¹Department of Biochemistry, RWTH Aachen, Pauwelsstr. 30, 52074 Aachen, Germany

²Department of Cancer Immunology, University School of Medical Sciences, Garbary St. 15, 61866 Poznan, Poland

³Max-Planck-Institute for Immunobiology, Spemann Laboratory, Stübeweg 51, 79108 Freiburg, Germany

⁴Department of Molecular Oncology, Biomedical Research Center, Osaka University Medical School, 2-2 Yamada-oka, Suita, Osaka 565, Japan

⁵Institute of Clinical Immunology and Transfusion Medicine, Section Molecular Immunology, University of Leipzig, Delitzscher Str. 141, 04129 Leipzig, Germany

Correspondence to: Iris Behrmann, Department of Biochemistry, RWTH Aachen, Pauwelsstr. 30, 52074 Aachen, Germany

Interleukin-6 (IL-6)-type cytokines lead to growth arrest of human A375 melanoma cells. The present study demonstrates that this effect depends on the activation of STAT transcription factors. We observed a correlation between the extent of growth inhibition exerted by IL-6, IL-6 plus soluble IL-6 receptor or oncostatin M (OSM) and the intensities of STAT3 and STAT1 signals. A truncated chimeric receptor retaining only the membrane-proximal region of gp130, the common signal transducer of IL-6-type cytokines, did neither activate STATs nor mediate growth arrest of stable transfectants. These functions were restored by the addition of short STAT recruitment modules comprising critical tyrosine residues from gp130 (Y767, Y814). A receptor carrying tyrosine module Y759 of gp130 effectively mediated activation of the phosphatase SHP-2 but did not alter cell growth. Overexpression of dominant negative forms of STAT3 but not STAT1 abrogated the inhibitory effect of OSM and IL-6 in A375 cells. In addition, we have identified the cyclin-dependent kinase inhibitor p27/Kip1 as a novel target to be regulated by IL-6-type cytokines. Stimulation-dependent upregulation of p27 mRNA occurred STAT3-dependently. Also p27 protein accumulated which coincided with the disappearance of hyperphosphorylated retinoblastoma protein in three human melanoma cell lines sensitive to IL-6-type cytokines.

melanoma; STAT; interleukin-6; oncostatin M; p27/Kip; growth arrest

Introduction

Cytokines are protein mediators playing an important role in defining the fate of cells, such as proliferation, differentiation or death. Distinct responses of target cells to cytokines are determined by signal transduction events which differ between various cell types, or even between separate phases of cell development. Local regulation of cell growth by intercellular cytokine signals may prevent tumor progression during the first in situ stages. Thus, the elucidation of the molecular basis of cytokine effects should allow the disclosure of mechanisms regulating tumor growth.

Interleukin-6 (IL-6) and oncostatin M (OSM) belong to the group of IL-6-type cytokines which further comprises leukemia inhibitory factor (LIF), IL-11, ciliary neurotrophic factor, and cardiotrophin-1 (for recent reviews see Heinrich et al., 1998; Hirano, 1998). Both, IL-6 and OSM are typical pleiotropic cytokines with a wide spectrum of functions. These include enhancement of megakaryocyte growth and platelet formation, promotion of hematopoietic progenitor cell proliferation, and regulation of hepatic acute phase protein synthesis. Furthermore, both cytokines mediate growth stimulation of myeloma and plasmacytoma cells, but growth inhibition and differentiation of mouse M1 myeloid leukemia cells. Also certain breast and lung cancer cells respond with growth arrest to IL-6 and OSM (Chen et al., 1988; Douglas et al., 1997; Horn et al., 1990; Takizawa et al., 1993). Besides that, an anti-proliferative effect of IL-6 and OSM has been described for melanocytes (Swope et al., 1991) and for melanoma cells from early phases of tumor growth (Lu et al., 1992, 1993, 1996; Swope et al., 1991). In contrast, cells from advanced stage melanomas are in most cases resistant to IL-6 and/or OSM, acquiring the so-called `multicytokine resistance' phenotype. Therefore, local paracrine growth inhibition may play an important role for the prevention of tumor development.

The redundant character of the biological actions of cytokines can be explained by the shared usage of signal transducing molecules. IL-6-type cytokines use gp130 as a common chain of their multimeric receptor complexes (Heinrich et al., 1998; Hirano, 1998). Upon binding of IL-6 to its ligand-specific -chain IL-6R (gp80), gp130 joins the receptor complex and, as a dimer, transduces the signal into the cell. Also a soluble form of the IL-6R in complex with IL-6 associates with gp130 and thereby promotes signal transduction (Mackiewicz et al., 1992; Taga et al., 1989). Human OSM utilizes two different types of receptor complexes. Both of them contain gp130 heterodimerized with either the LIFR (Gearing et al., 1992) or an OSM-specific receptor component (OSMR) (Mosley et al., 1996). The structures of LIFR and OSMR are closely related to gp130 and both molecules contribute to intracellular signaling.

IL-6-type cytokines, as many other cytokines and growth factors, utilize Jaks (Janus kinases) and STATs (signal transducers and activators of transcription) as major mediators of signal transduction (Heinrich et al., 1998; Hirano, 1998). Janus kinases Jak1, Jak2 and Tyk2 associated with the membrane-proximal region of gp130 become activated upon stimulation, thus leading to the tyrosine phosphorylation of cellular substrates, among them the cytoplasmic tail of gp130. The four membrane-distal tyrosine residues of gp130 are, when phosphorylated, docking sites for the transcription factors STAT3 and STAT1 (Gerhartz et al., 1996; Stahl et al., 1995). Subsequently, STATs also are tyrosine-phosphorylated, form homo- or heterodimers and translocate to the nucleus where they regulate the transcription of target genes. The tyrosine phosphatase SHP-2 also binds to a specific phosphotyrosine motif of gp130 (Stahl et al., 1995), thereby possibly forming a link to the Ras/Raf/MAP kinase pathway, also known to be activated by IL-6-type cytokines (Boulton et al., 1994; Kumar et al., 1994). Activation of the SHP-2 phosphatase may further lead to down-regulation of the signal (Kim et al., 1998; Schaper et al., 1998; Symes et al., 1997). In addition, other molecules such as vav (Lee et al., 1997) or tyrosine kinases Hck (Ernst et al., 1994), Tec, Btk (Matsuda et al., 1995b) and Fes (Matsuda et al., 1995a) have been reported to become phosphorylated upon gp130 stimulation, but their contribution to signal transduction is unclear.

The molecular mechanisms underlying the pleiotropy of IL-6-type cytokines have only recently started to become understood: IL-6-induced neurite outgrowth in rat pheochromocytoma cells crucially depended on the activation of MAP kinase (Ihara et al., 1997). Myeloid M1 cells required functional STAT3 to differentiate into macrophage-like cells (Minami et al., 1996; Nakajima et al., 1996). The inhibition of differentiation and maintenance of pluripotency of embryonic stem cells cultured in the presence of LIF similarly depended on STAT3 (Boeuf et al., 1997; Niwa et al., 1998). Growth of transfected BAF - B03 pro-B cells needed the contribution of both signals derived from gp130: STATs induced anti-apoptotic factors, whereas MAP kinase activation was necessary for proliferation (Fukada et al., 1996). Short-term DNA synthesis in these cells, however, had even been observed with a truncated gp130 lacking all tyrosine residues required for SHP-2 and STAT activation (Murakami et al., 1991).

In this study, we delineate the signaling mechanism of IL-6 and OSM in human A375 melanoma cells that have long been known to be growth inhibited by these cytokines (Morinaga et al., 1989; Zarling et al., 1986). We show with receptor chimeras and dominant negative forms of STATs that the gp130-mediated growth arrest of human A375 melanoma cells crucially depends on the activation of STATs. Specifically, we demonstrate that STAT3, rather than STAT1 or the tyrosine phosphatase SHP-2, plays a key role for the growth inhibitory effect. In addition, we investigated the effect of IL-6/sIL-6R and OSM on the expression of p27/Kip1, a member of the KIP family of cyclin-dependent kinase inhibitors that further comprises p21/Cip1/Waf1 and p57/Kip2. These factors prevent cell cycle progression by binding to and inhibiting the activity of cyclin D- and cyclin E-associated kinases (reviewed in Sherr and Roberts, 1995). The identification of p27/Kip1 as a novel target to be upregulated by IL-6-type cytokines thus leads to a further understanding of the molecular mechanisms ultimately leading to growth arrest.

Results

Effect of IL-6 and OSM on A375 cells: correlation between growth inhibition and STAT activation

We investigated the time and dose dependence of the growth inhibitory effect of IL-6 and OSM on A375 cells. It was also tested whether the soluble IL-6R would enhance the effect exerted by IL-6 alone, as described for other systems (Mackiewicz et al., 1992; Taga et al., 1989). After 2 days of incubation with the different cytokines a slight growth inhibition of the cells became visible. It increased thereafter and reached a plateau after 4 - 5 days of culture (Figure 1a). OSM induced the strongest growth arrest. The effect of IL-6 was less pronounced but could be enhanced significantly by addition of the soluble IL-6 receptor, indicating that the number of surface IL-6 receptors was limiting (Figure 1b).

A very early effect that can be observed upon stimulation of cells with IL-6-type cytokines is the activation of latent cytoplasmic STAT factors, which subsequently translocate to the nucleus. Therefore, we tested whether nuclear extracts of A375 cells contained stimulation-dependent DNA binding activities. For electrophoretic mobility shift assays the SIE-probe was used which specifically binds STAT3 and STAT1 (Zhong et al., 1994). As shown in Figure 2 all stimuli used activated STATs, albeit with different kinetics and intensities. Upon OSM treatment, after 1 min prominent bands corresponding to STAT3 and STAT1 homo- and heterodimers were already visible and were at maximum after 5 min of stimulation. At 60 min the signal was greatly decreased with just a faint band resulting from a STAT3 homodimer remaining (Figure 2a). The DNA-binding activity induced by IL-6/sIL-6R complexes appeared somewhat delayed compared to OSM, reaching a maximum after 15 min but similarly declined thereafter. In the absence of the soluble receptor IL-6 effects were considerably weaker. Thus, the intensities of the inducible STAT signals paralleled the antiproliferative potential of OSM, IL-6/sIL-6R and IL-6, respectively. Although the initial activation of both STAT3 and STAT1 is transient, significant STAT3 signals can be observed for at least 48 h when cells are cultured in presence of OSM or the IL-6/sIL-6R complex (Figure 2b). Also upon stimulation with suboptimal cytokine doses preferential stimulation of STAT3 becomes evident (Figure 2c).

Gp130-mediated growth inhibition requires the presence of STAT recruitment sites in the receptor

To investigate the role of the STATs for the growth inhibitory action of IL-6-type cytokines in greater detail, we took advantage of chimeric receptors, which enabled us to study signals elicited by the cytoplasmic part of gp130 or muteins thereof independently of endogenous wild-type gp130 in A375 cells. The suitability of this approach was demonstrated by the EG receptor construct containing the extracellular region of the mouse erythropoietin (EPO) receptor fused to the transmembrane and intracellular parts of gp130 (Gerhartz et al., 1996): clones stably expressing this chimera (Figure 3) were dose-dependently growth-inhibited by EPO (Figure 4, first panel). Also the STAT signal was comparable to the one observed after IL-6 stimulation: the signal became detectable after 1 min, a maximal stimulation of both STAT1 and STAT3 was observed after 5 min, and after 60 min only a faint STAT3 band remained detectable. Therefore, the use of these and similar chimeras should be appropriate to study in detail the growth inhibitory effect mediated by gp130 dimers.

Construct B contains only the membrane-proximal box1/box2 region of gp130, important for Jak association while lacking all tyrosine residues leading to STAT recruitment. Clones stably expressing this receptor (Figure 3) showed, as expected, no STAT activation when stimulated with EPO (Figure 4, second panel). Interestingly, their growth was not influenced by EPO. The same clones, however, were susceptible to the action of IL-6/sIL-6R, indicating that their signaling machinery leading to growth arrest was still functioning (data not shown).

We and others have previously shown that the addition of `tyrosine modules' from gp130 or from other receptors to the membrane proximal region of gp130 resulted in receptors capable of activating STATs (Gerhartz et al., 1996; Stahl et al., 1995). We therefore generated clones expressing chimeric receptor constructs with tyrosine modules Y767 or Y814 from gp130 (Figure 3). As shown in Figure 4 (third and fourth panel), these clones were growth-inhibited by EPO. Concomitantly, an EPO-induced STAT activation could be observed: Y814 showed a clear prevalence of STAT3 homodimers and STAT3/1 heterodimers, with very few STAT1 homodimers; Y767 showed STAT1 and STAT3 signals with comparable intensities. Compared to the full-length EG construct (first panel) the signal was somewhat prolonged: after 60 min no considerable decrease in band intensities was oberved.

SHP-2 activation without concomitant STAT activation does not influence growth of A375 transfectants

Although gp130-stimulation leads to a transient SHP-2 phosphorylation in A375 cells (Figure 5a), SHP-2 activation cannot be essential for the growth arrest of A375 cells since the chimeras Y767 and Y814 lack tyrosine residue Y759 of gp130 required for SHP-2 activation (Stahl et al., 1995) while still being able to mediate growth arrest (Figure 4). To investigate the effect of SHP-2 activation without concomitant STAT activation, transfectants were generated that expressed receptors containing tyrosine module Y759 (Figure 3, lower panel). Phosphorylation of SHP-2 could be observed upon stimulation of these transfectants but not of those expressing the B construct (Figure 5a), confirming that SHP-2 was recruited via the Y759 module. Y759 transfectants showed no significant STAT activation, and their growth behavior was neither positively nor negatively influenced by the addition of EPO (Figure 5b).

Dominant negative STAT3 abrogates the growth inhibitory effect of IL-6-type cytokines

The data presented above strongly indicate that STATs play an essential role for gp130-mediated growth arrest of A375 melanoma cells. To further corroborate these results we made use of dominant negative forms of both STAT1 and STAT3. STAT1F and STAT3F contain a phenylalanine instead of tyrosine residues Y701 or Y705, respectively, which become phosphorylated upon activation of the wild-type STATs. STAT3D has a mutation within its DNA binding domain. Consequently, while still being able to become phosphorylated it can no longer bind to DNA (Nakajima et al., 1996). Clones were generated that overexpressed mutated or wild-type STATs to a similar level (data not shown). Mock transfected cells served as controls. Interestingly, clones overexpressing STAT1F showed no different behavior towards IL-6/sIL-6R or OSM (Figure 6c) compared to cells overexpressing wild-type STAT1 (Figure 6b) or to mock transfected cells (Figure 6a). STAT1F, however, was proven to be functional: compared to mock transfected cells, overexpression of STAT1F greatly diminished IL-6/sIL-6R or IFN-inducible STAT1 DNA binding activity in a gel shift experiment (Figure 7a). Moreover, this dominant negative form of STAT1 prevented upregulation of surface MHC class I expression by IFN (Figure 7b). Overexpression of STAT3 and its two forms acting in a dominant negative fashion (Figure 7a) clearly influenced the growth of transfectants in the presence of IL-6-type cytokines. Whereas cells overexpressing wild-type STAT3 were somewhat more sensitive to the growth inhibitory effect (Figure 6d), STAT3D expressing clones were growth-inhibited only at the highest concentrations of IL-6/sIL-6R or OSM (Figure 6f). Cells overexpressing dominant negative STAT3F were completely resistant towards IL-6/sIL-6R or OSM (Figure 6e). These results strongly indicate that STAT3 is the key mediator of the growth arrest in A375 melanoma cells elicited by IL-6-type cytokines.

Upregulation of cyclin-dependent kinase inhibitor p27/Kip1 by IL-6-type cytokines occurs STAT-dependently

In search of potential gene targets influenced by IL-6-type cytokines we investigated the expression of the cyclin-dependent kinase inhibitor p27/Kip1. For this molecule a role in melanoma development and melanoma cell growth has been suggested (see Discussion). The level of p27 mRNA increased upon incubation of A375 cells with IL-6/sIL-6R or OSM (Figure 8a). The increase is visible after 6 h, maximal after 48 h and the level slightly decreases thereafter. We next examined whether the sensitivity of the clones overexpressing wild-type or dominant negative forms of STAT3 is reflected in their ability to upregulate p27. As shown in Figure 8b, cells overexpressing wild-type STAT3 show an increased upregulation of p27 mRNA compared to mock transfected cells (fourfold and fivefold after IL-6/sIL-6R and OSM stimulation, respectively). This effect is visible already at an early time point (4 h). However, cells expressing the dominant negative forms STAT3F and STAT3D show only a weak p27 mRNA increase (ca. 1.5-fold). Thus, there is a correlation between the sensitivity of the cells towards the growth inhibitory effect of IL-6-type cytokines and the levels of p27 mRNA induced.

Stimulation-dependent accumulation of p27/Kip1-protein in human melanoma cells sensitive to IL-6-type cytokines

The increase in p27 mRNA levels correlated with increased protein levels: incubation with IL-6/sIL-6R or OSM led to an accumulation of p27 protein (Figure 9a). The effect became visible after 8 - 16 h and continuously increased. The expression levels of Erk1, a control protein, remained unchanged. Expression of p27 was not induced in control cells left without cytokine for up to 3 days (not shown).

The retinoblastoma protein (pRb), a key regulator of the cell cycle, is known to be an important target of cyclin-dependent kinases. Hyperphosphorylation of pRb leads to inactivation thereby allowing cell cycle progression. Thus, the appearance of the cyclin-dependent kinase inhibitor p27 correlates well with the disappearance of the hyperphosphorylated form of pRb (pRb^P, Figure 9a). Moreover, p27 accumulation can also be observed in human WM902b and WM239 melanoma cells sensitive to IL-6/sIL-6R (Lu et al., 1992, and Figure 9b). Thus, upregulation of p27 is not confined to our model cell line A375 but may be a more general phenomenon for melanoma cells growth inhibited by IL-6-type cytokines.

Discussion

Growth inhibition of melanoma cells by IL-6-type cytokines depends on STAT activation

The first major finding of the present study is that STAT factors, in particular STAT3, play a crucial role for IL-6- and OSM-mediated growth inhibition of A375 melanoma cells. This conclusion is based on three lines of evidence: (1) For a given stimulus (IL-6, IL-6/sIL-6R, or OSM) we observed a strong correlation between the intensities of the short-term STAT signals and the extent of growth inhibition. (2) A truncated chimeric receptor lacking the STAT recruitment sites of gp130 was not able to mediate growth inhibition. However, addition of tyrosine modules Y767 and Y814 from gp130 restored this function. Stahl et al. (1995) have shown that the activation of Jaks associated with the membrane-proximal box1/2 receptor region is not affected by the presence or absence of added STAT recruitment modules. It therefore seems most likely that the different potency of the receptors to mediate anti-proliferative effects is due to their different ability to activate STATs. (3) STAT3 was identified to play a major role for the antiproliferative effects of IL-6/sIL-6R and OSM since stable clones overexpressing wild-type STAT3 exhibited a more pronounced effect than mock transfected cells. In addition, expression of dominant negative forms of STAT3 abrogated the growth inhibitory effect of IL-6-type cytokines. Since overexpression of the dominant negative forms of STAT3 also prevented activation of endogenous STAT1 (Figure 7a), we formally cannot exclude a potential contribution of STAT1 from these experiments. However, overexpression of STAT1F, the dominant negative form of STAT1, did not disturb signaling of IL-6/sIL-6R and OSM although IFN-signaling was suppressed. Thus, STAT1 seems to be dispensible for the growth inhibitory effect of IL-6-type cytokines, although a short-term STAT1 activation can be observed upon high-dose stimulation. Currently there is no evidence suggesting that dominant negative STAT3 might interfere with Jak synthesis or function. In M1 cells, at least, IFN-responses which depend on Jak1 and Jak2 are not affected by the overexpression of dominant negative STAT3 (Nakajima et al., 1996).

Also in other cellular systems STAT3 is essential for the function of IL-6-type cytokines: for the survival of transfected BAF - B03 pro-B-cells (Fukada et al., 1996), for the differentiation of promyelocytic M1 cells to macrophage-like cells (Minami et al., 1996; Nakajima et al., 1996; Yamanaka et al., 1996) and for the proliferation and maintenance of pluripotency of embryonic stem cells (Boeuf et al., 1997; Niwa et al., 1998). However, not all IL-6 effects depend on STAT3: neurite outgrowth of PC12 pheochromocytoma cells needs signals via Y759 leading to activation of the tyrosine phosphatase SHP-2 and in turn to MAP kinase activation. STAT3 activation rather inhibits this effect (Ihara et al., 1997). Also the gp130-mediated mitogenic response of BAF - B03 transfectants depends on SHP-2 activation (Fukada et al., 1996). Moreover, SHP-2 activation without concomitant STAT activation induces a slight proliferative signal in M1 cells (Yamanaka et al., 1996). Thus, also SHP-2 plays an important role in signaling of IL-6-type cytokines, in some systems leading to effects different from those induced by STATs so that the biological response may be determined by the balance between STAT and SHP-2 activation.

Although gp130-stimulation leads to a transient SHP-2 phosphorylation in A375 cells, SHP-2 activation does not seem to be essential for the growth arrest since the chimeras Y767 and Y814 lacking tyrosine residue Y759, important for SHP-2 activation, are functional. Moreover, activation of SHP-2 without concomitant STAT activation did not influence the growth behavior of A375 transfectants. It will be of interest to determine whether SHP-2 may play an inhibitory role for gp130-signaling in these cells, as has been shown for other cell types (Kim et al., 1998; Schaper et al., 1998; Symes et al., 1997). This could explain why transfectants expressing the receptors Y767 and Y814 showed no apparent decrease of the STAT signal within the first hour, while DNA binding activity induced by the full-length EG construct or by wild-type gp130 dramatically decreased between 30 and 60 min after stimulation. However, the decreased downregulation of the signal could also be due to the lack of the dileucine internalization motif in the truncated receptors, which is essential for endocytosis of gp130 (Dittrich et al., 1996; Thiel et al., 1998) and which may account for the lower surface expression of the full-length EG construct compared to the shorter constructs (Figure 3). Studies are in progress to answer this question.

The effect of IL-6 on A375 cells could be significantly enhanced by addition of the soluble IL-6 receptor which acts agonistically (Mackiewicz et al., 1992; Taga et al., 1989), indicating that the amount of surface IL-6R is limiting. Similar observations have been made for other tumor cells growth-inhibited by IL-6 such as murine B16 melanoma cells (Oh et al., 1997), human non-small cell lung carcinoma cells (Ganapathi et al., 1996) or human osteosarcoma cells (Bellido et al., 1998). Thus, loss of receptors may provide a possibility for tumor cells to evade the action of inhibitory cytokines. Loss of components of the signaling machinery may be regarded as an alternative strategy to attain a growth advantage: recently, melanoma cells resistant to the anti-proliferative effect of IFN were described to contain less STAT1 than IFN-sensitive cells. Overexpression of STAT1 improved the responsiveness to IFN (Wong et al., 1997). It will be of interest to find out whether defects in Jak/STAT signaling contribute to IL-6-resistance observed in several late-stage melanoma cell lines (Lu et al., 1992, 1993).

p27/Kip as a novel target of IL-6-type cytokines

It is a challenging question how IL-6-type cytokines mediate in a cell-type specific manner via the activation of the same STAT factors various, and in part even contrasting, responses such as growth inhibition, survival, proliferation or differentiation. To further understand the mechanism underlying this pleiotropy it is necessary to identify the cellular targets, since it is well conceivable that different sets of target genes are induced in a cell-type specific manner.

The second major finding of the present study is the identification of cyclin-dependent kinase inhibitor p27/Kip1 as a factor that is upregulated during growth inhibition of A375 cells. We observed a stimulation-dependent accumulation of p27 protein coinciding with the disappearance of hyperphosphorylated pRb. This effect is not confined to A375 cells but may be a more general phenomenon since it similarly occurred in two other melanoma cell lines sensitive to IL-6/sIL-6R. We also found an increase of p27 mRNA level induced by IL-6/sIL-6R or OSM in A375 cells which was enhanced by overexpression of wild-type STAT3 but abrogated by the presence of dominant negative forms of STAT3. Thus, the extent of the p27 mRNA upregulation correlates well with the sensitivity of the clones towards the growth inhibitory effect by IL-6-type cytokines. Intriguingly, BAF - B03-transfectants were recently shown to downregulate p27 expression during the proliferative response imposed onto these cells by gp130-stimulation (Fukada et al., 1998). If the p27 gene was a direct target of STAT3, this factor would play a positive regulatory role for p27 expression in A375 cells, but a negative one in BAF-cells. However, it is also possible that IL-6 induces different sets of genes in these two cell lines which then lead to the differential effects on p27 expression. It remains to be determined whether p27 upregulation plays a causative role for the induction of growth arrest or whether it just correlates with the growth state of the cells. With regard to the requirement of a 3 day span to detect a negative growth effect compared to the activation of STAT factors (within minutes) or the induction of p27 (within hours), we can not exclude a possible upregulation of a distinct autocrine factor with antiproliferative effects.

Other cyclin-dependent kinase inhibitors have been reported to be subject of regulation by IL-6-type cytokines: p21/Cip1/Waf1 was found to be upregulated by IL-6/sIL-6R complexes during growth inhibition and differentiation of murine B16.F12 melanoma cells (Oh et al., 1997) and of human MG63 osteosarcoma cells (Bellido et al., 1998). Induction of differentiation of B cells by IL-6 coincided with increased levels of p21 and p18 (Morse et al., 1997). During the differentiation of mouse M1 cells an upregulation of p19 was observed (Narimatsu et al., 1997). However, initial studies on cytokine effects on those cyclin-dependent kinase inhibitors did not provide evidence supporting a major role for them in growth arrested A375 cells (unpublished results). Thus, IL-6-type cytokines may target cell cycle regulators in a cell-type-specific manner during antiproliferative responses.

A possible importance of p27 for melanoma development has been implicated by a recent study: a relation between p27-levels and tumor thickness was found for nodular melanomas, with less protein being expressed in thicker lesions. Patients having tumors with few p27-staining cells had a significantly higher risk of early relapse of their disease compared with those expressing moderate or high levels (Florenes et al., 1998). Moreover, inhibition of cell cycle progression by a phosphatidylinositol-3 kinase inhibitor or by overexpression of MARCKS, a protein kinase C substrate, correlated with increased expression of p27 in human choroidal melanoma cells (Casagrande et al., 1998; Manenti et al., 1998). Our finding that p27 is induced by IL-6-type cytokines during growth inhibition of melanoma cells therefore agrees well with the possibility that dysregulation of p27 expression may contribute to melanoma development and to the growth of melanoma cells.

Materials and methods

Cells and cytokines

Human A375 melanoma cells were purchased from the ATCC (CRL-1619). WM902b and WM239 cells (Herlyn, 1990) were kindly provided by Dr RS Kerbel (Sunnybrook Health Science Centre, Toronto, Canada). Cells were grown in RPMI 1640 medium supplemented with 5% FCS, 50 g/ml penicillin, 100 g/ml streptomycin in a humidified atmosphere with 5% CO₂. Purified recombinant human IL-6 with a specific activity of 2´10⁶ B cell stimulatory factor-2 units/mg was prepared as described (Arcone et al., 1991). Soluble human IL-6 receptor was expressed in baculovirus-infected insect cells (Weiergräber et al., 1995). Recombinant human OSM (Sporeno et al., 1994) was a generous gift of Dr G Ciliberto (Instituto di Recerche di Biologia Molecolare, Roma, Italy). Human recombinant erythropoietin was kindly provided by Drs J Burg and K-H Sellinger (Boehringer Mannheim, Penzberg, Germany). IFN was purchased from PeproTech (London, UK).

Growth inhibition assay

For the estimation of cell growth inhibition we applied an XTT colorimetic assay from Boehringer Mannheim. Cells were seeded in triplicates into 96-microwell plates at a density of 3´10³ viable cells/well. For studies on dose-dependence, cytokines in various concentrations were added to the medium. After 4 days of culture the XTT test was performed. Cells were incubated for 3 h at 37°C with a tetrazolium salt-based reagent which is metabolized by mitochondrial enzymes. The absorbance of the colored formazane product was measured at 450 nm by an ELISA reader. The percentage of growth inhibition was calculated in relation to the growth of control cells in medium without cytokines. For the experiments on time-dependence, cells were seeded with or without cytokines present at a density of 1.5´10³ viable cells/well and XTT tests were performed after different periods of incubation. The measured absorbance correlated well with the cell number, as observed in initial experiments (not shown).

Electrophoretic mobility shift assays (EMSA)

Nuclear extracts were prepared as described (Wegenka et al., 1993). Protein concentrations were measured with the BioRad^TM protein assay. A double-stranded mutated SIE-oligonucleotide from the c-fos promoter (m67SIE: 5'-GAT CCG GGA GGG ATT TAC GGG AAA TGC TG-3') was labeled by filling in 5' protruding ends with the Klenow enzyme, using -³²P-dATP (10 mCi/ml, 3000 Ci/nmol). Nuclear extracts containing 5 g of protein were incubated with about 10 fmol (10 000 c.p.m.) of probe in gel shift incubation buffer: 10 mM HEPES pH 7.8, 1 mM EDTA, 5 mM MgCl₂, 10% glycerol, 5 M dithiothreitol, 0.7 M phenylmethylsulfonyl fluoride, 0.1 mg/ml of poly(dI-C) and 1 mg/ml bovine serum albumin for 10 min at room temperature. The protein-DNA complexes were separated on a 4.5% polyacrylamide gel containing 7.5% glycerol in 0.25-fold TBE at 20 V/cm for 4 h. Gels were fixed in a water solution of 10% methanol and 10% acetic acid for 30 min, dried autoradiographed. Data were further analysed with a Storm 840 PhosphorImager (Molecular Dynamics).

Generation of transfectants

The construction of chimeric receptors consisting of the extracellular part of the murine EPO receptor and the cytoplasmic part of gp130 or derivatives thereof has been described before (Gerhartz et al., 1996; May et al., 1996). Receptor B (retains only the membrane proximal box1/2 region) and receptors Y759, Y767, Y814 (contain additionally a dodecapeptide sequence comprising the indicated tyrosine residues) are equipped with a C-terminally located Flag-epitope. XbaI - BamHI fragments from the described pSVL-derivatives were subcloned into pBluescript II SK(+) (Stratagene), then excised with NotI and ApaI and inserted between the respective sites of the pRC/CMV expression plasmid (Invitrogen). The construction of pCAGGS plasmids encoding wild-type and dominant negative STATs has been reported previously (Nakajima et al., 1996).

For preparation of stable transfectants, 2´10⁶ cells in 0.8 ml medium with 30 g of the respective plasmid DNA were subjected to a double pulse (3.0 kV/99 sec) using an Electro Square Porator T820 from BTX (San Diego, USA). Transfectants were grown in the presence 1.2 mg/ml of G418 (GIBCO BRL, Eggenstein, Germany) and clones expressing exogenous protein were selected by Western blot analysis. Chimeric receptors were detected using anti-Flag monoclonal antibody M2 from Kodak (Fernwald, Germany) or a polyclonal antiserum against the cytoplasmic part of gp130 (Upstate Biotechnology, Eching, Germany). Expression of the receptors was later verified by flow cytometry. Heterologous expression of STATs was assessed with a monoclonal antibody against the hemagglutinin (HA)-tag from BAbCO (Richmond, CA, USA). At least two single clones from each transfection were chosen for further experiments and gave similar results.

Cell lysis and Western blot analysis

Cells were lysed on the plate with 500 l of lysis buffer containing 1% Triton X-100, 10% glycerol, 50 mM HEPES pH 7.5, 1 mM EGTA, 150 mM NaCl, 1.5 mM MgCl₂, 50 mM sodium fluoride, 1 mM sodium vanadate, 1 mM PMSF, 5 g/ml aprotinin and 5 g/ml leupeptin. Lysates were cleared by centrifugation at 12 000 g, and protein concentration was determined with the BioRad^TM protein assay using BSA as a standard. Equal amounts of protein were separated on the 7.5% SDS - PAGE gels, transferred to a PVDF membrane (Amersham, Braunschweig, Germany), probed with the respective antibodies and detected for signals using the ECL system (Amersham). Polyclonal anti-p27, anti-Rb and ant-ERK1 antibodies were obtained from Santa Cruz Biotechnology (Heidelberg, Germany).

Immunoprecipitation

Polyclonal antibodies against human SHP-2 (Santa Cruz Biotechnology, Heidelberg, Germany) were added to cell lysates. After overnight incubation at 4°C, immunoprecipitates were collected with protein A-Sepharose (1 h, 4°C), washed three times with lysis buffer and analysed further by SDS - PAGE. Western blot analysis was performed with the 4G10 monoclonal anti-phosphotyrosine antibody (Upstate Biotechnology). After stripping, the blot was redeveloped with the anti-SHP-2 antiserum.

Flow cytometry

Transfectants were released from the dishes by treatment with phosphate-buffered saline, 10 mM EDTA. For monitoring the expression of chimeric receptors, approximately 10⁶ cells were first incubated with a 1 : 100 dilution of a polyclonal rabbit antiserum raised against a glutathione-S-transferase fusion protein of the extracellular region of the murine EPO receptor. The fusion protein had been expressed in E. coli and purified by the sarcosyl method (Frangioni and Neel, 1993). As a secondary antibody, a phycoerythrin (PE)-conjugated donkey anti-rabbit IgG antibody (Dianova, Hamburg, Germany) was used. MHC class I expression was measured with monoclonal antibody W6/32 (Sigma-Aldrich, Steinheim, Germany) followed by incubation with PE-conjugated goat anti-mouse IgG antiserum (Dianova). Fluorescence data were collected with a FACSCalibur (Becton Dickinson) and further analysed using the CellQuest software.

RNA isolation and Northern blot analysis

Total RNA was isolated with the RNeasy kit from Qiagen (Hilden, Germany). Five g of RNA were denatured, size-fractionated by electrophoresis in a 1% agarose-formaldehyde gel and transferred onto GeneScreen Plus nylon membrane (DuPont NEN Research Products, Boston, MA, USA). Detections of p27 mRNA were performed using human p27 cDNA labeled with -³²P-dATP with a random primer labeling kit (Boehringer Mannheim). After stripping of the blot, hybridization with a GAPDH cDNA probe was performed to control loading and transfer of RNA. Signals were quantified using a Storm 840 PhosphorImager with the ImageQuant software (Molecular Dynamics) and normalized according to the GAPDH values.

Acknowledgements

We thank Claudia Gerhartz and Birgit Heesel for providing receptor constructs, Hildegard Schmitz-Van de Leur for excellent technical assistance, Susanne Esser for preparation of the GST fusion protein, Lutz Graeve, Gerhard Müller-Newen and Fred Schaper for critical reading of the manuscript. We are grateful to Gennaro Ciliberto for providing OSM and to Dirk Eick for providing human p27 cDNA. Human recombinant erythropoietin was a generous gift of Boehringer Mannheim, Penzberg, Germany. This work was supported by the Volkswagen-Stiftung, the State Committee for Scientific Research (Warsaw), the Deutsche Forschungsgemeinschaft (Bonn), and by Fonds der Chemischen Industrie (Frankfurt).

Arcone R, Pucci P, Zappacosta F, Fontaine V, Malorni A, Marino G and Ciliberto G. (1991). Eur. J. Biochem. 198, 541-547. MEDLINE

Bellido T, O'Brien CA, Roberson PK and Manolagas SC. (1998). J. Biol. Chem. 273, 21137-21144. MEDLINE

Boeuf H, Hauss C, De Graeve F, Baran N and Kedinger C. (1997). J. Cell. Biol. 138, 1207-1217. MEDLINE

Boulton TG, Stahl N and Yancopoulos GD. (1994). J. Biol. Chem. 269, 11648-11655. MEDLINE

Casagrande F, Bacqueville D, Pillaire MJ, Malecaze F, Manenti S, Breton-Douillon M and Darbon JM. (1998). FEBS Lett. 422, 385-390. MEDLINE

Chen L, Mory Y, Zilberstein A and Revel M. (1988). Proc. Natl. Acad. Sci. USA 85, 8037-8041. MEDLINE

Dittrich E, Renfrew Haft C, Muys L, Heinrich PC and Graeve L. (1996). J. Biol. Chem. 271, 5487-5494. MEDLINE

Douglas AM, Goss GA, Sutherland RL, Hilton DJ, Berndt MC, Nicola NA and Begley CG. (1997). Oncogene 14, 661-669. MEDLINE

Ernst M, Gearing DP and Dunn AR. (1994). EMBO J. 13, 1574-1584. MEDLINE

Florenes VA, Maelandsmo GM, Kerbel RS, Slingerland JM, Nesland JM and Holm R. (1998). Am. J. Pathol. 153, 305-312. MEDLINE

Frangioni JV and Neel BG. (1993). Anal. Biochem. 210, 179-187. Article MEDLINE

Fukada T, Hibi M, Yamanaka Y, Takahashi-Tezuka M, Fujitani Y, Yamaguchi T, Nakajima K and Hirano T. (1996). Immunity 5, 449-460. MEDLINE

Fukada T, Ohtani T, Yoshida Y, Shirogane T, Nishida K, Nakajima K, Hibi M and Hirano T. (1998). EMBO J. 17, 6670-6677. MEDLINE

Ganapathi MK, Weizer AK, Borsellino S, Bukowski RM, Ganapathi R, Rice T, Casey G and Kawamura K. (1996). Cell Growth Differ. 7, 923-929. MEDLINE

Gearing DP, Comeau MR, Friend DJ, Gimpel SD, Thut CJ, McGourty J, Brasher KK, King JA, Gillis S, Mosley B, Ziegler SF and Cosman D. (1992). Science 255, 1434-1437. MEDLINE

Gerhartz C, Heesel B, Sasse J, Hemmann U, Landgraf C, Schneider-Mergener J, Horn F, Heinrich PC and Graeve L. (1996). J. Biol. Chem. 271, 12991-12998. MEDLINE

Heinrich PC, Behrmann I, Müller-Newen G, Schaper F and Graeve L. (1998). Biochem. J. 334, 297-314. MEDLINE

Herlyn M. (1990). Cancer and Metastasis Rev. 9, 101-112.

Hirano T. (1998). Int. Rev. Immunol. 16, 249-284. MEDLINE

Horn D, Fitzpatrick WC, Gompper PT, Ochs V, Bolton-Hansen M, Zarling J, Malik N, Todaro GJ and Linsley PS. (1990). Growth Factors 2, 157-165. MEDLINE

Ihara S, Nakajima K, Fukada T, Hibi M, Nagata S, Hirano T and Fukui Y. (1997). EMBO J. 16, 5345-5352. Article MEDLINE

Kim H, Hawley TS, Hawley RG and Baumann H. (1998). Mol. Cell. Biol. 18, 1525-1533. MEDLINE

Kumar G, Gupta S, Wang S and Nel, AE. (1994). J. Immunol. 153, 4436-4447. MEDLINE

Lee IS, Liu Y, Narazaki M, Hibi M, Kishimoto T and Taga T. (1997). FEBS Lett. 401, 133-137. MEDLINE

Lu C, Vickers MF and Kerbel RS. (1992). Proc. Natl. Acad. Sci. USA 89, 9215-9219. MEDLINE

Lu C, Rak JW, Kobayashi H and Kerbel RS. (1993). Cancer Res. 53, 2708-2711. MEDLINE

Lu C, Sheehan C, Rak JW, Chambers CA, Hozumi N and Kerbel RS. (1996). Clin. Cancer Res. 2, 1417-1425. MEDLINE

Mackiewicz A, Schooltink H, Heinrich PC and Rose-John S. (1992). J. Immunol. 149, 2021-2027. MEDLINE

Manenti S, Malecaze F, Chap H and Darbon JM. (1998). Cancer Res. 58, 1429-1434. MEDLINE

Matsuda T, Fukada T, Takahashi-Tezuka M, Okuyama Y, Fujitani Y, Hanazono Y, Hirai H and Hirano T. (1995a). J. Biol. Chem. 270, 11037-11039. MEDLINE

Matsuda T, Takahashi-Tezuka M, Fukada T, Okuyama Y, Fujitani Y, Tsukada S, Mano H, Hirai H, Witte ON and Hirano T. (1995b). Blood 85, 627-633.

May P, Gerhartz C, Heesel B, Welte T, Doppler W, Graeve L, Horn F and Heinrich PC. (1996). FEBS. Lett. 394, 221-226. MEDLINE

Minami M, Inoue M, Wei S, Takeda K, Matsumoto M, Kishimoto T and Akira S. (1996). Proc. Natl. Acad. Sci. USA 93, 3963-3966. Article MEDLINE

Morinaga Y, Suzuki H, Takatsuki F, Akiyama Y, Taniyama T, Matsushima K and Onozaki K. (1989). J. Immunol. 143, 3538-3542. MEDLINE

Morse L, Chen D, Franklin D, Xiong Y and Chen-Kiang S. (1997). Immunity 6, 47-56. MEDLINE

Mosley B, De Imus C, Friend D, Boiani N, Thoma B, Park LS and Cosman D. (1996). J. Biol. Chem. 271, 32635-32643. MEDLINE

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

Nakajima K, Yamanaka Y, Nakae K, Kojima H, Ichiba M, Kiuchi N, Kitaoka T, Fukada T, Hibi M and Hirano T. (1996). EMBO J. 15, 3651-3658. MEDLINE

Narimatsu M, Nakajima K, Ichiba M and Hirano T. (1997). Biochem. Biophys. Res. Commun. 238, 764-768. MEDLINE

Niwa H, Burdon T, Chambers I and Smith A. (1998). Genes Dev. 12, 2048-2060. MEDLINE

Oh J-W, Katz A, Harroch S, Eisenbach L, Revel M and Chebath J. (1997). Oncogene 15, 569-577. MEDLINE

Schaper F, Gendo C, Eck M, Schmitz J, Grimm C, Anhuf D, Kerr IM and Heinrich PC. (1998). Biochem. J. 335, 557-565. MEDLINE

Sherr CJ and Roberts JM. (1995). Genes Dev. 9, 1149-1163. MEDLINE

Sporeno E, Barbato G, Graziani R, Pucci P, Nitti G and Paonessa G. (1994). Cytokine 6, 255-264. MEDLINE

Stahl N, Farruggella TJ, Boulton TG, Zhong Z, Darnell Jr JE and Yancopoulos GD. (1995). Science 267, 1349-1353. MEDLINE

Swope VB, Abdel-Malek Z, Kassem LM and Nordlund JJ. (1991). J. Invest. Dermatol. 96, 180-185. MEDLINE

Symes A, Stahl N, Reeves SA, Farruggella T, Servidei T, Gearan T, Yancopoulos G and Fink JS. (1997). Curr. Biol. 7, 697-700. MEDLINE

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

Takizawa H, Ohtoshi T, Ohta K, Yamashita N, Hirohata S, Hirai K, Hiramatsu K and Ito K. (1993). Cancer Res. 53, 4175-4181. MEDLINE

Thiel S, Behrmann I, Dittrich E, Muys L, Tavernier J, Wijdenes J, Heinrich PC and Graeve L. (1998). Biochem. J. 330, 47-54. MEDLINE

Wegenka UM, Buschmann J, Lütticken C, Heinrich PC and Horn F. (1993). Mol. Cell. Biol. 13, 276-288. MEDLINE

Weiergräber O, Hemmann U, Küster A, Müller-Newen G, Schneider J, Rose-John S, Kurschat P, Brakenhoff JP, Hart MH, Stabel S and Heinrich PC. (1995). Eur. J. Biochem. 234, 661-669. MEDLINE

Wong LH, Krauer KG, Hatzinisiriou I, Estcourt MJ, Hersey P, Tam ND, Edmondson S, Devenish RJ and Ralph SJ. (1997). J. Biol. Chem. 272, 28779-28785. Article MEDLINE

Yamanaka Y, Nakajima K, Fukada T, Hibi M and Hirano T. (1996). EMBO J. 15, 1557-1565. MEDLINE

Zarling JM, Shoyab M, Marquardt H, Hanson MB, Lioubin MN and Todaro GJ. (1986). Proc. Natl. Acad. Sci. USA 83, 9739-9743. MEDLINE

Zhong Z, Wen Z and Darnell Jr JE. (1994). Science 264, 95-98. MEDLINE

Figure 1 Effect of IL-6 and OSM on the growth of human A375 melanoma cells. (a) Kinetics of growth inhibition. 1500 cells/well were seeded into 96 well plates and cultured for different periods of time in the presence of IL-6 (100 ng/ml), IL-6/sIL-6R (100 and 500 ng/ml, respectively), OSM (50 ng/ml), or with no cytokine added. (b) Cells (initial density 3000 cells/well) were cultured for 4 days in the presence of different concentrations of cytokines. Soluble IL-6R was added to IL-6 in twofold molar excess. Growth was assessed by an XTT test, values in (b) are recalculated according to untreated controls. Error bars represent the standard deviation of triplicate samples

Figure 2 Time- and dose-dependence of STAT activation in A375. (a) Cells were stimulated for the times indicated with OSM (200 ng/ml), IL-6/sIL-6R (200 and 1000 ng/ml, respectively), or IL-6 (200 ng/ml). (b) Cells were stimulated for the times indicated with OSM (50 ng/ml) or IL-6/sIL-6R (200 and 1000 ng/ml, respectively). (c) Cells were stimulated for 15 min with cytokines at the indicated doses. sIL-6R was added to IL-6 in twofold molar excess. Nuclear extracts containing 5 g of protein were used for an EMSA with the SIEm67 probe. Positions of homo- and heterodimers of STAT1 and STAT3 are indicated by arrows

Figure 3 Expression of chimeric receptors on the surface of A375 transfectants. Cells were incubated with a polyclonal rabbit-anti-mouse EPOR antiserum (1 : 100 dilution) followed by staining with PE-conjugated donkey-anti-rabbit IgG antiserum. Fluorescence was measured by a FACSCalibur (Becton Dickinson) and analysed by the CellQuest software. Each panel contains the histograms of the indicated transfectants (black) in comparison to those of transfectants incubated with secondary antibody alone (dotted line) and of parental A375 cells (solid line) to show unspecific binding of the polyclonal antisera

Figure 4 Activation of STATs is necessary for gp130-mediated growth inhibition of A375 transfectants. Left: Schematic representation of the hybrid receptors expressed in A375 cells. EG: chimera with the full length gp130 cytoplasmic tail; B: truncated gp130 (box 1/2) construct; Y767, Y814: constructs with added tyrosine modules, the sequence downstream of the relevant tyrosine residue is indicated. Middle: Cells were stimulated with EPO (50 U/ml) for different periods of time as indicated, nuclear extracts were prepared and tested for DNA binding activity in an EMSA with the SIEm67 probe. The positions of homo- and heterodimers of STAT1 and 3 are indicated. Right: Cells were treated with EPO at various concentrations for 4 days before growth was assessed by an XTT assay recalculated according to untreated controls. Error bars represent the standard deviation of triplicate samples

Figure 5 SHP-2 activation without concomitant STAT activation does not influence growth of A375 transfectants. (a) EPO-inducible SHP-2 phosphorylation of stable transfectants. EG, B and Y579 transfectants were stimulated for the indicated times with EPO (50 U/ml). After lysis of the cells, immunoprecipitates of the anti-SHP-2 antiserum were separated on an SDS - polyacrylamide gel. A Western blot was performed, and an anti-phosphotyrosine antibody was used for detection. The blot was reprobed with the anti-SHP-2 antibody. (b) Gel retardation and proliferation assays of Y759 transfectants (shown schematically on the left) were performed as described in the legend of Figure 4

Figure 6 Dominant negative STAT3 abrogates IL-6/sIL-6R and OSM mediated growth inhibition. A375 cells transfected with the pCAGGS control vector (a) or with expression constructs for wild-type STAT1 (b), STAT1F (c), wild-type STAT3 (d), STAT3F (e), or STAT3D (f) were incubated with IL-6/sIL-6R (squares) or OSM (circles) at different concentrations. Cell growth was assessed by an XTT assay recalculated according to untreated controls. Error bars represent standard deviation of triplicate samples

Figure 7 Functionality of dominant negative STATs. (a) Interference with the DNA binding activity of endogenous STATs. Mock transfected cells or cells overexpressing STAT3F, STAT3D and STAT1F were left untreated or stimulated for 15 min with IL-6/sIL-6R (100 ng and 500 ng/ml, respectively) or with IFN (1000 U/ml). Nuclear extracts containing 5 g/ml protein were analysed for DNA binding activity in an EMSA with the SIEm67 probe. The positions of homo- and heterodimers of STAT1 and 3 are indicated. (b) STAT1F inhibits IFN mediated upregulation of MHC class I expression. Mock transfected cells or STAT1F expressing cells were treated with IFN (1000 U/ml) for 3 days or were left untreated. Cells were incubated first with a monoclonal antibody directed against human MHC class I molecules, and then with a PE-conjugated goat-anti-mouse IgG secondary antiserum. Fluorescence was measured by a FACS-Calibur (Becton Dickinson) and analysed by the CellQuest software. Histograms of untreated cells (thin line) were overlaid with the respective histograms of IFN-treated cells (bold line). Histograms of cells incubated only with the secondary antibody are depicted by a dotted line

Figure 8 IL-6/sIL-6R and OSM upregulate p27/Kip1 mRNA expression in a STAT3-dependent manner. Northern blot analysis of (a) A375 parental cells or (b) of transfectants stably expressing STAT3WT, STAT3D, STAT3F or of mock transfectants. Cells were cultured for the indicated times in the presence of IL-6/sIL-6R (200 and 1000 ng/ml, respectively) or OSM (50 ng/ml). 5 g of total RNA was used for Northern blot analysis. After p27 detection blots were stripped and hybridized with a GAPDH cDNA probe. Fold increase of p27 mRNA normalized according to GAPDH values was calculated using the ImageQuant software after scanning of the blots with a PhosphorImager (Molecular Dynamics)

Figure 9 Stimulation-dependent accumulation of p27 protein. (a) Western blot of A375 cells treated with IL-6/sIL-6R (200 and 1000 ng/ml, respectively) or with OSM (50 ng/ml) for the indicated times. Equal amounts of total cell lysates containing 50 g of protein were separated in a 6% (for pRb detection) or in a 15% (for p27 detection) SDS - polyacrylamide gel. Blots were developed with antibodies specific for pRB (upper panel), or for p27 (lower panel). After stripping, the blot of the lower panel was reprobed with an anti-Erk1 antibody to demonstrate equal loading. The position of molecular size markers, hyper- and hypophosphorylated retinoblastoma protein (pRb^P, pRb), p27 and Erk1 are indicated. (b) IL-6/sIL-6R induces p27 protein expression and pRB hypophosphorylation in WM902b and WM239 cells sensitive to the antiproliferative cytokine effect. Left panel: Western blot analysis was performed as in (a), right panel: WM902b cells (squares) and WM239 cells (circles) were incubated with IL-6/sIL-6R at different concentrations. Soluble IL-6R was added in twofold molar excess. Cell growth was assessed by an XTT assay recalculated according to untreated controls. Error bars represent standard deviation of triplicate samples