| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Cell and Tumor Biology |
1 Institut National de la Sante et de la Recherche Medicale U482, Hôpital Saint-Antoine, Paris, France; 2 Laboratory of Experimental Cancerology, University Hospital, Gent, Belgium; and 3 CEA, Service de Génomique Fonctionnelle, Evry, France
Requests for reprints: Attoub Samir, Institut National de la Sante et de la Recherche Medicale U482, Hôpital Saint-Antoine, 184 rue du Faubourg Saint-Antoine, 75571 Paris Cedex 12, France. Phone: 33-1-49284648; Fax: 33-1-44749318; E-mail: rivat{at}st-antoine.inserm.fr
| Abstract |
|---|
|
|
|---|
and STAT3ß isoforms. Blockade of STAT3 signaling by STAT3ß, depletion of the STAT3
/ß isoforms by RNA interference, and pharmacologic inhibition of STAT3
/ß phosphorylation by cucurbitacin or STAT3 inhibitory peptide abrogates TFF- and VEGF-induced cellular invasion and reduces the growth of HCT8/S11 tumor xenografts in athymic mice. Differential gene expression analysis using DNA microarrays revealed that overexpression of STAT3ß down-regulates the VEGF receptors Flt-1, neuropilins 1 and 2, and the inhibitor of DNA binding/differentiation (Id-2) gene product involved in the neoplastic transformation. Taken together, our data suggest that TFF3 and the essential tumor angiogenesis regulator VEGF165 exert potent proinvasive activity through STAT3 signaling in human colorectal cancer cells. We also validate new therapeutic strategies targeting STAT3 signaling by pharmacologic inhibitors and RNA interference for the treatment of colorectal cancer patients.
Key Words: TFF neuropilins VEGFR Id-2 cucurbitacin
| Introduction |
|---|
|
|
|---|
Recent studies have suggested a causal link between local inflammation and the occurrence of gastrointestinal tumors. Among the local mediators of inflammatory processes, trefoil factors (TFF) TFF1 (pS2), TFF2 (SP), and TFF3 (ITF) constitute a new family of regulatory peptides involved in mucosal protection and repair of the gastrointestinal tract (10). These peptides are overexpressed during tumor progression and promote tumor cell invasion and angiogenesis (11, 12). We have shown recently that theproinvasive activity of TFFs is controlled by src- and Rho-dependent signaling pathways, suggesting that the signal transducers and activators of transcription (STAT) 3 might be involved in TFF signaling (13). Indeed, STAT3 is activated by several cytokines, growth factors, and oncogenes involved in cellular transformation and cancer cell invasion, including hepatocyte growth factor, leptin and stem cell factor, the src family kinases, and the small GTPases Rac1 and RhoA (6, 1417).
STATs are latent transcription factors activated by phosphorylation of a conserved tyrosine residue. Experimental and clinical data have revealed the oncogenic potential of STAT3 through overexpression and constitutive activation in a variety of human malignancies, including leukemia, melanoma, head and neck squamous cell carcinoma, breast, prostate, ovarian, and colorectal carcinoma (18, 19). Accordingly, STAT3 exerts antiapoptotic and mitogenic effects (20, 21). The constitutively active STAT3C mutant displays a transforming activity scored by colony growth in soft agar and by tumor formation in nude mice (20). STAT3C also induces tumor angiogenesis by activating the promoter of the gene coding VEGF (22). Interestingly, VEGF and STAT3 function through a reciprocal signaling loop in endothelial cells because VEGF activates STAT3 (23). STAT3ß is a naturally occurring splice variant of STAT3, which lacks the COOH-terminal transactivation domain and functions as a dominant-negative element for STAT3-mediated transcription (24). Accordingly, src oncogenic transformation is impaired by STAT3ß, and overexpression of STAT3ß leads to cell and tumor growth inhibition, apoptosis, and suppression of colony formation (17, 25, 26).
In this study, we investigated the possible implication of STAT3 signaling in the proinvasive activity of TFF1, TFF3, and VEGF, because we have shown recently that several human colonic cancer cells, including HCT8/S11 cells, express specific and functional receptors and coreceptors for VEGF.4 In addition, we examined the impact of STAT3 inhibition on cell invasion and tumor growth after overexpression of the splice variant STAT3ß in human colon cancer cells, ablation of both isoforms of STAT3 by RNA interference, and use of pharmacologic agents targeting STAT3
and STAT3ß phosphorylation.
| Material and Methods |
|---|
|
|
|---|
, pYN3218-STAT3ß, pcDNA3-pS2, pCR-TFF3, and the GAS-luciferase vector were provided by Drs. Pfeffer, Schaefer, Emami, Podolsky, and Dusanter, respectively. Purified recombinant human TFF1 and TFF3 were provided by Drs. Thim and Westley. Human recombinant VEGF165 was from R&D Systems (Lille, France). STAT3 inhibitor peptide was from Calbiochem (France Biochem, Meudon, France). Cucurbitacin was provided by Dr. Schultz (National Cancer Institute, Bethesda, MD). The VEGF receptor (VEGFR) tyrosine kinase inhibitor ZD4190 was provided by Dr. Ryan (AstraZeneca, Macclesfield, United Kingdom).
Cell Culture, Transient Transfections, and Luciferase Reporter Gene Assays. Human colorectal cancer cells HCT8/S11 and embryonic kidney HEK-293T cells were cultured as described (27). HEK-293T and HCT8/S11 cells were transfected with 0.25 µg GAS-luciferase reporter construct using the LipofectAMINE Plus Reagent (Invitrogen, Cergy Pontoise, France). Where indicated, the transfection mixture was combined with 1 µg of TFF1, TFF3, STAT3
, or STAT3ß expression vectors, or cultures were pretreated for 1, 3, or 6 hours with cucurbitacin (1 µmol/L). Luciferase assays were done using the Luciferase Reporter Assay System kit (Promega, Charbonnières, France). For analysis of STAT3 phosphorylation, HEK-293T cells were transfected with either the empty control vector pcEFmyc, the STAT3
or STAT3ß expression plasmids, or the TFF1 or TFF3 plasmids. Cells were then grown for 24 hours in a medium containing 0.1% serum and lysed in SDS buffer. Alternatively, cells transfected with STAT3
or STAT3ß were grown for 24 hours in a 0.1% serum medium and then treated by TFF1 (0.1µmol/L), TFF3 (0.1 µmol/L), or VEGF (100 ng/mL) in fresh low-serum medium.
Construction of Small Interfering RNA Plasmids. Small interfering RNA oligonucleotide for STAT3 was designed using the Target Finder program (Ambion, Austin, TX). The following sequence was used to construct the small interference oligonucleotide for STAT3 vectors in pSUPER (Oligoengine, Seattle, WA): forward 5'-GATTGACCTAGAGACCCAC-3'.
Western Blot Analysis. Proteins were resolved on 8% SDS-PAGE gel and the blots were probed with the rabbit polyclonal antibodies directed against phospho-STAT3 (Cell Signaling, Ozyme, Saint-Quentin en Yvelines, France), the COOH-terminal domain of STAT3
(C-20), the conserved NH2-terminal domain of STAT3
and STAT3ß (K-15, Santa Cruz, Tebu, Le Perray en Yvelines, France), or the mouse monoclonal antibody (AC15) against ß-actin (Sigma, Saint-Quentin-Fallaviers, France).
Gene Expression Profiling in HCT8/S11-pcDNA3 and HCT8/S11-STAT3ß Cells by cDNA Microarray Analysis. Microarrays were prepared using a collection of 5,760 cDNA clones selected from a normalized infant brain library (28). The 3' and/or 5' ends of these cDNA clones have been sequenced previously (29). RNA was extracted from pcDNA3 and STAT3ß-transfected HCT8/S11 cells and reverse transcribed using the SuperScript II reverse transcriptase (Invitrogen) in the presence of oligo(dT) and eithercyanine 3-dUTP (red dye) or cyanine 5-dUTP (green dye) (Perkin-Elmer Life Science, Courtaboeuf, France). Hybridizations were done overnight at 42°C. Slides were scanned with an Axon 4000B fluorescence laser scanning instrument (Axon, Foster City, CA). Image analysis and the calculation of average foreground signal adjusted for local channel-specific background were done using the GenePix Pro version 4.0 software (Axon). Cy5/Cy3 intensity ratios from each gene were calculated and globally normalized to make the median value of the log2 ratio equal to 0. Genes were considered as significantly modulated according to the criteria of 2-fold cutoff ratio.
RNA Isolation and Reverse Transcription-PCR Amplification. Total RNA was extracted using the Trizol reagent (Invitrogen). Reverse transcription-PCR (RT-PCR) analyses were done with the SuperScript One-Step RT-PCR kit (Invitrogen). Fragments of the human Flt-1, neuropilin-1 (np-1), neuropilin-2 (np-2), Id-2, and STAT3 cDNAs were amplified using the following primers: Flt-1: 5'-GCACCTTGGTTGTGGGTGAC-3' and 5'-CGTGCTGCTTCCTGGTCC-3', np-1: 5'-ACGATGAATGTGGCGATACT-3' and 5'-AGTGCATTCAAGGCTGTTGG-3', np-2: 5'-GTGACTGGACAGACTCCAAG-3' and 5'-CGAACACAATCTGGTACTCC-3', Id-2: 5'-TCCTTGCAGGCTTCTGAA-3' and 5'-CCATTCAACTTGTCCTCC-3', and STAT3: 5'-ATTCAAACACTTGACCCTGA-3' and 5'-ATTGTTGGTCAGCATGTTGT-3'. PCR products were resolved on a 2% agarose gel stained with ethidium bromide. Basal mRNA level of each gene was normalized to 1. Results presented are representative of three independent experiments.
Cell Invasion, Proliferation, Apoptosis, and Tumor Growth. Invasion assays were done as described previously (27). For cell proliferation assays, HCT8/S11 cells were seeded in six-well plates at a density of 100,000 cells per well and treated with cucurbitacin. DNA synthesis was measured with a 4-hour pulse with 1 µCi/mL [3H]thymidine and liquid scintillation counting. Alternatively, cells were seeded for 24 hours in 60-mm dishes at a density of 400,000 cells per well. Control and treated cells were then harvested and either counted using a cell counter (Beckman Coulter, Villepinte, France) or fixed with 70% ethanol in PBS and submitted to fluorescence-activated cell sorting analysis for the determination of apoptosis.
Parental and stably transfected HCT8/S11 cells were injected s.c. into the lateral flank of 6-week-old athymic nu/nu mice (3 x 106 cells in each xenograft, 8 animals in each group, Elevage Janvier, Le Genest, France). Tumor volume (V) was calculated using the formula: V = 0.4 x a x b2 where a is the length and b is the width of the tumor. Where indicated, animals were randomized into two groups, and one group was treated with cucurbitacin (1 mg/kg/d) diluted in DMSO 20% and water. Throughout this study, procedures were conducted in conformity with institutional guidelines (European Economic Community's Council Directive 86/609, OJ L 358, 1, December 12, 1987; NIH Guide for Care and Use of Laboratory Animals, NIH Publication No. 85-23, 1985).
Statistical Analysis. Data are means ± SEM of three independent experiments. The differences between experimental values was assessed by the unpaired Student's t test and P < 0.05 versus control was considered statistically significant.
| Results |
|---|
|
|
|---|
or its splice variant STAT3ß. Cotransfection of TFF3, but not TFF1, vector increased basal levels of STAT3 phosphorylation at Tyr705 in both isoforms with no impact on the total amount of STAT3 protein (Fig. 1A). We next examined the transcriptional activity of STAT3 in TFF1- and TFF3-transfected HEK-293T cells using the GAS reporter construct containing three STAT consensus elements upstream of the luciferase gene. As shown in Fig. 1B, TTF3 increased STAT3-dependent transcription 2.3-fold, whereas TFF1 had no significant effect (Fig. 1B). We further examined the effects of the addition of the trefoil peptides TFF1 and TFF3 on the activation status of STAT3 as immediate cellular responses. For this purpose, STAT3-transfected HEK-293T cells were treated for 5 to 30 minutes with 0.1 µmol/L TFF1 or TFF3. TFF3 induced a time-dependent activating phosphorylation of STAT3
, reaching maximal levels at 5 to 10 minutes, and a persistent activation of STAT3ß (Fig. 1C). In agreement with our data in Fig. 1A, no STAT3
/ß activation was observed in response to TFF1 stimulation. We also treated HCT8/S11 cells with VEGF165 (100 ng/mL), because we detected the expression of functional proinvasive VEGFR in human colonic cancer cells.5 Moreover, a recent report showed that STAT3 is essential for VEGF-induced migration and morphogenesis in human endothelial cells (22). In a manner identical to TFF3, VEGF165 induced prompt activation of STAT3
in HCT8/S11 cells (510 minutes) that returned to basal levels 15 to 30 minutes after stimulation (Fig. 1C). Interestingly, phosphorylation of the STAT3ß isoform by VEGF165 was also induced at 5 minutes and persisted up to 30 minutes after stimulation (Fig. 1C, bottom). We next investigated the impact of STAT3
or STAT3ß transient expression on STAT3 transcriptional activity. As expected, STAT3
activated the GAS-luciferase reporter gene 2.2-fold, whereas STAT3ß reduced by 50% STAT3-mediated transcription in HCT8/S11 cells (Fig. 1D).
|
or STAT3ß. Expression of the STAT3 transgenes was detected by Western blot analyses in individual clones and pools of clones (Fig. 2A). As shown in Fig. 2B, cellular invasion is not induced or potentiated byoverexpression of STAT3
in control cells or HCT8/S11 cells exposed to the proinvasive agents TFF1, TFF3, or VEGF. Most interestingly, blockade of STAT3 signaling by STAT3ß, which lacks the STAT3 transactivation domain, abrogated the invasive behavior of HCT8/S11-STAT3ß cells treated by these proinvasive agents. These encouraging results led us to investigate the possible impact of STAT3ß on the growth of human tumor xenografts. For this purpose, parental and STAT3
- or STAT3ß-transfected HCT8/S11 cells were xenografted in nude mice. Activation of STAT3 signaling by stable expression of STAT3
in HCT8/S11 cells substantially increased tumor size (2.0-fold) after 6 weeks following xenografts (Fig. 2C, left). Conversely, disruption of STAT3 signaling by STAT3ß led to a remarkable reduction (67%) of tumor growth in HCT8/S11-STAT3ß xenografts (Fig. 2C, right).
|
|
/ß Proteins by RNA Interference Abrogates TFF- and VEGF-Induced Cellular Invasion. We next investigated the impact of STAT3 depletion on the invasive properties of parental HCT8/S11 cells using RNA interference. Our STAT3 interference vector was introduced into HCT8/S11 cells, generating a stable HCT8/S11-siSTAT3 cell line. RT-PCR experiments revealed that STAT3 transcripts decreased by 28% and 67% in HCT8/S11-siSTAT3 pool and clone 2, respectively (Fig. 4A). Both isoforms of STAT3 proteins were depleted by 65% and 80% in silenced STAT3 cells pool and clone 2, respectively (Fig. 4B). As expected, STAT3 transcriptional activity was significantly reduced by 31% and 50% in pool and clone 2, respectively, compared with the stably transfected control HCT8/S11-pSUPER cell line (Fig. 4C). We next investigated the functional impact of STAT3 protein depletion on cellular invasion. Figure 4D shows that siSTAT3 abrogates TFF1-, TFF3-, and VEGF-induced invasiveness in the HCT8/S11-siSTAT3 cell line.
|
|
and ß isoforms. As expected, the STAT3 inhibitory peptideinhibited both TFF3- and VEGF-induced HCT8/S11 cell invasiveness (Fig. 5D). Impact of Cucurbitacin on HCT8/S11 Cell Survival, Proliferation, and Growth of HCT8/S11 Tumor Xenografts. Inhibitionof STAT3 activity by pretreatment of HCT8/S11 cells for 1 to 6 hours with cucurbitacin (1 µmol/L) is associated with a very weak induction of apoptosis from 3.7% (control HCT8/S11 cells) to 5.3% to 7.8% in cucurbitacin-treated cells and inhibition of cell proliferation (Fig. 6A). Similarly, both thymidine incorporation and HCT8/S11 cell counts indicate that 6-hour pretreatment with cucurbitacin had a modest impact on HCT8/S11 cell proliferation, which was reduced by 37% to 38% (Fig. 6B and C). Prolonged treatment of HCT8/S11 cells for 24 hours with 1 µmol/L cucurbitacin results in a significant induction of cell death (23% of apoptotic cells) and inhibition of cell proliferation by 93% (Fig.6D and E). Consistent with these data, we observed that cucurbitacin (1 mg/kg/d) reduced the growth of HCT8/S11 xenografts in the nude mice (Fig. 6F). When HCT8/S11 tumors grew to 70 to 100 mm3 volume at day 21 after s.c. injections, athymic nude mice were treated daily with cucurbitacin for an additional period of 3 weeks. Treatment of mice with cucurbitacin (1 mg/kg/d) resulted in a significant decrease (60%; P < 0.05) of tumor volume.
|
| Discussion |
|---|
|
|
|---|
In human colorectal cancer cells, both TFF3 and VEGF are connected to STAT3 signaling as proven by their ability to induce phosphorylation and transcriptional activation of STAT3. This conclusion is also supported by the ability of several STAT3 inhibitors, including STAT3ß, STAT3 interfering RNAs, and cucurbitacin, to reverse the invasive phenotype determined by TFF and VEGF. In contrast, no phosphorylation of STAT3 was induced by TFF1 in response to external addition of the TFF1 peptide and forced expression of TFF1. Interestingly, TFF1-induced cellular invasion was also abolished by inhibition of STAT3 signaling. One can postulate that the invasive phenotype promoted by TFF1 is STAT3 dependent, although TFF1 does not directly activate STAT3. Moreover, both TFF3 and VEGF genes are directly up-regulated by STAT3 (22, 36). It is possible that TFF3 activates STAT3 phosphorylation and transcriptional activity, thereby inducing transcription of its own gene. Earlier reports also indicated that VEGF activates STAT3, which in turn up-regulates VEGF gene transcription. Interestingly, we found that blockade of STAT3-mediated transcription by overexpression of STAT3ß leads to a dramatic decrease in the transcript levels of VEGFR-1 (Flt-1) and VEGFR coreceptors np-1 and np-2, suggesting that these genes are also functioning as transcriptional targets of STAT3 in human colorectal cancer cells. Inhibition of STAT3 transcription and consequent loss of VEGFR and coreceptors renders cells insensitive to the addition of VEGF as shown by the blockade of VEGF-induced cellular invasion in HCT8/S11-STAT3ß cells. Moreover, thepowerful action of cucurbitacin against tumor growth in the present study can be related to a direct inhibition of cancer cell proliferation, invasiveness, and tumor neovascularization by adjacent endothelial cells.
Consistent with the observations made by Schaefer et al. (37) regarding the activation of STAT3 by EGF, we found that phosphorylation of the splice variant STAT3ß on Tyr705 on cell stimulation by TFF3 or VEGF is more stable than phosphorylation of the STAT3
isoform. It is conceivable that STAT3
undergoes a more rapid dephosphorylation than STAT3ß. Because STAT3
is largely predominant (
90%) over STAT3ß in colon cancer cell lines (data not shown), it can be hypothesized that on ligand stimulation both isoforms are phosphorylated and dimerize, producing preferentially STAT3
homodimers for the transcriptional activation of STAT3-dependent genes. Subsequently, after dephosphorylation of STAT3
, the persistence of phosphorylated forms of STAT3ß results in the inhibition of STAT3-dependent transcription. Indeed, overexpression of STAT3ß in HCT8/S11 cells favors the formation of STAT3ß-containing dimers and results in a 50% decrease in STAT3 transcription levels. Consequently, forced expression of STAT3ß and depletion of both STAT3
and STAT3ß isoforms by RNA interference are associated with reduced cellular invasion or tumor growth. In addition, down-regulation of the Id-2 mRNA levels was observed in the HCT8/S11-STAT3ß cell line. The Id-2 gene, a target of the oncogenic Wnt/ß-catenin pathway, is found overexpressed in colon cancer (38). Id-2 in turn inhibits the helix-loop-helix transcription factors through heterodimeric interactions (39). The balance between Id proteins and other helix-loop-helix factors determines the commitment to either proliferation ordifferentiation. Thus, a high level of Id proteins inhibits cell differentiation and promotes proliferation by inducing the G1-S transition. Conversely, depletion of Id-2 in colorectal cancer cells leads to growth arrest (40). It is therefore reasonable to consider that Id-2 down-regulation induced by forced expression of STAT3ß can be responsible for the tumor growth inhibition observed here when STAT3 signaling is inhibited.
In conclusion, we propose to consider STAT3 signaling and STAT3-regulated genes as potential molecular targets for the design of new therapeutic agents against colon cancer progression. The anticancer activity of drugs and RNA interference vectors targeting TFF3 and VEGFR signaling should be reexamined in view of their potential impact on human colonic cancer cells themselves besides their classic actions on morphogenesis of the tumor vasculature, cancer growth, and metastasis.
| Acknowledgments |
|---|
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.
| Footnotes |
|---|
5 Nguyen et al., in preparation, and Abstract 2669 at the AACR 94th Annual Meeting (2003). ![]()
6 Nguyen et al., in preparation, and Abstract 2669 at the AACR 94th Annual Meeting (2003). ![]()
7 Nguyen et al., in preparation, and Abstract 2669 at the AACR 94th Annual Meeting (2003). ![]()
8 Nguyen et al., in preparation, and Abstract 2669 at the AACR 94th Annual Meeting (2003). ![]()
Received 7/ 9/04. Revised 10/ 5/04. Accepted 10/27/04.
| References |
|---|
|
|
|---|
-catenin in human colon cancer cells. Cancer Res 1995;55:47228.
and Stat3ß. Mol Cell Biol 1997;17:530716.[Abstract]
This article has been cited by other articles:
![]() |
Z. Dubeykovskaya, A. Dubeykovskiy, J. Solal-Cohen, and T. C. Wang Secreted Trefoil Factor 2 Activates the CXCR4 Receptor in Epithelial and Lymphocytic Cancer Cell Lines J. Biol. Chem., February 6, 2009; 284(6): 3650 - 3662. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Fan, M. J. Gray, N. A. Dallas, A. D. Yang, G. Van Buren II, E. R. Camp, and L. M. Ellis Effect of chemotherapeutic stress on induction of vascular endothelial growth factor family members and receptors in human colorectal cancer cells Mol. Cancer Ther., September 1, 2008; 7(9): 3064 - 3070. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. H. Aziz, H. T. Manoharan, D. R. Church, N. E. Dreckschmidt, W. Zhong, T. D. Oberley, G. Wilding, and A. K. Verma Protein Kinase C{varepsilon} Interacts with Signal Transducers and Activators of Transcription 3 (Stat3), Phosphorylates Stat3Ser727, and Regulates Its Constitutive Activation in Prostate Cancer Cancer Res., September 15, 2007; 67(18): 8828 - 8838. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. LeSimple, I. van Seuningen, M.-P. Buisine, M.-C. Copin, M. Hinz, W. Hoffmann, R. Hajj, S. L. Brody, C. Coraux, and E. Puchelle Trefoil Factor Family 3 Peptide Promotes Human Airway Epithelial Ciliated Cell Differentiation Am. J. Respir. Cell Mol. Biol., March 1, 2007; 36(3): 296 - 303. [Abstract] [Full Text] [PDF] |
||||
![]() |
Q.-D. Nguyen, S. Rodrigues, C. M. Rodrigue, C. Rivat, C. Grijelmo, E. Bruyneel, S. Emami, S. Attoub, and C. Gespach Inhibition of vascular endothelial growth factor (VEGF)-165 and semaphorin 3A-mediated cellular invasion and tumor growth by the VEGF signaling inhibitor ZD4190 in human colon cancer cells and xenografts. Mol. Cancer Ther., August 1, 2006; 5(8): 2070 - 2077. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. J. Cohen, R. B. Cohen, and N. J. Meropol Targeting Signal Transduction Pathways in Colorectal Cancer--More Than Skin Deep J. Clin. Oncol., August 10, 2005; 23(23): 5374 - 5385. [Abstract] [Full Text] [PDF] |
||||
![]() |
STAT3, TFF3, VEGFR in Colon Cancer Cells Cancer Res., March 15, 2005; 65(6): 2505 - 2505. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |