Decreased CHK protein levels are associated with Src ... - Nature

2 downloads 49 Views 339KB Size Report
Oct 15, 2007 - colon cancer cells results in inactivation of Src without phosphorylating Y530 at ..... the Cancer Council, Victoria, Australia (HCC). References.
Oncogene (2008) 27, 2027–2034

& 2008 Nature Publishing Group All rights reserved 0950-9232/08 $30.00 www.nature.com/onc

ORIGINAL ARTICLE

Decreased CHK protein levels are associated with Src activation in colon cancer cells S Zhu1, JD Bjorge1, HC Cheng2 and DJ Fujita1 1 Department of Biochemistry and Molecular Biology, and Southern Alberta Cancer Research Institute, University of Calgary, Calgary, Alberta, Canada and 2Department of Biochemistry and Molecular Biology, Bio21 Molecular Science and Biotechnology Institute, University of Melbourne, Parkville, Victoria, Australia

Src activation has been associated with colon cancers but the mechanism underlying Src activation is largely unknown. Csk-homologous kinase (CHK) can inhibit the kinase activity of certain Src kinase family members in vitro by phosphorylating the C-terminal tyrosine and by a non-catalytic mechanism. CHK was previously reported to be expressed primarily in brain and hematopoietic cells. We report herein that CHK is also expressed in normal colon cell lines. Furthermore, CHK protein levels are significantly decreased in various colon cancer cell lines and the decrease correlates with the increased specific activity of Src in these cell lines, while the level of the other Src inhibitory kinase, C-terminal Src kinase, is not significantly changed. CHK is also expressed in normal colon tissues but its expression level is decreased in colon cancer tissues collected from the same patients. Immunofluorescence microscopy shows that CHK colocalizes with Src in normal colon FHC cells. Overexpression of CHK in colon cancer cells results in inactivation of Src without phosphorylating Y530 at its C-terminus. In addition, CHK suppresses anchorage-independent cell growth and cell invasion of colon cancer cells. These results reveal a potentially important role for CHK in Src activation and tumorigenicity in colon cancer cells. Oncogene (2008) 27, 2027–2034; doi:10.1038/sj.onc.1210838; published online 15 October 2007 Keywords: CHK; src; colon cancer

Introduction Cancer of the colon and rectum is the second leading cause of cancer death in North America. The development of colon cancer has been characterized as a multistep process starting from the transformation of normal colonic epithelium to an adenomatous polyp Correspondence: Dr DJ Fujita, Department of Biochemistry and Molecular Biology, University of Calgary Health Sciences Centre, 3330 Hospital Dr, NW, Calgary, Alberta, Canada T2N 4N1. E-mail: [email protected] Received 23 October 2006; revised 30 August 2007; accepted 3 September 2007; published online 15 October 2007

and subsequently to an invasive cancer (Kinzler and Vogelstein, 1996). Genetic alterations in many genes have been linked to this neoplastic progression, and include the proto-oncogenes K-ras and tumor suppressor genes APC, DCC, P53 (reviewed in Fearon and Vogelstein, 1990) and MLH1 (Kuismanen et al., 2000). However, the mechanism of colon tumorigenesis still remains largely unknown. Src is a protein tyrosine kinase that plays important roles in cell survival, growth and proliferation. v-Src, a mutationally activated form of c-Src, is sufficient to transform cells (reviewed in Jove and Hanafusa, 1987). In breast cancer, colon cancer and pancreatic cancer, Src kinase activity has been found to be elevated (reviewed in Irby and Yeatman, 2000). Downregulating Src activity with antisense RNA in the colon cancer cell line HT-29 (Staley et al., 1997) or with siRNA in the colon cancer cell line SW48 (Zhu et al., 2007) reduces colony formation numbers in soft agar, demonstrating that elevated Src activity plays an important role in colon cancer tumorigenicity. The mechanisms underlying the disregulation of Src activity in colon cancers are largely unknown, except in a subset (12%) of advanced human colon cancers, in which a mutation at the C-terminus, causing an early translational termination at Y530, has been found to contribute to elevated Src activity (Irby et al., 1999). Src kinase activity is primarily regulated through the phosphorylation at a C-terminal negative regulatory tyrosine residue (Y530) (reviewed in Hunter, 1987). Phosphorylation at Y530 is catalysed by C-terminal Src kinase (Csk), which reduces Src kinase activity, while dephosphorylation is catalysed by several protein tyrosine phosphatases, which increase kinase activity (reviewed in Bjorge et al., 2000). Phosphorylation at other sites can also regulate Src kinase activity (reviewed in Levin, 2004). Binding to phosphorylated receptors including EGFR, Neu/HER2 and PDGFR or interaction proteins such as Sin, FAK and p130Cas via the Src SH2 domain and/or SH3 domain can also lead to Src activation (reviewed in Bjorge et al., 2000). Csk homologous kinase (CHK) is a protein tyrosine kinase that is reported to be expressed primarily in brain and hematopoietic cells (Chow et al., 1994b). CHK shares 53% amino acid identity with Csk. In vitro, CHK can phosphorylate Y530 of Src (Avraham et al., 1995;

Regulation of Src by CHK in colon cancer cells S Zhu et al

2028

Chong et al., 2004) and CHK can inhibit Src activity by a non-catalytic mechanism (Chong et al., 2004). The main objective of this current study was to explore the mechanisms of Src activation in colon cancer cells. We found that while Csk levels in various colon cancer cell lines did not differ significantly as compared with those in normal colon cells, CHK was expressed in normal colon cells and its protein levels were significantly decreased in colon cancer cells, correlating with the increased specific activity of Src in these cancer cells. CHK colocalized with Src in the cytoplasm, and inactivated Src in vivo without phosphorylating Y530 of Src. These results indicate that CHK protein levels are downregulated in colon cancer cells and this contributes to Src activation in these cells.

Results and discussion Csk protein levels in colon cancer cell lines The mechanism underlying Src activation in human colon cancers is largely unknown, except in a small subset (12%) of advanced human colon cancer specimens in which Src activation has been attributed by a truncation mutation in Src (Irby et al., 1999). We found that the specific activity of Src kinase (activity per molecule) was greatly increased in all seven of the colon cancer cell lines tested, including SW48, DLD-1, HCT 116, HT-29, Caco-2, COLO 201 and LS 174T, as compared to the normal colon cell lines FHC and CCD18Co, ranging from 5.2- to 18.7-fold (Zhu et al., 2007), indicating that Src is activated in these cancer cell lines. Since Csk is known to be an important regulator of Src kinase activity, we examined cellular Csk protein levels to determine if variations in Csk protein levels in the colon cancer cells might contribute to this Src activation (increased Src specific activity). Figure 1 shows that Csk protein levels in the cancer cells examined were very similar as compared with those in the normal cells (FHC and CCD-18Co). Hence, there was no correlation between Csk protein level and activation of Src in these colon cancer cells. Csk levels in colon cancer cell lines were examined by other groups previously, which reported different results. Cam et al. (2001) observed reduced Csk levels in several colon cancer cell lines, while Watanabe et al. (1995) reported

Figure 1 C-terminal Src kinase (Csk) protein levels in human colon cell lines. Five micrograms of lysate protein was analysed by western blotting with anti-Csk antibody and anti-tubulin antibody, respectively. The representative blots of three independent experiments are shown. Oncogene

elevated Csk levels in several colon cancer cell lines. These differences in Csk levels might be due to the use of different cell lines or in some cases technical differences in methodology. It is possible that in a subset of colon cancer cell lines that we have not examined, modulation of Csk protein levels might be involved in regulating Src kinase activity. However, our result did not rule out the possibility that other aspects of Csk modulation might be involved in the activation of Src in these colon cancer cell lines. These aspects include changes in kinase activity of Csk and subcellular localization of Csk. CHK is expressed in colon tissues Unlike CSK, which is ubiquitously expressed, CHK mRNA expression has been reported primarily in brain tissue and hematopoietic cells (Bennett et al., 1994; Chow et al., 1994b; Kuo et al., 1994; Sakano et al., 1994), and thus has been thought to be limited to these tissues and cells. Using reverse transcription (RT)–PCR, we examined CHK expression in mouse colon tissues as compared with brain and muscle tissues, as CHK is expressed in brain but not muscle (Chow et al., 1994b; Sakano et al., 1994). There are three splicing variants of murine CHK (Chow et al., 1994a; Ershler et al., 1995). We designed primers to amplify regions spanning multiple exons to prevent potential chromosomal DNA contamination in the RNA samples from affecting the results (Figure 2a, upper panel). Splicing variant 2 (798 bp) and splicing variant 3 (834 bp) were absent from colon tissues (Figure 2a, middle panel). However, surprisingly, a significant amount of CHK mRNA of splicing variant 1 (650 bp) was detected in the colon tissues (Figure 2a, middle panel). Western blotting using an antibody against CHK (CHK (C89)), made against the C-terminal segment (89 amino acids) of CHK, showed that the protein level of CHK in mouse colon tissue was comparable to that in mouse brain (Figure 2b). This newly described expression pattern of CHK suggests that CHK may have a wider tissue distribution than initially thought and may play a role in signaling in colon tissues/cells. CHK protein levels are downregulated in human colon cancer cell lines and human colon cancer tissues There are also three splicing variants of human CHK in the National Center for Biotechnology Information GenBank (Figure 3a), all sharing high amino-acid homology. We examined the levels of CHK protein in the human colon cell lines, using different CHK antibodies. Western blotting using Matk (N20), an antibody generated against the N-terminal segment of CHK splicing variants 1 (Figure 3a), revealed that the 56 kDa splicing variant 1 was expressed in normal colon CCD18-Co and FHC cells, with expression levels even higher than that found in K-562 cells (Figure 3b), a CHK-positive hematopoietic cell line (Sakano et al., 1994). However, its level was significantly reduced in all the colon cancer cell lines (Figure 3b). Lsk (C-20) from Santa Cruz Biotech. (Santa Cruz, CA, USA) was raised against a peptide corresponding to the C-terminal

Regulation of Src by CHK in colon cancer cells S Zhu et al

2029

Figure 2 Csk-homologous kinase (CHK) is expressed in mouse colon tissues. (a) Upper panel: the structure (Chow et al., 1994a; Ershler et al., 1995) of the mouse CHK splicing variants (arrows indicate the various locations of the different sense primers and the single antisense primer that were used to reverse transcription (RT)–PCR fragments of splicing variants 1 þ 2 and 3, respectively). Middle panels: mRNA levels of CHK in mouse colon, brain (positive control) and skeletal muscle tissues (negative control) were analysed by RT–PCR. Colon#1, RNA isolation using standard protocol; Colon#2, RNA isolation using adapted protocol including a proteinase K digestion. Lower panel: mRNA levels of actin (as loading control) were analysed by RT–PCR. These results are representatives of three independent experiments. (b) CHK protein level in mouse colon tissues. Twenty and ten micrograms of lysate protein were analysed by western blotting with anti-CHK antibody CHK (C89) and anti-tubulin antibody, respectively. A blot representative of three experiments is shown.

segment (16 amino acids) of splicing variants 1, 2 and 3 (Figure 3a). The Lsk (C-20) blot gave a similar expression pattern of bands and confirmed the above observation that CHK was expressed in normal colon cell lines and was greatly downregulated in colon cancer cell lines (Figure 3b). The Lsk (C-20) blot also indicated that the p52 isoform (splicing variant 3) was absent in all these colon cells. We also used the antibody CHK (C89) and obtained similar results as using Matk (N20) and Lsk (C-20) (data not shown), which further confirmed the above observation. RT–PCR shows that CHK mRNA was present in normal colon cell lines, FHC and CCD-18Co, but undetectable in all the colon cancer cell lines (Figures 3c and d). We also examined the expression levels of CHK in human colon cancer tissues. Of the six randomly

Figure 3 Csk-homologous kinase (CHK) is downregulated in colon cancer cell lines. (a) Recognition of human CHK splicing variants by various antibodies. Light gray represents identical amino-acid sequence while dark gray represents non-identical amino acids. (b) CHK protein levels in human colon cancer cells and normal colon cells (FHC and CCD-18Co). Ten micrograms of lysate protein was analysed by western blotting with CHK antibodies Lsk (C-20) and Matk (N20) and tubulin antibody, respectively. K-562 is a positive control of CHK (Bennett et al., 1994). This result is representative of three independent blots. (c) Structure of human CHK splicing variants. The National Center for Biotechnology Information identification of mRNA of splicing variants 1, 2 and 3 are NM139355, NM002378 and NM139354, respectively. Arrows indicate the conserved locations of sense and antisense primers used to reverse transcription (RT)–PCR all the splicing variants. (d) Analysis of the mRNA levels of CHK in colon normal and cancer cells by RT–PCR. This is representative of results from two independent experiments.

selected cancer samples, CHK levels were significantly lower than those in the adjacent normal tissues in five samples (Figure 4). These data further confirmed the expression of CHK in normal colon cells. Furthermore, the reduction of CHK protein levels in human colon cancer cells Oncogene

Regulation of Src by CHK in colon cancer cells S Zhu et al

2030

Figure 4 Csk-homologous kinase (CHK) is downregulated in human colon cancers. Ten and thirty micrograms of lysate protein of normal (N) colon tissues and cancer (C) colon tissues from six patients were analysed by western blotting with anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) antibody (control) and anti-CHK antibody Lsk (C-20), respectively. The representative blots of three independent experiments are shown.

suggested that CHK downregulation might contribute to Src activation in these cancer cells. CHK inhibits Src in vivo in colon cancer cells without affecting Y530 phosphorylation As a homologue of Csk, CHK has been shown to phosphorylate the C-terminal negative regulatory tyrosine of Lck (Chow et al., 1994b; Klages et al., 1994) and Lyn (Hirao et al., 1997, 1998; Chong et al., 2004), hematopoietic members of the Src family. However, Yes is neither phosphorylated nor inhibited in presence of CHK overexpression, demonstrating that the specificity of CHK for different Src family kinase members may vary (Yamaguchi et al., 2001). Recently, CHK was found to be able to inactivate Hck, another hematopoietic Src family member, by a novel non-catalytic mechanism that involves binding of CHK to Hck (Chong et al., 2004, 2006). CHK also inactivates Src (Chong et al., 2004; Davidson et al., 1997) and Fyn (Davidson et al., 1997) in vitro. When expressed in Cskdeficient mouse embryo fibroblasts, CHK is able to downregulate Src kinase activity (Davidson et al., 1997). To study whether CHK inactivates Src in vivo where Csk is not deficient, such as colon cells, and to find out whether this inactivation is via phosphorylation at Src Y530 or the non-catalytic mechanism, we examined the effect of CHK overexpression in colon cancer cells on activated Src. Under optimal conditions, DLD-1 was transfected with a relatively high efficiency (approximately 60%, determined by fluorescence activated cell sorting analysis of cells expressing Green Fluorescent Protein, following transfection of EGFP C1 vector in the control experiments) among several cancer cell lines we tested. As shown in Figure 5, overexpression of CHK reduced Src activity by 47%. Considering the transfection efficiency in these experiments, Src probably would be further inactivated upon complete transfection. Src Y419 phosphorylation was decreased, while CHK protein levels in Src immunoprecipitates and Src protein levels in CHK immunoprecipitates were significantly increased in CHK-overexpressed cells (Figure 5). As the overexpression of CHK did not cause a detectable change in Src protein level or the phosphorylation level at Y530 of Src (Figure 5), this decrease in Src activity was likely a result of the non-catalytic inhibition mechanism (Chong et al., 2004). This conclusion was further supported by the detection of stable Src–CHK complexes in co-immunoprecipitation experiments (Figure 5). Oncogene

Figure 5 Overexpression of Csk-homologous kinase (CHK) inactivates Src in DLD-1 cells. Transfected cells were lysed with NP-40 lysis buffer, and 20 and 5 mg of lysate protein were analysed by western blotting with anti-CHK antibody Lsk (C-20) and antitubulin, respectively. One hundred micrograms of lysate protein was immunoprecipitated with anti-Src antibody MAb327 and assayed for kinase activity using Src optimal peptide (a). Twenty micrograms of lysate protein was immunoprecipitated with MAb327 and then immunoblotted with anti-Src antibody Src2, anti-Y530-P antibody, anti-Y419-P antibody and anti-CHK antibody Lsk (C-20) or immunoprecipitated with Lsk (C-20) and then immunoblotted with MAb327 (b). The kinase assays were from three independent experiments and each blot is representative of three independent experiments.

CHK inhibits Src in normal colon cells To examine if CHK also affects Src kinase activity in normal colon cells, we used an RNA interference approach. The transfection efficiency of siRNA was very low (o20%) for FHC cells using various siRNA transfection reagents including Oligofectamine (Invitrogen, Carlsbad, CA, USA), Lipofectamine 2000 (Invitrogen), TransMessenger (Qiagen, Valencia, CA, USA) and siLentFect (Bio-Rad, Hercules, CA, USA), based on the fluorescence of cells transfected with Alexa488control siRNA and analysed by fluorescence activated cell sorting. However, in CCD-18Co cells, transfection efficiency with siLentFect was higher (95% cell fluorescence) and resulted in an approximately 50% reduction of CHK levels. As shown in Figure 6, this caused a 63% increase in Src kinase activity. siRNA targeting Csk knocked down Csk levels by approximately 50%, and caused a 70% increase in Src kinase activity (Figure 6). Knocking down both CHK and Csk in CCD-18Co cells caused a 97% increase in Src kinase activity. Therefore, CHK suppresses Src kinase activity and appears to act additively with Csk in the normal colon cells.

Regulation of Src by CHK in colon cancer cells S Zhu et al

2031

Figure 6 Reduction of Csk-homologous kinase (CHK) activates Src in CCD-18Co cells. Transfected cells were lysed using NP-40 lysis buffer, and 200 mg of lysate protein was immunoprecipitated with anti-Src antibody MAb327 and assayed for kinase activity using Src optimal peptide. Ten micrograms of lysate protein (5 mg for anti-Csk) was analysed by western blotting with anti-CHK antibody Lsk (C-20), Mab327, anti-tubulin and anti-Csk antibodies. The kinase assays were from three independent experiments and the blots representative of three independent experiments are shown.

CHK colocalizes with Src in normal colon cell FHC To understand further the mechanism of CHK inhibition on Src, we examined the subcellular localization of each protein. In epithelial FHC cells, CHK was expressed in the cytoplasm as well as in the nucleus (Figure 7a). Nuclear localization of ectopic CHK has previously been observed in monkey kidney fibroblast COS-1 cells, and overexpression of CHK induced multinucleation, which in combination with other data, suggested a potential role of CHK in chromosome dynamics (Yamaguchi et al., 2001). Nuclear staining of endogenous CHK has also been reported in normal human brain tissues (Kim et al., 2004). It appears that nuclear CHK induces nuclear multilobulation in late S-phase and inhibition of proliferation (Nakayama and Yamaguchi, 2005). Our observation of nuclear staining in normal colon epithelial cells is consistent with these reports. However, our deconvoluted microscopy largely separated DNA staining from CHK staining in the nucleus. This suggested that the impact of CHK on chromosomes might be indirect (Figure 7a, inset). In particular, CHK appeared to be localized with relatively

Figure 7 Subcellular localization of Csk-homologous kinase (CHK) and Src in FHC cells. FHC cells were immunostained for CHK and Src and microscopic fluorescence images were deconvoluted. (a) Superimposing of CHK (red) and 4,6-diamidino-2phenylindole (DAPI) (blue) staining. Inset, same staining of the nucleus with high magnification. (b) Superimposing of Src (green) and DAPI (blue) staining. (c) Superimposing of CHK (red), Src (green) and DAPI (blue) staining. This shows a representative cell from three independent experiments.

high density in some unknown nuclear body that has yet to be identified. Figure 7b shows the subcellular localization of Src in the same cell. Src was localized primarily in the cytoplasm, with a small percentage nucleus-bound as well as membrane-bound, which is different from that of CHK. While well-known as a membrane-bound and cytoplasmic protein, Src has also been reported expressed in the nucleus of some cells (Isozumi et al., 1997; Krueger et al., 1980; Trubert et al., 1997). It would be interesting to determine the nature of the functional complexes containing Src in the nucleus in the colon FHC cells. The majority of CHK, as well as Src, were cytoplasmic in FHC cells (Figures 7a and b). Both cytoplasmic pools of CHK and Src appeared to be filamentous and perinuclear. This is consistent with a previous biochemical observation that CHK accumulates in the detergentinsoluble cellular fractions (Chow et al., 1994a), Oncogene

Regulation of Src by CHK in colon cancer cells S Zhu et al

2032

suggesting its possible association with the cytoskeleton. While the filamentous structures are yet to be identified, superimposing the high-resolution deconvoluted image of CHK and Src revealed their strong colocalization (Figure 7c). These data indicate that CHK probably binds to Src and inhibits its activity by the non-catalytic mechanism in normal colon cells. Perinuclear and filamentous localization of CHK was also observed in CCD-18Co, and staining of CHK was significantly weaker in control colon cancer cells that have low levels of CHK (Supplementary Figure 1). Since inactive Src has been shown to be tightly localized to the perinuclear region of the cell (Fincham and Frame, 1998; Fincham et al., 2000), it is possible that CHK is the main contributor that keeps Src in the inactive state in the perinuclear region. Src appears to be activated during stimulus-induced subcellular transit from the perinuclear region to the plasma membrane (Sandilands et al., 2004); it is possible this is due to an disruption of the CHK/Src interaction progressively along the route of subcellular transit. The Src-associated filamentous structure might be associated with RhoB and actin (Sandilands et al., 2004). We do not know the exact percentage of cytoskeleton-associated CHK in the cytoplasm of normal colon cells, thus experiments for quantitative determination of subcellular localization of free CHK versus cytoskeleton–CHK–Src complex will be helpful for further clarification of the model. CHK affects anchorage-independent growth and invasion of colon cancer cells While the functional role of Csk has been relatively well defined, less is known about CHK. Mice lacking the CHK gene have no apparent defects in the hematopoietic system (Samokhvalov et al., 1997). Our results indicate that both anchorage-independent cell growth (colony formation in soft agar) and cell invasion (Matrigel, San Jose, CA, USA) of CHK-overexpressed DLD-1 cells were significantly decreased as compared to control transfected cells (Figure 8). These results thus reveal a potential important role of CHK in the malignant properties of colon cancers. CHK has been studied in breast cancer cells (Zrihan-Licht et al., 1997; Bougeret et al., 2001; Kim et al., 2002). However, interestingly, CHK expression was detected in human breast cancer specimens, but not in normal human breast tissues, just the opposite to what we have observed in colon normal and cancer cells. Thus it is possible that CHK functions differently in breast cancer cells as compared with colon cancer cells. In fact, CHK does not negatively regulate Src in platelets (Hirao et al., 1997), suggesting the effects of CHK on Src might be tissue-/cell-type specific. In summary, we provide the first evidence that CHK is expressed and colocalizes with Src in normal human colon cells, and that CHK levels are downregulated in colon cancer cells, contributing to Src activation, possibly via a non-conventional allosteric mechanism, and contributing to the development of malignant properties of colon cancer cells. However, this does Oncogene

Figure 8 Csk-homologous kinase (CHK) overexpression inhibits anchorage-independent cell growth and invasion of DLD-1 cells. Soft agar colony assay (a) and Matrigel invasion assay (b) were initiated 12 h after transfection. The number of colonies formed in agar and cells invaded through Matrigel were counted under microscope and plotted against control. Over 300 colonies appeared on each soft agar plate, and over 100 cells appeared on each invasion membrane lower surface. Results were from experiments done in triplicate. Representative microscopic fields of soft agar plates are shown.

not exclude other possible mechanisms of Src activation in colon cancer cells. Other data from our lab suggest that elevated PTP1B activity might also contribute to Src activation in some colon cancer cells (unpublished data). Recently, the traditional multistep model of colon cancer development (reviewed in Fearon and Vogelstein, 1990) has been revised to include many more gene mutations than before (reviewed in Ilyas et al., 1999). The heterogeneous pattern of tumor mutations suggests that multiple alternative genetic pathways and/or other mechanisms that lead to the development of colorectal cancer exist, and that the traditional model of cancer development might not be fully representative of the majority of colorectal tumors (Smith et al., 2002). Among the potential mechanisms, elevation of Src activity has been consistently associated with colon cancers. Here we have shown that CHK loss may be

Regulation of Src by CHK in colon cancer cells S Zhu et al

2033

responsible for the Src activation in the colon cancer cells. This new discovery will both improve our understanding of Src regulation and help in search for therapeutic strategies for the cure and/or prevention of colon cancers.

Materials and methods Cell culture, reagents and tissue samples See Supplementary Information for details. Western blotting Cells were lysed in RIPA lysis buffer supplemented with phosphatase and protease inhibitors, unless stated otherwise. Proteins were resolved by sodium dodecyl sulfate–polyacrylamide gel electrophoresis, transferred to membranes and immunoblotted with various antibodies. See Supplementary Information for details. RT–PCR Reverse transcription–PCR was performed using an OneStep RT–PCR kit (Qiagen) according to the manufacturer’s specification. See Supplementary Information for more details. Src kinase assay and co-immunoprecipitation Cells were lysed in NP-40 lysis buffer supplemented with phosphatase and protease inhibitors. Src kinase activity was measured in Src immmunoprecipitates using Src optimal peptide. Src immunoprecipitates and CHK immunoprecipitates were analysed by western blotting to examine co-immunoprecipitation of CHK and Src. See Supplementary Information for details.

Transfection of DNA and siRNA DLD-1 cells were transfected with plasmid using Lipofectamine 2000 (Invitrogen). CCD-18Co cells were transfected with siRNAs using siLentFect lipid reagent (Bio-Rad). See Supplementary Information for more details. Immunofluoresence Cells were fixed, permeabilized and immunostained with CHK and Src antibodies, followed by Cy3 and Alexa Fluor488 immunostaining. Images were deconvoluted using ImageJ software. See Supplementary Information for details. Soft agar colony assay Transfected DLD-1 cells were grown in growth medium containing 0.3% agarose for 10 days and the colonies containing more than 100 cells were counted with an inverted microscope. See Supplementary Information for more details. Cell matrigel invasion assay This was done by using BioCoat Matrigel Invasion Chamber (BD Biosciences, San Jose, CA, USA) and stained with DiffQuik stain, as per the manufacturer’s instructions. Acknowledgements We thank Joan Brugge for MAb327 antibody. This work was supported by grants from the Alberta Cancer Board (DJF), the Canadian Institutes for Health Research (DJF) and the National Cancer Institute of Canada (DJF), the National Health and Medical Research Council of Australia (HCC) and the Cancer Council, Victoria, Australia (HCC).

References Avraham S, Jiang S, Ota S, Fu Y, Deng B, Dowler LL et al. (1995). Structural and functional studies of the intracellular tyrosine kinase MATK gene and its translated product. J Biol Chem 270: 1833–1842. Bennett BD, Cowley S, Jiang S, London R, Deng B, Grabarek J et al. (1994). Identification and characterization of a novel tyrosine kinase from megakaryocytes. J Biol Chem 269: 1068–1074. Bjorge JD, Jakymiw A, Fujita DJ. (2000). Selected glimpses into the activation and function of Src kinase. Oncogene 19: 5620–5635. Bougeret C, Jiang S, Keydar I, Avraham H. (2001). Functional analysis of Csk and CHK kinases in breast cancer cells. J Biol Chem 276: 33711–33720. Cam WR, Masaki T, Shiratori Y, Kato N, Ikenoue T, Okamoto M et al. (2001). Reduced C-terminal Src kinase activity is correlated inversely with pp60(c-src) activity in colorectal carcinoma. Cancer 92: 61–70. Chong YP, Chan AS, Chan KC, Williamson NA, Lerner EC, Smithgall TE et al. (2006). C-terminal Src kinase-homologous kinase (CHK), a unique inhibitor inactivating multiple active conformations of Src family tyrosine kinases. J Biol Chem 281: 32988–32999. Chong YP, Mulhern TD, Zhu HJ, Fujita DJ, Bjorge JD, Tantiongco JP et al. (2004). A novel non-catalytic mechanism employed by the C-terminal Src-homologous kinase to inhibit Src-family kinase activity. J Biol Chem 279: 20752–20766. Chow LM, Davidson D, Fournel M, Gosselin P, Lemieux S, Lyu MS et al. (1994a). Two distinct protein isoforms are encoded by ntk, a csk-related tyrosine protein kinase gene. Oncogene 9: 3437–3448. Chow LM, Jarvis C, Hu Q, Nye SH, Gervais FG, Veillette A et al. (1994b). Ntk: a Csk-related protein-tyrosine kinase expressed in brain and T lymphocytes. Proc Natl Acad Sci USA 91: 4975–4979.

Davidson D, Chow LM, Veillette A. (1997). Chk, a Csk family tyrosine protein kinase, exhibits Csk-like activity in fibroblasts, but not in an antigen-specific T-cell line. J Biol Chem 272: 1355–1362. Ershler MA, Samokhvalov IM, Belyavsky AV, Visser JW. (1995). Genomic structure and alternative splicing of the murine bhk/ctk/ ntk gene. FEBS Lett 375: 50–52. Fearon ER, Vogelstein B. (1990). A genetic model for colorectal tumorigenesis. Cell 61: 759–767. Fincham VJ, Brunton VG, Frame MC. (2000). The SH3 domain directs acto-myosin-dependent targeting of v-Src to focal adhesions via phosphatidylinositol 3-kinase. Mol Cell Biol 20: 6518–6536. Fincham VJ, Frame MC. (1998). The catalytic activity of Src is dispensable for translocation to focal adhesions but controls the turnover of these structures during cell motility. EMBO J 17: 81–92. Hirao A, Hamaguchi I, Suda T, Yamaguchi N. (1997). Translocation of the Csk homologous kinase (Chk/Hyl) controls activity of CD36anchored Lyn tyrosine kinase in thrombin-stimulated platelets. EMBO J 16: 2342–2351. Hirao A, Huang XL, Suda T, Yamaguchi N. (1998). Overexpression of C-terminal Src kinase homologous kinase suppresses activation of Lyn tyrosine kinase required for VLA5-mediated Dami cell spreading. J Biol Chem 273: 10004–10010. Hunter T. (1987). A tail of two src’s: mutatis mutandis. Cell 49: 1–4. Ilyas M, Straub J, Tomlinson IP, Bodmer WF. (1999). Genetic pathways in colorectal and other cancers. Eur J Cancer 35: 1986–2002. Irby RB, Mao W, Coppola D, Kang J, Loubeau JM, Trudeau W et al. (1999). Activating SRC mutation in a subset of advanced human colon cancers. Nat Genet 21: 187–190. Irby RB, Yeatman TJ. (2000). Role of Src expression and activation in human cancer. Oncogene 19: 5636–5642. Oncogene

Regulation of Src by CHK in colon cancer cells S Zhu et al

2034 Isozumi T, Yamada E, Hazama F, Handa J. (1997). Altered expression of src tyrosine kinase after axotomy. Acta Neuropathol (Berl) 93: 438–442. Jove R, Hanafusa H. (1987). Cell transformation by the viral src oncogene. Annu Rev Cell Biol 3: 31–56. Kim S, Zagozdzon R, Meisler A, Baleja JD, Fu Y, Avraham S et al. (2002). Csk homologous kinase (CHK) and ErbB-2 interactions are directly coupled with CHK negative growth regulatory function in breast cancer. J Biol Chem 277: 36465–36470. Kim SO, Avraham S, Jiang S, Zagozdzon R, Fu Y, Avraham HK. (2004). Differential expression of Csk homologous kinase (CHK) in normal brain and brain tumors. Cancer 101: 1018–1027. Kinzler KW, Vogelstein B. (1996). Lessons from hereditary colorectal cancer. Cell 87: 159–170. Klages S, Adam D, Class K, Fargnoli J, Bolen JB, Penhallow RC. (1994). Ctk: a protein-tyrosine kinase related to Csk that defines an enzyme family. Proc Natl Acad Sci USA 91: 2597–2601. Krueger JG, Wang E, Garber EA, Goldberg AR. (1980). Differences in intracellular location of pp60src in rat and chicken cells transformed by Rous sarcoma virus. Proc Natl Acad Sci USA 77: 4142–4146. Kuismanen SA, Holmberg MT, Salovaara R, de la Chapelle A, Peltomaki P. (2000). Genetic and epigenetic modification of MLH1 accounts for a major share of microsatellite-unstable colorectal cancers. Am J Pathol 156: 1773–1779. Kuo SS, Moran P, Gripp J, Armanini M, Phillips HS, Goddard A et al. (1994). Identification and characterization of Batk, a predominantly brain-specific non-receptor protein tyrosine kinase related to Csk. J Neurosci Res 38: 705–715. Levin VA. (2004). Basis and importance of Src as a target in cancer. Cancer Treat Res 119: 89–119. Nakayama Y, Yamaguchi N. (2005). Multi-lobulation of the nucleus in prolonged S phase by nuclear expression of chk tyrosine kinase. Exp Cell Res 304: 570–581. Sakano S, Iwama A, Inazawa J, Ariyama T, Ohno M, Suda T. (1994). Molecular cloning of a novel non-receptor tyrosine kinase, HYL

(hematopoietic consensus tyrosine-lacking kinase). Oncogene 9: 1155–1161. Samokhvalov I, Hendrikx J, Visser J, Belyavsky A, Sotiropolous D, Gu H. (1997). Mice lacking a functional chk gene have no apparent defects in the hematopoietic system. Biochem Mol Biol Int 43: 115–122. Sandilands E, Cans C, Fincham VJ, Brunton VG, Mellor H, Prendergast GC et al. (2004). RhoB and actin polymerization coordinate Src activation with endosome-mediated delivery to the membrane. Dev Cell 7: 855–869. Smith G, Carey FA, Beattie J, Wilkie MJ, Lightfoot TJ, Coxhead J et al. (2002). Mutations in APC, Kirsten-ras, and p53–alternative genetic pathways to colorectal cancer. Proc Natl Acad Sci USA 99: 9433–9438. Staley CA, Parikh NU, Gallick GE. (1997). Decreased tumorigenicity of a human colon adenocarcinoma cell line by an antisense expression vector specific for c-Src. Cell Growth Differ 8: 269–274. Trubert CL, Bernard F, Hivroz C, Carlioz A, Fischer A, Cournot G. (1997). PP60c-src expression in osteoclasts from osteopetrotic children and in giant tumor cells. Eur J Histochem 41: 169–176. Watanabe N, Matsuda S, Kuramochi S, Tsuzuku J, Yamamoto T, Endo K. (1995). Expression of C-terminal src kinase in human colorectal cancer cell lines. Jpn J Clin Oncol 25: 5–9. Yamaguchi N, Nakayama Y, Urakami T, Suzuki S, Nakamura T, Suda T et al. (2001). Overexpression of the Csk homologous kinase (Chk tyrosine kinase) induces multinucleation: a possible role for chromosome-associated Chk in chromosome dynamics. J Cell Sci 114: 1631–1641. Zhu S, Bjorge JD, Fujita DJ. (2007). PTP1B contributes to the oncogenic properties of colon cancer cells through Src activation. Cancer Res 67 (in press). Zrihan-Licht S, Lim J, Keydar I, Sliwkowski MX, Groopman JE, Avraham H. (1997). Association of csk-homologous kinase (CHK) (formerly MATK) with HER-2/ErbB-2 in breast cancer cells. J Biol Chem 272: 1856–1863.

Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc).

Oncogene