occur normally in D-R cells (Fig. 6A). However, IRF-1 57:362. function is antagonized by the closely related factor IRF-2, the 23. Grimley, P., H. Fang, H. Rui, E. Petricoin, S. Ray, F. Dong, K. Fields, R. Hu, K. Zoon, S. Audet, and J. Beeler. 1998. Prolonged Stat1 activation related to the levels of which are normally higher than those of IRF-1 and the growth arrest of malignant lymphoma cells by interferon-␣. Blood 91:3017. growth arrest of which has been shown to be related to changes in 24. Kessler, D., R. Pine, L. Pfeffer, D. Levy, and J. J. Darnell. 1988. Cells resistant to interferon are defective in activation of a promoter-binding factor. EMBO J. the ratio of these two factors (56). Therefore, the submaximal 7:3779. IRF-1 induction displayed by D-R cells may be unable to over- 25. Kohler, I., and E. Rieber. 1993. Allergy-associated I⑀ and Fc⑀ receptor II come the antagonistic effects of IRF-2, thus providing a link be- (CD23b) genes activated via binding of an interleukin-4-induced transcription factor to a novel responsive element. Eur. J. Immunol. 23:3066. tween defects in Stat6 activation and protection against the anti- 26. Kotanides, H., and N. C. Reich. 1993. Requirement of tyrosine phosphorylation proliferative actions of IFN-␣. for rapid activation of a DNA binding factor by IL-4. Science 262:1265. The multiple differences in STAT activation displayed by IFN- 27. Schindler, C., H. Kashleva, A. Pernis, R. Pine, and P. Rothman. 1994. STF-IL4: A novel IL-4 induced signal transducing factor. EMBO J. 13:1350. resistant cell lines are likely not to be mutually exclusive and may 28. Hou, J., U. Schindler, W. Henzel, T. Ho, M. Brasseur, and S. McKnight. 1994. stem from a common mechanism. Differential IFN-␥ responsive- An interleukin-4-induced transcription factor: IL-4 Stat. Science 265:1701. ness in T cell subsets has previously been ascribed to changes in 29. Evans, S., R. Collea, J. Leasure, and D. Lee. 1993. IFN-␣ induces homotypic adhesion and Leu-13 expression in human B lymphoid cells. J. Immunol. 150: receptor components (57, 58). It will thus be interesting to deter- 736. mine whether modulation of specific receptor components, e.g., by 30. Meurs, E., and A. Hovanessian. 1988. ␣-Interferon inhibits the expression of the employment of alternatively spliced forms of the receptor heavy chain ␮ messenger RNA in Daudi cells. EMBO J. 7:1689. ␣ 31. Pernis, A., S. Gupta, J. Yopp, H. Kashleva, C. Schindler, and P. Rothman. 1995. chains, may allow IFN- -resistant cell lines to selectively extin- ␥ chain-associated cytokine receptors signal through distinct transducing factors. guish the antiproliferative effects of IFN-␣. J. Biol. Chem. 270:14517. The Journal of Immunology 3841

32. Decker, T., P. Kovarik, and A. Meinke. 1997. GAS elements: a few nucleotides 46. Witthuhn, B., O. Silvennoinen, O. Miura, K. Lai, C. Cwik, E. Liu, and J. Ihle. with a major impact on cytokine-induced gene expression. J. Interferon Cytokine 1994. Involvement of the Jak-3 Janus kinase in signalling by interleukins 2 and Res. 17:121. 4 in lymphoid and myeloid cells. Nature 370:153. 33. Levy, D., D. Kessler, R. Pine, and J. J. Darnell. 1989. Cytoplasmic activation of 47. Johnston, J., M. Kawamura, R. Kirken, Y. Chen, T. Blake, K. Shibuya, J. Ortaldo, ISGF-3, the positive regulator of interferon-␣-stimulated transcription, reconsti- D. McVicar, and J. O’Shea. 1994. Phosphorylation and activation of the Jak-3 tuted in vitro. Genes Dev. 3:1362. Janus kinase in response to interleukin-2. Nature 370:151. 34. Gupta, S., D. Xia, M. Jiang, S. Lee, and A. Pernis. 1998. Signaling pathways 48. Oakes, S., F. Candotti, J. Johnston, Y. Chen, J. Ryan, N. Taylor, X. Liu, mediated by the TNF- and cytokine-receptor families target a common cis-ele- L. Hennighausen, L. Notarangelo, W. Paul, et al. 1996. Signaling via IL-2 and ment of the IRF-1 promoter. J. Immunol. 161:5997. IL-4 in JAK3-deficient severe combined immunodeficiency lymphocytes: JAK- 35. Fu, X.-Y., D. Kessler, S. Veals, D. Levy, and J. J. Darnell. 1990. ISGF3, the 3-dependent and independent pathways. Immunity 5:605. transcriptional activator induced by IFN-␣, consists of multiple interacting 49. Kammer, W., A. Lischke, R. Morigg, B. Groner, A. Ziemiecki, C. Gurniak, polypeptide chains. Proc. Natl. Acad. Sci. USA 87:8555. L. Berg, and K. Friedrich. 1996. Homodimerization of interleukin-4 receptor ␣ 36. Fu, X., C. Schindler, T. Improta, R. Aebersold, and J. E. Darnell. 1992. The chain can induce intracellular signaling. J. Biol. Chem. 271:23634. proteins of ISGF-3, the IFN-a induced transcription activator, define a new gene 50. Lai, S., J. Molden, K. Puck, M. White, and M. Goldsmith. 1996. Interleukin-4- family of signal transducers. Proc. Natl. Acad. Sci. USA 89:7840. specific signal transduction events are driven by homotypic interactions of the 37. Stancato, L., M. David, C. Carter-Su, A. Larner, and W. Pratt. 1996. Preasso- interleukin-4 receptor ␣ subunit. EMBO J. 15:4506. ciation of STAT1 with STAT2 and STAT3 in separate signaling complexes prior 51. Chen, X., B. Patel, L. Wang, M. Frankel, N. Ellmore, R. Flavell, W. LaRochelle, to cytokine stimulation. J. Biol. Chem. 271:4134. and J. Pierce. 1997. Jak1 expression is required for mediating interleukin-4-in- 38. Domanski, P., M. Witte, M. Kellum, M. Rubinstein, R. Hackett, P. Pitha, and duced tyrosine phosphorylation of insulin receptor substrate and Stat6 signaling O. R. Colamonici. 1995. Cloning and expression of a long form of the ␤ subunit molecules. J. Biol. Chem. 272:6556. of the interferon ␣␤ receptor that is required for signaling. J. Biol. Chem. 270: 52. Fujiwara, H., S. Hanissian, A. Tsytsykova, and R. Geha. 1997. Homodimeriza- 21606. tion of the human interleukin 4 receptor ␣ chain induces Cepsilon germline tran- 39. Li, X., S. Leung, I. M. Kerr, and G. R. Stark. 1997. Functional subdomains of scripts in the absence of the interleukin 2 receptor ␥ chain. Proc. Natl. Acad. Sci. STAT2 required for preassociation with the ␣ interferon receptor and for signal- 94:5866. ing. Mol. Cell. Biol. 17:2048. 53. Reichel, M., B. Nelson, P. Greenberg, and P. Rothman. 1997. The IL-4 receptor 40. Hovanessian, A., E. Meurs, A. Laurent, and J. Svab. 1986. Specific inhibition of ␣-chain cytoplasmic domain is sufficient for activation of JAK-1 and Stat6 and IgM synthesis in interferon treated Daudi cells: an effect correlated with anti- the induction of IL-4-specific gene expression. J. Immunol. 158:5860. Downloaded from growth action of interferon. In UCLA Symposia: Interferons as Growth Factors 54. Petricoin, E. F., S. Ito, B. L. Williams, S. Audet, L. F. Stancato, A. Gamero, and Antitumor Factors. R. M. Friedman, T. Streevalsan, and T. Merigan, eds. K. Clouse, P. Grimley, A. Weiss, J. Beeler, et al. 1997. Antiproliferative action Liss, New York, pp. 48–66. of interferon-␣ requires components of T-cell-receptor signaling. Nature 390: 41. Pfeffer, L. M., N. Stebbing, and D. B. Donner. 1987. Cytoskeletal association of 629. human ␣-interferon-receptor complexes in interferon-sensitive and -resistant 55. Sun, W., C. Pabon, Y. Alsayed, P. Huang, S. Jandeska, S. Uddin, L. Platanias, lymphoblastoid cells. Proc. Natl. Acad. Sci. USA 84:3249. and S. Rosen. 1998. Interferon-␣ resistance in a cutaneous T-cell lymphoma cell 42. Scarozza, A. M., T. J. Collins, and S. S. Evans. 1992. DNA synthesis in nuclei line is associated with lack of STAT1 expression. Blood 91:570.

isolated from Daudi B cells: a model to study the antiproliferative mechanisms of 56. Harada, H., M. Kitagawa, N. Tanaka, H. Yamamoto, K. Harada, M. Ishihara, and http://www.jimmunol.org/ interferon-␣. J. Interferon Res. 12:35. T. Taniguchi. 1993. Anti-oncogenic and oncogenic potentials of transcriptional 43. Pine, R., A. Canova, and C. Schindler. 1994. Tyrosine phosphorylated p91 binds regulators IRF-1 and -2. Science 259:971. to a single element in the ISGF2/IRF-1 promoter to mediate induction by IFN␣ 57. Pernis, A., S. Gupta, K. Gollob, E. Garfein, R. Coffman, C. Schindler, and and IFN␥, and is likely to autoregulate the p91 gene. EMBO J. 13:158. P. Rothman. 1995. Lack of interferon ␥ receptor ␤ chain and the prevention of 44. Sarfati, M. 1993. CD23 and chronic lymphocytic leukemia. Blood Cells 19:591. interferon ␥ signaling in TH1 cells. Science 269:245. 45. Horvath, C., G. Stark, I. Kerr, and J. J. Darnell. 1996. Interactions between STAT 58. Bach, E., S. Szabo, A. Dighe, A. Ashkenazi, M. Aguet, K. Murphy, and and non-STAT proteins in the interferon-stimulated gene factor 3 transcription R. Schreiber. 1995. Ligand-induced autoregulation of IFN-␥␤chain expression complex. Mol. Cell. Biol. 16:6957. in T helper cell subsets. Science 270:1215. by guest on October 2, 2021