¹Sbarro Institute for Cancer Research & Molecular Medicine, College of Science & Technology, Temple University, Philadelphia, Pennsylvania, PA 19122, USA

²Istituto di Anatomia e Istologia Patologica, Via delle Scotte, 6-53100 Siena, Italy

³Dipartimento di Morfologia ed Embriologia, Sezione di Anatomia Umana, Università di Ferrara, via Fossato di Mortara 66, 44100 Ferrara, Italy

⁴Istituto di Citomorfologia Normale e Patologica del CNR, c/o IOR, via di Barbiano 1/10, 40137 Bologna, Italy

⁵Dipartimento di Biologia Evolutiva e Comparata, Università Federico II di Napoli, Italy

⁶Jefferson Medical College, KCI Room 376, 1020 Locust Street, Philadelphia, Pennsylvania, PA 19107, USA

⁷Istituto di Anatomia Topografica, Seconda Università di Napoli, Italy

Correspondence to: A Giordano, Sbarro Institute for Cancer Research & Molecular Medicine, College of Science & Technology, Temple University, BioLife Science Building, Suite 333, 1900 N. 12th Street, Philadelphia, Pennsylvania, PA 19122, USA;E-mail: giordano@temple.edu

Cdk9 is a member of the Cdc2-like family of kinases. It binds to members of the family of cyclin T (T1, T2a and T2b) and to cyclin K. The Cdk9/cyclin T complex appears to be involved in regulating several physiological processes. In fact Cdk9 is the kinase of the P-TEFb complex, involved in basal transcription. Cdk9 has also been described as the kinase of the TAK complex, homologous to P-TEFb and involved in HIV replication. Here we show that Cdk9 interacts with gp130, the receptor of the Interleukin-6 (IL-6) family of cytokines, which includes Leukemia Inhibitory Factor (LIF), Oncostatin M (OSM), Ciliary Neurotrophic Factor (CNTF), Interleukin-11 (IL-11) and Cardiotrophin (CT-1). IL-6 is a key regulator of hematopoiesis, immunological responses and inflammation. In addition, IL-6 plays a major role in the endocrine and nervous systems. Signal transduction by gp130 is mediated by physical interaction of the cytoplasmic region of gp130 with cellular kinases and results in the transcriptional activation of cellular and viral genes. We found that Cdk9 interacts in vitro with the cytoplasmic region of gp130 and we succeded in reproducing this interaction in vivo. Cdk9 expression was found both in the nucleus and in the cytoplasm. The binding occurring between Cdk9 and gp130 increased upon IL-6 stimulation. We also observed that Cdk9 synergized with IL-6 in inducing the activation of an IL-6-responsive reporter plasmid. In summary, these results point to a previously undisclosed role for Cdk9 in signal transduction.

Oncogene (2002) 21, 7464-7470. doi:10.1038/sj.onc.1205967

signal transduction; cytokine receptors; kinases; phosphatases; growth factors; AIDS; HIV

Introduction

Cdk9 is a cdc2-like serine/threonine kinase, isolated by screening a human cDNA library, using degenerated oligonucleotides, in order to identify Cdk-related proteins. It was originally referred to as PITALRE because of its homology with the PSTAIRE motif of cdc2 (Graña et al., 1994; for a review see de Falco and Giordano, 2002). Cdk9 is a 43 kDa protein; its deduced amino acid sequence contains the 11 conserved regions characteristic of protein kinases' catalytic domain (Hanks et al., 1988). Its expression is ubiquitous, and is greater in terminally differentiated tissues, suggesting a role for this kinase in specialized functions in the cell (Graña et al., 1994). Cdk9 maps to chromosome 9q34.1 (Bullrich et al., 1995), a region involved in non-random chromosome alterations, such as T-cell non-Hodgkin's lymphomas, bladder tumors and several myeloproliferative disorders (Offit et al., 1993; Orlow et al., 1994). Cyclin partners of Cdk9 are members of the family of the T cyclins (T1, T2a and T2b) (Peng et al., 1998) and cyclin K (Fu et al., 1999).

The complex Cdk9/cyclin T belongs to the positive-acting transcription elongation factor (P-TEFb), which supports basal transcription elongation (Zhu et al., 1997). This complex was isolated from Drosophila (Zhu et al., 1997); it functions by hyperphosphorylating RNA Pol II Carboxyl-Terminal Domain (CTD) and preventing polymerase arrest (Marshall et al., 1996). The kinase activity of the complex is due to Cdk9, because when the kinase is immuno-depleted from cell extracts, basal transcription is abolished and can be restored after addition of Cdk9 (Zhu et al., 1997; Zhou et al., 1998). Interestingly, P-TEFb has been shown to interact with the HIV protein Tat, suggesting that it may be a direct target of Tat (Zhu et al., 1997; Herrmann et al., 1998) and thus opening new possibilities for better understanding HIV replication.

In order to identify new partners for Cdk9, we screened a mouse embryonic library in the yeast two-hybrid system; this led to the identification of TRAF-2, the signal transducer of tumor necrosis factor (TNF), as a partner of Cdk9 (Maclachlan et al., 1998), suggesting a role for Cdk9 in signal transduction. We started analysing whether Cdk9 may interact with other receptors involved in signal transduction. Cdk9 is involved in neural (Bagella et al., 1998) and lymphocytic (Herrmann et al., 1998) differentiation programs, and it has an active role in HIV replication (Ghose et al., 2001). In addition, Cdk9 protein level and TAK activity increase in response to cytokines, such as TNF and IL-6, in resting CD4⁺ T lymphocytes isolated from infected individuals (Ghose et al., 2001). Therefore among the cytokines we focused our attention on IL-6, which is involved in neural development, in the proliferation of the transformed cells such as myeloma and plasmacytoma cells, and in viral infections, including human immunodeficiency virus (HIV) (Nakajima et al., 1989; Akira et al., 1993). We identified gp130, the transducer of the IL-6 family of cytokines, as a possible candidate for Cdk9 interaction and we tested whether such an interaction may occur both in vitro and in vivo.

We found that Cdk9 interacts with gp130, the signal transducer shared by several cytokines, including LIF, OSM, IL-11, CNTF and CT-1, which comprise the IL-6 family of cytokines (Kishimoto et al., 1994). IL-6 is a key regulator of hematopoiesis, immunological responses and inflammation. In addition, IL-6 plays a major role in the endocrine and nervous systems. After the release of IL-6, the cytokine interacts with its receptor (IL-6R); the IL-6/IL-6R complex recruits gp130 (Taga et al., 1989) and induces homodimerization of gp130 via disulphide bonds (Murakami et al., 1993). Dimerization itself is not enough for the signaling (Murakami et al., 1991). No obvious enzymatic motifs have been found in the structure of gp130 (Taga and Kishimoto, 1992; Ulrich and Schlesinger, 1990). Gp130 utilizes receptor-associated tyrosine kinases of the Janus family (Jak) (Stahl et al., 1994). A few downstream mediators of gp130 signaling have been identified, including adapter molecules which link gp130/Jak2 activation with the Ras/MAP/ERK pathway and which may be involved in the threonine phosphorylation of C/EBP/NF-IL-6 induced by IL-6 (Boulton et al., 1994).

In addition, signal transduction and activation of transcription (STAT) factors are activated following IL-6-gp130 signaling and directly regulate the transcription of a specific set of genes (Ihle, 1996). We found that Cdk9 binds to gp130 both in vitro and in vivo. In addition, Cdk9/gp130 complexes increased upon IL-6 stimulation and resulted in the activation of an IL-6-responsive reporter plasmid. These results suggest that Cdk9 may be involved in signaling by IL-6, thus pointing to an undisclosed role for Cdk9 in gp130-mediated signal transduction.

Results

Cdk9 interacts in vitro with gp130

We identified gp130 as a candidate for interaction with Cdk9. We investigated whether Cdk9 was able to interact with gp130, in vitro and in vivo. Cdk9 was in vitro translated and tested for binding to GST-gp130 or to the GST portion alone, used as a negative control. As shown in Figure 1a, an interaction between gp130 and Cdk9 was observed, while no interaction with GST alone was detected. To further assess the in vitro binding of gp130 and Cdk9, we used a baculovirus expression plasmid that allows the in vitro expression of Cdk9 in an eukaryotic system. Cell extracts expressing the baculovirus-derived Cdk9 were used in pull-down experiments with GST-gp130, using GST as a negative control. In these experiments, a specific interaction between Cdk9 and gp130 was also observed (Figure 1b). In addition, cell extracts obtained from a human B-cell line, previously engineered to constitutively express high levels of both of the IL-6 receptor and gp130 (Giordano et al., 1997), was used in a pull-down experiment with GST-Cdk9, followed by a Western blot with anti-gp130. As shown in Figure 1c, a strong signal of interaction between Cdk9 and gp130 was observed, while no interaction occurred with GST alone.

Cdk9 and gp130 interact in living cells

To confirm the binding between Cdk9 and gp130 in living cells, we checked for endogenous interactions in 293 cell extracts. After pre-clearing with normal rabbit serum, proteins were immunoprecipitated with anti-Cdk9 Ab and analysed by Western blotting with anti-gp130. As shown in Figure 2, endogenous gp130/Cdk9 complexes were detected in these experiments, indicating that this interaction normally occurs in cells.

Cdk9 is expressed both in the nucleus and in the cytoplasm

Cdk9 has been described previously as a nuclear protein (Graña et al., 1994). If Cdk9 has an exclusive nuclear localization, it would be difficult to explain its interaction with gp130, which is a trans-membrane protein. Cell fractionation was performed and the integrity of the nuclei was analysed by staining with Trypan blue or by electron microscopy. By these criteria, the isolated nuclei were intact with no detectable cytoplasmic fraction (data not shown). Both nuclear and cytoplasmic fractions were analysed by Western blot with anti-Cdk9 Ab. As shown in Figure 3, Cdk9 is expressed in both cell compartments, consistent with its binding to gp130 in the cytoplasm.

The binding between Cdk9 and gp130 is increased upon IL-6 stimulation

We next tested whether the binding between gp130 and Cdk9 could be enhanced after IL-6 stimulation. 293 cells were demonstrated to normally express IL-6 receptor and to be responsive to IL-6 (von Laue et al., 2000). gp130 and either HA-Tagged Cdk9 or its dominant negative form were overexpressed in 293 cells. Thirty-six hours after transfection, cells were stimulated with IL-6 at 200 U/ml, for time periods ranging between 0 and 10 min. Cell extracts were precleared with a normal mouse serum and were subjected to immunoprecipitation with an anti-gp130 antibody, followed by Western blot with an anti-HA-Tag.

Figure 4 shows a constitutive binding between Cdk9 and gp130 which increased after IL-6 stimulation, peaking at 5 min. No significant variation in the binding was detectable in the case of a dominant negative mutant of Cdk9, suggesting that the observed interaction does not depend on the enzymatic function of the kinase.

Cdk9 synergizes with IL-6 in inducing the activation of a NF-B reporter plasmid

To understand the physiological meaning of the interaction occurring between Cdk9 and gp130, we tested the possibility that Cdk9 could participate in the IL-6-induced activation of a NF-B reporter plasmid. A plasmid carrying a NF-B-driven luciferase reporter gene was co-transfected with either Cdk9 or its dominant negative form in 293 cells, followed by IL-6 stimulation. In these experiments, Cdk9 was able to activate the NF-B reporter gene in the absence of IL-6 stimulation (Figure 5). Moreover, a substantial increase of luciferase activity was induced by IL-6 in the presence of Cdk9, suggesting that the kinase plays a role in IL-6-mediated activation of NF-B/rel transcription factors (Figure 5). In addition, expression of a dominant negative form of Cdk9 led to a decrease in the base-line luciferase activity and dramatically reduced the luciferase gene expression in IL-6-stimulated cells (Figure 5).

Discussion

Cdk9 has been described as a multifunctional kinase, exerting a leading role in transcription by acting as the kinase subunit of the p-TEFb complex (Zhu et al., 1997), promoting transcriptional elongation by phosphorylating the carboxyl terminus of RNA Pol II, and converting it into the active form (Marshall et al., 1996). Data support an involvement of Cdk9 in cell differentiation. Cdk9 and cyclin T2a appear to be involved in myogenic differentiation, as demonstrated by the fact that their expression levels and activity are not down-regulated during muscle differentiation (Simone et al., 2002). Overproduction of cdk9/cycT2a enhances MyoD function and promotes myogenic differentiation, while inhibition of cdk9 kinase activity by a dominant negative form prevents the activation of the myogenic program (Simone et al., 2002). In addition, Cdk9 seems to participate in the differentiative program of neurons (De Falco, personal communication).

Cdk9 seems to be required also for the differentiation of monocytes. In these cells, in fact, upon induction with PMA, there is a strong induction of mRNA for cyclin T1 (Herrmann et al., 1998). In monocytes, there is an induction of TAK activity upon PMA induction (Herrmann et al., 1998). This suggests that cyclin T1 may act as a regulatory element for TAK activity in these cells and may be involved in the differentiation program of monocytes (Herrmann et al., 1998).

In addition, Cdk9 seems to have an anti-apoptotic effect in specific cell lines. Monocytes overexpressing the dominant negative form of Cdk9 undergo apoptosis, especially after induction with PMA (Foskett et al., 2001). This suggests that Cdk9 may have an anti-apoptotic function during monocyte differentiation, due to a direct role of Cdk9 in an apoptosis pathway or to a block in the differentiation program of monocytes by the loss of kinase activity of Cdk9 (Foskett et al., 2001). This may suggest that P-TEFb regulates the expression of genes that oppose apoptosis or that Cdk9 kinase activity is necessary for monocyte differentiation (Foskett et al., 2001). Cdk9 and cyclin T1 mRNA and protein levels, in fact, increase following peripheral blood lymphocyte activation (Herrmann et al., 1998). When monocytes differentiate to macrophages Cdk9 expression level remains constantly high, while cyclin T1 is regulated at post-transcriptional level (Herrmann et al., 1998). In addition, Cdk9 participates in the differentiation program of lymphocytes (Bellan et al., 2002). Taken together, this information establishes a link between Cdk9 and hematopoiesis.

Cdk9 has been identified as the kinase required for HIV-1 Tat to efficiently promote HIV-1 gene expression (Wei et al., 1998; Yang et al., 1997). In this regard, our finding that the gp130/Cdk9 complex is activated by IL-6 stimulation may shed light on the known capacity of IL-6 to stimulate HIV-1 replication in latently infected cells (Chun et al., 1998).

Cdk9 expression has been reported to be very high in hematopoietic lineages (de Luca et al., 1997). We describe here for the first time the Cdk9 interaction with gp130, the transducer of the IL-6 family of cytokines, which suggests a role for the observed expression of Cdk9 in these lineages. Consistent with the above observations, a deregulated gp130 signal transduction induced by cytokines, such as IL-6, leads to a tumorigenic phenotype. Signal transduction occurs through interaction of gp130 with several proteins, such as Shc proteins (Giordano et al., 1997), Jak kinases (Stahl et al., 1994) and STAT factors (Ihle, 1996), and depends upon different incoming stimuli. The pleiotropic function of gp130 may require the recruitment of additional transducer proteins. We found that Cdk9 interacts with gp130, both in vitro and in vivo in human cell lines under physiological conditions.

We used several approaches to demonstrate the occurrence of this binding. Up to this time, serine-threonine kinase Cdk9 has been described as a nuclear protein, which is able to enter the cytosol only after a differentiating stimulus (Maclachlan et al., 1998). Our results extend these observations by showing the presence of Cdk9 in the cytosol in unstimulated cells. We also observed an increased recruitment of Cdk9 to gp130 following IL-6 stimulation in other experiments, suggesting that gp130/Cdk9 complexes may participate in the IL-6-activated metabolic pathways. Consistent with this possibility, we took advantage of plasmids expressing either a wild-type or a negative transdominant Cdk9 to show that expression of Cdk9 mediates the IL-6 activation of an NF-B-reporter plasmid. The Cdk9/gp130 complex formation does not require the involvement of cyclin T1, the cyclin partner of Cdk9: cyclin T1, in fact, has an exclusively nuclear localization, where it is complexed to Cdk9 (Napolitano et al., 2002). Cdk9, on the contrary, is also present in the cytoplasm, where binding with gp130 occurs.

Interleukin-6 (IL-6) is a key regulator of hematopoiesis, immunological responses and inflammation. In addition, IL-6 plays a major role in the endocrine and nervous systems. IL-6 is involved in regulation of biological activities, like stimulation of bone marrow precursor cells, immunoglobulin production, stimulation of T and B cell proliferation, and induction of acute-phase protein synthesis (Kishimoto et al., 1992). An additional role has been described for IL-6 in neural development, in proliferation of transformed cells such as myeloma and plasmacytoma cells, and in viral infections, including human immunodeficiency virus (HIV) infection (Nakajima et al., 1989; Akira et al., 1993).

The IL-6 receptor complex shares a signal transducer subunit, gp130, with a set of cytokines which include Leukemia Inhibitory Factor (LIF), Oncostatin M (OSM), Ciliary Neurotrophic Factor (CNTF), Interleukin-11 (IL-11) and Cardiotrophin (CT-1).

The metabolic pathways activated by gp130 engagement lead to the activation of transcription factors, such as NF-B/rel and NF-IL-6/CEBP required for cell survival and for proper organ development, including bone marrow, liver and heart; thus, mice with a homozygous deletion of gp130 are not viable.

gp130 contains no obvious enzymatic motifs in its structure (Taga and Kishimoto, 1992; Ulrich and Schlesinger, 1990), and uses receptor-associated tyrosine kinases of the Janus family (Jak) (Stahl et al., 1994). Jak2 kinase is constitutively pre-associated with gp130. After IL-6 release, Jak2 becomes phosphorylated in tyrosine, is activated, and then phosphorylates tyrosine residues of the gp130 cytoplasmic domains (Stahl et al., 1994).

The data present new insights about the signal transduction by Cdk9, whose activity is mediated by physical interaction with gp130, resulting in the transcriptional activation of cellular and viral genes.

We previously showed that Cdk9 interacts with TRAF-2 (Maclachlan et al., 1998). This interaction occurs with a conserved WKI amino acid motif present in the TRAF domain of TRAF-2, a domain essential for both kinase interaction and for NF-B activation (Maclachlan et al., 1998). In addition, Cdk9 showed a mild synergism with TNF in activation of NF-B. The dominant negative form of the kinase decreased the TRAF-2 and TNF-induced NF-B activity. These results suggested a novel role for Cdk9 as being involved in specific pathways of signal transduction.

Collectively, our observations establish a novel functional link between a major cytokine signal transducer, gp130, and the emerging role of Cdk9 in signal transduction. Moreover, because of the pleiotropic activities of the gp130-utilizing cytokines, such as IL-6, LIF, OSM, CNTR, IL-11 and CT-1, the results point to an extensive role for Cdk9 as a regulator of immune response, inflammation and cell differentiation.

Materials and methods

In vitro binding

In vitro binding conditions have already been described (de Falco et al., 2000).

Construction of the baculovirus construct

The construct for baculovirus expression was made in PVL1393, originating from the vector pAC-93SV (Invitrogen, CA, USA), cloning Cdk9 in the BamHI/PstI sites. After the infection, cells were washed in PBS, collected and protein expression was checked by Western blot.

In vitro binding with extracts from baculovirus

For the binding, baculovirus extracts prepared in PLC/LB (50 mM HEPES pH 7.5, 150 mM sodium chloride, 10% glycerol, 1% Triton X-100, 1 mM EGTA, 1.5 mM magnesium chloride, 100 mM sodium fluoride, 10 mM sodium pyrophosphate and 10 g/ml aprotinin, leupeptin and phenylmethylsulphonyl (PMSF)) were incubated with GST-gp130 O/N, then washed 5´ with PBST, boiled and run on SDS-PAGE. The gel was transferred to a nitrocellulose filter and subjected to Western blotting.

Immunoprecipitation and Western blot

Monoclonal anti-gp130 was kindly provided by IRBM. The antibody (Ab) was used at a dilution of 1 : 1000 in Western blot. Anti-Cdk9 has already been described (de Falco et al., 2000). The antibody was affinity purified and used at a dilution 1 : 300 in Western blot. The polyclonal anti-HA Tag (Santacruz) was used at a dilution 1 : 3000 in Western blot. Immunoprecipitation and Western blotting conditions have been described (de Falco et al., 2000).

Cell fractionation

To separate the cytosolic fraction from the nuclear, cells were washed in PBS and resuspended in RSB buffer (10 mM HEPES pH 7.9, 10 mM KCl, 1.5 mM MgCl₂). After 20-30 min in ice, cells were homogenized using a 27G needle syringe and were centrifuged for 10 s at 14 000 r.p.m. Supernatant constituted the cytosolic fraction. Nuclei were washed in PBS and were resuspended in buffer C (20 mM HEPES pH 7.9, 420 mM NaCl, 1.5 mM MgCl₂, 0.2 mM EDTA, 25% glycerol). Samples were incubated for 40 min in ice, then centrifuged for 5 min at 14 000 r.p.m. Supernatant constituted the nuclear fraction.

Cell lines

293 cells (obtained from ATCC) were grown in DMEM supplemented with 10% FBS, L-glutamine. De Few IL6R/gp130 cells, grown in RPMI supplemented with 10% FCS, have already been described (Giordano et al., 1997).

Plasmids

The plasmid pZipNeo/gp130 has already been described (Giordano et al., 1997). pcDNA3.1/Cdk9-HATag and pcDNA3.1/dnCdk9-HATag have been described (de Falco et al., 2000).

Overexpression in 293 of Cdk9 and gp130

Transfections were performed using the calcium phosphate kit of Invitrogen following the conditions suggested by the manufacturer.

IL-6 stimulation of the cells

Transfected cells were stimulated 36 h from transfection with human IL-6 (Boehringer, IN, USA), at a concentration of 200 U/ml, for 0, 2, 5 and 10 min. Cells then were pelleted, and extracts were prepared as described above.

Luciferase assays

The conditions used for the assay have already been described (de Falco et al., 2000).

Acknowledgements

We acknowledge G Scala for his contribution the early stage of this work. G De Falco and C Bellan are supported by 'Dottorato in Patologia Diagnostica Quantitativa', University of Siena, Italy. This work was supported by grants from the National Institute of Health and the Sbarro Health Research Organization (A Giordano). G De Falco would like to thank Giovanni Sorrentino and LM Neri is grateful to Paola Ziccone for their continuous encouragement, support and understanding. We also thank Alan and Cecilia Bergstein for their support.

Akira S, Taga T, Kishimoto T. (1993). Adv. Immunol., 54: 1-78. MEDLINE

Boulton TG, Stahl N, Yancopoulos GD. (1994). J. Biol. Chem., 269: 648-655.

Bagella L, MacLachlan TK, Buono RJ, Pisano MM, Giordano A, De Luca A. (1998). J. Cell Physiol., 177: 206-213. Article MEDLINE

Bellan C, De Falco G, Lazzi S, Scherfeld K, Micheli P, Bartolommei S, Vestri M, Nyongo A, Pileri S, Leoncini L, Tosi P, Giordano A. (2002). J. Clin. Pathol., 55: (Suppl 1) A29.

Bullrich F, MacLachlan T, Sang N, Druck T, Veronese ML, Allen S, Chiorazzi N, Koff A, Heubner K, Croce CM, Giordano A. (1995). Cancer Res., 55: 1199-1205. MEDLINE

Chun TW, Engel D, Mizell SB, Ehler LA, Fauci AS. (1998). J. Exp. Med., 188: 83-91. Article MEDLINE

De Falco G, Bagella L, Claudio PP, De Luca A, Fu Y, Calabretta B, Sala A, Giordano A. (2000). Oncogene, 19: 373-379.

De Falco G, Giordano A. (2002). Cancer Biol. Ther., 1: 66-69.

De Luca A, Esposito V, Baldi A, Claudio PP, Fu Y, Caputi M, Pisano MM, Baldi F, Giordano A. (1997). J. Cell. Physiol., 172: 265-273. MEDLINE

Foskett SM, Ghose R, Tang DN, Lewis DE, Rice AP. (2001). J. Virol., 75: 1220-1228. MEDLINE

Fu TJ, Peng J, Lee GA, Price DH, Flores O. (1999). J. Biol. Chem., 274: 34527-34530. MEDLINE

Ghose R, Liou LY, Herrmann CH, Rice AP. (2001). J. Virol., 75: 11336-11343.

Giordano V, De Falco G, Ambrosino C, Ruocco MR, Quinto I, Pelicci PG, Del Mastro P, Scala G. (1997). J. Immun., 158: 4097-4103.

Graña X, De Luca A, Sang N, Fu Y, Claudio PP, Rosenblatt J, Morgan DO, Giordano A. (1994). Proc. Nat. Acad. Sci. USA, 91: 3834-3838.

Hanks SK, Quinn AM, Hunter T. (1988). Science, 241: 42-52. MEDLINE

Herrmann CH, Carroll RG, Wei P, Jones KA, Rice AP. (1998). J. Virol., 72: 9881-9888. MEDLINE

Ihle JN. (1996). Cell, 84: 331-334. MEDLINE

Kishimoto T, Akira S, Taga T. (1992). Science, 258: 593-597. MEDLINE

Kishimoto T, Taga T, Akjira S. (1994). Cell, 76: 253-262. MEDLINE

MacLachlan TK, Sang N, Puri PL, Levrero M, Giordano A. (1998). J. Cell. Biochem., 71: 467-478. Article MEDLINE

Marshall NF, Peng J, Xie P, Price DH. (1996). J. Biol. Chem., 271: 27176-27183. MEDLINE

Murakami M, Narazaki M, Hibi M, Yawata H, Yasukawa K, Hamaguchi M, Taga T, Kishimoto T. (1991). Proc. Natl. Acad. Sci. USA, 88: 11349-11353. MEDLINE

Murakami M, Hibi M, Nakagawa T, Yasukawa K, Yamanishi K, Taga T, Kishimoto T. (1993). Science, 260: 1808-1810. MEDLINE

Nakajima N, Martinez-Maza O, Hirano T, Breen EC, Nishanian PG, Sazar-Gonzalez JF, Fahey JL, Kishimoto T. (1989). J. Immunol., 142: 531-536.

Napolitano G, Licciardo P, Carbone R, Majello B, Lania L. (2002). J. Cell Physiol., 192: 209-215.

Offit K, Parza NZ, Jhanwar SC, Filippa D, Wachtel M, Chagenti RS. (1993). Genes Chrom. Cancer, 7: 1-7.

Orlow I, Lianes P, Lacombe L, Dalbagni G, Reuter VE, Cordon-Cardo C. (1994). Cancer Res., 54: 2848-2851. MEDLINE

Peng J, Zhu Y, Milton JT, Price DH. (1998). Genes Dev., 12: 755-762. MEDLINE

Simone C, Stiegler P, Bagella L, Pucci B, Bellan C, De Falco G, De Luca A, Guanti G, Puri PL, Giordano A. (2002). Oncogene, 21: 4137-4148. Article MEDLINE

Stahl N, Boulton TG, Farruggella T, Ip NY, Davis S, Witthuhn BA, Quelle FW, Silvenoinen O, Barbieri G, Pellegrini S, Ihle JN, Yancopoulos GD. (1994). Science, 263: 92-95. MEDLINE

Taga T, Kishimoto T. (1992). FASEB J., 6: 3387-3396. MEDLINE

Taga T, Hibi M, Hirata Y, Yamasaki K, Yasukawa K, Matsuda T, Hirano T, Kishimoto T. (1989). Cell, 58: 573-581. MEDLINE

Ulrich A, Schlesinger J. (1990). Cell, 61: 203. MEDLINE

Von Laue S, Finidori J, Maamra M, Shen XY, Justice S, Dobson PR, Ross RJ. (2000). J. Endocrinol., 165: 301-311. MEDLINE

Wei P, Garber ME, Fang SM, Fischer WH, Jones K. (1998). Cell, 92: 451-462. MEDLINE

Yang X, Gold MO, Tang DN, Lewis DE, Aguilar-Cordova E, Rice AP, Herrmann C. (1997). Proc. Natl. Acad. Sci. USA, 94: 12331-12336. MEDLINE

Zhou Q, Chen D, Pierstorff E, Luo K. (1998). EMBO J., 17: 3681-3691. Article MEDLINE

Zhu Y, Pe'hery T, Peng J, Ramanathan Y, Marshall N, Amendt B, Mathews MB, Price DH. (1997). Genes Dev., 11: 2622-2632. MEDLINE

Figure 1 (a) Cdk9 and gp130 interact in vitro. Cdk9 was in vitro translated and 2 l were used for in vitro binding with GST-gp130. The GST portion was used as a negative control. One l of the in vitro translated protein was loaded as a control. It is possible to see a specific binding between GST-gp130 and Cdk9, while no signal is observed with the GST portion alone. (b) gp130 interacts with Cdk9 from baculovirus. The in vitro binding between Cdk9 and gp130 was further confirmed using a Cdk9 from baculovirus. Cell extracts containing baculovirus Cdk9 were subjected to a pull-down with GST-gp130, using GST as a negative control. The Western blotting with anti-Cdk9 shows that the two proteins interact in vitro. CE, Cell extracts. (c) GST-Cdk9 interacts with gp130 from cell extracts. Cell extracts from De Few IL6R/gp130, a cell line overexpressing both chains of the IL6 receptor, were used for a pull-down with GST-Cdk9, using GST as a negative control. The pull-down was followed by Western blotting with anti-gp130. The result shows an interaction between gp130 and GST-Cdk9, while no signal is observed in the GST lane. CE, Cell extracts

Figure 2 Cdk9 and gp130 interact in living cells. The in vivo binding was performed under physiological conditions using 293 cells. Cell extracts were immunoprecipitated with anti-Cdk9 and subjected to Western blotting with anti-gp130. A clear signal of the in vivo interaction between the two proteins is visible. The bottom part shows the Western blot with anti-Cdk9 of the immunoprecipitated sample. CE, Cell extracts; NRS, Normal rabbit serum; IP, immunoprecipitation

Figure 3 Cdk9 is present both in the nucleus and in the cytoplasm. 293 cells were fractionated into cytosolic and nuclear fractions. The extracts were subjected to a Western blot with anti-Cdk9. The result shows that Cdk9 is present also in the cytosolic fraction. The bottom part shows the Western blotting of the two fractions with anti-tubulin, used as a control of the purity of the extracts

Figure 4 The binding between Cdk9 and gp130 increases upon IL-6 stimulation. 293 cells, transfected with gp130, Cdk9-HA Tag and dnCdk9-HA Tag, were stimulated with IL-6. The extracts were immunoprecipitated with anti-gp130 and subjected to Western blotting with anti-HA Tag. The figure shows a constitutive binding between Cdk9 and gp130 that increases upon IL-6 stimulation, with a maximum at 5'. No significant difference in the binding was observed using the dominant negative form of the kinase (dnCdk9). The lower part shows the Western blot of the immunoprecipitated sample with anti-gp130. NMS, normal mouse serum; CE, cell extracts

Figure 5 Cdk9 synergizes with IL-6 in inducing the activation of an NF-B reporter plasmid. 293 cells were co-transfected with a plasmid carrying a NF-B-driven luciferase reporter gene, and either Cdk9, or its dominant negative form, followed by IL-6 stimulation. In the absence of IL-6 stimulation Cdk9 was able to activate the NF-B reporter gene. After stimulation with IL-6, a substantial increase of luciferase activity was observed in cells transfected with Cdk9. Instead, expression of a dominant negative form of Cdk9 led to a decrease in the base-line luciferase activity and dramatically reduced the luciferase gene expression in IL-6-stimulated cells. Two g of each plasmid were transfected, the assay was performed in triplicate and was repeated five times. Standard deviation for each transfection was less than 3%