Activation of protein kinase A inhibits interferon induction of the Jak/Stat pathway in U266 cells.

Activation of early response genes by interferons (IFNs) requires tyrosine phosphorylation of the Stat transcription factors and is mediated by the Jak family of tyrosine kinases. Recent evidence suggests that ERK2 serine/threonine kinase modulates the IFN-stimulated Jak/Stat pathway. In this report we show that in the myeloma cell line U266 protein kinase A specifically interacts with the cytoplasmic domain of the IFNalpha/beta receptor. Treatment of cells with the adenylate cyclase activator forskolin inhibits IFNbeta-, IFNgamma-, and hydrogen peroxide/vanadate-induced formation of complexes that bind to enhancers known to stimulate the expression of IFN-regulated genes. Immunoprecipitations followed by anti-phosphotyrosine immunoblots indicate that tyrosine phosphorylation of the alpha chain of the IFNalpha/beta receptor, Jak1, Tyk2, as well as Stat1 and Stat2 is reduced as a consequence of incubation of cells with forskolin. In contrast, dideoxyforskolin, which fails to activate adenylate cyclase, has no effect on IFN induction of the Jak/Stat pathway. These results indicate a novel regulatory mechanism by which protein kinase A can modulate the Jak/Stat signaling cascade.

The Janus family of tyrosine kinases (Jak1, Jak2, Jak3, and Tyk2) is an integral component of many cytokine-activated signaling cascades, which regulate tyrosine phosphorylation of the Stat proteins such that they can translocate to the nucleus and bind DNA (1). These kinases are associated with one or more chains of a given cytokine receptor. For instance, Tyk2 interacts with the ␣ subunit of the type I IFN 1 receptor. Along with Tyk2, we have determined that Jak1 and the SH2 domain containing tyrosine phosphatase PTP1C, which acts as a negative regulator of the Jak/Stat pathway, also selectively bind to a GST fusion protein, which consists of the membrane proximal 50 amino acids of the ␣ subunit of the IFN␣/␤ receptor (2). During the course of these experiments we showed that the 42-kDa form of MAPK (ERK2) binds the IFN␣/␤ receptor and is enzymatically activated by IFN␤, probably leading to the phosphorylation of Stat1 on serine 727 (3). This serine phosphorylation has been proven crucial for the transactivation of IFN␤regulated genes without being necessary for DNA binding of the Stat proteins (4). As it has been previously shown that in some cells the activation of MAPK through various growth factors can be inhibited by increased levels of intracellular cAMP (5)(6)(7)(8), we wanted to explore the possible involvement of protein kinase A, one of the main targets of elevated cAMP, in the regulation of the IFN␤-induced Jak/Stat pathway.

MATERIALS AND METHODS
Cells-U266 cells were grown as a suspension culture in RPMI 1640 (Life Technologies, Inc.) supplemented with 10% fetal calf serum (Upstate Biotechnology, Inc.).
Electrophoretic Mobility Shift Assay (EMSA) and Measurement of Intracellular Cyclic AMP-EMSAs were performed using whole cell extracts prepared with Triton-containing buffers as described (9). Probes consisting of the GRR found within the promoter of the Fc␥R1 gene (IFN␥-induced gene) (5Ј AGCATGTTTCAAGGATTTGAGATG-TATTTCCCAGAAAAG 3Ј) and the ISRE of the ISG15 gene (5Ј GATC-CATGCCTCGGGAAAGGGAAACCGAAACTGAAGCC 3Ј) were end-labeled using polynucleotide kinase and [␥-32 P]ATP. Intracellular cAMP was assayed by radioimmunoassay (10).
Precipitation with Antibodies or GST Fusion Proteins-Lysates were incubated with the indicated antibodies for 2 h prior to the addition of Protein G-Sepharose beads (Pharmacia Biotech Inc.) and incubation for an additional hour. All antibodies used for immunoprecipitation have been described (3). GST fusion proteins representing the cytoplasmic domain of the IFN␣/␤ receptor (11) bound to agarose beads were incubated with the extracts at 4°C for 2-12 h. In either case the beads were pelleted at 15,000 ϫ g for 2 min and washed three times with ice-cold lysis buffer (1 ml).
In Vitro Kinase Assays-Cellular extracts were prepared as described above and incubated with GST fusion proteins. The beads were washed and resuspended in a buffer containing 1.4 mg/ml myelin basic protein (MBP), 20 M ATP, 20 mM Hepes (pH 7.4), 10 mM MgCl 2 , 10 mM ␤-glycerophosphate, 1 mM vanadate, 5 mM EGTA (all from Sigma), and 1 Ci of [␥-32 P]ATP (Amersham Corp.). The samples were incubated at 30°C for 20 min, SDS sample buffer was added to terminate the reaction, the proteins were separated by SDS-PAGE and transferred to Immobilon, and autoradiography was done.
Western Blotting-Proteins were separated on 4 -16% SDS-PAGE gels (Novex) and transferred to Immobilon (Millipore). Membranes were probed with monoclonal antibodies against Tyk2, Jak1, Stat1, Stat2, pan-MAPK, the R1 subunit of protein kinase A (all from Transduction Laboratories), or the ␣ chain of the IFN␣/␤ receptor (Biogen) or phosphotyrosine (PY20, ICN) using concentrations and conditions recommended by the manufacturer. Immunoblots were developed using appropriate secondary antibodies and enhanced chemiluminescence (ECL, Amersham Corp.).

RESULTS AND DISCUSSION
The antagonistic actions of cAMP on growth factor activation of MAPK have been shown to be cell-specific (12). We therefore * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
wanted to determine whether forskolin, a direct stimulator of adenylate cyclase, could inhibit IFN␤ activation of MAPK in the myeloma cell line U266. Cells were pretreated with forskolin for 20 min prior to the addition of IFN␤ for 5 min. Extracts were prepared and incubated with the IFN␣/␤ receptor GST fusion protein as described (2,3). MAPK activity was then assayed with myelin basic protein and [␥-32 P]ATP to GST fusion protein beads that had been incubated with the extracts. After SDS-PAGE, the proteins were blotted to Immobilon and either exposed for 32 P incorporation into MBP (Fig. 1, upper  panel), probed with anti-phosphotyrosine antibody (middle panel) or with anti-MAPK antibody (lower panel). Incorporation of 32 P into MBP was enhanced about 5-fold when cells were treated with IFN␤ (lanes 1 and 2). Prior treatment of U266 cells with forskolin inhibited activation of MAPK bound to the GST fusion protein by IFN␤, as well as IFN␤-stimulated tyrosine phosphorylation of MAPK (compare lanes 2 and 4 of the upper and middle panels). The amount of MAPK bound to the beads was not altered by either forskolin or IFN␤ treatment of cells (lower panel).
MAPK activity is presumably only required for Stat1␣ serine phosphorylation, which augments the transactivation of IFN␤induced genes, but not for the tyrosine phosphorylation that mediates DNA binding. In this context it also has been shown that the serine/threonine kinase inhibitor H7 can block the transcription of IFN␤-regulated genes without affecting the formation of the ISGF3 complex (13,14). To determine whether the inhibitory actions of forskolin were limited to MAPK, activation of Stat proteins was examined by their ability to bind known enhancers using EMSA. IFN-activated Stat1␣ binds specifically to the GRR, an enhancer present in promoter of the high affinity Fc␥R1 gene (15). Stat1␣ and Stat2 together with the DNA binding component ISGF3␥ form the multiprotein complex ISGF3, which interacts with the ISRE present in many IFN␣/␤-activated early response genes (1).
Surprisingly, forskolin treatment of U266 cells abrogated IFN␤-stimulated formation of DNA binding complexes interacting with both the GRR ( Fig. 2A, lanes 2 and 3) and the ISRE enhancers (lanes 10 and 11) by 70 -90%. Furthermore, forskolin also prevented the activation of GRR binding complexes by treatment of cells with either IFN␥ (lanes 5 and 6) or with H 2 O 2 and vanadate (lanes 7 and 8). In order to exclude other possible effects of forskolin besides activation of adenylate cyclase as the mechanism of inhibition of the Jak/Stat pathway, U266 cells were also exposed to dideoxyforskolin, a derivative of forskolin that fails to activate adenylate cyclase. As expected, this compound had no effect on the activation of the Jak/Stat pathway (compare lanes 2, 3, and 4).
To determine whether there was a correlation between forskolin-stimulated accumulation of intracellular cAMP and inhibition of IFN␤-stimulated GRR binding activity, cells were incubated with increasing concentrations of forskolin prior to treatment with IFN␤. Cell extracts were prepared and intracellular cAMP concentrations were measured, and IFN␤-stimulated GRR binding complexes were assayed by EMSA and subsequently quantitated on the PhosphorImager (Fig. 2B). Inhibition in IFN␤-stimulated GRR binding correlated with dose-dependent forskolin-induced increases in intracellular cAMP, confirming that increases in cAMP in these cells could be inhibiting IFN␤ activation of the Jak/Stat pathway.
To determine directly whether IFN␤-stimulated tyrosine phosphorylation of Stat1␣ and Stat2 was affected by forskolin treatment of U266 cells, cellular extracts were subjected to immunoprecipitation with anti-Stat1␣ and anti-Stat2 antisera, and immunoprecipitates were analyzed using anti-phosphotyrosine antibodies (Fig. 2, C and D, upper panels). These results clearly demonstrated that IFN␤-stimulated tyrosine phosphorylation of these proteins is drastically decreased in U266 cells incubated with forskolin ( compare lanes 2 and 3). Reprobing of the blots with Stat1␣ or Stat2 antisera indicated that equal amounts of protein were immunoprecipitated from all samples (lower panels).
Tyk2 and Jak1 are required for IFN␣/␤-stimulated tyrosine phosphorylation of Stat1␣ and Stat2 and become activated and tyrosine-phosphorylated as a result of incubation of cells with IFN␣/␤. In order to determine whether forskolin was exerting its inhibitory effects on activation of Tyk2 and Jak1, U266 cells were incubated with IFN␤ in the presence or absence of forskolin, and Tyk2 and Jak1 were immunoprecipitated from cell lysates. The immunoblots from this experiment were probed with anti-phosphotyrosine antibody (Fig. 2, E and F). IFN␤stimulated tyrosine phosphorylation of both Tyk2 and Jak1 was also significantly inhibited with forskolin treatment of U266 cells (compare lanes 2 and 3) while the total amount of these proteins in the immunoprecipitates is not altered (lower panels).
It has been previously shown that the ␣ chain of the IFN␣/␤ receptor becomes tyrosine-phosphorylated in response to IFN treatment (16). In order to determine whether the IFN␣/␤ receptor itself was also a target for the inhibitory effects of elevated cAMP levels, monoclonal antibodies were used to immunoprecipitate the ␣ chain of the IFN␣/␤ receptor from cells that had been exposed to IFN␤ with or without prior treatment with forskolin (2,3). Proteins were resolved on SDS-PAGE and transferred to Immobilon, and the blots were probed with antiphosphotyrosine antibody. The results shown in Fig. 2G clearly demonstrate that IFN-stimulated tyrosine phosphorylation of the IFN␣/␤ receptor is decreased in U266 cells incubated with forskolin and IFN␤ (compare lanes 2 and 3) compared with cells treated with IFN␤ alone.
It has been previously shown that many of the key signaling components required for IFN␣/␤ stimulation of the Jak/Stat pathway constitutively associate with a GST fusion protein containing the membrane-proximal 50 amino acids of the ␣ chain of the IFN␣/␤ receptor (3,11). In order to explore a possible association of protein kinase A with the receptor, GST fusion proteins were expressed, which contained either the  2 and 4) for 10 min at 37°C. Forskolin (Forsk, 30 M) was added to cells for 20 min at 37°C prior to the addition of IFN␤ (lanes 3 and 4). MAPK activity was visualized by autoradiography of 32 P incorporated into MBP (top panel), and the blot was then probed either with anti-phosphotyrosine (middle panel) or pan-MAPK antibody (lower panel). The membranes were developed using enhanced chemiluminescence (ECL). All monoclonal antibodies except PY20 (ICN) were obtained from Transduction Laboratories. WB, Western blot; CTL, control. membrane-proximal 50 amino acids of the cytoplasmic domain of the ␣ subunit of the IFN␣/␤ receptor or GST alone. U266 cells were either left untreated or treated with IFN␤ for 5 min, and cell lysates were prepared and incubated with the GST or the GST fusion proteins coupled to glutathione-agarose. Complexed proteins were resolved on SDS-PAGE and transferred to Immobilon, and the blots were probed with a monoclonal antibody to the type 1 regulatory subunit of protein kinase A.

FIG. 2. Forskolin treatment of U266 cells inhibits IFN␤ activation of the Jak/Stat pathway.
A, extracts were prepared from U266 cells incubated with IFN␤ or with combinations of forskolin (Forsk) and IFN␤. Electrophoretic mobility shift assays were performed using equal amounts of protein and a 32 P-labeled oligonucleotide probe corresponding to the ␥ response region in the high affinity Fc␥R1 gene (lanes 1-8) or the ISRE (lanes 9 -11). The IFN␤-induced complex which binds to the GRR is labeled IFN-SF and that which binds to the ISRE is termed ISGF3. CTL, control; VAN, vanadate; DDO, dideoxyforskolin. B, forskolin-induced intracellular cAMP correlates with inhibition of IFN-induced IFN-SF. U266 cells were either untreated or incubated with 0, 1, 3, 30, or 100 M forskolin for 15 min prior to the addition of IFN␤ for 15 min. Cell extracts were prepared, and cAMP was assayed (right abscissa) or IFN-SF formation was quantitated using a PhosphorImager (Molecular Dynamics) (left abscissa). C and D, forskolin inhibits IFN␤-stimulated tyrosine phosphorylation of Stat1␣ and Stat2. Lysates from U266 cells were incubated with anti-Stat1␣ or anti-Stat2 antiserum for 2 h. Immune complexes were collected on protein G beads, subjected to SDS-PAGE, and immunoblotted with PY20 anti-phosphotyrosine antibody and detected with ECL. Membranes were reprobed with anti-Stat1␣ or anti-Stat2 antibodies (lower panel). WB, Western blot; IP, immunoprecipitate; ␣-PY, anti-phosphotyrosine. E, F, and G, forskolin inhibits IFN␤-stimulated tyrosine phosphorylation of Jak1, Tyk2, and the IFN␣/␤ receptor. Treatment of U266 cells, preparation of cell lysates, immunoprecipitations, and Western blots with PY20 were identical to B and C. Membranes were reprobed with anti-Tyk2, anti-Jak1, or anti-IFN␣/␤ receptor monoclonal antibodies (lower panels). The antibodies used to immunoprecipitate Tyk2, Jak1, and the IFN␣/␤ receptor have been previously described (2).
Protein kinase A was found to associate specifically but in a ligand-independent manner with the GST fusion protein representing the membrane-proximal 50 amino acids of the cytoplasmic tail of the IFN␣/␤ receptor (Fig. 3, lanes 2 and 3) but not with GST alone (lane 1).
Although cross-talk between signaling networks is a well described phenomenon, until recently modulation of the Jak/ Stat pathway has been restricted to covalent modification of its components by tyrosine phosphorylation. Recent reports indicate that serine phosphorylation of Stat1 and Stat3 presumably by activation of MAPK can enhance IFN␣/␤ and IFN␥ induction of early response genes (3,4). The results presented here implicate the serine/threonine kinase protein kinase A as a regulator of Jak/Stat activation. The effects of forskolin described here are also cell-specific with regard to inhibition of tyrosine phosphorylation of the Jak/Stat proteins (data not shown). Protein kinase A may directly or indirectly be responsible for modification of one or several of the Jak tyrosine kinases such that they cannot become tyrosine-phosphorylated (and presumably activated). Alternatively, the target for the actions of protein kinase A is upstream of Tyk2 and Jak1. The fact that forskolin also inhibits both IFN␥and vanadate-induced GRR binding complexes in U266 cells suggests that the IFN␣/␤ receptor is not the primary target for the inhibitory actions of forskolin. There are other key control points in the Jak/Stat pathway though, which could also be regulated by protein kinase A, such as the protein tyrosine phosphatase activity, which appears to be required to initiate tyrosine phosphorylation of Stat1 and Stat2 by IFN␣/␤ but not to maintain the activation cascade (17). Other signaling proteins such as insulin substrate 1 or protein kinase C, which are known to be regulated by IFN treatment of cells but not at present implicated in IFN stimulation of the Jak/Stat pathway, must also be considered targets for the actions of protein kinase A (18,19). Definition of the cell-specific substrates modified by protein kinase A, which inhibit Jak/Stat activation, is clearly an important piece of information to understand the mechanisms by which these distinct signaling networks influence each other's activities.