T-Lymphokine–Activated Killer Cell–Originated Protein Kinase Functions as a Positive Regulator of c-Jun-NH₂-Kinase 1 Signaling and H-Ras–Induced Cell Transformation

Mitogen-activated protein kinases (MAPK), including extracellular signal–regulated kinases, c-Jun-NH₂-kinases (JNK), and p38 MAPKs, function as critical regulatory effectors in many cellular processes, including cell proliferation, migration, survival, and death (1–4). Particularly, JNKs, also known as stress-activated MAPKs, contain the dual phosphorylation motif Thr-Pro-Tyr (5), and have been shown to bind and phosphorylate c-Jun on Ser⁶³ and Ser⁷³ in response to UV irradiation (6, 7). JNKs are activated by many diverse stimuli such as proinflammatory cytokines [e.g., tumor necrosis factor (TNF) and interleukin 1], UV irradiation, and tumor promoters (8–13). Activated JNKs can phosphorylate numerous substrates, including several transcription factors such as activator protein-1 (AP-1), thereby affecting gene expression and subsequent physiologic cellular functions (14). The JNK signaling pathway consists of JNKs/stress-activated MAPKs, MAP2Ks (e.g., MKK4 or MKK7), and MAP3Ks, which include the mixed lineage kinase (MLK) 3 or MAPK kinase kinase 1. Some of these kinases form signaling complexes mediated by scaffold proteins, such as the JNK-interacting protein 1 (JIP1; refs. 15–17). Various scaffold proteins, including JIP1, were suggested to form complexes with MAPK signaling components, thereby creating a functional signaling module (18). JIP1 physically associates with JNKs, MKK7, and members of the MLK group of MAP3Ks functioning to stimulate JNK activity (16).

T-lymphokine–activated killer cell–originated protein kinase (TOPK), a MAPKK-like protein kinase, is a serine/threonine kinase that is highly expressed in many tumors such as leukemias, lymphomas, and myelomas (19–21). TOPK is also up-regulated in testicular tissue and is believed to participate in spermatogenesis (22). Structurally, TOPK comprises kinase subdomains, the conserved dual specificity active site sequence (D-X-K-X-X-N) spanning amino acids 174 to 179, and a COOH-terminal ETDV motif that binds PDZ domains, which might associate with the tumor suppressor protein hDlg (20). TOPK is known to be phosphorylated during mitosis and is a substrate of Cdc2/cyclin B (20, 23). Furthermore, p38 MAPK has also been reported to be a substrate of TOPK (19). More recently, TOPK was shown to phosphorylate histone H3 at Ser¹⁰ in vitro and in vivo, and TOPK kinase activity seems to be involved in its oncogenic function (24). However, the detailed signaling pathways and biological functions associated with TOPK, particularly in tumor cells, still remain to be determined.

Here, we identified TOPK as a novel upstream activator of JNKs. We showed that TOPK regulates JNK activity in response to UVB irradiation and further showed that TOPK has a major role in H-ras oncogene–induced cell transformation. We also provided evidence showing that TOPK could associate with and phosphorylate JNK1 and that TOPK regulation of JNK1 activity enhanced the ability of JNKs to mediate H-Ras–induced cell transformation.

Cells and reagents. Mouse epidermal skin (JB6 Cl41), human epithelial kidney (HEK293), human malignant melanoma (RPMI 7951), and mouse embryo fibroblasts (3T3) were purchased from American Type Culture Collection. JNK wild-type murine embryonic fibroblasts (MEF), JNK1 knockout (JNK1^−/−) MEFs, and JNK2 knockout (JNK2^−/−) MEFs were kind gifts from Dr. Roger J. Davis (University of Massachusetts Medical School, Worcester, MA). JB6 Cl41 cells and RPMI 7951 cells stably transfected with TOPK cDNA and TOPK small interfering RNA (siRNA), respectively, were selected with G418, identified, and established as described (12). Cells were cultured in DMEM or MEM supplemented with 5 or 10% fetal bovine serum, 2 mmol/L of l-glutamine, and 1% penicillin/streptomycin. Transfection of cells was done using jetPEI (Qbiogene) according to the instructions of the manufacturer. 12-O-Tetradecanoylphorbol-13-acetate (TPA) or TNFα was purchased from Sigma and the DNA ligation kit and glutathione-Sepharose 4B were from Takara Bio, Inc. and Amersham Biosciences, respectively. Antibodies against HA, V5, Myc, or FLAG epitopes were purchased from Santa Cruz Biotechnology, Inc., Invitrogen, BD Biosciences, Inc., or Sigma, respectively. Antibodies against c-Jun, phospho-c-Jun (Ser⁶³), JNK, phospho-JNKs (Thr¹⁸³/Tyr¹⁸⁵), and TOPK were from Cell Signaling Technology, Inc. Texas red– or FITC-conjugated IgG was purchased from Invitrogen and active TOPK or MKK7β1 was from Invitrogen or Upstate, respectively.

Plasmids. The plasmids for the expression of pcDNA3.1/V5-tagged JNK1, JNK2, or JNK3 and deletion mutants of V5-tagged JNK1or JNK3 were described previously (10). The plasmid pcDNA3/HA-TOPK encoding TOPK cDNA was a gift from Dr. Yasuhito Abe (Ehime University School of Medicine, Ehime, Japan). The glutathione S-transferase (GST)-TOPK construct was generated by cloning the PCR product into the BamHI/EcoRI sites of pGEX-5X1 (Amersham) and the cDNA insert was verified by sequencing (Genewiz, Inc.). Expression vectors for MKK7, JIP1, and MLK3 were kindly provided by Dr. Roger J. Davis, Dr. Lawrence B. Holzman (University of Michigan Medical School, Ann Arbor, MI), and Dr. James R. Woodgett (Ontario Cancer Institute, Toronto, Canada), respectively. Expression vectors encoding GST-c-Jun (1–79) and c-Jun dominant-negative TAM-67 were gifts from Dr. Lynn E. Heasley (University of Colorado Health Sciences Center, Denver, CO) and Dr. Michael J. Birrer (National Cancer Institute, Bethesda, MD), respectively.

In vitro GST pull down. The plasmids for expression of pcDNA3.1/V5-JNK1, 2, 3, or deletion mutants of JNK1 or JNK3 were translated in vitro with L-[³⁵S]methionine using the T_NT Quick coupled transcription/translation system (Promega). Briefly, GST-TOPK proteins immobilized on glutathione-Sepharose beads (Amersham Biosciences) were incubated at 4°C for 3 h with 300 μg of each cell lysate or ³⁵S-labeled translated JNK1, JNK2, JNK3, or deletion mutants. The bound proteins were washed extensively, boiled in sample buffer, and separated by SDS-PAGE and subsequently analyzed by immunoblotting with appropriate antibodies or by autoradiography.

Kinase assay. The in vitro kinase assays and GST-c-Jun pull-down kinase assays were done as described previously (12). Briefly, active TOPK or MKK7 was incubated at 30°C for 30 min with 1 μg of GST-JNK1 and 10 μCi of γ-³²p(ATP). The reaction mixture was separated with SDS-PAGE and exposed to X-ray film for autoradiography. The JNK kinase assays were done using c-Jun immobilized on glutathione-Sepharose beads. Each 300 μg of cell lysate was incubated at 4°C for 2 h with 20 μL of GST-c-Jun beads, washed extensively, and subjected to immunoblotting with phospho-c-Jun (Ser⁶³). For the TOPK kinase assay, HEK293 cells were transfected with control vector or pcDNA3/HA-TOPK. At 48 h after transfection, cells were irradiated with UVB (4 kJ/m²), and cell lysates were subjected to immunoprecipitation using protein G Sepharose beads with anti-HA followed by a kinase assay with GST-JNK1 as substrate.

Immunoblot analysis and immunoprecipitation. Immunoblotting or immunoprecipitation was done as described (25). To check the level of phosphorylation of JNKs in TOPK-expressing cells, TOPK siRNA cells, or their respective control cells after exposure to UVB (4 kJ/m²), immunoblotting was carried out using a phospho-JNK antibody recognizing Thr¹⁸³/Tyr¹⁸⁵ residues (Cell Signaling Technology).

Immunofluorescence assay. To determine the localization of endogenous JNK or TOPK in vivo, 70% confluent RPMI 7951 cells were starved in serum-free medium for 24 h, and then were or were not irradiated with UVB (4 kJ/m²). An immunofluorescence assay was done as described (12). Briefly, cells were fixed in 4% paraformaldehyde and incubated with anti–phospho-JNK and FITC-conjugated IgG or anti-TOPK and Texas red–conjugated IgG. The phosphorylated JNK or TOPK was stained for green or red fluorescence, respectively. Nuclei were stained with Hoechst and samples were analyzed with a fluorescence microscope system (Leica).

Reporter gene assays. For reporter gene assays, transient transfections were carried out using jetPEI (Qbiogene) and different combinations of each expression construct together with the AP-1 reporter gene. Cotransfection with 0.5 μg of the pRL-SV40 gene (Promega) for each transfection was done and Renilla luciferase activity was normalized for transfection efficiency. At 48 h after transfection, cells were harvested and luciferase activity was assessed using the Luminoskan Ascent (Thermo Electron, Corp.).

Focus-forming assay. A transformation assay of 3T3 Swiss cells was carried out according to standard protocols (26). Briefly, 3T3 Swiss cells were plated in six-well plates at a density of 2 × 10⁵ cells 24 h before transfection. Cells were transiently transfected with various combinations of plasmids, H-Ras^G12V (100 ng), pcDNA3.1/V5-tagged JNK1 (1 μg), pcDNA3/HA-TOPK (1 μg), or JNK1 siRNA (1 μg). At 48 h posttransfection, cells were transferred to 100 mm plates, and then cultured in DMEM with 5% bovine calf serum for 2 weeks. Foci were fixed with methanol, stained with 0.4% crystal violet, and then counted under a microscope.

TOPK associates with and phosphorylates JNK1. Signal transduction cascades related to TOPK still remain undiscovered. Here, we found that TOPK associated with JNK. To confirm that JNK directly interacted with TOPK, we cotransfected HEK293 cells with HA-TOPK and V5-JNK1, V5-JNK2, or V5-JNK3 constructs and then immunoprecipitation experiments were done. Results indicated that ectopically expressed JNK1 or JNK3, but not JNK2, bound to TOPK (Fig. 1A), indicating that JNK1 or JNK3 could associate with TOPK in vivo. We also carried out GST pull-down assays to further confirm the interaction of JNK and TOPK in vitro. GST or GST-TOPK, coupled to glutathione-Sepharose beads, was incubated with ³⁵S-methionine–labeled JNKs using an in vitro transcription and translation system (Promega). Consistent with the immunoprecipitation results, the GST pull-down assay showed that JNK1 or JNK3 bound to TOPK in vitro, supporting a direct interaction between TOPK and JNK (Fig. 1B,, top). To identify the domains that are required for the association of the two kinases, a GST pull-down assay was done, using constructs of several deletion mutants of JNK1 or JNK3. Results revealed that fragments containing the glycine-rich region of JNK1 or JNK3 bound to TOPK, implying that this portion of the NH₂-terminal region is important for binding of JNK1 (Fig. 1B,, middle) or JNK3 (Fig. 1B,, bottom) to TOPK. We next investigated whether the MAPKK-like TOPK could phosphorylate JNK. A kinase assay using GST-JNK1 revealed that active TOPK phosphorylated JNK1 in vitro (Fig. 1C,, top, lane 4). Active MKK7 phosphorylation of JNK1 was used as a positive control (Fig. 1C,, top, lane 5). To further verify that this TOPK-mediated JNK1 phosphorylation occurred in vivo, we examined the possibility that UVB irradiation (4 kJ/m²) elicits phosphorylation of TOPK and subsequent JNK1 phosphorylation. Results revealed that UVB irradiation induced the phosphorylation of serine and threonine (Ser/Thr) residues on TOPK (Fig. 1C,, middle), and that activated TOPK phosphorylated JNK1 (Fig. 1C,, bottom). We then examined the effect of the TOPK-JNK1 association on JNK1 activity using GST-c-Jun as substrate and found that JNK1 activity was enhanced by the presence of the recombinant His-tagged TOPK protein (data not shown). These results suggested that the TOPK interaction with JNK1 might stimulate JNK1 activity to phosphorylate c-Jun. To next determine the cellular location of phosphorylated JNK1 and TOPK, immunofluorescence staining was done and results showed that following UVB, phosphorylated JNK1 was translocated into the nucleus where it colocalized with TOPK (Fig. 1D). Collectively, these data indicated that TOPK associates with and phosphorylates JNK1 in vitro and in vivo.

TOPK participates in UVB-induced JNK activation. We next investigated whether TOPK was involved in TPA-, UV- or TNFα-induced JNK activation. For this purpose, HEK293 cells were transiently transfected with HA-TOPK, and then were treated with TPA, UVB, or TNFα. Results of GST-c-Jun pull-down kinase assays showed that TOPK expression resulted in a strong increase in JNK activity to phosphorylate c-Jun in response to UVB irradiation (4 kJ/m²; Fig. 2A). Also, TNFα treatment of cells transfected with TOPK exhibited an increase in JNK activity. However, TOPK expression did not affect TPA-induced JNK activity, indicating that TOPK might contribute specifically to JNK activation signaling pathways depending on the stimulus. We also confirmed that TOPK expression enhanced UVB-induced JNK activity in JB6 Cl41 cells (Fig. 2B). In another experiment, epidermal JB6 Cl41 cells, which have almost no TOPK endogenous expression, and malignant melanoma RPMI 7951 cells, which express high levels of TOPK, were stably transfected with TOPK cDNA or TOPK siRNA, respectively. The results of a kinase assay revealed that the TOPK-expressing JB6 cells exhibited dramatically increased UVB-induced JNK activity (Fig. 2C), whereas knockdown of TOPK by siRNA in RPMI cells resulted in substantially decreased activity (Fig. 2D). These data suggested that TOPK contributes to JNK activation and phosphorylation of c-Jun in response to UVB.

TOPK-mediated JNK1 activation induced by UVB requires phosphorylation of Thr¹⁸³/Tyr¹⁸⁵ on JNK1. We next determined whether TOPK regulated JNK activity through its phosphorylation of the Thr¹⁸³/Tyr¹⁸⁵ residues of JNK, a posttranslational modification that is required for JNK activation in response to UVB irradiation. Immunoblot analysis revealed that UVB-induced phosphorylation of Thr¹⁸³/Tyr¹⁸⁵ was increased by ectopic expression of TOPK (Fig. 3A). In addition, JNK phosphorylation was also increased in cells stably expressing TOPK compared with cells expressing empty vector (Fig. 3B), whereas the phosphorylation was markedly impaired in TOPK knockdown cells (Fig. 3C). Interestingly, phosphorylation of c-Jun induced by UVB irradiation was augmented in TOPK-expressing cells (Fig. 3B), suggesting that TOPK-mediated UVB-induced JNK activation culminates in the phosphorylation of c-Jun. Taken together, these data indicated that TOPK plays a key role in UVB-induced JNK activation and phosphorylation, and c-Jun phosphorylation.

TOPK associates with JIP1 and modulates JIP1 scaffolding activity. JIP1 was shown to be closely involved in the regulation of JNK and p38 kinase activities, and especially JNKs (27). We thus investigated whether TOPK regulation of JNKs might also affect JIP1 scaffolding activity. HEK293 cells were transfected with HA-TOPK and Myc-JIP1 constructs and an immunoprecipitation experiment was done. Results revealed that TOPK associated with JIP1 in vivo (Fig. 4A). To further confirm the interaction of TOPK and JIP1, total lysates from cells transfected with Myc-JIP1 were incubated with GST or GST-TOPK. Data showed that TOPK directly interacted with JIP1 in vitro, indicating that a TOPK association with JIP1 could be implicated in JIP1 function (Fig. 4B). To assess whether TOPK regulates JIP1-mediated JNK activity, we next examined the effect of TOPK on JIP1 scaffolding activity. JIP1 has been shown to physically interact with JNKs, MKK7, and members of the MLK group of MAP3Ks, resulting in an enhancement of the JNK signaling pathway (17, 27). As expected, coexpression of JIP1 together with MLK3, MKK7, and JNK1 increased JIP1-mediated JNK1 activity (Fig. 4C). However, JIP1 scaffolding activity was also substantially elevated with the coexpression of TOPK together with JIP1, MLK3, MKK7, and JNK1. Furthermore, immunoblot analysis using anti–phospho-JNK1/2 revealed that coexpression of TOPK together with MLK3, MKK7, JNK1, and JIP1 strongly increased JNK phosphorylation (Fig. 4D), implying that the interaction between TOPK and MLK3 or MKK7 might be critical for JIP1 scaffolding activity. These data suggested that the interaction of TOPK with JIP1 might play a key role in the regulation of JNK activity.

UVB-induced AP-1 activity is mediated by the TOPK-JNK1-c-Jun signaling cascade. UV is suggested to induce JNK activation that results in the phosphorylation of c-Jun, which plays a pivotal role in UV-induced AP-1 activation (28). To explore the role of TOPK in UVB-induced AP-1 activity, we next assessed the UVB-induced AP-1 promoter–driven luciferase activity in JB6 Cl41 cells expressing TOPK or RPMI 7951 cells expressing TOPK siRNA. The UVB-induced AP-1 transcriptional activity was increased by ∼2-fold in cells expressing TOPK compared with the level of increase of control cells expressing an empty vector (Fig. 5A), whereas the activity was diminished in TOPK siRNA cells compared with control siRNA cells (Fig. 5B). We next determined the effect of the TOPK-JNK1 association on UVB-induced AP-1 activity. Coexpression of TOPK and JNK1 increased the UVB-induced JNK1-c-Jun–mediated transcriptional activity of AP-1. In contrast, expression of a dominant-negative mutant of c-Jun, TAM-67, strongly attenuated AP-1 activity (Fig. 5C). These results indicated that TOPK is implicated in the UVB-induced JNK1-c-Jun–dependent signaling pathway leading to AP-1 activation.

JNK is required for TOPK-mediated H-Ras–induced cell transformation. The oncogenic H-ras is known to enhance the ability of JNKs to phosphorylate c-Jun and to influence cell transformation and tumorigenesis (29, 30). We therefore investigated whether TOPK was involved in JNK1-mediated AP-1 activity induced by H-Ras. AP-1 transcriptional activities were evaluated in several cell lines. As expected, expression of H-Ras^G12V increased AP-1 activity, and coexpression of JNK1 or TOPK together with H-Ras^G12V resulted in a greater increase in AP-1 activity (Fig. 6A). Moreover, AP-1 transcriptional activity was even more augmented by coexpression of H-Ras^G12V, JNK1, and TOPK, suggesting that TOPK and JNK1 function as mediators in H-Ras^G12V signaling leading to AP-1 activation. The role of TOPKs' association with JNK1 in AP-1 activation was next examined in TOPK-expressing cells or JNK1 knockout (JNK1^−/−) MEFs. Results indicated that H-Ras^G12V–induced AP-1 activity was impaired in JNK1^−/− MEFs compared with wild-type MEFs (Fig. 6B). More importantly, coexpression of H-Ras^G12V and TOPK did not greatly augment AP-1 activity in JNK1^−/− MEFs compared with wild-type MEFs. In contrast, AP-1 activity in JNK2^−/− MEFs was similarly increased by the expression of H-Ras^G12V and TOPK compared with wild-type MEFs, indicating that JNK1, but not JNK2, mediated the H-Ras^G12V-TOPK signaling pathway leading to AP-1. Similarly, H-Ras^G12V–induced AP-1 activity was greatly increased in TOPK-expressing cells, compared with control cells (Fig. 6C). But again, coexpression of H-Ras^G12V and JNK1 caused an even greater augmentation in the activity, indicating that TOPK also mediated the H-Ras^G12V signaling pathway associated with AP-1 activation. Taken together, these results suggested that TOPK and JNK1 were critical effectors for H-Ras^G12V signaling to AP-1. Oncogenic H-ras has been shown to strongly augment JNK activity and c-Jun phosphorylation, suggesting that JNKs act as important mediators of tumorigenesis (10). We hypothesized that regulation of JNK activity by TOPK might affect the ability of JNKs to mediate H-Ras–induced cell transformation. Thus, we transfected 3T3 cells with H-Ras^G12V and/or various combinations of TOPK and JNK1, and then subjected these cells to a focus-forming assay. Results indicated that transformed foci were increased in cells transfected with H-Ras^G12V compared with mock-transfected cells (Fig. 6D,, top, plate 1 versus 2). As expected, coexpression of JNK1 together with H-Ras^G12V caused a further increase in foci formation (Fig. 6D,, top, plate 2 versus 3). Interestingly, the number of foci was substantially augmented by coexpression of TOPK and H-Ras^G12V (Fig. 6D,, top, plate 2 versus 4); and furthermore, coexpression of TOPK and JNK1 with H-Ras^G12V elicited even more foci (Fig. 6D,, middle, plate 1 versus top, plates 2, 3, and 4). Importantly, the number of transformed foci induced by H-Ras^G12V and TOPK or JNK1 was substantially diminished in cells cotransfected with combinations of H-Ras^G12V, TOPK siRNA, or JNK1 siRNA (Fig. 6D , middle, plates 2, 3, and 4). Collectively, these results suggested that TOPK might function as a key effector in H-Ras signaling, leading to cell transformation and tumorigenesis, and that JNK1 is an important component of H-Ras-TOPK signaling to cell transformation.

TOPK, a serine/threonine MAP2K, has been suggested to be activated in response to TPA, and Cdc2/cyclin B was reported to phosphorylate Thr⁹ on TOPK in vitro (19, 20). Also, p38 MAPK and c-Myc have been considered to be potential substrates of TOPK (19, 31). A possible association between Raf and TOPK was also found by yeast two-hybrid screening analysis (32). These previous studies implied that TOPK might be closely involved in the MAPK signaling pathways. However, unequivocal signal transduction cascades involving activators and downstream targets of TOPK remain elusive. On the other hand, the upstream effectors of the JNK signaling pathways have been suggested to include Rac-Cdc42, MAPK kinase kinase, and MKK4/MKK7 (33). Recently, another upstream signaling component, protein kinase C, has also been implicated in the regulation of JNKs (34, 35). Here, we reported that TOPK functions as an upstream regulator of UVB-induced JNK activation, and showed that TOPK mediates H-Ras–induced cell transformation. These data identify TOPK as a novel kinase for regulating JNK activity with a possible role in oncogenesis. Our findings also suggested that the cross-talk between TOPK and JIP1 might play a key role in regulating JNK activity.

Of interest is that TOPK specifically regulated the JNK signaling pathway leading to the phosphorylation of c-Jun at Ser⁶³ and Ser⁷³ in response to UVB or TNFα, but not TPA (5, 36, 37). The upstream kinases or effectors of JNK activation signaling cascades can be different based on the specific physiologic stimuli (2, 14, 28). Thus, these findings suggested that TOPK might function as a mediator of JNK1 signaling in a stimuli-dependent manner. Moreover, results of experiments with TOPK-overexpressing or knockdown cells supported the notion that TOPK might contribute to UVB-mediated JNK signaling cascades. These results implied that TOPK might be considered as a key mediator regulating downstream MAPK signaling depending on specific stimuli, especially UVB.

The JIP1 scaffolding complex is believed to consist of JNK, MKK7, and members of the MLK group of MA3PKs (16). Previous studies indicated that the JIP1 scaffolding complex enhanced JNK activation and JIP1 deficiency impaired stress-induced JNK activation in hippocampal neurons (16, 17). The results of the present study indicated that TOPK associated with JIP1 and thus seemed to enhance the ability of JIP1 scaffolding to augment JNK activity. These findings suggested that TOPK might, at least in part, contribute to JNK signaling through its interaction with JIP1. Thus, although TOPK directly interacted with, and phosphorylated JNK1, the TOPK association with the JIP1-dependent scaffolding process seemed to be required for the full activation of JNK1. Intriguingly, JIP1 scaffolding activity was enhanced when TOPK was coexpressed with MLK3 or MKK7. In addition, TOPK did not seem to compete with MLK3, MKK7, or JNK1 for JIP1 binding (data not shown), suggesting that TOPK might synergistically function with MLK3 or MKK7 as one component of the JIP1 scaffolding complex. Given the role of MLK3 or MKK7 in the activation of JNKs, these results provided important evidence for the interplay between TOPK and MLK3 or MKK7 in the JNK activation pathway. On the other hand, JNKs might be activated by alternative signaling pathways independently of JIP1 scaffolding (38–40). Therefore, we cannot exclude the possibility that TOPK might be involved in JIP1 scaffold-independent signaling pathways leading to JNK activation.

Skin cancer is known to be one of the most common human cancers, and solar UV irradiation such as UVA and UVB has been suggested to function as an important carcinogen in the development of skin cancers, including nonmelanoma skin cancers, squamous cell carcinomas, and cutaneous malignant melanomas (41, 42). TOPK has been shown to be highly expressed in proliferating malignant cell lines including melanomas, and was suggested to be a molecular marker of breast cancer (21, 24). The H-ras oncogene has been proposed to strongly increase JNK activity and c-Jun phosphorylation (10, 29) and AP-1 components, particularly c-Jun, which play a key role in H-Ras–induced cell proliferation and transformation (43). These findings, showing that TOPK acts as a key effector in UVB-H-Ras-JNK signaling, might provide a mechanistic basis for a detailed role of TOPK in tumorigenesis. Notably, TOPK apparently mediated H-Ras-JNK signaling and the UVB-induced JNK activation pathway in mouse epidermal JB6 Cl41 cells or human RPMI 7951 melanoma cells, suggesting a role for TOPK in skin cancers, a finding that agrees with a previous report that TOPK might have an oncogenic effect in breast cancers (24).

Based on these observations, we propose that TOPK is a novel effector in oncogenic signaling and might be especially implicated in UVB-Ras-JNK–mediated tumorigenesis leading to skin cancers. Taken together, we conclude that the TOPK-JNK interaction is crucial for their H-Ras–mediated oncogenic potential and suggest that TOPK might be a potential molecular and therapeutic target for skin cancers.

Grant support: Hormel Foundation and NIH grants CA77646, CA81064, CA88961, CA27502, and CA111356.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.