Cross-Talk Between Phospho-STAT3 and Plcg1 Plays a Critical Role in Colorectal Tumorigenesis

Published OnlineFirst August 12, 2011; DOI: 10.1158/1541-7786.MCR-11-0147

Molecular Cancer Signaling and Regulation Research

Cross-talk between Phospho-STAT3 and PLCg1 Plays a Critical Role in Colorectal Tumorigenesis

Peng Zhang1, Yiqing Zhao1,5, Xiaofeng Zhu2, David Sedwick3, Xiaodong Zhang6, and Zhenghe Wang1,4

Abstract Hyperphosphorylation at the Y705 residue of signal transducer and activator of transcription 3 (STAT3) is implicated in tumorigenesis of leukemia and some solid tumors. However, its role in the development of colorectal cancer is not well defined. To rigorously test the impact of this phosphorylation on colorectal tumorigenesis, we engineered a STAT3 Y705F knock-in to interrupt STAT3 activity in HCT116 and RKO colorectal cancer cells. These STAT3 Y705F mutant cells fail to respond to cytokine stimulation and grow slower than parental cells. These mutant cells are also greatly diminished in their abilities to form colonies in culture, to exhibit anchorage- independent growth in soft agar, and to grow as xenografts in nude mice. These observations strongly support the premise that STAT3 Y705 phosphorylation is crucial in colorectal tumorigenesis. Although it is generally believed that STAT3 functions as a transcription factor, recent studies indicate that transcription-independent functions of STAT3 also play an important role in tumorigenesis. We show here that wild-type STAT3, but not STAT3 Y705F mutant protein, associates with phospholipase Cg1 (PLCg1). PLCg1 is a central signal transducer of growth factor and cytokine signaling pathways that are involved in tumorigenesis. In STAT3 Y705F mutant colorectal cancer cells, PLCg1 activity is reduced. Moreover, overexpression of a constitutively active form of PLCg1 rescues the transformation defect of STAT3 Y705F mutant cells. In aggregate, our study identifies previously unknown cross- talk between STAT3 and the PLCg signaling pathways that may play a critical role in colorectal tumorigenesis. Mol Cancer Res; 9(10); 1418–28. 2011 AACR.

Introduction tumorigenesis, because its mitochondrial activity appears to be required for Ras-mediated tumor transformation Signal transducer and activator of transcription 3 (5, 6). Third, transgenic mice with keratinocytes expressing (STAT3) is thought to be an oncogene (1). Several lines constitutively active STAT3 develop hyperproliferative of evidence support such a premise. First, persistent STAT3 dermatologic disorders in vivo (7). Fourth, targeted deletion activation has been detected in leukemia and in a variety of of STAT3 in skin cells prevents epithelial cancers in mice solid tumors including breast, brain, pancreas, ovarian, and (8), and targeting STAT3 specifically in B and T cells squamous cell carcinomas of head and neck (SCCHN) prevents development of lymphomas and myelomas (9). cancers, and melanomas (2). Second, constitutively active Latent cytoplasmic STAT3 becomes activated through STAT3 transforms rat and mouse cells and dominant phosphorylation of Y705 by cytoplasmic nonreceptor negative STAT3 blocks Src-induced transformation in vitro tyrosine kinases including Janus-activated kinase (JAK) (3, 4). Interestingly, recent studies show that the mitochon- and Src (10). Phosphorylated STAT3 dimerizes through drial functions of STAT3 may be important factors in reciprocal Src Homology 2 (SH2)-phosphotyrosine interaction and accumulates in the nucleus (2). Therefore, STAT3 activates the transcription of a wide array of genes Authors' Affiliations: Departments of 1Genetics and Case Comprehen- 2 3 Bcl-XL sive Cancer Center, Epidemiology and Biostatistics, Medicine, School of including B-cell lymphoma-extra large ( ) and sup- Medicine, Case Western Reserve University; 4Genomic Medicine Institute, pressor of cytokine signaling 3 (SOCS3; ref. 2). Although Cleveland Clinic Foundation, Cleveland, Ohio; 5The Second Huaxi Hos- pital, Sichuan University, Chengdu; and 6College of Life Sciences, Wuhan the kinases that phosphorylate the Y705 residue of STAT3 University, Wuhan, China are well defined in epithelial and hematopoietic cells, the Note: Supplementary data for this article are available at Molecular Cancer phosphatases that specifically dephosphorylate pY705 have Research Online (http://mcr.aacrjournals.org/). received little attention. Corresponding Authors : Zhenghe Wang, Case Western Reserve Univer- We previously identified STAT3 as a direct substrate of sity, 10900 Euclid Avenue, WRB 3120, Cleveland, OH 44106-7285. Phone: protein tyrosine phosphatase receptor T (PTPRT; ref. 11). 216-368-0446; Fax: 216-368-8919; E-mail: [email protected] or Xiaodong Zhang, College of Life Sciences, Wuhan University, Luojia Hill, PTPRT is mutated in colon, lung, stomach, and skin Wuhan, P. R. China 430072. Phone/Fax: 86-27-68756606; E-mail: (melanoma) cancers (12). Moreover, PTPRT knockout mice [email protected] are highly susceptible to azoxymethane-induced colon doi: 10.1158/1541-7786.MCR-11-0147 tumors (13), indicating that PTPRT normally functions 2011 American Association for Cancer Research. as a tumor suppressor. Our finding that PTPRT specifically

1418 Mol Cancer Res; 9(10) October 2011

Cross-talk between STAT3 and PLCg1

dephosphorylates STAT3 at the Y705 residue supports a days. The geneticin resistant clones were then screened for critical role for regulation of STAT3 Y705 phosphorylation homologous recombination by 35 cycles of genomic PCR in colorectal tumorigenesis. Although STAT3 is implicated with primers derived from the neomycin resistance gene in oncogenesis of leukemia, skin, and head and neck cancers 50-GTTGTGCCCAGTCATAGCCG-30 and the up- (1), the impact of STAT3 Y705 phosphorylation in colo- stream region of the left homologous arm 50-TCA- rectal tumorigenesis has not heretofore been well defined. GTTTCTTGGCCCAAAGT-30. Confirmatory genomic Here we show that successful knock-in (KI) of the STAT3 PCR was also carried out with positive clones identified by Y705F mutant allele into 2 different colorectal cancer cell primers derived from the neomycin resistant gene (50- lines results in mutant colorectal cancer cells that are less TCTGGATTCATCGACTGTGG-30) and the down- tumorigenic both in vitro and in vivo. The results of this stream region of the right homologous arm (50-TA- study further suggest that modulation of tumorigenicity is at GGCGCCTCAGTCGTATCT-30). The DNA fragments least partially dependent on STAT3 cross-talk with phos- from the confirmatory PCRs were then sequenced to pholipase Cg1 (PLCg1) through effects on S1248 phos- ensure the presence of the mutant Y705F alleles. To phorylation. PLCg1 is a key signaling molecule that target the second allele with the same targeting virus, hydrolyzes phosphatidylinositol-4,5-biophosphate to gener- correctly targeted clones were infected with adenoviruses ate inositol-1,4,5-triophosphate (IP3) and 1,2-diacylglycerol expressing the Cre-recombinase to delete the drug selec- þ (DAG), which, in turn, activate intracellular Ca2 and tion marker. To select clones with successful deletion of protein kinase C (PKC) signaling pathways that are impli- the drug selection marker, 30 cycles of genomic PCR were cated in tumorigenesis (14). In support, we show that carried out to amplify an approximately 200-bp genomic colorectal cancer cells carrying STAT3 mutated inY705 also fragment in which the Lox P site was inserted (using exhibit reduced PKC activities. primers 50-GCAGATGGAGCTTTCCAGAC-30 and 50- CGCCTGGGAAGAAGAAAAC-30). The heterozygous Materials and Methods KI clones were infected with the same targeting virus to target the second allele and the neomycin resistance gene Cell culture was excised as described earlier. HCT116, RKO, and HEK 293T cells were obtained from the American Type Culture Collection. HCT116 and Plasmid transfection RKO colorectal cancer cells were maintained in McCoy 5A Cells were plated 1 day prior to transfection to achieve a media plus 10% FBS. HEK 293T cells were maintained in 70% confluence at the time of transfection. Plasmids were Dulbecco's modified Eagle's medium media plus 10% FBS. transfected with the Lipofectamine Transfection Reagent (Invitrogen; catalogue no. 18324-020) according to the Somatic cell gene targeting manufacturer's instructions. To transfect a T75 flask of Somatic cell gene targeting was conducted as described HEK 293 cells, 9 mg of plasmids were mixed with 54 mLof (15, 16). Briefly, a 1.3-kb fragment from intron 21 to intron Lipofectamine Transfection Reagent and 1.5 mL of Opti- 22 of the STAT3 locus containing the exon 22 sequences MEM (Invitrogen; catalogue no. 31985070). The trans- was amplified 25 cycles from genomic DNA, using primers fection mixture was then incubated with cells at 37C for 4 50-TGACCAACTAGTCTGCTTACTGAATGCGAAG- hours. Cells were subsequently cultured in normal medium TCACAG-30 and 50-TGACCACCGCGGAGGGTCCT- after washing once with Hank's balanced buffer. TTCTCATTCCCACCTTA-30. The coding sequences for Y705 were then mutated from TAC (Tyr) to TTC Western blot (Phe) by site-directed mutagenesis, using primers 50-TC- Cells were lysed in radioimmunoprecipitation assay CCAGGCGCTGCCCCATTCCTGAAGACCAAGTT- (RIPA) buffer with complete protease inhibitor mixture TATC-30 and 50-AATGGGGCAGCGCCTGGGA-30. and phosphatase inhibitors [50 mmol/L Tris-HCl (pH 8.0), This mutated fragment was used as the right homologous 0.5% triton X-100, 0.25% sodium deoxycholate, 150 arm and cloned in to an pAAV-Neo-Lox P vector with mmol/L sodium chloride, 1 mmol/L EDTA, 1 mmol/L restriction enzymes SpeI (left cloning site) and SacII (right sodium orthovanadate, 50 mmol/L NaF, 80 mmol/L cloning site). Another 1.2-kb fragment from intron 21 b-glycerophosphate, and 20 mmol/L sodium pyro- of the STAT3 gene was also amplified 25 cycles from phosphate]. Western blots were carried out essentially as genomicDNAastheleftarm,usingprimers50-TG- described (17). Antibodies used included anti-pY705 ACCACTCGAGTCCCGTCAACGCATTTCTAACT- STAT3 (catalogue no. 9139) antibody, anti-STAT3 anti- GTA-30 and 50-TGACCAGAATTCATCTGC CTCGG- body (catalogue no. 9145; Cell Signaling Technology), and CAGGACTGATTTGA-30. The targeting adeno-associat- the antibodies listed in the Supplementary Table S1. ed viruses (AAV) were packaged in 293T cells (a T75 flask at 70% confluence) by transfecting equal amounts of the Immunofluorescence staining targeting vector, pHelper and pRC plasmids (3 mgeach). Cells were seeded on glass cover slips and grown to 50% Viruses were harvested 72 hours posttransfection. confluence and serum starved for 18 hours. A subset of cells HCT116 and RKO cells were infected with the STAT3 was treated with 10 ng/mL of interleukin-6 (IL-6) for 30 KI targeting viruses and selected with geneticin for 20 minutes following by fixation with 4% paraformaldehyde

www.aacrjournals.org Mol Cancer Res; 9(10) October 2011 1419

Zhang et al.

for 30 minutes at room temperature. The fixed cells were Boyden chamber cell migration and invasion assay permeabilized with 0.2% Triton X-100 at room tempera- Cell migration and invasion assay was carried out as ture for 5 minutes and then blocked with Image-iT FX previously described (18). Transwell membranes (pore size signal enhancer (Invitrogen) at room temperature for 30 8.0 mm; Corning Inc.) were coated with either 50 mg/mL minutes. Immunofluorescent staining was done with anti- fibronectin, 12.5 mg/mL of collagen IV or 1 mg/mL of STAT3 antibody (Santa Cruz Biotechnology) and the Alexa Matrigel. The matrices were coated on membranes accord- 488–conjugated anti-rabbit secondary antibody (Invitro- ing to the manufacturer's instructions. Specifically, Matrigel gen). Nuclei were stained with 40,6-diamidino-2-phenylin- (BD Biosciences) was thawed overnight at 4C and then dole (DAPI; 1 mg/mL) at room temperature for 20 minutes. diluted in serum-free medium to 1 mg/mL, and 100 mLof Images were captured with a LSM 510 META confocal the diluted Matrigel was added to upper chamber of 24-well microscope (Carl Zeiss MicroImaging). and incubated at 37C for 2 hours. Cells were detached with 2 mmol/L EDTA and resuspended in serum-free medium Flow cytometry with 0.1% of bovine serum albumin. Five hundred thou- Cell were harvested during the log phase of growth and sand cells were added to the upper compartment of the fixed with methanol. Cells were then incubated at 37C for transwell chamber in the wells of a 24-well plate and allowed 30 minutes in 5% normal goat serum diluted in PBS. to migrate to the underside of the inserts for 24 hours. Propidium iodide (PI) solution (100 mg/mL; 0.1% Complete medium containing 10% FBS was added in the NP40; 0.1% sodium azide) was used to stain cells at lower compartment as a chemoattractant. Nonmigrating 4C for 1 hour. Cells were analyzed on an Epics XL flow cells on the upper membrane were removed with a cotton cytometer (Beckman Coulter). WinMDI2.9 was used for swab, and cells that had migrated and become attached to data analysis. Cell debris and aggregates were excluded on PI the bottom surface of the membrane were fixed and stained – gating. Percentages of G1, S, and G2 M populations were with crystal violet. Migrated cells were counted microscop- determined by histograms generated by WinMDI2.9. ically at 200 magnification. Five randomly chosen fields were counted for each transwell membrane. The experi- Reverse transcriptase-PCR ments were repeated 3 times with 2 replicates for each cell Total RNAs were isolated from 1 million cells by using line. Average numbers of migrated cells per field from each the Qiagen RNeasy Mini Kit according to the man- experiment were plotted. ufacturer's instructions and cDNAs were synthesized by the SuperScript III First-Strand Synthesis System (Invitro- Xenograft gen). The 50-TCCCAGAAAGGATACAGCTGG-30 and Five million cells were injected subcutaneously and 50-ACTGAAGAGTGAGCCCAGCAG-30 primers were bilaterally into 4- to 6-week-old female nude mice (5 nude used to PCR amplify Bcl-XL. The 50-AGCTGG- mice in each group). Tumor formation and size were TACTCGCTCTTGGA-30 and 50-AGGCTCCTTTGT- assessed by weekly caliper measurements of the length GGACTTCA-30 primers were used to amplify SOCS3. and width of the tumors. Tumor volumes were calculated The PCR products were resolved on 1% agarose gel. PCR by the formula: Volume ¼ (width)2 length/2. After 21 conditions: Platinum Taq polymerase (Invitrogen; cata- days, the mice were sacrificed and tumors were harvested. logue no. 10966083) was used and the PCR mixtures were prepared according to the manufacturer's instructions; the Construction of PLCg1-myc tag and PLCg1 mutant mixtures were run on a GeneAmp PCR system 9700 plasmid (Applied Biosystems) at 94C for 30 seconds, 55C for To facilitate molecular cloning, we constructed a pCMV- 30 seconds, 70C for 30 seconds for 25 cycles. USER-3xMyc vector by inserting a USER cassette into the pCMV-3Tag-2A plasmid (Agilent Technologies). Briefly, Colony formation assay primers 50-AATTCGATATCGCTGAGGTCCCATCTA- HCT116 and RKO cells were trypsinized, counted twice GAGGATCCTCTAGACTATGCCTCAGC-30 and 50-T- by using a hemocytometer, and placed into 6-well plates at CGAGCTGAGGCATAGTCTAGAGGATCCTCTAG- 400 cells per well. Cells were grown for 14 days before ATGGGACCTCAGCGATATCG-30 were annealed staining with Crystal Violet (Sigma). The experiment was together and cloned into the pCMV-3Tag-2A plasmid, repeated 3 times with 2 replicates each. Average numbers of using EcoRI and XhoI restriction enzymes. The open read- colonies from each experiment were plotted. ing frame of PLCg1 was PCR amplified from a human PLCg1 cDNA purchased from Open Biosystems, using Focus formation assay in soft agar primers 50-GGTCCCA(d)Utggcgggcgtcgcgaccccct-30 and HCT116 and RKO clones were trypsinized, counted 50-GGCATAG(d)Uttaaagagagcacttccaca-30 and then twice by a hemocytometer, and plated at 5,000 cells/mL in cloned into the pCMV-USER-3xMyc vector by using top plugs consisting of 0.4% SeaPlaque agarose (FMC the USER cloning system (New England Biolabs) according Bioproducts) and McCoy's 5A medium. After 30 days, to the manufacturer's instructions. The construct was the colonies were photographed and counted. The exper- sequenced to ensure no mutation was introduced by PCR. iment was repeated 3 times with 2 replicates each. Average Constitutively active PLCg1 Y509A, F510A, and numbers of colonies from each experiment were plotted. D1019L mutations were then introduced into the

1420 Mol Cancer Res; 9(10) October 2011 Molecular Cancer Research

Cross-talk between STAT3 and PLCg1

Myc-tagged PLCg1 plasmid by 2-step site-directed muta- amount of glutathione S-transferase (GST) beads were genesis. The first step created the PLCg1 Y509A and F510A treated identically as a control. double mutant, using primers 50-TGACCACCATGGG- CACACTCTCACCA-30 and 50-TGACCAAGATCTT- Statistical analyses GCTGCTAGTCAGAACGGCGGCGTGGGGATACC- All statistical analyses were carried out by the SAS AC-30. The second step generated the D1019L mutation in software (SAS Institute). We applied the t test to compare the above double-mutant plasmid, using primers 50-GGG- the means between 2 groups assuming unequal variances. TCATAATTGGAGGAGAGTAGCCTCTGGCCCTTA- For xenograft growth, we carried out MANOVA for GGG-30 and 50-CTCTCCTCCAATTATGACCCT. The repeated measurements to test whether there is an overall details of site-directed mutagenesis methods are described in difference in the tumor sizes by testing group difference as ref. 19. well as whether there was a difference in development of tumor sizes over time between the 2 groups by testing the Glutathione S-transferase–fusion protein pull-down interaction between time and group. Sequences encoding the 2 tandem SH2 domains of PLCg1 were amplified by PCR from PLCg1 cDNA, using Results primers 50-TGACCAGAATTCGAGAAGTGGTTCCA- CGGGAAG-30 and 50-TGGTCACTCGAGGTACAGG- Engineering STAT3 Y705F knock-in colorectal cancer GGGTGCTTCTCA-30. The PCR product was then cell lines cloned into pGEM-6p-1 (GE Healthcare), using EcoRI To rigorously test whether regulation of STAT3 Y705 and XhoI restriction enzymes. Recombinant proteins were phosphorylation is critical to colorectal tumorigenesis, we expressed and purified from Escherichia coli. Twenty mil- set out to engineer STAT3 Y705F KI colorectal cancer cell lion HCT116 cells were lysed in RIPA buffer with complete lines. The AAV targeting system was used to engineer the KI protease inhibitor mixture and phosphatase inhibitors (see cell lines because of its high homologous recombination Western blot section for the recipe) for STAT3 pull-down. frequency in somatic cells (20, 21). We first chose to knock GST–PLCg1–SH2 domain fusion protein (1 mg) bound in the Y705F STAT3 mutant allele into the human colon beads were incubated with the cell lysate at 4C for 1 hour. cancer cell line HCT116, because we had shown that The beads were washed and then boiled and the aliquots STAT3 can be activated by IL-6 in HCT116 cells (11) were analyzed by SDS-PAGE and Western blotting. Equal and because HCT116 had been widely used for successful

Figure 1. Engineering Y705F STAT3 KI colorectal cancer cell lines. A, diagram of the KI construct. B and C, PCR products of the KI clones after excision of the neomycin resistance gene by Cre-recombinase. P indicates parental cells; WT/KI indicates Y705F STAT3 heterozygous KI cells; KI/KI indicates Y705F STAT3 homozygous KI cells; arrow indicates the KI allele. D and E, parental, Y705F heterozygous and Y705F homozygous cells were starved for 18 hours and treated with or without IL-6 for 30 minutes. Western blots were carried out with pSTAT3 or STAT3 antibodies.

www.aacrjournals.org Mol Cancer Res; 9(10) October 2011 1421

Zhang et al.

gene targeting by homologous recombination (15, 20). The targeting strategy is outlined in the schematic diagram in Figure 1A. After the first round of gene targeting, 4 targeted clones were identified out of 192 geneticin resistant clones screened. To ensure the presence of the mutant allele, STAT3 exon 22 genomic PCR products of the targeted clones were DNA sequenced. Three of the 4 targeted clones harbored STAT3 Y705F mutant allele (Supplementary Fig. S1). Two clones were infected with adenovirus expressing Cre-recombinase to excise the neomycin resistance gene (Fig. 1B) and targeted for the second allele to generate homozygous KI clones. To confirm that there was no wild-type (WT) STAT3 allele expression in the homozygous Y705F STAT3 clones, the RT-PCR products of STAT3 were sequenced. As expected, both WT STAT3 alleles were replaced by the STAT3 Y705F mutants in the homozygous KI clones (Supplementary Fig. S1). Furthermore, Western blot analyses showed that STAT3 proteins were expressed in the homozygous KI cells but they remained unphosphorylated at residue 705 after IL-6 stimulation, whereas STAT3 proteins in the parental and heterozygous cells were heavily phosphorylated post–IL-6 stimulation (Fig. 1D). These data indicated that we had successfully engineered STAT3 Y705F KI cells. We chose 2 independently derived heterozygous and homozygous KI clones for in-depth analyses and both clones behaved similarly in all the studies described in the following text. To ensure that what we observe with HCT116 cells is not cell line specific, we used the same method to generate STAT3 Y705F mutant RKO cells (Fig. 1C and E). Figure 2. STAT3 Y705F mutant proteins fail to activate its target genes. STAT3 Y705F mutant failed to activate its target genes A, WT and STAT3 Y705F homozygous KI HCT116 colorectal cancer cells were starved for 18 hours and stimulated with IL-6 for 30 minutes. Our previous studies showed that IL-6 induces STAT3 Cells were fixed and stained with anti-STAT3 antibodies. Scale bar, 10 mm. translocation from the cytoplasm to the nucleus and B, WT and STAT3 Y705F homozygous KI HCT116 colorectal cancer cells activates transcription of its target genes, Bcl-XL and were starved for 18 hours and stimulated with IL-6 for the indicated times. SOCS3 in colorectal cancer cells (11). To delineate the Gene expression of STAT3 target genes, Bcl-XL and SOCS3, were role of STAT3 phosphorylation in the translocation pro- assayed by RT-PCR. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) served as the control. cess, current experiments examined whether IL-6 induces STAT3 Y705F mutant protein translocation. This translocation process was directly monitored in parental and homozygous STAT3 Y705F mutant HCT116 and RKO period, both HCT116 and RKO STAT3 Y705F homo- cells using immunofluorescence staining. As shown in zygous KI cells grew slower than their parental cells Figure 2A and Supplementary Figure S2A, STAT3 (Fig. 3A). Although the doubling times of the parental Y705F mutant proteins remain diffused on IL-6 stimu- HCT116 and STAT3 Y705F mutant cells were not lation, whereas WT STAT3 proteins accumulate in the significantly different (Supplementary Fig. S3A), the nucleus in parental cells. Then, we utilized gene expres- average doubling times of the RKO STAT3 Y705F sion analyses by RT-PCR to show that IL-6 fails to mutant clones increased by 4 hours in comparison with activate gene transcription of Bcl-XL and SOCS3,2of the parental cells (Supplementary Fig. S3A). Consistently, the STAT3 target genes, in STAT3 Y705F homozygous compared with WT cells, the STAT3 Y705 mutant RKO KI cells (Fig. 2B; Supplementary Fig. S2B). Taken to- cells had an elevated G1 population, whereas no cell cycle gether, these data showed that STAT3 Y705F mutant is profile difference was observed among the HCT116 unresponsive to IL-6 stimulation, and strongly suggest parental and mutant clones (Supplementary Fig. S3B). that STAT3 Y705 phosphorylation is critical for its Given that STAT3 is also involved in tumor invasion and activation in colorectal cancer cells. metastasis (22), we set out to determine whether the STAT3 Y705 mutation affects colorectal cancer cell mi- STAT3 Y705F mutant colorectal cancer cells grew gration and invasion. As shown in Supplementary Figure slower than parental cells in tissue culture S4, no significant difference was found between the When grown under normal tissue culture conditions parental and the homozygous KI cells on fibronectin, (McCoy's 5A supplemented with 10% FBS) over a 4-day collagen IV, and Matrigel matrices.

1422 Mol Cancer Res; 9(10) October 2011 Molecular Cancer Research

Cross-talk between STAT3 and PLCg1

Figure 3. STAT3 Y705F mutant colorectal cancer cells are less tumorigenic in vitro. A, 1 105 cells of indicated clones were plated in 6-well plates. Duplicate wells of each clone were counted daily for 4 consecutive days by using a hemocytometer. Each well was counted 3 times. Average cell numbers of each time point from 3 independent experiments are plotted. B, cells from indicated clones were plated in 6-well plates in triplicates. Cells were grown for 14 days and stained with crystal violet. Colony numbers were counted and plotted for each of the clones. C, colorectal cancer cells of the indicated clones were mixed in 0.4% soft agar and plated in 6-well plates in triplicates. Cells were grown for 30 days. Colony foci were counted and plotted for each of the clones. *, P < 0.05; **, P < 0.001, t test.

In vitro, STAT3 Y705F mutant colorectal cancer cells STAT3 Y705F mutant colorectal cancer cells were less are reduced in properties predicative of in vivo tumorigenic in vivo tumorigenicity Tumorigenicity of the KI cells was also tested in a more To test whether STAT3 Y705F mutant affects tumor- stringent in vivo model. For these studies, STAT3 Y705F igenicity-correlated responses in vitro,wecarriedout homozygous, heterozygous clones or the parental HCT116 colony formation and soft agar assays with the STAT3 and RKO cells were injected subcutaneously into nude mutant KI cells. Compared with the parental cells, mice. Tumor formation and size were assessed by weekly homozygous STAT3 Y705F KI HCT116 and RKO cells caliper measurements. After 21 days of growth, WT cells exhibited 3- to 5-fold (P < 0.001) reduced abilities to formed tumors in all mice injected, whereas each of the form colonies in colony-formation assay (Fig. 3B). Sim- STAT3 Y705F homozygous KI clones failed to form ilarly, homozygous STAT3 mutant colorectal cancer tumors in at least 1 of the 5 mice injected (Fig. 4A). cell clones formed approximately 3-fold (P < 0.001) less The average tumor volumes of STAT3 Y705F homozygous foci in soft agar assay than their WT counterparts KI clones were 10-fold smaller than those produced by the (Fig. 3C). Interestingly, all of the heterozygous KI clones parental cells for both the HCT116 (P < 0.001) and RKO also displayed significant (P < 0.05) reduction in colony (P < 0.0001) colorectal cancer cell lines (Fig. 4B). Further- numbers and soft agar foci with respect to WT cells more, the development of tumor sizes over time for STAT3 (Fig. 3B and C). Y705F homozygous KI clones was significantly slower than

www.aacrjournals.org Mol Cancer Res; 9(10) October 2011 1423

Zhang et al.

that of the parental cells for both the HCT116 also reduced in the STAT3 Y705F mutant cells (Fig. 5A; (P < 0.001) and RKO (P < 0.0001). Notably, the average Supplementary Fig. S5). tumor sizes of STAT3 Y705F heterozygous KI clones were We speculated that STAT3 might physically interact with also significantly (P < 0.05) less than those of parental cells PLCg1 leading to modulation of its activity. In support, (Fig. 4B). STAT3 was shown to form complexes with PLCg1 under both overexpression and physiologic conditions, as shown STAT3-modulated PLCg1 activity by reciprocal immunoprecipitations (Fig. 5B and C). In- The studies above showed that phosphorylation of the terestingly, the WT, but not the STAT3, Y705F mutant STAT3 Y705 residue plays a critical role in colorectal proteins interact with PLCg1. Furthermore, the STAT3 tumorigenesis. It was also desirable to gain insights into proteins immunoprecipitated by PLCg1 were phosphory- the effects of this phosphorylation on downstream signaling. lated (Fig. 5C). These results suggest that STAT3 phos- In these studies, we examined how the STAT3 Y705F KI phorylation may play a role in its physical interaction with affects phosphorylation of other signaling molecules in PLCg1. In support, the STAT3-PLCg1 interaction was colorectal cancer cells after IL-6 stimulation. It is well readily detected when cells were stimulated with IL-6 documented that IL-6 activates multiple well-characterized (Fig. 5D), but was barely detectable under starvation signaling pathways including Ras-MAPK, PI3K-AKT, and conditions. Moreover, this interaction is mediated by the PLCg (23). Therefore, we tested the phosphorylation status 2 SH2 domains of PLCg1, as recombinant PLCg1 SH2 of 26 sites on 17 proteins on the basis of their ability to domains pulled STAT3 down from HCT116 lysates respond to IL-6 stimulation (Supplementary Table S1). (Fig. 5E). Among these candidates, PLCg1 S1248 phosphorylation was consistently elevated in STAT3 Y705F mutant cells in Constitutively active PLCg1 rescued the colony comparison with the parental cells (Fig. 5A; Supplementary formation defect of STAT3 Y705F mutant cells Fig. S5). As S1248 phosphorylation negatively regulates The functional significance of the cross-talk between PLCg1 activity (24), this result suggests that PLCg1 is less STAT3 and PLCg1 was then tested by determining wheth- active in STAT3 Y705F mutant cells. In support, pPKC er constitutively active PLCg1 mutant could phenotypically levels, an immediately downstream target of PLCg1, were rescue STAT3 Y705 mutant cells. A PLCg1 triple mutant

Figure 4. STAT3 Y705F mutant colorectal cancer cells are less tumorigenic in vivo. Athymic nude mice were injected subcutaneously with cells from the indicated clones. Tumors sizes were measured weekly for 3 weeks. Mice were then sacrificed and tumors were harvested. A, tumors grown from HCT116 clones. B, average sizes of the indicated clones were plotted.

1424 Mol Cancer Res; 9(10) October 2011 Molecular Cancer Research

Cross-talk between STAT3 and PLCg1

Figure 5. STAT3 cross-talks with PLCg1. A, parental and STAT3 Y705F KI cells were stimulated with IL-6 for the indicated time. Cell lysates were blotted with the indicated antibodies. See Supplementary Figure S5 for quantification. B, HEK 293 cells were transfected with the indicated plasmids. Cell lysates were immunoprecipitated with either anti-Flag or anti-Myc antibodies and blotted with the indicated antibodies. Arrow indicates co-immunoprecipitated Flag-STAT3. C, cell lysates from parental and STAT3 Y705F KI cells were immunoprecipitated with antibodies against either STAT3 or PLCg1and blotted with the indicated antibodies. D, HCT116 cells were serum-starved for 16 hours and then stimulated with or without 10 ng/mL of IL-6 for 30 minutes. Cells were then lysed and immunoprecipitated with either anti-PLCg1 antibodies or control immunoglobulin G (IgG). Immunocomplexes were resolved on 8% SDS-PAGE and Western blots were carried out with the indicated antibodies. E, HCT116 cell lysates were incubated with 1 mg of either GST–PLCg1–SH2 domains or GST alone for 1 hour at 4C. The pull-down mixtures were resolved on SDS-PAGE and blotted with an anti-STAT3 antibody. F, STAT3 Y705F KI cells were transfected with either an empty vector or a vector expressing PLCg1 triple mutant (D1019L,Y509A and F510A). Colony formation assay was performed. *, P < 0.05; **, P < 0.001, t test. Western blot was carried out to show expression of myc-tagged PLCg1 mutant protein. construct (D1019L, Y509A, and F510A), which is well pectedly, extension of studies of the STAT3 Y705F mutant documented to be constitutively active (25), was introduced also shows that cross-talk between STAT3 and PLCg1 into STAT3 Y705F mutant HCT116 cells. As shown in signaling pathways may provide further mediating mechan- Figure 5F, the mutant PLCg1 partially rescued the colony isms that modulate colorectal tumorigenesis. formation defect of the STAT3 Y705F mutant cells, sug- PLCg1 is a major signal transducer of growth factor and gesting that PLCg1 is a critical downstream mediator of cytokine signaling (26). PLC hydrolyzes phosphatidylino- STAT3 oncogenic signaling. sitol-4,5-biophosphate to generate IP3 and DAG. Although þ IP3 modulates intracellular Ca2 signaling, DAG activates Discussion PKCs (26). Results of our studies show, for the first time, that STAT3 modulates PLCg1 activity. In this Using genetically engineered STAT3 Y705 KI colorectal regard, decreased PKC activities in STAT3 Y705F mutant cancer cells, we show that STAT3 phosphorylation can colorectal cancer cells suggest that STAT3 activates PLCg1 regulate the efficiency of colorectal tumorigenesis. Unex- (Fig. 5A; Supplementary Fig. S5). However, it remains to be

www.aacrjournals.org Mol Cancer Res; 9(10) October 2011 1425

Zhang et al.

þ determined whether STAT3 can regulate intracellular Ca2 mutated in colorectal cancers (11), this study clearly shows signaling. Furthermore, our data suggest that activation of that regulation of STAT3 Y705 phosphorylation plays a PLCg1 by STAT3 requires physical interaction between the critical role in colorectal tumorigenesis, because STAT3 2 proteins and that this interaction seems to be mediated by Y705F homozygous KI colorectal cancer cells grow slower STAT3 pY705 and the SH2 domains on PLCg1. than the parental cells both in tissue culture and in tumor Given that numerous studies suggest that PLCg1is xenograft models. Interestingly, STAT3 Y705F heterozy- involved in tumor progression (14), it is not surprising that gous KI cells also exhibit reduced tumorigenicity, suggesting overexpression of constitutively active PLCg1 can partially that STAT3 Y705F protein may act as either a dominant rescue the colony formation defect of STAT3 Y705F KI negative or haploid-insufficiency mutant. Although our colorectal cancer cells. Data presented here strongly suggest data cannot distinguish these 2 possibilities, several previous that a transcription-independent function of STAT3 is at studies suggest that STAT3 Y705F mutant has a dominant least partially involved in critical control of tumor trans- negative effect (4, 32, 33). formation and that the general belief that STAT3 mainly Finally, our data indicate that inactivation of STAT3 in functions through its transcriptional activity is an oversim- colorectal cancer cells slows down tumor growth both in plification. In this regard, recent studies show that the tissue culture and in tumor xenograft models (Figs. 3A and mitochondrial functions of STAT3 may be important 4). Interestingly, RKO STAT3 Y705F KI cells progress factors in tumorigenesis, because its mitochondrial activity slower through cell cycle, but no cell cycle defect was appears to be required for Ras-mediated tumor transfor- observed in HCT116 STAT3 Y705F mutant cells. It is mation (5, 6). Notably, however, our studies show that possible that HCT116 STAT3 mutant cells grow slower PLCg1 only partially rescues the defect in the STAT3 through other mechanisms (e.g., The cells may be less Y705F mutant cells suggesting that STAT3 transcriptional responsive to growth factor stimulation). Our study suggests activity plays an important role in regulating STAT30s that the STAT3 signaling pathway may be a good target for oncogenic functions. colorectal cancer therapy. It is conceivable that interference Our data indicate that STAT3 Y705F mutant proteins with STAT3 phosphorylation, dimerization, and DNA fail to accumulate in the nucleus on IL-6 stimulation (Fig. 2; binding processes or inhibition of its upstream kinase(s) Supplementary Fig. S2). However, the unphosphorylated could be exploited to inhibit the tumorigenic growth of STAT3 proteins can still be imported into the nucleus, as colorectal cancers. Indeed, peptides, peptidomimetics and shown by the observation that a significant portion of small chemical compounds that target STAT3 dimerization STAT3 Y705F mutant protein is localized in the nucleus have shown promising inhibition of in vitro growth of breast in both HCT116 and RKO colorectal cancer cells (Fig. 2). cancer cells as well as cancer cells from other tissue types (34, This observation is consistent with a previous study showing 35, 36). Recently, a combination of in silico approaches and that STAT3 nuclear importation is independent of tyrosine chemical screens led to identification of several small mole- phosphorylation and that the N-terminal coiled-coil do- cules that potently inhibit STAT3 activities and cause breast main of STAT3 is important for its nuclear importation cancer cell death in both in vitro and in vivo models (37, 38). (27). In fact, numerous studies show that STAT3 proteins Further, platinum compounds that block the DNA binding translocate and accumulate in the nucleus after cytokine and of phosphorylated STAT3 were also found to inhibit cancer growth factor stimulation (1). In this regard, the data from cell growth (39). Along the same line, Dr. Jennifer Grandis’ our laboratory show unequivocally that the nuclear accu- group has successfully used a decoy oligonucleotide that mulation of STAT3 is dependent on STAT3 Y705 phos- blocks the transcription activity of STAT3 to inhibit tumor phorylation. Although we showed that STAT3 Y705F growth of SCCHN xenografts which harbor persistently mutant proteins failed to activate IL-6–induced target active STAT3 (40, 41). Moreover, targeting STAT3 genes, Bcl-XL and SOCS3, our data do not exclude the activation by inhibiting upstream kinases with chemical possibility that the mutant STAT3 can activate transcrip- compounds also caused potent antitumor effect (42). Our tion of other non-canonic STAT3 target genes. Interest- data provide a strong rationale for exploring existing and ingly, recent studies showed that overexpression of the emerging STAT3 inhibitors alone and in combination as STAT3 Y705F mutant induces expression of genes that targeted therapy for colorectal cancers. Furthermore, our are distinct from those induced by pSTAT3 (28). STAT3 Y705F mutant clones and their parental cells should STAT3 activity is upregulated in various cancers includ- provide ideal reagents for testing the specificity of STAT3 ing colorectal cancers (2). Although STAT3 has not yet inhibitory compounds that target STAT3 Y705 phosphor- been found to be mutated in human cancers, kinases and ylation. phosphatases that regulate STAT3 Y705 phosphorylation are mutated in a variety of cancers that result in STAT3 Disclosure of Potential Conflicts of Interests activation. Activating mutations of JAK2 are found in the majority of myeloproliferative neoplasms (29), whereas No potential conflicts of interest were disclosed. inactivation mutations of phosphatases PTPRT and Acknowledgments PTPRD that dephosphorylate STAT3 are found in various solid tumors (12, 30, 31). Although our previous study We thank Drs. Sanford Markowitz and Chao Wang for helpful discussions and showed that STAT3 is a direct substrate of PTPRT that is critical reading of this manuscript.

1426 Mol Cancer Res; 9(10) October 2011 Molecular Cancer Research

Cross-talk between STAT3 and PLCg1

Grant Support Received March 31, 2011; revised August 1, 2011; accepted August 4, 2011; published OnlineFirst August 12, 2011.

This research was supported by grants from the NIH (R01-CA127590, R01-HG004722, R01-HG003054) and the V foundation.

References 1. Levy DE, Inghirami G. STAT3: a multifaceted oncogene. Proc Natl 21. Hirata R, Chamberlain J, Dong R, Russell DW. Targeted transgene Acad Sci U S A 2006;103:10151–2. insertion into human chromosomes by adeno-associated virus vec- 2. Bowman T, Garcia R, Turkson J, Jove R. STATs in oncogenesis. tors. Nat Biotechnol 2002;20:735–8. Oncogene 2000;19:2474–88. 22. Devarajan E, Huang S. STAT3 as a central regulator of tumor metas- 3. Bromberg JF, Wrzeszczynska MH, Devgan G, Zhao Y, Pestell RG, tases. Curr Mol Med 2009;9:626–33. Albanese C, et al. Stat3 as an oncogene. Cell 1999;98:295–303. 23. Hodge DR, Hurt EM, Farrar WL. The role of IL-6 and STAT3 in 4. Bromberg JF, Horvath CM, Besser D, Lathem WW, Darnell JE Jr. Stat3 inflammation and cancer. Eur J Cancer 2005;41:2502–12. activation is required for cellular transformation by v-src. Mol Cell Biol 24. Park DJ, Min HK, Rhee SG. Inhibition of CD3-linked phospholipase C 1998;18:2553–8. by phorbol ester and by cAMP is associated with decreased phos- 5. Wegrzyn J, Potla R, Chwae YJ, Sepuri NB, Zhang Q, Koeck T, et al. photyrosine and increased phosphoserine contents of PLC-gamma 1. Function of mitochondrial Stat3 in cellular respiration. Science J Biol Chem 1992;267:1496–501. 2009;323:793–7. 25. Everett KL, Bunney TD, Yoon Y, Rodrigues-Lima F, Harris R, Driscoll 6. Gough DJ, Corlett A, Schlessinger K, Wegrzyn J, Larner AC, Levy DE. PC, et al. Characterization of phospholipase C gamma enzymes with Mitochondrial STAT3 supports Ras-dependent oncogenic transfor- gain-of-function mutations. J Biol Chem 2009;284:23083–93. mation. Science 2009;324:1713–6. 26. Patterson RL, van Rossum DB, Nikolaidis N, Gill DL, Snyder SH. 7. Sano S, Chan KS, Carbajal S, Clifford J, Peavey M, Kiguchi K, et al. Phospholipase C-gamma: diverse roles in receptor-mediated calcium Stat3 links activated keratinocytes and immunocytes required for signaling. Trends Biochem Sci 2005;30:688–97. development of psoriasis in a novel transgenic mouse model. Nat 27. Corvinus FM, Orth C, Moriggl R, Tsareva SA, Wagner S, Pfitzner EB, Med 2005;11:43–9. et al. Persistent STAT3 activation in colon cancer is associated with 8. Chan KS, Sano S, Kiguchi K, Anders J, Komazawa N, Takeda J, et al. enhanced cell proliferation and tumor growth. Neoplasia 2005;7: Disruption of Stat3 reveals a critical role in both the initiation and the 545–55. promotion stages of epithelial carcinogenesis. J Clin Invest 2004; 28. Yang J, Chatterjee-Kishore M, Staugaitis SM, Nguyen H, Schlessinger 114:720–8. K, Levy DE, et al. Novel roles of unphosphorylated STAT3 in onco- 9. Chiarle R, Simmons WJ, Cai H, Dhall G, Zamo A, Raz R, et al. Stat3 is genesis and transcriptional regulation. Cancer Res 2005;65:939–47. required for ALK-mediated lymphomagenesis and provides a possible 29. Quintas-Cardama A, Kantarjian H, Cortes J, Verstovsek S. Janus therapeutic target. Nat Med 2005;11:623–9. kinase inhibitors for the treatment of myeloproliferative neoplasias 10. Darnell JE. Validating Stat3 in cancer therapy. Nat Med 2005;11: and beyond. Nat Rev Drug Discov 2011;10:127–40. 595–6. 30. Veeriah S, Brennan C, Meng S, Singh B, Fagin JA, Solit DB, et al. The 11. Zhang X, Guo A, Yu J, Possemato A, Chen Y, Zheng W, et al. tyrosine phosphatase PTPRD is a tumor suppressor that is frequently Identification of STAT3 as a substrate of receptor protein tyrosine inactivated and mutated in glioblastoma and other human cancers. phosphatase T. PNAS 2007;104:4060–4. Proc Natl Acad Sci U S A 2009;106:9435–40. 12. Wang Z, Shen D, Parsons DW, Bardelli A, Sager J, Szabo S, et al. 31. Solomon DA, Kim J-S, Cronin JC, Sibenaller Z, Ryken T, Rosenberg Mutational analysis of the tyrosine phosphatome in colorectal can- SA, et al. Mutational inactivation of PTPRD in glioblastoma multiforme cers. Science 2004;304:1164–6. and malignant melanoma. Cancer Res 2008;68:10300–6. 13. Zhao Y, Zhang X, Guda K, Lawrence E, Sun Q, Watanabe T, et al. 32. Kaptein A, Paillard V, Saunders M. Dominant negative Stat3 mutant Identification and functional characterization of paxillin as a target of inhibits interleukin-6-induced Jak-STAT signal transduction. J Biol protein tyrosine phosphatase receptor T. Proc Natl Acad Sci U S A Chem 1996;271:5961–4. 2010;107:2592–7. 33. Rivat C, Wever OD, Bruyneel E, Mareel M, Gespach C, Attoub S. 14. Wells A, Grandis JR. Phospholipase C-gamma1 in tumor progression. Disruption of STAT3 signaling leads to tumor cell invasion through Clin Exp Metastasis 2003;20:285–90. alterations of homotypic cell-cell adhesion complexes. Oncogene 15. Zhang X, Guo C, Chen Y, Shulha HP, Schnetz MP, LaFramboise T, 2004;23:3317–27. et al. Epitope tagging of endogenous proteins for genome-wide 34. Turkson J, Ryan D, Kim JS, Zhang Y, Chen Z, Haura E, et al. ChIP-chip studies. Nat Methods 2008;5:163–5. Phosphotyrosyl peptides block Stat3-mediated DNA binding activity, 16. Du Z, Song J, Wang Y, Zhao Y, Guda K, Yang S, et al. DNMT1 stability gene regulation, and cell transformation. J Biol Chem 2001;276: is regulated by proteins coordinating deubiquitination and acetyla- 45443–55. tion-driven ubiquitination. Sci Signal 2010;3:ra80. 35. Turkson J, Kim JS, Zhang S, Yuan J, Huang M, Glenn M, et al. Novel 17. Yu J, Becka S, Zhang P, Zhang X, Brady-Kalnay SM, Wang Z. Tumor- peptidomimetic inhibitors of signal transducer and activator of tran- Derived Extracellular Mutations of PTPRT/PTP{rho} Are Defective in scription 3 dimerization and biological activity. Mol Cancer Ther Cell Adhesion. Mol Cancer Res 2008;6:1106–13. 2004;3:261–9. 18. Miao H, Li DQ, Mukherjee A, Guo H, Petty A, Cutter J, et al. EphA2 36. Yu H, Jove R. The STATs of cancer–new molecular targets come of mediates ligand-dependent inhibition and ligand-independent pro- age. Nat Rev Cancer 2004;4:97–105. motion of cell migration and invasion via a reciprocal regulatory loop 37. Song H, Wang R, Wang S, Lin J. A low-molecular-weight compound with Akt. Cancer Cell 2009;16:9–20. discovered through virtual database screening inhibits Stat3 func- 19. Zhang P, Becka S, Craig SE, Lodowski DT, Brady-Kalnay SM, Wang tion in breast cancer cells. Proc Natl Acad Sci U S A 2005;102: Z. Cancer-derived mutations in the fibronectin III repeats of PTPRT/ 4700–5. PTPrho inhibit cell-cell aggregation. Cell Commun Adhes 2009;16: 38. Siddiquee K, Zhang S, Guida WC, Blaskovich MA, Greedy B, Lawr- 146–53. ence HR, et al. Selective chemical probe inhibitor of Stat3, identified 20. Kohli M, Rago C, Lengauer C, Kinzler KW, Vogelstein B. Facile through structure-based virtual screening, induces antitumor activity. methods for generating human somatic cell gene knockouts using Proc Natl Acad Sci U S A 2007;104:7391–6. recombinant adeno-associated viruses. Nucleic Acids Res 2004; 39. Turkson J, Zhang S, Palmer J, Kay H, Stanko J, Mora LB, et al. 32:e3. Inhibition of constitutive signal transducer and activator of transcrip-

www.aacrjournals.org Mol Cancer Res; 9(10) October 2011 1427

Zhang et al.

tion 3 activation by novel platinum complexes with potent antitumor 41. Xi S, Gooding WE, Grandis JR. In vivo antitumor efficacy of STAT3 activity. Mol Cancer Ther 2004;3:1533–42. blockade using a transcription factor decoy approach: implications for 40. Leong PL, Andrews GA, Johnson DE, Dyer KF, Xi S, Mai JC, et al. cancer therapy. Oncogene 2005;24:970–9. Targeted inhibition of Stat3 with a decoy oligonucleotide abrogates 42. Nam S, Buettner R, Turkson J, Kim D, Cheng JQ, Muehlbeyer S, et al. head and neck cancer cell growth. Proc Natl Acad Sci U S A Indirubin derivatives inhibit Stat3 signaling and induce apoptosis in 2003;100:4138–43. human cancer cells. Proc Natl Acad Sci U S A 2005;102:5998–6003.

1428 Mol Cancer Res; 9(10) October 2011 Molecular Cancer Research

Cross-talk between Phospho-STAT3 and PLCγ1 Plays a Critical Role in Colorectal Tumorigenesis

Peng Zhang, Yiqing Zhao, Xiaofeng Zhu, et al.

Mol Cancer Res 2011;9:1418-1428. Published OnlineFirst August 12, 2011.

Updated version Access the most recent version of this article at: doi:10.1158/1541-7786.MCR-11-0147

Supplementary Access the most recent supplemental material at: Material http://mcr.aacrjournals.org/content/suppl/2011/08/12/1541-7786.MCR-11-0147.DC1

Cited articles This article cites 42 articles, 22 of which you can access for free at: http://mcr.aacrjournals.org/content/9/10/1418.full#ref-list-1

Citing articles This article has been cited by 2 HighWire-hosted articles. Access the articles at: http://mcr.aacrjournals.org/content/9/10/1418.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mcr.aacrjournals.org/content/9/10/1418. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.