Cancer Research Meeting Calendar EMT and Cancer Progression and Treatment
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
Cancer Research 67, 178, January 1, 2007. doi: 10.1158/0008-5472.CAN-06-2801
© 2007 American Association for Cancer Research
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nielsen, C.
Right arrow Articles by Kallunki, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nielsen, C.
Right arrow Articles by Kallunki, T.

Cell, Tumor, and Stem Cell Biology

c-Jun NH2-Terminal Kinase 2 Is Required for Ras Transformation Independently of Activator Protein 1

Christina Nielsen, Jacob Thastrup, Trine Bøttzauw, Marja Jäättelä and Tuula Kallunki

Apoptosis Department and Centre for Genotoxic Stress, Institute of Cancer Biology, Danish Cancer Society, Copenhagen, Denmark

Requests for reprints: Tuula Kallunki, Apoptosis Department and Centre for Genotoxic Stress, Institute of Cancer Biology, Danish Cancer Society, Strandboulevarden 49, DK-2100 Copenhagen, Denmark. Phone: 45-3525-7345; Fax: 45-3525-7721; E-mail: tk{at}cancer.dk.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Active Ras oncogene is expressed in ~30% of human cancers. Yet, very little is known about the molecular mechanisms responsible for its transforming potential. Here, we show that H-Ras-mediated transformation requires isoform 2 of the c-Jun-NH2-terminal kinase (JNK). H-Ras-transduced JNK2-deficient (Jnk2–/–) murine embryonic fibroblasts (MEFs) were severely inhibited in colony formation and growth in soft agar in vitro as well as in tumor formation in immunodeficient mice as compared with corresponding Jnk1–/– and wild-type MEFs. Accordingly, the RNA interference–based depletion of JNK2 form wild-type MEFs also resulted in defective Ras transformation. The extra barrier against H-Ras transformation in Jnk2–/– MEFs was not due to their inability to inactivate p53 signaling because all JNK2-deficient MEF lines had lost p19Arf. Furthermore, expression of the E6 protein of the human papilloma virus failed to overcome the transformation defect. It could, however, be overcome by coexpression of H-Ras with the SV40 large T antigen or c-Myc. Surprisingly, the H-Ras-transduced JNK2-deficient MEFs exhibited higher activity of activator protein-1 and higher levels of c-Jun expression compared with H-Ras-transduced JNK1-deficient or wild-type cells, indicating that the key target of JNK2 during Ras transformation was divergent from activator protein-1. These results clearly show that a single kinase, JNK2, could control Ras transformation and thus point out a vulnerable control point that may prove important for the tumor development in general. [Cancer Res 2007;67(1):178–85]


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
The c-Jun NH2-terminal kinase (JNK) belongs to a family of mitogen-activated protein kinases (MAPK), together with the p38 and extracellular regulated kinases (ERK). The JNK subfamily consists of three isoforms: JNK1, JNK2, and JNK3, which can altogether give rise to 10 splice variants. The splice variants differ in their ability to bind substrates like c-Jun (1) and ATF-2, and in the length of their COOH terminus (2, 3). JNK has been shown to be involved in various very different cellular processes. The data on the functional differences of the specific JNK isoforms or splice variants is just beginning to emerge. Further complexity to JNK signaling is provided by a plethora of JNK upstream kinases and scaffold proteins (4, 5).

Ras is a small membrane-bound monomeric G protein that transmits and links membrane receptor–mediated extracellular signals to the intracellular signaling network. The Ras family consists of three members: H-Ras, K-Ras, and N-Ras. Mutations resulting in the constitutive activation of all three Ras family members have been detected. Constitutively active mutated Ras oncogenes can be found in 30% of malignant human tumors, which makes it the most common oncogene involved in human cancer (6, 7). Ras can activate various cellular signaling pathways that promote cell survival and proliferation including Raf, ERK, phosphoinositide-3-kinase (PI3K)–protein kinase B, JNK, and nuclear factor {kappa}B pathways. Abrogation of Ras activity by dominant negative effector molecules or Ras pathway inhibitors results in decreased cell survival and proliferation capacity, showing that these cancer-promoting effects of Ras could be reversed (8–11). Thus, various strategies to inactivate Ras are currently under investigation for their effectiveness in anticancer therapy.

Very little is known of the actual molecular mechanisms by which Ras transforms cells and what the central mediators for this process are. It has been suggested that the growth-stimulating and -transforming effect of Ras could be largely mediated via c-Jun/activator protein-1 (AP-1), which is also a central target of JNK signaling (12). c-Jun is one of the components of the transcription factor AP-1. Supportively, active oncogenic Ras can promote c-Jun phosphorylation and AP-1 activation, whereas c-Jun knockout murine embryonic fibroblasts (MEFs) are completely deficient in Ras transformation (13–15). Interestingly, MEFs carrying c-Jun alleles with major JNK phosphorylation sites (serines 63 and 73) mutated to alanines (JunAA) exhibit only ~50% reduction in Ras-induced tumorigenesis in vivo as compared with the corresponding wild-type MEFs (16), suggesting that the c-Jun phosphorylation by JNK is not a requirement for Ras transformation. Another recent study has shown that Jnk-null MEFs exhibit ~50% reduction in Ras-mediated transformation in vitro as compared with wild-type MEFs, but are surprisingly clearly more tumorigenic than wild-type cells in vivo (17). These data point to a rather complicated role for the JNK signaling in Ras transformation.

In order to thoroughly explore the role of JNK and its isoforms in Ras-mediated transformation and tumorigenesis, we used genetic inactivation and RNA interference technology (RNAi). Interestingly, we observed that the Jnk2–/– MEFs were markedly impaired in their ability to be transformed by H-Ras as compared with the corresponding wild-type and Jnk1–/– cells both in vitro and in vivo. Surprisingly, the activation of the transcription factor AP-1, which is the most conventional Ras and JNK signaling target, was not impaired in Jnk2–/– MEFs, revealing the existence of a transformation supporting JNK2 downstream signaling component divergent from AP-1. The H-Ras-expressing Jnk2–/– MEFs could only be transformed by coexpression of either SV40 large T antigen (SV40-LT) or c-Myc, the combination of which is known to be needed for the transformation of primary MEFs, suggesting that the immortalized MEFs lacking JNK2 have gained an extra barrier to protect them from Ras-induced tumorigenicity.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cell culture and retroviral gene transfer. C57BL/6 mice heterozygous for Jnk1 and Jnk2 and their corresponding mutated alleles were crossed (18, 19). Embryos were harvested at E12 or E14, fibrous tissues were collected, trypsinized, and cultured in DMEM as described previously (20). Wild-type, Jnk1–/–, and Jnk2–/– fibroblasts were further passaged and allowed to immortalize spontaneously according to the NIH3T3 protocol. The PhoenixEco packaging cells were cultured in DMEM supplemented with 10% FCS, 100 µg/mL of streptomycin, and 100 units/mL of penicillin. The retroviral gene transfer was done as described previously (20).

Plasmid construction. The viral H-Ras was cloned into retroviral pBabe (pB)-Puro or pB-Hygro expression vectors at the BamHI site. The JNK2 short hairpin RNA constructs, J2sh-1 (GATCCCCGTGAACTCGTCCTCTTAAATTCAAGAGATTTAAGAGGACGAGTTCACTTTTTGGAAA/AGCTTTTCCAAAAAGTGAACTCGTCCTCTTAAATCTCTTGAATTTAAGAGGACGAGTTCACGGG), J2sh-2 (GATCCCCGTACCCTGGAATCAAGTTTTTCAAGAGAAAACTTGATTCCAGGGTACTTTTTGGAAA/AGCTTTTCCAAAAAGTACCCTGGAATCAAGTTTTCTCTTGAA AAACTTGATTCCAGGGTACGGG), and J2sh-3 (GATCCCCGTGATGGACTGGGAAGAAATTCAAGAGATTTCTTCCCAGTCCATCACTTTTTGGAAA/AGCTTTTCCAAAAAGTGATGGACTGGGAAGAAATCTCTTGAATTTCTTCCCAGTCCATCACGGG) were annealed and ligated to the HindIII-BglII linearized pRetro-puro vector using the Rapid ligation kit (Roche, Basel, Switzerland). Four mouse JNK2 splice variants J2{alpha}1, J2{alpha}2, J2ß1, and J2ß2 in puromycin-resistant pB vectors (20) were mutated in positions 2, 8, 14, and 17 according to the binding site for J2sh-1. Mutations were introduced two at a time using the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA).

In vitro tumorigenicity assays. For focus formation assays, 200 or 500 cells of the retrovirally transduced MEFs were plated with 8.3 x 105 NIH3T3 as feeder cells per well in a six-well plate. The cells were grown for 1 to 2 weeks and the medium was changed every 3 to 4 days. The plates were washed with PBS and fixed in methanol for 10 min. The fixed cells were then stained with prefiltered 10% Giemsa for 15 min. Ras-transduced wild-type and Jnk1–/– cells were stained after 7 to 10 days and Jnk2–/– and mock controls were stained after a minimum of 14 days. Foci numbers were calculated per 200 infected cells.

For soft agar assay 1.5 mL of 0.5% agar (electrograde ultra pure; Invitrogen, Carlsbad, CA) supplemented with DMEM were plated in six-well plates. Fifteen thousand cells were mixed with 1.5 mL of 0.35% agar supplemented with DMEM and plated on the solidified bottom agar. Media was added on top of the agar layers and the colonies were allowed to grow for 2 to 3 weeks. The colonies were quantified using a light microscope at 10x magnification.

In vivo tumorigenicity assay. Five million H-Ras-transduced MEFs (in 100 µL of PBS) were s.c. inoculated into the right flank of female BALB/c nude mice. The tumor size was measured each 3rd day in the beginning, and later when the tumors started to grow, more frequently. The mice were sacrificed when the tumor size reached the maximum tolerated size with a diameter of 12 mm, or alternatively, after 3 months if no tumors had developed. The tumor diameter was measured using a caliper.

Western blot analysis. The cell lysates and Western blots were prepared as described before (20) except for the c-Jun, Fra-1, and phospho-c-Jun blot, for which the cells were harvested in 2x Laemmli sample buffer and sonicated for 10 s in ice prior to SDS-PAGE separation.

Mouse monoclonal antibody Jnk (666) was purchased from PharMingen (San Diego, CA) and anti {alpha}-tubulin was from Sigma (St. Louis, MO). p53 antibody was purchased from Oncogene (Cambridge, MA) and p19Arf antibody ab-80 was from Abcam (Cambridge, United Kingdom). Ras antibody (MIB-39) was a kind gift from Berthe Willumsen (Department of Molecular Biology, University of Copenhagen, Denmark). c-Jun H-79 and Fra-1 R-20 antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA) and the phospho-c-Jun (Ser63) antibody was purchased from Cell Signaling Technology (Danvers, CA).

Cell growth assay. Cell density was evaluated with the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide reduction assay. The cells were plated at a density of 1,000 cells/well in a 96-well plate 24 h prior to the first time point. Cells were analyzed at the indicated time points. The assay was done as described previously (21), and all time points were assayed as triplicates.

Transient transfection and reporter assays. MEFs were cotransfected with pRL-null renilla luciferase (Promega, Madison, WI) and 4xAP-1 firefly luciferase plasmid using the FuGeneHD transfection reagent (Roche). Cells were lysed 48 h after transfection and the luciferase activity was measured according to the manufacturer's protocol (Dual-Luciferase Assay System; Promega).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Oncogenic Ras signaling requires the JNK2 isoform of JNK. To investigate the role of the JNK isoforms in Ras transformation, MEFs were isolated from several individual wild-type, Jnk1–/–, and Jnk2–/– embryos and immortalized with the NIH3T3 protocol. Immortalized cells were subsequently subjected to an additional 30 passages and transduced with retroviral pB-H-Ras expression construct. Because JNK3 expression was undetectable in MEFs (22),1 the Jnk1–/– and Jnk2–/– MEFs represented single JNK isoform–expressing cell lines. Interestingly, all three different Jnk2–/– MEFs analyzed were severely deficient in multilayer growth as compared with the Jnk1–/– and wild-type MEFs (Fig. 1A, top ), although they expressed similar amounts of H-Ras (Fig. 1A, middle and bottom). In spite of their clearly distinct behavior, JNK1- and JNK2-deficient MEFs expressed practically equal amounts of the remaining JNK isoform (Fig. 1A, middle). The slower migrating Ras form is due to the inactivating Thr59 phosphorylation of Ras (23). The growth of the H-Ras-expressing Jnk2–/– cells was considerably slower than the growth of the corresponding Jnk1–/– and wild-type cells (Fig. 1B). Interestingly, it was previously reported that the Jnk2–/– primary MEFs grow faster than the corresponding wild-type and Jnk1–/– MEFs (24). This is exactly what we have also observed with the primary Jnk2–/– MEFs.2 Ras-transduced Jnk2–/– MEFs were also deficient in their ability to grow both anchorage-independently in soft agar and anchorage-dependently in culture dishes (Fig. 1C and D). Whereas the H-Ras-transduced Jnk1–/– MEFs exhibited thin, elongated, and spindle-like morphology and loss of contact inhibition characteristic of transformed fibroblasts, the H-Ras-transduced Jnk2–/– MEFs retained the nontransformed morphology comparable to that of the vector-transduced control cells and were not capable of multilayer growth when seeded at a higher density (Fig. 1D). Accordingly, Jnk2–/– MEFs exhibited severely reduced tumorigenic potential when injected s.c. into BALB/c nude mice. Although the wild-type and the Jnk1–/– MEFs produced tumors with a 12 mm diameter at an average of 13.1 and 10.9 days, respectively, one of the H-Ras-transduced Jnk2–/– MEF lines (J2-2, n = 4) did not form any measurable tumors during the 3-month observation period, and the other Jnk2–/– MEF line (J2-1, n = 4) took 26.3 days to form 12 mm tumors (Table 1 ). Taken together, these data show that H-Ras failed to transform the Jnk2–/– MEFs.


Figure 1
View larger version (25K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Oncogenic Ras signaling requires JNK2. A, two wild-type, two Jnk1–/–, and three Jnk2–/– MEF lines (passage 30) transduced with H-Ras were tested for their ability to form multilayer foci. Columns, means from experiments done in triplicate; bars, SD. Experiments were repeated thrice with similar results (top). Western blot analysis of H-Ras and JNK expression in indicated MEF lines. {alpha}-Tubulin was used as a loading control (middle and bottom). B, the growth of the indicated H-Ras MEF lines was evaluated by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide reduction assay. Points, means from one experiment done in triplicate; bars, SD. The experiment was repeated twice with similar results. C, the ability of the indicated H-Ras-transduced MEF lines to grow anchorage-independently is shown as the number of colonies which grew in soft agar. Columns, means from one experiment done in triplicate; bars, SD. The experiment was repeated twice with similar results. D, representative phase contrast images of the indicated vector-transduced or H-Ras-transduced MEF lines.

 

View this table:
[in this window]
[in a new window]

  		Table 1. H-Ras expressing Jnk2–/– MEFs exhibit reduced tumorigenicity in vivo compared with the wild-type and Jnk1–/– cells

 
Down-regulation of JNK2 by RNAi in wild-type MEFs impairs the capability of Ras to transform them. In order to study if the down-regulation of JNK2 in the wild-type MEFs would have similar effects to the H-Ras-induced transformation as the genetic inactivation, we prepared three retroviral short hairpin constructs against JNK2 (J2-sh1, J2-sh2, and J2-sh3) and tested their ability to down-regulate JNK2 protein levels. Both J2-sh1 and J2-sh3 clearly down-regulated the 55 kDa form of JNK in the wild-type cells (Fig. 2A, left blot ), which is the major JNK2 form in the MEFs (see Fig. 1A, middle). The expression of these constructs in Jnk1–/– cells further showed that they were very efficient in down-regulating JNK2 (Fig. 2A, middle blot). Furthermore, the two most efficient hairpins (J2-sh1 and J2-sh3) had no detectable effect on the expression of the JNK1, as shown in the Jnk2–/– cells (Fig. 2A, right blot). To test the ability of these hairpins to reduce the tumorigenicity of the wild-type MEFs, they were first used to down-regulate JNK2 in the wild-type cells, after which the cells were transduced with H-Ras. Down-regulation of JNK2 prior to H-Ras transduction clearly reduced the tumorigenicity of wild-type cells (Fig. 2B). The ability of the JNK2 down-regulation to affect focus formation of the H-Ras-transformed wild-type cells was also tested. Down-regulation of JNK2 after Ras transformation did not have any measurable effects on the tumorigenicity of the wild-type MEFs (Fig. 2C). Coexpression of all four J2-sh1 recognition site–mutated JNK2 splicing variants in the J2-sh1 and H-Ras-transduced wild-type cells rescued the transformation-deficient phenotype (Fig. 2D). The expression levels of the mutated JNK2 splicing variants vary somewhat, probably due to the fact that ectopic Ras could up-regulate the expression of the retroviral LTR promoter. The ectopic expression of the four JNK2 splicing variants in Jnk2–/– MEFs failed, however, to augment the H-Ras-mediated transformation (data not shown). Overall, we can conclude from the RNAi data that the inactivation of the JNK2 has an effect on the tumorigenicity when it occurs prior to Ras transformation, further supporting the role of JNK2 in the early phase of Ras transformation.


Figure 2
View larger version (21K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Down-regulation of JNK2 by RNAi prior to H-Ras transformation reduces the tumorigenicity of the wild-type MEFs. A, the specificity of three prepared short hairpin RNAs against JNK2 (J2sh-1, J2sh-2, and J2sh-3) was evaluated by Western blotting. JNK expression was investigated in wild-type, Jnk1–/–, and Jnk2–/– MEFs stably expressing J2sh-1, J2sh-2, and J2sh-3. {alpha}-Tubulin was used as a loading control. B, focus formation assay of wild-type MEFs from JNK2 down-regulated by J2sh-1 and J2sh-3 and subsequently transformed with H-Ras (left). Columns, means from one experiment done in triplicate; bars, SD (***, P < 0.001). The experiment was repeated twice with similar results. Western blot analysis was used to evaluate the expression of JNK and H-Ras in wild-type MEFs infected with J2sh-1 and J2sh-3 and with H-Ras. {alpha}-Tubulin was used as a loading control. C, focus formation assay of wild-type MEFs infected with H-Ras and subsequently with J2sh-1 or J2sh-3. Columns, means from one experiment done in triplicate; bars, SD. The experiment was repeated twice with similar results. Expression of JNK and Ras in H-Ras-transformed wild-type MEFs infected with J2sh-1 and J2sh-3 was evaluated by Western blot analysis. {alpha}-Tubulin was used as a loading control. D, focus formation assay of wild-type MEFs infected with a pool of all four mutated Jnk2 splice variants (mut-Jnk2) or vector control before infection with J2sh-1 and subsequently transformed with H-Ras (left). Columns, means from one experiment done in triplicate; bars, SD. The experiment was repeated twice with similar results. The JNK and H-Ras expression was evaluated by Western blot. {alpha}-Tubulin was used as a loading control (right).

 
Deficiency for Ras transformation in Jnk2–/– MEFs is not due to the lack of Ras-responsive c-Jun phosphorylation or AP-1 activation. Because c-Jun–/– MEFs cannot form multilayer foci upon expression of oncogenic Ras (14), we next tested if the deficient Ras transformation of the Jnk2–/– MEFs would be due to the inability of these cells to activate c-Jun and AP-1. We tested their ability to accumulate AP-1 components, c-Jun and Fra-1, a typical feature for Ras-transformed fibroblasts due to the fact that Ras can induce AP-1 activation, and both c-Jun and Fra-1 genes are regulated by AP-1 (25). H-Ras-transduced Jnk2–/– MEFs were just as capable of accumulating both c-Jun and Fra-1 as the corresponding wild-type and Jnk1–/– MEFs (Fig. 3A ). Furthermore, phosphorylation of c-Jun was just as efficient in Ras-transduced Jnk2–/– as it was in corresponding wild-type and Jnk1–/– MEFs (Fig. 3A). We also tested the AP-1 reporter activity in these cell lines and found that the Jnk2–/– MEFs were slightly more efficient in activating AP-1 than the wild-type and Jnk1–/– MEFs both in response to H-Ras and MAPK kinase kinase (MEKK1), an efficient JNK and AP-1 activator (Fig. 3B). We conclude from these data that the Jnk2–/– MEFs are capable of activating c-Jun and AP-1, and that the deficiency of Ras transformation in the Jnk2–/– MEFs is not due to defective activation of c-Jun/AP-1.


Figure 3
View larger version (44K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. AP-1 and c-Jun activation is not impaired in Jnk2–/– MEFs. A, Western blot of the expression of c-Jun, Fra-1, phospho-c-Jun, and H-Ras in H-Ras-transduced wild-type, Jnk1–/–, and Jnk2–/– MEFs. {alpha}-Tubulin was used as a loading control. B, AP-1 luciferase reporter activities of the wild-type, Jnk1–/–, and Jnk2–/– MEFs. MEFs were transfected as quadruplicates with 4xAP-1 luciferase construct either alone or in combination with H-Ras or MEKK1. Luciferase activity was measured 48 h posttransfection. Columns, means from one experiment done in quadruplicate; bars, SD. The experiments were repeated thrice with similar results.

 
The deficient Ras transformation of the Jnk2–/– cells does not involve p53 activity or incomplete immortalization. The spontaneous immortalization (NIH3T3 method) of mouse fibroblasts involves inactivation of the p53 tumor suppressor pathway. Thus, the introduction of a single transforming oncogene, like Ras, is sufficient to transform them (26). Because the immortalized Jnk2–/– MEFs cannot be transformed with H-Ras alone, we studied if these cells had failed to immortalize and still had active p53 pathways. For this purpose, we tested the expression of p53 and p19Arf in these cells. One of the tested wild-type (wt-1) and two Jnk1–/– (J1-1 and J1-2) MEF lines had clearly accumulated p53 expression indicative of an inactivating mutation in p53 (Fig. 4A, top ). Because none of the Jnk2–/– MEFs exhibited strong p53 accumulation, we also evaluated the expression of the p19Arf by Western blotting. We found out that all the Jnk2–/– MEFs had lost the expression of p19Arf, and could therefore be classified as "immortal" according to these criteria (Fig. 4A, middle). Furthermore, we also "forced" the p53 pathway inactivation in Jnk2–/– and wild-type MEFs by overexpression of the E6 protein of the human papillomavirus E16, which is known to efficiently inactivate the p53 pathway (27). Coexpression of the E6 protein together with Ras was still not sufficient for transforming Jnk2–/– MEFs (Fig. 4B), further suggesting that the defect in Ras transformation in the Jnk2–/– MEFs is not due to the defective inactivation of the p53 pathway. Because all of the above described experiments were done at passages 30 to 35, we passaged the cells 30 more times to ensure that they were immortal, and tested their capability for transformation by H-Ras to find out if it would improve on passage number, time, and handling. At passage 65, Jnk2–/– MEFs grew slightly faster than on passages 30 to 35 (data not shown), but were still deficient in Ras transformation (Fig. 4C). Thus, we conclude that the deficiency in Ras transformation in Jnk2–/– MEFs was not due to incomplete immortalization.


Figure 4
View larger version (35K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. The defective transformation of H-Ras-transduced Jnk2–/– MEFs was not due to p53 activity or incomplete immortalization. A, Western blot analysis of p53 and p19Arf expression in wild-type, Jnk1–/–, and Jnk2–/– MEFs. {alpha}-Tubulin was used as a loading control. B, focus formation assay with wild-type and Jnk2–/– MEFs infected with E16E6 alone or in combination with H-Ras. Columns, means from one experiment done in triplicate; bars, SD. The experiment was repeated thrice with similar results. C, focus formation of the wild-type, Jnk1–/–, and Jnk2–/– MEFs that were passaged for an additional 65 times, and subsequently transduced with H-Ras. Columns, means from one experiment done in triplicate; bars, SD.

 
Jnk2–/– MEFs can be transformed by Ras when cotransduced with a Ras-cooperating oncogene. Because immortalized Jnk2–/– MEFs could not be transformed with Ras, which is a strongly transforming oncogene and is normally sufficient by itself to transform immortalized murine fibroblasts, we tested if the coexpression of a Ras-cooperating oncogene would be able to transform them similar to primary fibroblasts (28, 29). Indeed, coexpression of SV40-LT and H-Ras in Jnk2–/– MEFs efficiently transformed these cells as analyzed by the focus formation assay (Fig. 5A, left and B ). Similarly, another Ras-cooperating oncogene, c-Myc, assisted H-Ras-expressing Jnk2–/– MEFs to form foci (Fig. 5A, right and B). In addition to the ability to form foci, coexpression of SV40-LT or c-Myc together with H-Ras in Jnk2–/– MEFs resulted in clearly transformed cellular morphologies similar to that of the corresponding wild-type cells (Fig. 5C). The phospho-c-Jun Western blot analysis also failed to show any clear correlation between the transformation capability and c-Jun phosphorylation in these cells (Fig. 5D). This data strongly suggests that the lack of JNK2 created an extra protective barrier in these cells against Ras transformation that could be overcome by coexpression of cooperating oncogenes and was not dependent on c-Jun phosphorylation.


Figure 5
View larger version (44K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5. Transformation of Jnk2–/– MEFs requires cooperating oncogenes. A, focus formation assay with wild-type and Jnk2–/– MEFs infected with SV40-LT or c-Myc and H-Ras. Columns, means from one experiment done in triplicate; bars, SD. The experiment was repeated thrice with similar results. B, representative images of focus formation in wild-type and Jnk2–/– MEFs transduced with vector or H-Ras either alone or in combination with SV40-LT or c-Myc. C, phase contrast pictures of wild-type and Jnk2–/– MEFs transduced with vector control or H-Ras in combination with SV40-LT or c-Myc. D, Western blot analysis of phospho-c-Jun in wild-type and Jnk2–/– MEFs transduced with vector control or H-Ras in combination with c-Myc or SV40-LT. {alpha}-Tubulin was used as a loading control.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The data presented above clearly shows that the JNK2 isoform of JNK is indispensable for the Ras transformation process in MEFs. The fact that the Ras-transduced Jnk1–/– MEFs expresses similar amounts of JNK compared with the corresponding Jnk2–/– MEFs (Fig. 1A, middle), and the fact that they do not have any defect in Ras transformation, indicates that the JNK2 isoform of JNK is specifically supporting Ras transformation and that the results reported above are not just due to a possible overall lower level of JNK expression in Jnk2–/– MEFs. Furthermore, because retrovirally transduced H-Ras was expressed in the Jnk2–/– cells as efficiently as in the corresponding wild-type and Jnk1–/– cells, the defect in Ras transformation in Jnk2–/– MEFs was not due to lower Ras levels either, which have previously been shown to influence Ras transformation efficiency (23). Thus, our data clearly points out that selective inactivation of the JNK signaling pathway by preventing the expression of the JNK2 isoform of JNK impairs Ras transformation.

It is very surprising that inactivation of only one isoform of JNK can have so drastic an effect in the Ras transformation process. In order to rule out the possibility that the MEFs used in this study might have abnormal Ras signaling, we transduced them with the dominantly active Ras effector loop mutants; 12V35S, 12V37G, and 12V40C. These mutants have different preferences in the activation of Ras downstream signaling pathways in NIH3T3 murine fibroblasts, 12V35S being the most prominent in activating the Raf-MEK pathway, 12V37G the Ral-JNK, and 12V40C the PI3K-protein kinase B pathways (30). Ras signaling in MEFs did not exhibit differences in the pattern of activation of these three effector pathways in comparison with NIH3T3 fibroblasts, but instead were activated in an identical manner,3 showing that the MEFs used in this study exhibit normal Ras signaling. Supporting our data, earlier in vivo studies on murine 7,12-dimethylbenz(a)anthracene (DMBA)/12-O-tetradecanoylphorbol-13-acetate (TPA) induced skin tumorigenesis showed that the Jnk2–/– mice were resistant to tumor formation, whereas Jnk1–/– mice were not (31, 32). Because DMBA/TPA treatment could induce H-Ras-activating mutations in mouse skin (33), these findings and the data presented in this article provide strong evidence for the active role of the JNK2 isoform of JNK, and that it is one of the key players in the Ras transformation process in the mouse.

A previous partially contradictory study involving the ectopic expression of oncogenic H-Ras in a Jnk-null MEF line showed only ~50% reduction in the transformation efficiency in vitro. Furthermore, "s.c. tumors were detected in all mice injected with H-Ras-transformed wild-type or Jnk-null cells" but strikingly, the tail vein injections of the H-Ras-expressing Jnk-null cells resulted in "lung tumor nodules with substantially greater tumor mass" than the wild-type cells. Interestingly, in that study, expression of JNK1 or JNK2 in the Jnk-null cells resulted in decreased tumor mass, complementing the effect of JNK deficiency and leading to the conclusion that JNK could act as a tumor suppressor in vivo (17). It is notable that we have not observed decreased tumorigenicity in our Jnk1–/– MEFs in vitro or in vivo even though we also use an immunodeficient mouse xenograft model. Furthermore, the DMBA/TPA treatment of Jnk1–/– mice led to increased rather than decreased tumorigenicity in vivo (32). The actual reason for this clear difference may be very difficult to pinpoint, but some of it could be due to differences in the knockout mouse backgrounds. It could also be influenced by the fact that the Jnk-null MEFs are extremely difficult to establish because JNK is important for cellular survival, and the Jnk-null mice die at embryonic days 10 to 11 (34, 35), suggesting that the Jnk-null MEFs may have undergone major changes in order to survive, which might also affect their Ras-inducible tumorigenicity.

Several studies suggest an active role for JNK signaling in human cancer. For example, down-regulation of JNK2 with antisense oligos has been shown to cause growth inhibition in human T98G glioblastoma and prostate carcinoma cells (36, 37), as well as inhibition of epidermal growth factor–induced proliferation of human A549 lung carcinoma cells (38). Thus far, however, constituent JNK activation in cancer cells has only been found in human glioblastomas (39, 40). Additionally, contrary to the above findings, an upstream kinase of JNK, MAPK kinase 4 (MKK4), has been suggested to be a metastasis suppressor because some human breast, pancreatic, biliary, ovarian, and prostate carcinomas have been found to exhibit no or reduced expression levels of MKK4 compared with corresponding nonmalignant tissue (41–43). These data, altogether, point towards a rather complicated role for the JNK signaling pathway in cancer.

Utilization of the Ras effector loop mutants that can preferentially activate the Raf-MEK-ERK, Ral-JNK, or PI3K pathways has shown that in human cells, the activation of the Ral-JNK pathway was the most transforming and the Raf-MEK-ERK the second most transforming, suggesting that the Ras transformation process may differ between human and mouse cells (44). However, one should notice that these mutants function preferentially, but not exclusively, on the designated effector pathways, and furthermore, that they do not cover all signaling pathways downstream of Ras, such as, for example, the Rac-PAK pathway whose activation leads to JNK activation in Ras-transformed Rat-1 cells (45). Interestingly, our data showing that JNK2 has a crucial role in Ras transformation of MEFs, together with the previous study (44), supports the existence of a transformation model in which the importance of the JNK pathway could be more universal between human and mouse in Ras-induced tumorigenesis than previously suggested.

Very little data exists on how Ras actually transforms cells. MEFs immortalized by the NIH3T3 method have been widely used to study the Ras transformation processes because ectopic expression of active oncogenic Ras alone is capable of transforming these cells (26). Because c-Jun–/– MEFs are fully resistant to Ras transformation, and because activation of the transcription factor AP-1 and consequent accumulation of the AP-1 components, c-Jun and Fra-1, could be detected in Ras-transformed wild-type but not in corresponding c-Jun–/– MEFs, it has been suggested that the c-Jun/AP-1 activation is one of the key events leading to transcriptional changes required for the Ras transformation process (14, 25). However, an AP-1 target gene, whose protein product would be indispensable for Ras transformation, is yet to be identified. One of the suggested candidates has been cyclin D1 because transcription from the cyclin D1 promoter could be regulated by AP-1 (46). However, cyclin D1–/– MEFs are not impaired in Ras transformation (47). Moreover, by showing that the Jnk2–/– MEFs are not deficient in Ras-induced c-Jun phosphorylation or AP-1 activation, our data challenges the role of JNK in the phosphorylation of c-Jun as well as the central role of AP-1 in the Ras transformation process. Thus, as such, it also supports the recent in vivo study showing that JunAA MEFs are only partially resistant to Ras-induced tumor formations (16). Furthermore, our data clearly points out that JNK2 substrates other than c-Jun or any other AP-1 components are crucial for regulating Ras transformation.

Ras-transduced Jnk2–/– MEFs show several phenotypes that are typical for nontransformed primary embryonic fibroblasts: Ras-transduced Jnk2–/– MEFs are contact-inhibited, they have a growth disadvantage, and they enter an initial crisis upon Ras transduction which they overcome in 2 to 3 weeks (data not shown). In addition, they exhibit impaired migration (data not shown), multilayer formation, anchorage-independent growth, and tumor formation in mice. Compiling all of these phenotypes suggests that JNK2 has a very central role in the regulation of Ras transformation. Furthermore, akin to primary MEFs, Jnk2–/– MEFs can be fully transformed by coexpression of H-Ras with SV40-LT or c-Myc. A similar phenotype was observed with the CDK4–/– MEFs, which could not be transformed with Ras, suggesting that a critical threshold for CDK4 activity was required for oncogenic cell cycling (48). However, Ras expression in CDK4–/– MEFs rendered them into a state resembling premature senescence (48), which does not seem to be the case in Jnk2–/– MEFs. It remains to be seen if JNK2 can control the activities of CDK or whether the mechanism involving JNK2 in Ras transformation is something very different.

The data obtained in this study gives novel important information on the Ras transformation process. It clearly points out that the JNK signaling pathway, and especially the JNK2 isoform of JNK, can mediate the transforming signal of Ras and is required for the development of Ras-induced tumorigenicity. Furthermore, our data shows that Ras transformation of the immortalized fibroblasts which carry genetically inactivated JNK2 requires the coexpression of a cooperating oncogene, showing that the JNK2 inactivation in these cells created an extra protective barrier against Ras transformation. Importantly, our data also strongly suggests that the activation of a JNK2 target other than c-Jun is crucial for the Ras transformation process.


   Acknowledgments
 
Grant support: Bøje Benzon stipend of the Alfred Benzon Foundation (T. Kallunki), the Danish Cancer Society (all authors), the Danish Medical Research Council (T. Kallunki), the Danish National Research Foundation (T. Kallunki and M. Jäättelä), the Meyer Foundation (M. Jäättelä), and the Novo Foundation (T. Kallunki).

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.

We thank G. Nolan for the pB-puro and pB-hygro constructs, J. Rohn for the SV40-LT pB-puro construct, H. Land for c-Myc pB-puro construct, C.J. Der for the Ras effector loop mutants, D.A. Galloway for the E16E6 pB-neo construct, B. Willumsen for the H-Ras plasmid and the Ras antibody, and K. Yoshioka for 4xAP-1 luciferase reporter construct. Additionally, we thank J. Westermarck and N. Fehrenbacher for stimulating discussions and K.G. Henriksen, T. Chaaban, I.F. Larsen, and the personnel at the animal facilities of the Danish Cancer Society for excellent assistance.


   Footnotes
 
1 J. Thastrup and T. Kallunki, unpublished observations. Back

2 T. Kallunki and N. Dietrich unpublished observations. Back

3 C. Nielsen, T. Bøttzauw, M. Jäättelä, and T. Kallunki, manuscript in preparation. Back

Received 7/28/06. Revised 10/ 2/06. Accepted 11/ 2/06.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Kallunki T, Su B, Tsigelny I, et al. JNK2 contains a specificity-determining region responsible for efficient c-Jun binding and phosphorylation. Genes Dev 1994;8:2996–3007.[Abstract/Free Full Text]
  2. Gupta S, Barrett T, Whitmarsh AJ, et al. Selective interaction of JNK protein kinase isoforms with transcription factors. EMBO J 1996;15:2760–70.[Medline]
  3. Sanchez I, Hughes RT, Mayer BJ, et al. Role of SAPK/ERK kinase-1 in the stress-activated pathway regulating transcription factor c-Jun. Nature 1994;372:794–8.[Medline]
  4. Minden A, Karin M. Regulation and function of the JNK subgroup of MAP kinases. Biochim Biophys Acta 1997;1333:F85–104.[Medline]
  5. Weston CR, Davis RJ. The JNK signal transduction pathway. Curr Opin Genet Dev 2002;12:14–21.[CrossRef][Medline]
  6. Adjei AA. Blocking oncogenic Ras signaling for cancer therapy. J Natl Cancer Inst 2001;93:1062–74.[Abstract/Free Full Text]
  7. Friday BB, Adjei AA. K-ras as a target for cancer therapy. Biochim Biophys Acta 2005;1756:127–44.[Medline]
  8. Finco TS, Westwick JK, Norris JL, Beg AA, Der CJ, Baldwin AS, Jr. Oncogenic Ha-Ras-induced signaling activates NF-{kappa}B transcriptional activity, which is required for cellular transformation. J Biol Chem 1997;272:24113–6.[Abstract/Free Full Text]
  9. Pruitt K, Pruitt WM, Bilter GK, Westwick JK, Der CJ. Raf-independent deregulation of p38 and JNK mitogen-activated protein kinases are critical for Ras transformation. J Biol Chem 2002;277:31808–17.[Abstract/Free Full Text]
  10. Roberts TM. Cell biology. A signal chain of events. Nature 1992;360:534–5.[CrossRef][Medline]
  11. Rodriguez-Viciana P, Warne PH, Khwaja A, et al. Role of phosphoinositide 3-OH kinase in cell transformation and control of the actin cytoskeleton by Ras. Cell 1997;89:457–67.[CrossRef][Medline]
  12. Eferl R, Wagner EF. AP-1: a double-edged sword in tumorigenesis. Nat Rev Cancer 2003;3:859–68.[CrossRef][Medline]
  13. Binetruy B, Smeal T, Karin M. Ha-Ras augments c-Jun activity and stimulates phosphorylation of its activation domain. Nature 1991;351:122–7.[CrossRef][Medline]
  14. Johnson R, Spiegelman B, Hanahan D, Wisdom R. Cellular transformation and malignancy induced by ras require c-jun. Mol Cell Biol 1996;16:4504–11.[Abstract]
  15. Smeal T, Binetruy B, Mercola DA, Birrer M, Karin M. Oncogenic and transcriptional cooperation with Ha-Ras requires phosphorylation of c-Jun on serines 63 and 73. Nature 1991;354:494–6.[CrossRef][Medline]
  16. Behrens A, Jochum W, Sibilia M, Wagner EF. Oncogenic transformation by ras and fos is mediated by c-Jun N-terminal phosphorylation. Oncogene 2000;19:2657–63.[CrossRef][Medline]
  17. Kennedy NJ, Sluss HK, Jones SN, Bar-Sagi D, Flavell RA, Davis RJ. Suppression of Ras-stimulated transformation by the JNK signal transduction pathway. Genes Dev 2003;17:629–37.[Abstract/Free Full Text]
  18. Sabapathy K, Hu Y, Kallunki T, et al. JNK2 is required for efficient T-cell activation and apoptosis but not for normal lymphocyte development. Curr Biol 1999;9:116–25.[CrossRef][Medline]
  19. Sabapathy K, Kallunki T, David JP, Graef I, Karin M, Wagner EF. c-Jun NH2-terminal kinase (JNK)1 and JNK2 have similar and stage-dependent roles in regulating T cell apoptosis and proliferation. J Exp Med 2001;193:317–28.[Abstract/Free Full Text]
  20. Dietrich N, Thastrup J, Holmberg C, et al. JNK2 mediates TNF-induced cell death in mouse embryonic fibroblasts via regulation of both caspase and cathepsin protease pathways. Cell Death Differ 2004;11:301–13.[CrossRef][Medline]
  21. Mosmann T. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J Immunol Methods 1983;65:55–63.[CrossRef][Medline]
  22. Tournier C, Hess P, Yang DD, et al. Requirement of JNK for stress-induced activation of the cytochrome c-mediated death pathway. Science 2000;288:870–4.[Abstract/Free Full Text]
  23. Jacobsen K, Groth A, Willumsen BM. Ras-inducible immortalized fibroblasts: focus formation without cell cycle deregulation. Oncogene 2002;21:3058–67.[CrossRef][Medline]
  24. Sabapathy K, Hochedlinger K, Nam SY, Bauer A, Karin M, Wagner EF. Distinct roles for JNK1 and JNK2 in regulating JNK activity and c-Jun-dependent cell proliferation. Mol Cell 2004;15:713–25.[CrossRef][Medline]
  25. Mechta F, Lallemand D, Pfarr CM, Yaniv M. Transformation by ras modifies AP1 composition and activity. Oncogene 1997;14:837–47.[CrossRef][Medline]
  26. Simons PJ, Dourmashkin RR, Turano A, Phillips DE, Chesterman FC. Morphological transformation of mouse embryo cells in vitro by murine sarcoma virus (Harvey). Nature 1967;214:897–8.[CrossRef][Medline]
  27. Scheffner M, Werness BA, Huibregtse JM, Levine AJ, Howley PM. The E6 oncoprotein encoded by human papillomavirus types 16 and 18 promotes the degradation of p53. Cell 1990;63:1129–36.[CrossRef][Medline]
  28. Cavender JF, Conn A, Epler M, Lacko H, Tevethia MJ. Simian virus 40 large T antigen contains two independent activities that cooperate with a ras oncogene to transform rat embryo fibroblasts. J Virol 1995;69:923–34.[Abstract]
  29. Land H, Parada LF, Weinberg RA. Tumorigenic conversion of primary embryo fibroblasts requires at least two cooperating oncogenes. Nature 1983;304:596–602.[CrossRef][Medline]
  30. Khosravi-Far R, White MA, Westwick JK, et al. Oncogenic Ras activation of Raf/mitogen-activated protein kinase-independent pathways is sufficient to cause tumorigenic transformation. Mol Cell Biol 1996;16:3923–33.[Abstract]
  31. Chen N, Nomura M, She QB, et al. Suppression of skin tumorigenesis in c-Jun NH(2)-terminal kinase-2-deficient mice. Cancer Res 2001;61:3908–12.[Abstract/Free Full Text]
  32. She QB, Chen N, Bode AM, Flavell RA, Dong Z. Deficiency of c-Jun-NH(2)-terminal kinase-1 in mice enhances skin tumor development by 12-O-tetradecanoylphorbol-13-acetate. Cancer Res 2002;62:1343–8.[Abstract/Free Full Text]
  33. Kemp CJ. Multistep skin cancer in mice as a model to study the evolution of cancer cells. Semin Cancer Biol 2005;15:460–73.[CrossRef][Medline]
  34. Kuan CY, Yang DD, Samanta Roy DR, Davis RJ, Rakic P, Flavell RA. The Jnk1 and Jnk2 protein kinases are required for regional specific apoptosis during early brain development. Neuron 1999;22:667–76.[CrossRef][Medline]
  35. Sabapathy K, Jochum W, Hochedlinger K, Chang L, Karin M, Wagner EF. Defective neural tube morphogenesis and altered apoptosis in the absence of both JNK1 and JNK2. Mech Dev 1999;89:115–24.[CrossRef][Medline]
  36. Potapova O, Gorospe M, Bost F, et al. c-Jun N-terminal kinase is essential for growth of human T98G glioblastoma cells. J Biol Chem 2000;275:24767–75.[Abstract/Free Full Text]
  37. Yang YM, Bost F, Charbono W, et al. c-Jun NH(2)-terminal kinase mediates proliferation and tumor growth of human prostate carcinoma. Clin Cancer Res 2003;9:391–401.[Abstract/Free Full Text]
  38. Bost F, McKay R, Bost M, Potapova O, Dean NM, Mercola D. The Jun kinase 2 isoform is preferentially required for epidermal growth factor-induced transformation of human A549 lung carcinoma cells. Mol Cell Biol 1999;19:1938–49.[Abstract/Free Full Text]
  39. Antonyak MA, Kenyon LC, Godwin AK, et al. Elevated JNK activation contributes to the pathogenesis of human brain tumors. Oncogene 2002;21:5038–46.[CrossRef][Medline]
  40. Tsuiki H, Tnani M, Okamoto I, et al. Constitutively active forms of c-Jun NH2-terminal kinase are expressed in primary glial tumors. Cancer Res 2003;63:250–5.[Abstract/Free Full Text]
  41. Kim HL, Vander Griend DJ, Yang X, et al. Mitogen-activated protein kinase kinase 4 metastasis suppressor gene expression is inversely related to histological pattern in advancing human prostatic cancers. Cancer Res 2001;61:2833–7.[Abstract/Free Full Text]
  42. Su GH, Hilgers W, Shekher MC, et al. Alterations in pancreatic, biliary, and breast carcinomas support MKK4 as a genetically targeted tumor suppressor gene. Cancer Res 1998;58:2339–42.[Abstract/Free Full Text]
  43. Yamada SD, Hickson JA, Hrobowski Y, et al. Mitogen-activated protein kinase kinase 4 (MKK4) acts as a metastasis suppressor gene in human ovarian carcinoma. Cancer Res 2002;62:6717–23.[Abstract/Free Full Text]
  44. Hamad NM, Elconin JH, Karnoub AE, et al. Distinct requirements for Ras oncogenesis in human versus mouse cells. Genes Dev 2002;16:2045–57.[Abstract/Free Full Text]
  45. Tang Y, Chen Z, Ambrose D, et al. Kinase-deficient Pak1 mutants inhibit Ras transformation of Rat-1 fibroblasts. Mol Cell Biol 1997;17:4454–64.[Abstract]
  46. Albanese C, Johnson J, Watanabe G, et al. Transforming p21ras mutants and c-Ets-2 activate the cyclin D1 promoter through distinguishable regions. J Biol Chem 1995;270:23589–97.[Abstract/Free Full Text]
  47. Yu Q, Ciemerych MA, Sicinski P. Ras and Myc can drive oncogenic cell proliferation through individual D-cyclins. Oncogene 2005;24:7114–9.[CrossRef][Medline]
  48. Zou X, Ray D, Aziyu A, et al. Cdk4 disruption renders primary mouse cells resistant to oncogenic transformation, leading to Arf/p53-independent senescence. Genes Dev 2002;16:2923–34.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
B. Bjorkblom, J. C. Vainio, V. Hongisto, T. Herdegen, M. J. Courtney, and E. T. Coffey
All JNKs Can Kill, but Nuclear Localization Is Critical for Neuronal Death
J. Biol. Chem., July 11, 2008; 283(28): 19704 - 19713.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nielsen, C.
Right arrow Articles by Kallunki, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nielsen, C.
Right arrow Articles by Kallunki, T.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online