The MEK/ERK Pathway Acts Upstream of NF (cid:1) B1 (p50) Homodimer Activity and Bcl-2 Expression in a Murine B-Cell Lymphoma Cell Line MEK INHIBITION RESTORES RADIATION-INDUCED APOPTOSIS*

we described the NF (cid:1) B status for two murine B-cell lymphoma cell lines, LY-as (apoptosis-sensitive) and LY-ar (apoptosis-refractory) and provided evidence that NF (cid:1) B1 (p50) homodimers contribute to the expression of Bcl-2 in the LY-ar line. In the present study, we investigated the upstream signals leading to p50 homodimer activation and Bcl-2 expression. We found that in LY-ar cells, ERK1 and ERK2 were constitutively phosphorylated, whereas LY-as cells had no detectable ERK1 or ERK2 phosphorylation. Treatment of LY-ar cells with the MEK inhibitors PD 98059, U0126, and PD 184352 led to a loss of phosphorylated ERK1 and ERK2, a reversal of nuclear p50 homodimer DNA binding, and a decrease in Bcl-2 protein expression. Similarly, activation of the MEK/ERK pathway in LY-as cells by phorbol ester led to Bcl-2

Nuclear factor B (NFB) is traditionally described as a family of transcription factors that are activated during immune and inflammation responses, regulating the genes encoding cytokines such as IL-2, 1 IL-6, IL-8, and granulocyte mac-rophage-colony stimulating factor (1). NFB has since been dubbed a central mediator of the human immune response and has been shown to control hundreds of genes in addition to those for cytokines, including cell adhesion molecule and immunoreceptor genes (2). In the last few years, NFB has additionally been recognized as a key regulator of anti-apoptotic genes, including the genes coding for the inhibitor of apoptosis proteins and members of the Bcl-2 family (3,4). This control of anti-apoptotic genes has implicated aberrant NFB activity as a cell survival signal for many cancer cell types. Cancers classified as having aberrant NFB activity include Hodgkin's disease, chronic myelogenous leukemia, adult T-cell leukemia, acute lymphoblastic leukemia, melanoma, and solid tumors such as breast, colon, ovarian, pancreatic, thyroid, bladder, and prostate carcinomas (5,6).
Often, chromosomal rearrangement or overexpression of Rel family members leads to aberrant NFB activity, particularly in hematopoietic tumors (6). However, not all aberrant NFB activity in cancer cells is a result of Rel family member overexpression, and it has been recognized that NFB activity may also arise by the activation of signaling kinases that converge on the activation of I B kinase (IKK). IKK, in turn, phosphorylates IB, initiating the degradation of IB through the ubiquitin/proteasome pathway. This results in the release of bound NFB subunits that then translocate to the nucleus and activate gene transcription (7). NFB-inducing kinase, a mitogenactivated protein kinase kinase kinase (M3K) that is constitutively activated in melanoma cells (8), exemplifies a signaling kinase that leads to aberrant NFB activity in tumor cells upstream of IKK.
There are a number of other kinases known to act upstream of IKK. These include members of the M3K family such as MEKK-1, MEKK-2, MEKK-3, and Tpl-2/COT (9 -13). MEKK-1 activity has been associated with pancreatic cancer (14), MEKK-3 with hepatocellular carcinoma (15), and Tpl-2 with breast, colon, and gastric cancers (16,17). Recently, Tpl-2 has also been found in malignancies associated with Epstein-Barr virus infection, where it acts as a mediator of latent membrane protein-1-induced NFB activation (18). Thus, cancers that have aberrant kinase activity upstream of IKK would be expected to have constitutively nuclear NFB; this activity could, in turn, contribute to the expression of genes important to the survival of those cancers.
As mentioned above, several M3K family members have been found to act upstream of IKK. These kinases additionally activate members of the mitogen-activated protein kinase (MAPK) cascade, leading to the activation of Jun N-terminal kinase (JNK), p38 MAPK, and/or extracellular signal-regulated protein kinase (ERK) (19). In melanoma cells, the NFB activity downstream of NFB-inducing kinase (NIK) appears to be dependent not only on the activation of IKK but also on NIK-regulated activation of the MEK/ERK cascade (8). Other reports implicating downstream MAPK signaling kinases in the activation of NFB include the demonstration that ERK5 and ERK2 cooperatively regulate NFB activity in NIH 3T3 cells and the suggestion that persistent activation of NFB by IL-1 is mediated by the MEK/ERK pathway (20,21). The activation of JNK may also contribute to NFB activity through the induction of ␤-transducin repeat-containing protein, which mediates ubiquitination of phosphorylated IB (22). Thus, several different MAPK proteins may be capable of activating NFB independently of IKK. It is possible, in terms of cell survival, that this parallel pathway plays an important and redundant role to IKK-induced NFB activity.
We previously described the NFB status in two murine B-cell lymphoma cell lines, LY-as and LY-ar. The parent line, LY-as, was found to be lacking in NFB activity, whereas the derived line, LY-ar, had constitutively nuclear NFB1 (p50) homodimers that apparently contribute to the expression of the bcl-2 gene (23). In the present study, we used MEK inhibitors to establish a role for the MEK/ERK pathway upstream of p50 homodimer activity and Bcl-2 expression and demonstrate that MEK inhibitors sensitize normally radioresistant LY-ar cells to radiation-induced apoptosis. This study provides evidence that the activation of the MEK/ERK pathway could be an important step in the progression of lymphoma from an apoptosis-sensitive to an apoptosis-resistant phenotype. These data may therefore have clinical implications for the treatment of advanced cancers with therapeutic agents that induce apoptosis.

EXPERIMENTAL PROCEDURES
Cell Culture and Treatments-LY-ar and LY-as murine B-cell lymphoma cell lines were maintained in RPMI 1640 culture medium supplemented with 10% (v/v) fetal bovine serum, 1% (v/v) 200 mM Lglutamine, and 1% (v/v) 10,000 units/ml penicillin/streptomycin in an atmosphere of 5% CO 2  Fluorescence-activated Cell Sorter (FACS) Sorting and Analysis to Establish Stable LY-as Cell Lines Expressing a bcl-2 Construct-A retroviral infection strategy was adopted to deliver both human bcl-2 and the gene encoding green fluorescent protein (GFP) on a single construct containing an internal ribosomal entry site. Target LY-as cells were infected with retrovirus produced by PHOENIX-Ampho cells using standard polybrene-enhanced retroviral infection. Virus-treated LY-as cells were grown for 2-3 days and bulk-sorted to enrich for GFP expression. After several growth and sorting cycles, GFP-expressing cells were cloned through single-cell sorting by fluorescence-activated cell sorter. Stable clones that expressed both GFP and Bcl-2 (as detected by Western blot) were chosen for future experiments.
Electrophoretic Mobility Shift Assay (EMSA)-Nuclear extracts were prepared as described previously (24), and 15 g of nuclear protein was incubated with 32 P-labeled NFB oligonucleotide as described by the manufacturer (Promega) except that .07 pmol of radiolabeled consensus oligonucleotide probe per reaction was used rather than .035 pmol. Bound probe was resolved from free probe using 4% native gels run at 180 V at 4°C for ϳ2.5 h. Gels were visualized by phosphorimaging.
DNA Fragmentation-Enzymatically induced DNA fragmentation as a result of apoptosis was quantified as described previously (25). Briefly, cells were prelabeled with 10 nCi/ml of [ 14 C]thymidine and then incubated for 72 or 96 h in the presence of the MEK inhibitors described above. The cells were then exposed to 5 Gy of ␥ irradiation from a high dose-rate 137 Cs unit (4 -5 Gy/min) at room temperature. Cells were collected 4 h later, washed in phosphate-buffered saline, and lysed by incubation in 0.5 ml of lytic buffer (10 mM Tris, 1 mM EDTA, and 0.2% Triton X-100) for 20 min on ice. Insoluble chromatin was separated from soluble DNA fragments by centrifugation at 13,000 ϫ g for 10 min. Soluble and insoluble fractions were transferred to scintillation vials containing 1 ml of Soluene 350 (Packard) and incubated overnight at 60°C. Ten milliliters of Hionic Fluor (Packard) was added, and samples were counted in a liquid scintillation counter. DNA fragmentation was quantified as the percentage of the total radioactivity that appeared in the soluble fraction.

LY-ar Cells Have Constitutively Phosphorylated ERK1 and ERK2 (p44 and p42), whereas LY-as Cells Lack Phosphorylated ERK1 and ERK2
Proteins-Two murine B-cell lymphoma cell lines, isolated from a mouse lymphoma (LY-TH) and grown in vitro, were used for these studies. LY-as and LY-ar cells have been described previously for their differences in radiationinduced apoptotic response (25). To determine whether the MEK1/MEK2 pathway was activated in these cell lines, Western blot analysis was performed using anti-phospho-p44 and -p42 antibodies. The two cell lines had equal expression of total p44 and p42 proteins, but only the LY-ar line possessed constitutively phosphorylated p44 and p42 (Fig. 1) 1. The ERK1 and ERK2 MAP kinase pathway is activated in LY-ar cells but not in LY-as cells. Western blots were performed using 25 g of protein from whole-cell lysates prepared from LY-as and LY-ar cells for phospho-and total-p44/p42. cells were treated with the MEK inhibitor PD 98059, which has been shown along with the inhibitors U0126 and PD 184352 to be highly specific for MEK kinases based on its inability to inhibit over 30 other kinases as part of a specificity screen (26). Treatment with PD 98059 for 24 and 48 h led to decreases in phospho-p44 and -p42 as detected by Western blot (Fig. 2A). Similar results were obtained when LY-ar cells were treated with U0126 or PD 184352 for 48 h (Fig. 2B). To determine how rapidly p44 and p42 phosphorylation was diminished with MEK inhibitor treatment, we also treated LY-ar cells with PD 184352 for 1 h. This treatment also led to decreased p44 and p42 phosphorylation (Fig. 2C).
It should be noted that the MEK inhibitors PD 98059 and U0126 additionally have been reported to inhibit ERK5 phosphorylation by inhibiting MEK5 activity (27); however, Western blot demonstrated that whole-cell lysates from LY-as and LY-ar cells do not differ in their amounts of phosphorylated ERK5. Moreover, treatment of LY-ar cells with PD 184352, which has been shown to be more specific for inhibiting the phosphorylation of ERK1/ERK2 than ERK5 (28), did not result in a loss of ERK5 phosphorylation (data not shown). Thus, the ERK1/ERK2 pathway that is activated in LY-ar cells is more likely to contribute to their radioresistant phenotype than the ERK5 pathway. Taken together, the data in Fig. 2 demonstrate that MEK inhibitors very rapidly reverse the phosphorylation of ERK1 and ERK2 in LY-ar cells and maintain MEK inhibition for long periods of time.
Treatment of LY-ar Cells with MEK Inhibitors Leads to Reversal of Nuclear p50 Homodimer DNA Binding Activity-Based on reports suggesting that the MEK/ERK pathway acts upstream of NFB activation (8,20,21), we performed an EMSA on nuclear lysates prepared from PD 98059-treated LY-ar cells to determine whether MEK inhibition would lead to a loss of nuclear p50 homodimer activity. DNA binding was lost by 24 h of treatment with PD 98059, and binding remained inhibited at 48 and 72 h (Fig. 3A). Similar results were obtained using nuclear lysates prepared from LY-ar cells treated for 24 h with U0126 or PD 184352 (Fig. 3B). Because ERK1/ ERK2 phosphorylation was lost very rapidly when LY-ar cells were treated with PD 184352 (Fig. 2C), we also performed EMSA to determine whether the loss of p50/p50 binding correlated directly with the loss of ERK1/ERK2 activity. Indeed, DNA binding was diminished at both 1 and 3 h of treatment with PD 184352 with losses in p50/p50 DNA binding intensity of 20 and 40%, respectively (Fig. 3C). These data suggest that the MEK/ERK pathway acts upstream of p50 homodimer activation in LY-ar cells.
Treatment of LY-ar Cells with MEK Inhibitors Leads to Decreases in Bcl-2 Protein Expression-Because our previous data suggested that p50 homodimer activity in LY-ar cells contributed to the expression of the bcl-2 gene (23), we investigated whether the reversal of nuclear p50 homodimer DNA binding by MEK inhibitors correlated with a loss in Bcl-2 protein expression. Because of the long half-life of Bcl-2 protein, Western blot was performed on whole-cell lysates prepared from LY-ar cells treated with PD 98059 for up to 120 h. Treatment of LY-ar cells with PD 98059 led to decreases in Bcl-2 protein expression with roughly 2-and 4-fold decreases in expression at 48 and 120 h, respectively (Fig. 4A). Similar results were obtained when Western blot was performed on LY-ar cells treated for 72 h with U0126 or PD 184352, with roughly 2-fold decreases for each (Fig. 4B).
Treatment of LY-as Cells with TPA Induces ERK1 and ERK2 Phosphorylation as well as Bcl-2 Protein Expression That Can Be Blocked by PD 98059 -We previously demonstrated that the treatment of LY-as cells with the phorbol ester TPA led to an increase in Bcl-2 protein expression (23). To determine whether TPA treatment would also lead to an activation of ERK1/ERK2, whole-cell lysates were prepared from TPAtreated LY-as cells and analyzed by Western blot for phospho-p44 and -p42. Treatment of LY-as cells with TPA led to phosphorylation of ERK proteins, and this activation was blocked by simultaneous treatment with PD 98059 (Fig. 5A). TPAinduced Bcl-2 protein expression was also blocked by simultaneous treatment with PD 98059 as detected by Western blot (Fig. 5B).

LY-as Cells Retrovirally Infected with Human bcl-2 Lack Phosphorylated ERK1/ERK2 and Nuclear NFB but Are Resistant to Radiation-induced Apoptosis-To directly test the role of Bcl-2 expression in blocking radiation-induced apoptosis, LY-as cells were infected with retroviral vectors containing human bcl-2 cDNA. Virally infected cells were selected and cloned into stable cell populations by detection of GFP expression downstream of the bcl-2 open reading frame and an inter-
nal ribosome entry site (IRES). These cells are referred to as the BIG cell line (Bcl-2-IRES-GFP). As a control, LY-as cells were also infected with a retroviral vector expressing GFP alone. These cells are referred to simply as the GFP cell line. To determine whether Bcl-2 expression in LY-as cells caused activation of the MEK/ERK pathway, Western blot for phospho-p44 and p42 was performed using whole-cell lysates from GFP and BIG cells. Neither GFP nor BIG cells possessed phosphorylated p44 and p42 proteins (Fig. 6A).
In contrast to reports by ourselves and others (23, 29, 30) suggesting NFB family members directly regulate transcription of the bcl-2 gene, some researchers have suggested that Bcl-2 can act upstream of NFB, providing evidence that bcl-2 transfection causes NFB activation (31)(32)(33). To determine whether Bcl-2 expression in LY-as cells led to NFB activation, EMSA was performed using nuclear lysates from GFP and BIG cells. Both cell types were treated with TPA as a positive control for NFB binding. Like their parent cell type LY-as, neither GFP nor BIG cells had constitutively nuclear NFB DNA binding; however, NFB activation by TPA remained intact in both cell lines (Fig. 6B).
Although Bcl-2 expression did not lead to the activation of ERK1/ERK2 or NFB in the LY-as cell background, we performed DNA fragmentation assays on irradiated BIG cells to determine the effect of Bcl-2 expression on apoptosis. As expected, the BIG cell line was resistant to radiation-induced apoptosis (Fig. 6C).
These data are consistent with a model in which activation of the MEK/ERK and NFB1 pathways is upstream of Bcl-2 expression in LY-ar cells and that Bcl-2 expression is sufficient for resistance to radiation-induced apoptosis in this B-cell lymphoma cell system.
Treatment of LY-ar Cells with TNF␣ in the Presence of PD 98059 Suggests That the MEK/ERK Pathway, Not IKK, Is Sufficient for p50 Homodimer Activity-We previously demonstrated that TNF␣ treatment led to the nuclear translocation of the p50/p65 NFB dimer in LY-ar cells (23). In this case, TNF␣ did not affect the level of constitutive p50 homodimers, suggesting that TNF␣ activated p50/p65 via a separate pathway from that responsible for p50 homodimer activation. TNF␣ is the prototypal agent for inflammatory cytokine-induced activation of NFB through activation of IKK (7). Because p50 homodimers were sensitive to MEK inhibitors (Fig. 3, A-C) and unaffected by a known IKK activator, TNF␣, p50 homodimer activity in LY-ar cells appears IKK-independent. To further test this possibility, LY-ar cells were maintained in the presence of PD 98059 for 24 h, treated for an additional 1 h with TNF␣ while in the presence of PD 98059, and prepared for EMSA. PD 98059 inhibited p50 homodimer DNA binding in both PD 98059-treated LY-ar cells and TNF␣/PD 98059-treated LY-ar cells (Fig. 7). Similar results were obtained in TNF␣/PD 184352-treated LY-ar and TPA/PD 98059-treated LY-as cells (data not shown). These data indicate that MEK inhibition blocks p50 homodimer activity, even in the presence of a presumably activated IKK signal, suggesting that p50 homodimer activity in LY-ar cells is IKK-independent. It should be noted that TNF␣-induced p50/p65 heterodimer activity was partially abrogated by PD 98059 (Fig. 7), suggesting that, in addition to its control of p50 homodimers, the MEK/ERK pathway contributes to the activation of p50/p65 downstream of IKK in LY-ar cells.
MEK Inhibitors Sensitize LY-ar Cells to Radiation-induced Apoptosis-Because of the known changes to cell survival pathways caused by such treatments, we investigated whether MEK inhibitors could sensitize LY-ar cells to radiation-induced apoptosis. LY-ar cells were pretreated for 72 or 96 h with the MEK inhibitor PD 980589, U0126, or PD 184352, irradiated with 5 Gy ionizing radiation, and analyzed for apoptosis on the basis of DNA fragmentation 4 h later. It should be noted that within the 4 h time frame that apoptosis is induced, phosphorylated ERK1/ERK2 and Bcl-2 protein levels are not elevated by irradiation (data not shown). All three inhibitors sensitized normally apoptosis-resistant LY-ar cells to radiation-induced apoptosis by 72 h of treatment, reaching DNA fragmentation levels that were similar to those observed for the parent cell type, LY-as, by 96 h (Fig. 8). These results provide evidence that MEK inhibitors radiosensitize LY-ar cells by restoring apoptosis propensity. DISCUSSION In this study, we examined the role of the MEK/ERK signal transduction pathway in the activation of NFB1 (p50) homodimers and expression of Bcl-2 in two murine B-cell lymphoma cell lines. Using highly specific MEK inhibitors, we obtained evidence that activation of the MEK/ERK pathway is associated with p50 homodimer activity and Bcl-2 protein ex- Western blots were performed using 25 g of protein from whole-cell lysates, phospho-and total-p44/p42 (A) and Bcl-2 (B). pression in LY-as and LY-ar cells. Further evidence suggested that the MEK/ERK pathway and not IKK mediates p50 homodimer activation. Lastly, we demonstrated that MEK inhibitors sensitize LY-ar cells to radiation-induced apoptosis.
Our finding that the MEK/ERK pathway affects p50 homodimer DNA binding is, to our knowledge, the first indication of this activity. Although the MEK/ERK pathway has been implicated in the activation of other NFB dimer forms (8,20,21), the constitutive activity of p50 homodimers is considered to be the result of a rearrangement of p50-p105 heterodimers into p50 homodimers by the IB family member Bcl-3 (34). IL-9and granulocyte macrophage-colony stimulating factor-induced Bcl-3 protein expressions have been linked to enhanced p50 homodimer activity by a similar mechanism (35,36). The ability of Bcl-3 to interact with and control p50 DNA-binding activity appears to be phosphorylation-dependent (37,38). We have found no difference in the amount of Bcl-3 protein expression between LY-as and LY-ar cells by Western blot (data not shown), but it is possible that the MEK/ERK pathway controls Bcl-3 phosphorylation, thereby affecting p50 homodimer formation and/or DNA binding in LY-ar cells.
An alternative mechanism by which p50 homodimers become activated is through IKK activation. The p50 precursor, p105, has been shown to be a target for IKK phosphorylation during cytokine-induced responses that lead to the activation of both p50/p65 and p50/50 dimer forms (39). By a similar mechanism, the constitutive p50 homodimer signal in LY-ar cells may not be a result of MEK/ERK-induced Bcl-3 activity but rather a result of increased p105 processing directly or indirectly controlled by activation of the MEK/ERK pathway.
LY-as cells and LY-ar cells are characterized by profound differences in their apoptotic responses. Changes in NFB signaling and Bcl-2 expression are apparently responsible for those differences. Indeed, Bcl-2 expression alone can render LY-as cells resistant to radiation-induced apoptosis (Fig. 6C). The data presented here establish a role for the MEK/ERK pathway upstream of both NFB and Bcl-2 cell survival pathways, implicating the MEK/ERK pathway as a key mediator of apoptosis propensity for these lymphoma cells. These data therefore suggest that activation of the MEK/ERK pathway may be an excellent marker of disease progression, and our FIG. 6. LY-as cells expressing human Bcl-2 protein do not possess phosphorylated p44/p42 but are resistant to radiation-induced apoptosis. A, LY-as cells were infected with retroviral vectors to express GFP alone or Bcl-2 and GFP. Western blots were performed using 25 g of protein from wholecell lysates for total-and phospho-p44/ p42 as well as human and mouse Bcl-2. B, EMSA was performed with 32 P-labeled NFB oligonucleotide using 15 g of protein from nuclear extracts prepared from LY-ar, LY-as, GFP, and BIG cells as described. As a positive control for NFB binding, nuclear extracts were also prepared from GFP and BIG cells incubated in the presence of 50 ng/ml TPA for 5 h. C, cells were treated with 5 Gy of ionizing radiation; DNA fragmentation, as a measure of apoptosis, was determined 4 h later. Results are based on the average of at least three independent experiments Ϯ S.E. GFP, LY-as cells expressing green fluorescent protein alone. BIG, LY-as cells expressing both Bcl-2 and green fluorescent protein. data demonstrating that MEK inhibitors sensitize LY-ar cells to radiation-induced apoptosis may be relevant to other advanced cancers displaying aberrant MEK/ERK activity.
MEK inhibitors have been used by other groups to successfully sensitize cells to anti-cancer therapies. For example, PD 98059 and PD 184352 have been used to impair growth, abrogate clonogenicity, and sensitize acute myelogenous leukemia (AML) cells to chemotherapy-induced apoptosis (40,41). Similar to our observations, decreases in the expression of members of the Bcl-2 family were observed. In addition to AML, the MEK/ERK pathway has also been implicated in Bcl-2 family member expression and survival in pancreatic and breast cancer cells (42,43), and Sebolt-Leopold et al. (44) have demonstrated that PD 184352 administered either intraperitoneally or orally inhibited growth of colon tumors in vivo by as much as 80% with no signs of toxicity. Therefore, small molecule approaches to inhibiting MEK activity may be useful clinically in the treatment of a variety of tumor types. In fact, PD 184352, also known as CI-1040 (45), is currently in clinical trials for patients with advanced cancer.
Although previous studies have suggested that MEK inhibitors suppress the growth of various cancers and enhance the effectiveness of certain treatment modalities, some researchers have suggested that MEK inhibitors are not useful for sensitizing carcinoma cells to ionizing radiation (46 -48). However, Vrana et al. (49) have used PD 98059 in combination with ionizing radiation to treat HL-60 cells, which resulted in a large increase in apoptosis and a large decrease in clonogenicity when compared with either treatment alone. That study, in combination with the work presented here and recent findings from Shonai et al. (50) implicating the MEK/ERK pathway in radioresistance of lymphocytic leukemia cells, suggests that MEK inhibitors may be primarily useful for radiosensitizing hematopoietic cancers.
Our study defines a novel upstream signal to NFB1 activation, i.e. the MEK/ERK pathway; however, the upstream signals leading to MEK/ERK activation in LY-ar cells remain to be elucidated. We have performed a Western blot analysis to demonstrate that the M3K Tpl-2 is overexpressed in LY-ar cells as compared with LY-as cells (data not shown). Tpl-2 directly interacts with p105, and its overexpression leads to increased turnover of p105 into p50 subunits (51). Here, p50 homodimer activity appeared to be strictly under control of the MEK/ERK pathway. In addition to its interaction with p105, Tpl-2 has been shown to activate MEK (52). Therefore, Tpl-2 is a good candidate for the upstream signal leading to the sole activation of p50 homodimers in LY-ar cells.
It would be interesting to identify a distal upstream signal in LY-ar cells. In T-cells, Tpl-2 has been implicated in NFB signaling derived from cell surface interactions such as CD28 stimulation (13). It is possible that Tpl-2 has a similar function in LY-ar cells, and the identification of a receptor on LY-ar cells that may be engaged upstream of the observed changes in the MEK/ERK and NFB signaling pathways is an aim we are actively pursuing.
In summary, the MEK/ERK pathway is a subject receiving a lot of attention at the basic science level as well as in the clinic. The identification of signaling pathways associated with its activity will undoubtedly shed light on its role in cancer and will hopefully lead to novel strategies that reverse its effects.