SEARCH

SEARCH BY CITATION

Keywords:

  • colorectal cancer;
  • cell lines;
  • mutations;
  • wnt signaling;
  • signal therapies

Abstract

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Tumor-derived cell lines are indispensable tools for understanding the contribution of activated signaling pathways to the cancer phenotype and for the design and testing of targeted signal therapies. In our study, we characterize 10 colorectal carcinoma cell lines for the presence of mutations in the wnt, Ras/MAPK, PI3K and p53 pathways. The mutational spectrum found in this panel of cell lines is similar to that detected in primary CRC, albeit with higher frequency of mutation in the β-catenin and B-Raf genes. We have monitored activation of the wnt and Ras/MAPK pathways in these cells and analyzed their sensitivity to selective signaling inhibitors. Using β-catenin subcellular distribution as a marker, we show that cells harboring APC mutations have low-level activated wnt signaling, which can be blocked by the extracellular wnt inhibitor DKK-1, suggesting autocrine activation of this pathway; proliferation of these cells is also blocked by DKK-1. In contrast, cells with β-catenin mutations are unresponsive to extracellular wnt inhibition. Constitutive phosphorylation of MAPK is present in the majority of the cell lines and correlates with B-Raf but not K-Ras mutations; correspondingly, the proliferation of cells harboring mutations in B-Raf, but not K-Ras, is exquisitely sensitive inhibition of the MAPK pathway. We find no correlation between PI3K mutation or loss of PTEN expression and increased sensitivity to PI3K inhibitors. Our study discloses clear-cut differences in responsiveness to signaling inhibitors between individual mutations within an activated signaling pathway and suggests likely targets for signal-directed therapy of colorectal carcinomas. © 2009 UICC

The progression of colorectal cancer from adenomas to carcinomas is characterized by a series of genetic changes, which lead to the activation of oncogenes and to the inactivation of tumor suppressor genes, perturbing the self-renewal, proliferative, adhesive and survival characteristics of the cells.1, 2 Colorectal cancers are characterized by mutations in 4 major pathways: wnt, Ras/MAPK, p53 and DNA repair.

The APC gene is a key regulator of the wnt pathway. APC mutations are detected early in the adenoma-carcinoma progression,3 are the hallmark of Familial Adenomatous Polyposis (FAP) and also occur with high frequency (up to 80%) in sporadic CRC. Mutations of the APC gene cluster in the central region of the protein and result in truncations, which delete most of the β-catenin binding sites, the axin-binding domain and the microtubule-binding regions. These deletions have been associated with activation of the canonical wnt signaling pathway and cytosolic/nuclear accumulation of β-catenin4 but may also play an important, β-catenin independent, role in the loss of cell polarity observed in many colorectal carcinomas.5 Underlying the relevance of the wnt pathway in the aetiogenesis of CRC, mutations of other components in the canonical wnt pathway also occur with significant frequency in primary CRC: β-catenin is mutated in 1–3%6, 7 and Axin-2 in 3–30% of CRC, with the highest frequency in MSI tumors.8, 9 The most frequent mutation of β-catenin found in CRC occurs at codon 45, a serine critical for priming phosphorylation, but mutations of the other phosphorylation sites also occur; these mutations abolish phosphorylation-mediated degradation of β-catenin leading to its cytosolic and nuclear accumulation.

Components of the Ras/MAPK and Ras/PI3K pathway are also frequently mutated in CRC. Mutations in the proto-oncogene K-Ras are present in ∼50% of colorectal cancers,10, 11 occurring with similar frequency in both familial and sporadic colorectal cancers.12 The proto-oncogene B-Raf, a serine/threonine kinase activated by Ras-GTP, is mutated in 10–15% of sporadic colorectal cancers.1, 13, 14 The most common B-Raf mutation, accounting for ∼90% of all B-Raf mutations in human cancers, occurs at codon 600, resulting in an amino acid change from the neutral valine (V) to the negatively charged amino acid glutamic acid (E) and in the constitutive activation of B-Raf.15 Although both K-Ras and B-Raf activate the MEK/MAPK pathway, K-Ras also directly activates the PI3K pathway. The enzyme PI3Kinase comprises a p110 catalytic subunit and a p85 regulatory subunit16 and mutations can occur in both subunits.17, 18 Mutations of PI3Kinase occur with relatively high frequency in CRC (2–30% [Refs.19 and20]) and can be coincident with K-Ras mutations.21 The PI3K pathway has been linked to tumor invasion and metastasis22 as well as to suppression of apoptosis.23

Finally, TP53, a gatekeeper gene that controls cell cycle arrest and apoptosis, is mutated in 20–40% of colorectal cancers.24, 25 Expression of functional p53 is important in the prevention of cancer development from precancerous lesions,26 and mutations of p53 frequently occur together with mutations of tumor-suppressor genes or of proto-oncogenes.24 Thus, in colorectal cancer, the 4 signaling pathways—wnt, Ras/MAPK, PI3K and p53—each contribute to important aspects of tumorigenesis and act in concert during tumor development.

Our increased understanding of the processes that lead to cancer development has shifted the focus of new cancer therapies to targeted interventions, which attempt to block selectively the pathways activated by the genetic mutations present in individual cancers.27, 28 Given the complexity of the genetic changes present in cancer, and the still incomplete knowledge of how different pathways interact to establish and maintain tumors, much of the development phase relies on “in vitro” model systems. Cell lines derived from primary or metastatic colorectal cancers can provide invaluable experimental tools for understanding the molecular events leading to tumor formation and for the manipulation of relevant signaling pathways. For these experimental systems to be meaningful, however, we need to assess whether the cell lines reflect the genetic characteristics of human CRC, whether the genetic mutations lead to “addiction” of the cells to constitutively active signaling pathways and whether the tumorigenic potential of the cells can be manipulated by interfering with the activated pathways.

CRC lines from the LIM series, which were established over a long period of time from primary colorectal cancers, have been previously characterized phenotypically29–31 and have been used extensively as models for colorectal cancer, with emphasis on the responsiveness to growth factors,32 epithelial-to-mesenchymal transition33 and induction of differentiation.34 We now report a detailed analysis of the mutations present in these cells, focusing on the pathways known to be altered in primary CRC. We have used available inhibitors to determine the role of signaling pathway activation in the proliferation and survival of the cells. Our results indicate that it is possible to pharmacologically revert inappropriate activation of signaling pathways driven by particular mutations, with concomitant reduction in the proliferation, survival and/or malignancy of the cells; however, not all the pathways mutated in CRC are responsive to specific signal therapeutics.

Material and methods

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Cell lines

LIM colorectal carcinoma cell lines were established from tumor biopsies at the Ludwig Institute for Cancer Research, Melbourne Branch (Ludwig Institute Melbourne) as described previously.17, 29, 30, 35, 36 The cell lines, their derivation and morphology are presented in Table I. LIM2463B cells were derived from the original LIM2463 cell line36 as described in Supporting Information. All LIM cell lines are routinely passaged in RPMI containing hydrocortisone (1 μg/ml), thyoglycerol (0.01 μg/ml), insulin (0.025 U/ml) and 10% fetal calf serum (FCS). LIM 1863 was cultured in medium containing 5% FCS.

Table I. Origin and Morphology of LIM Cell Lines
Cell lineSourceMS statusMorphologyRef
  1. Cells were derived from biopsies of colorectal adenocarcinomas or adenomas, as described in Supporting Information. Images were obtained with a Nikon TE2000-E, 20× lens.

LIM 1215Omental metastasis, adenocarcinoma of the ascending colonMSILoosely adherentchemical structure image29
LIM 1863Adenocarcinoma of the ileocaecal valveMSSFloating organoidschemical structure image30
LIM1899Columnar cell adenocarcinoma of the colonMSSAdherentchemical structure image35
LIM 2099Liver metastasis, moderately differentiated sclerosing adenocarcinomaMSSAdherentchemical structure image31
LIM 2405Poorly differentiated adenocarcinoma of the cecumMSIAdherentchemical structure image31
LIM 2408Moderately differentiated adenocarcinoma of the splenic flexureMSIAdherentchemical structure image31
LIM 2463BTubulovillous adenoma of the rectumMSSAdherent and Floating organoidschemical structure image36
LIM 2537Dysplastic polyp, transverse colonMSIAdherent and roundedchemical structure image17
LIM 2550Colon carcinomaMSIAdherent and loose aggregateschemical structure imageThis work
LIM2551Adenocarcinoma of the transverse colonMSILoose aggregateschemical structure imageThis work

Microsatellite status

Slides were prepared from 3-mm sections of cell blocks. Immunohistochemistry was performed on slides for MLH1 (BD Pharmingen, Franklin Lakes, NJ), MSH2 (Oncogene, Cambridge, MA) and MSH6 (BD Transduction Laboratories, Franklin Lakes, NJ) with Vision Biosystems, Norwell, MA, Bond-Max automated immunohistochemistry stainer using Vision Biosystems peroxidase polymer kit DS9713. Nuclear staining for each antibody was deemed normal; absence of nuclear staining was deemed an abnormal (MSI) phenotype.

DNA preparation, PCR and sequencing

The APC gene's exons 15 and 16 were sequenced in full to assess unequivocally the sites of mutation. Similarly, all β-catenin exons were sequenced. For each of the K-RAS, BRAF, PIK3CA, PIK3R1 and TP53 genes, we sequenced the exons which contain the common activating mutation: exons 2 and 3 for K-RAS, exons 11 and 15 for B-RAF, exons 1, 2, 4, 7, 9, 18 and 20 for PIK3CA (p110), exons 12 and 13 for PIK3R1 (p85) and exons 4–9 for TP53. Genomic DNA was extracted from ∼5 × 106 cells of each cell line using the DNeasy Blood & Tissue kit (Qiagen, Germantown, MD) as per manufacturer's instructions. The primer sequences for each region of each gene are listed in Supporting Information Tables I–IV. All primers were synthesized by Sigma Genosys (Castle Hill, NSW, Australia). PCR products were purified with QIAEX Gel Extraction Kit (Qiagen, Germantown, MD). DNA sequencing was conducted by the Welcome Trust sequencing Centre at Monash University, Melbourne, Australia, or in-house as detailed later.

APC gene expression and promoter methylation

Gene expression level of APC in each cell line was monitored by qRT-PCR. For each cell line, RNAs were prepared from 5 × 106 cells with RNeasy Plus mini kit (Qiagen, Germantown, MD). cDNA was synthesized from 2 μg RNA per sample using a high-capacity cDNA reverse transcription kit (Applied Biosystems, Foster City, CA). qRT-PCR was performed using the primers listed in Supporting Information Table IV. GAPDH was used as internal control. PCRs were carried out in a reaction volume of 25 μl using Power SYBR Green PCR Master Mix (Applied Biosystems). Samples were amplified in a 7300 Real-Time PCR system (Applied Biosystems) and data analyzed with ABI SDS software version 4.0 using the ΔΔCT method.

For the determination of APC promoter methylation, 6 μg of genomic DNA per sample was modified by sodium bisulfite treatment with EzWay DNA methylation detection express kit (TrendBio, Fitzroy North, VIC, Australia). APC Promoter methylation was monitored by real-time PCR using primers and probes described by others.37 Fluorogenic probes were custom synthesized by Applied Biosystems. PCR conditions and data were analyzed as described earlier. The data are presented as ratios between the gene of interest and the internal reference, myogenic differentiation antigen 1 (MYOD1).

TP53 high-resolution melting mutation detection and direct sequencing

PCR cycling and high-resolution melting (HRM) analysis were performed on the Rotor-Gene 6000 (Corbett Research, Sydney, Australia). Each sample was analyzed in duplicate. HRM analysis of exons 5–8 of TP53 was performed, as described previously.38 The latter part of Exon4, coding for beginning of the DNA binding domain, was analysed using primers TP53-Exon4-DBD-F, 5′-CCCCTGCACCAGCA GCTCCTA-3′ and TP53-Exon4-DBD-R, 5′-CAGCCCCTCAGGGCAACTGA-3′. To confirm the mutation positive HRM results, the HRM products of exons 4 and 7 were purified, sequenced in both directions and analyzed as described previously.38

Mutational analysis and sequencing of PI3K

Cell lines were screened for mutations in PIK3R1 (p85α) exons 12 and 13, and for PIK3CA (p110α) exons 1, 2, 7, 9, 18 and 20 by sequencing genomic DNA. Primers are listed in Supporting Information Table IV. PCR products were treated with ExoSap-IT (GE Healthcare, Buckinghamshire, England) according to the manufacturer's instructions and sequenced directly with the BigDye terminator method (Applied Biosystems) on an autosequencer (ABI PRISM 3100).

Immunofluorescence analysis of β-catenin distribution

Cells plated onto microchamber slides (LabTek, Brisbane, QLD, Australia) were cultured for 3 days in complete medium. On Day 4, the medium was replaced with 200 μl/chamber of medium supplemented with L-cell conditioned medium (CM), 30% v/v (control); wnt 3a L-cell CM, 30% v/v (wnt 3a); human recombinant DKK-1 (R&D systems), 1 μg/ml. Cells were incubated overnight before fixation in 4% formaldehyde/PBS, followed by permeabilization in 0.2%Triton X-100 in PBS. Cells were incubated in PBS containing 2% bovine serum albumin (BSA) for 1 hr at RT to block nonspecific binding. Cells were stained with anti β-catenin antibody (rabbit polyclonal H-102 from Santa Cruz Biotechnology) and anti E-cadherin mouse monoclonal antibody (BD Transduction Laboratories, Franklin Lakes, NJ), followed by incubation with Alexa-488 anti-rabbit Ig and Alexa-563 anti-mouse Ig (Invitrogen Molecular Probes, Carlsbad, CA). DAPI (0.1 μg/ml; Molecular Probes) was added in the last 10 min of incubation. Slides were dehydrated sequentially in ethanol and xylene and mounted with DPX mounting medium. Images were obtained on Nikon C1 confocal microscopes. Single and triple-label images were detected using standard filter sets and laser lines. Cells were imaged with Nikon Plan Apo 60× (NA1.4) oil immersion lens. Attribution of β-catenin distribution was made visually on an average of 60 cells from 3 randomly chosen fields.

Gel electrophoresis and Western blots

Cells in log-phase growth were harvested and lysed in deoxycholate lysis buffer (HEPES 20 mM, NaCl 150 mM, EDTA 5 mM, Triton X-100 1%v/v, sodium deoxycholate 1% w/v, β-glycerophosphate 5 mM and Na Vanadate 1 mM, protease inhibitor cocktail) at 4°C for 45 min. Cellular proteins (10 μg/lane) were separated on 4–12% Bis-Tris Novex gels (Invitrogen, Carlsbad, CA) and transferred electrophoretically to nitrocellulose membranes using an iBlot apparatus (Invitrogen, Carlsbad, CA). Membranes were blocked in 5% skim milk powder (5% w/v in PBS) for 1 hr before incubation with the antibodies: anti phospho-MAPK (Cell Signaling Technology, Danvers, MA), anti PTEN (Cell Signaling Technology) or anti-tubulin (Sigma, Castle Hill, NSW, Australia). Secondary antibodies were Odyssey anti-rabbit Ig IRDye 800 and Odyssey anti-mouse Ig IRDye 800 (Li-Cor Bioscience, Lincoln, NE). Nitrocellulose membranes were scanned on an Odyssey IR scanner and band intensities were quantitated by wide-line integration in ImageQuant.

Pathway inhibitors

Human recombinant DKK-1 was purchased from R&D (Minneapolis, MN) and resuspended at 100 μg/ml in RPMI medium containing human serum albumin (1% w/v). LY294002 (Sigma) was diluted to 50 mM in DMSO, and UO126 (Sigma) was diluted to 10 mM in DMSO.

Proliferation and inhibition assays

Cells in complete medium (RPMI+Adds, 10% FCS) were plated at 2 × 104/well in 96-well plates. An equal volume of inhibitors, serially diluted from the stock solutions in complete medium, was added to the cells. After 3 days in culture, MTT (Sigma) was added to each well; cells were harvested 4 hr later and processed according to the manufacturer's instructions. Optical density was measured on a spectrophotometer at 560λ. Data are plotted as percent inhibition, relative to control, after subtraction of the background.

Results

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Phenotype of LIM CRC cell lines

All cell lines were derived from tissue biopsies as previously described.29, 36 Table I summarizes the origin and characteristics of the cell lines chosen for our study. This panel contains a high proportion of cell lines with microsatellite instability (MSI). Although LIM 1215 and LIM 2551 were derived from HNPCC patients,39 the other cell lines were derived from sporadic colon cancer specimens. The original LIM 2463 cell line consists of floating organoids surrounded by large amount of mucus,36 which in the original studies could not be disaggregated without loss of viability and thus were not cloned. We now have derived viable single cells from LIM 2463 (as described in Supporting Information), which grow well in liquid culture and have a high cloning efficiency in soft agar, but no longer produce mucus. This derivative cell line (referred to as LIM 2463B) forms an adherent monolayer of spindly cells, with focal high-density patches from which arise spherical organoids (Supporting Information Fig. 1). Identity between the parental LIM 2463 and the derivative LIM 2463B has been confirmed by qRT-PCR on a panel of 82 wnt-responsive genes (data not shown) and the two cell lines have identical mutations for all genes tested (see later).

thumbnail image

Figure 1. APC mRNA expression and APC promoter methylation. (a) Expression of APC relative to housekeeper genes was measured by qRT-PCR in all the LIM cell lines. Data were analyzed with the ABI RQstudy software. LIM 1215 was arbitrarily chosen as the equalizer sample (expression level = 1). (b) APC promoter methylation was assessed by qRT-PCR as described in Material and methods. LIM 1215 was arbitrarily chosen as the equalizer sample (expression level = 1).

Download figure to PowerPoint

The lineage of the LIM cell lines was directly assessed by monitoring expression of the epithelial marker E-cadherin and of the mesenchymal marker Vimentin.40, 41 Consistent with the morphological evaluation of their epithelial nature, LIM 1215, 1863, 1899, 2405, 2408, 2537 and 2550 cells express E-cadherin (Supporting Information Fig. 2). LIM 2099 and LIM 2551 do not express either E-cadherin or vimentin, whereas LIM 2463B expressed vimentin, as detected by immunoblotting (Supporting Information Figs. 2a and 2b) and immunofluorescence (data not shown). The morphology, growth characteristics and expression of vimentin in LIM 2463B suggest that these cells have acquired a mesenchymal phenotype.

thumbnail image

Figure 2. Coincidence of mutations in CRC cell lines. Concurrence of mutations in APC (a) or β-catenin (b) with mutations of K-Ras and B-Raf; and of p53 mutations with APC/β-catenin (c) or with K-Ras/B-Raf (d). APC and β-catenin mutations are mutually exclusive, as are K-Ras and B-Raf mutations.

Download figure to PowerPoint

All but one of the LIM cell lines have relatively high cloning efficiency in soft agar (Supporting Table V). The exception is LIM 2099, which, as previously reported, fails to form colonies even at high cell density.31

Mutational analysis of LIM CRC cell lines

APC and β-catenin

The results of the genomic DNA sequencing for these 2 genes are summarized in Table II. Five of 10 cell lines have a mutated APC gene. We did not specifically address LOH in our samples; however, only 2 of the carcinoma-derived cell lines were homozygous for APC mutations (see Table II); the remainder carried both a wt and a mutated allele at the APC locus. The presence of full-length APC protein in cells wt for exons 15 and 16, and of both a truncated and full-length APC proteins in the heterozygous cell lines, was confirmed by SDS-PAGE and Western blotting. The APC mutations are single-base deletions in 3 cases, a 2-base insertion and a base change in 1 case each, all resulting in a premature stop codon and truncation of the APC protein. Most base changes occur within the mutation cluster region causing the characteristic loss of axin binding sites and maintenance of at least two 20 aa, β-catenin-binding repeats. Two cell lines (LIM 2405 and LIM 2408) carry an identical mutation which truncates the APC protein just before the microtubule-binding region but preserves the 2 SAMP repeats, allowing binding of axin. Cell lines LIM 2463B and LIM 2551 have a wt APC gene; however, we could not detect the APC protein by immunoblotting (data not shown) or at the mRNA level (Fig. 1, panel a). APC promoter methylation levels, measured using the methylation-specific qRT-PCR test,37 were strikingly elevated in these 2 cell lines (Fig. 1, panel b), suggesting epigenetic silencing of APC.

Table II. Mutational Analysis of LIM Cell Lines
 LIM 1215LIM 1863LIM 1899LIM 2099LIM 2405LIM 2408LIM 2463BLIM 2537LIM 2550LIM 2551
  1. Genomic DNAs encoding for APC, β-catenin, K-Ras, B-Raf, p53 and both subunits of PI3K were amplified by PCR and directly sequenced as described in Material and methods. The table shows the mutations detected and the expected alterations in the protein product.

APC
 DNA4464ΔAwt and 6579ΔAwt and 6579ΔAwt and 4421_4422Ins AG4630G >T
 ProteinwtStop at aa 1,506wtwtwt/stop at aa 2,198wt/stop at aa 2,198wtwt/ stop at aa 1,472stop at aa 1544wt
β-catenin
 DNA121A>G 350A/ 350A>C133T>CΔ133–135_______________Δ 244–455 (exon 3)
 ProteinT41A, Q177/Q177PwtS45PΔS45wtwtwtwtwtΔA5to A76
p53
 DNA700T>CDel734G>A734G>Awt/267_268 InsC
 ProteinwtY234HwtwtwtwtNullG245DG245Dwt/stop at aa 49
K-Ras
 DNAwt/G35CG34Twt/A183Cwt/A183C
 Proteinwtwtwt/G12AG12Cwtwtwt/Q61Hwtwt/Q61Hwt
B-Raf
 DNAT1799AT1799AT1799AT1799A
 ProteinwtwtwtwtV600EV600EwtV600EwtV600E
PI3K
 DNAPIK3R1 AG>AC int12/ex13 junctionPIK3CA A3140G
 Proteinwtwtwtwtwtwtwt23aa deletion in p85aH1047R p110awt

We found a surprisingly high incidence (4/10) of β-catenin mutations in our panel of CRC cell lines: this unusual frequency is unlikely to be due to the high proportion of MSI + samples in our series (Table I), since β-catenin mutations occur at a similar rate in MSS and MSI LIM cell lines. As expected, all mutations occurred in exon 3 and affected at least one of the key phosphorylation sites (Table II). The cell line LIM 1215 harbored a second, heterozygous mutation of β-catenin at Q177. This mutation occurs in the H3 helix of the first ARM repeat, and is of unknown consequence to β-catenin function. The β-catenin gene in LIM 2551 had an extensive deletion at the beginning of exon 3 and running into the following intron, with loss of the acceptor splice site resulting in ablation of exon 3. cDNA sequencing confirmed that the sequence, which fuses exon 2 to exon 4, is in frame (data not shown). The resultant protein has correspondingly smaller molecular mass and lacks all the critical phosphorylation sites. Mutations in APC and β-catenin were mutually exclusive. Cumulatively, genetic alterations of the canonical wnt pathway occur in 9 of the 10 LIM cell lines.

TP53

Five of the LIM cell lines carried TP53 mutations (Table II), i.e., at the upper limit of the reported frequency of TP53 mutations in colorectal cancer.42 There was no correlation between TP53 mutation and MSI status. Three of the TP53 mutations occurred in exon 7, in the region coding for the L3 loop of the DNA binding domain; one (LIM 2463B) was a homozygous deletion of TP53; and the 5th (LIM 2551) had a heterozygous frameshift in exon 4, which abolishes the DNA-binding domain.

K-Ras, B-Raf and PI3K

Sequencing results for these 3 genes are presented in Table II. All K-RAS mutations detected are activating mutations and occur in codons G12 and Q61, which have the highest reported mutation frequency in CRC. All B-Raf mutations involve codon 600, resulting in the activating change 600V > E. Four of the 10 cell lines had activating K-Ras mutations, and four had activating B-Raf mutation; K-Ras and B-Raf mutations were mutually exclusive (Table II). There is no clear-cut correlation between MSI status and Ras mutation; however, all B-Raf mutants are MSI+. Two LIM cell lines have PI3K mutations (Table II): LIM2537 has a point mutation in the splice recognition site of intron 12/exon 13 of PIK3R1, leading to loss of exon 13. This mutation in LIM 2537 has been previously reported17 and results in a 23-aa deletion (M582-N605) within the p110-binding region of p85. The H1047R mutation in LIM 2550 occurs at a mutational “hot spot” in PIK3CA (Catalogue of somatic mutations in cancer, Wellcome Trust: http://www.sanger.ac.uk/) and is likely to mimic RAS-GTP binding, thus activating the kinase independently of Ras.43

Coincidence of mutations in individual cell lines

Although the number of samples analyzed is far too small for a statistical analysis, there were interesting correlations between mutations. For example, APC mutations were often coincident with B-Raf mutation, whereas β-catenin mutation associated with K-Ras (Figs. 2a and 2b). Both APC and B-Raf mutations were more prevalent in MSI cell lines (4/5 and 4/4, respectively), whereas β-catenin and K-Ras mutations were equally represented in MSI and MSS cell lines. TP53 mutations were slightly more prevalent in our MSI cell lines, tended to be associated with mutations in APC rather than β-catenin and occurred both with K-Ras and with B-Raf mutations (Figs. 2c and 2d).

Inhibition of activated signaling pathways

We measured the functional effects of some of the mutations detected in LIM cell lines by monitoring baseline activation of the wnt, Ras and PI3K-mediated pathways. Because of the low levels of transfectability of many LIM cell lines, β-catenin subcellular distribution rather than a Tcf reporter assay was used as readout for wnt signaling activity. Activation of the Ras/Raf pathway and the PI3K pathway was assessed by immunoblotting for p-MAPK and p-Akt, respectively. We then utilized specific signaling inhibitors to assess the relevance of activated pathways for cell proliferation.

Differential activation of Wnt signaling in APC or β-catenin mutant cells

Subcellular distribution of β-catenin in control cells and in cells exposed to the stimulating ligand wnt3a44 or the extracellular wnt inhibitor DKK-145, 46 was assessed by immunofluorescence and confocal microscopy. Examples of immunofluorescent staining are presented in Figure 3a. Within each set, an overlay of the 3 stains (β-catenin: green; E-cadherin: red and DAPI: blue) is shown in panel a, and β-catenin staining alone in panel b. A summary of the results for all cell lines is shown in Figure 3b.

thumbnail image

Figure 3. Activation and inhibition of wnt canonical pathway. The effects of activation and inhibition of canonical wnt signaling were tested by immunofluorescence and proliferation assays. (a) β-catenin subcellular distribution was used as a surrogate marker of wnt signaling. Cells grown in multichamber slides were exposed to control medium, human recombinant DKK-1 (1 μg/ml) or wnt 3a (L-cell conditioned medium, 30% v/v) for 16 hr. Cells were fixed and stained with anti-E-cadherin (red), anti β-catenin (green) and DAPI (blue) as described in Material and methods. Slides were analyzed on a Nikon C1confocal microscope with a 60× oil lens. Within each group, the left panel (a) shows the 3-color composite image and the right panel (b) the β-catenin staining only. Size bars: 200 μM (top row), 20 μM (second and third row), 40 μM (bottom row). (b) Subcellular localization of β-catenin was visually assessed as membrane, cytosolic or nuclear in 3 randomly chosen fields for each slide, on a minimum of 20 cells per field. The table lists the relative abundance of membrane-associated (M), cytosolic (C) and nuclear (N) β-catenin staining. (c) The effects of DKK-1 on cell proliferation were monitored by the MTT assay. Cells were seeded in 96-well plates in full growth medium and exposed to increasing concentrations of hrDKK-1. Plates were harvested after 3 days and processed as described in Material and methods. At each point, the ratios of OD[560] of test/control were calculated and plotted. A Bolzman nonlinear fit to each set of data is shown. Each data point is the average of 3 replicate wells. The experiment is representative of 3 independent tests.

Download figure to PowerPoint

Cells carrying β-catenin mutations showed marked cytoplasmic β-catenin distribution, which was generally refractory to both wnt3a and DKK-1 (Fig. 3a). The exception in this set was LIM 2099, which, despite a deletion of the β-catenin Ser 45 phosphorylation site, showed a redistribution of β-catenin from diffusely cytosolic to perinuclear when exposed to DKK-1, and clear nuclear localization when stimulated with wnt3a (Fig. 3a, bottom panels). It is unclear whether the deletion, unlike the more common amino acid substitutions at this site, allows some phosphorylation of β-catenin to still occur. In cells carrying APC mutations, β-catenin was localized predominantly at the plasma membrane, albeit with varying degrees of cytosolic and (rarely) some accumulation in the nucleus (Figs. 3a and 3b). APC mutant cells were more amenable to both stimulation and inhibition of wnt signaling, as shown by the redistribution of β-catenin to the cytosol and nuclei with wnt3a, and the marked increase in membrane association with DKK-1. Exposure to DKK-1 also caused a clear change in morphology, with increased adhesion of the cells to the slides, increased formation of adherence junctions, and concomitant redistribution of E-cadherin (red) to the cell periphery (see for example Fig. 3a, centre panels). These data suggest that, in cells carrying an APC mutation, there is autocrine wnt signaling, which can be blocked by DKK-1; interestingly, cells with homozygous APC mutation show stronger evidence of autocrine wnt activation compared to cells with heterozygous mutations, at least as assessed by β-catenin distribution (cf. Fig. 3b and Table II).

We next tested whether autocrine, canonical wnt signaling played a role in the maintenance of cell proliferation. Cell proliferation was monitored by the MTT assay in control cultures and in cultures exposed to DKK-1 (Fig. 3c). The cell lines could clearly be divided in 2 sets: DKK-sensitive (Fig. 3c, left panel) and DKK-resistant (Fig. 3c, right panel). There was a clear correlation between the ability of DKK to redistribute β-catenin as monitored by immunofluorescence, and its effectiveness in reducing cell proliferation (cf. Figs. 3b and 3c); furthermore, the most sensitive cell lines were LIM 1863 and LIM 2550, which carry a homozygous APC mutation and show clear evidence of autocrine wnt signaling, as evidenced by cytosolic and nuclear β-catenin and a reversal of the distribution by DKK-1 (Fig. 3b). LIM 2463B, which have undetectable levels of APC protein due to APC promoter methylation, are also sensitive to DKK-1, implying that loss of APC expression is functionally equivalent in this context to homozygous APC truncaction. In contrast, cells carrying a wt as well as a mutant APC allele appear to be relatively resistant to DKK inhibition in both assays. Surprisingly, one of the cell lines with β-catenin mutation (LIM 2099) was also sensitive to DKK inhibition. In these cells, DKK-1 does not induce redistribution of β-catenin to the plasma membrane, possibly because of the low levels of E-cadherin (Supporting Information Fig. 2), but significantly reduces its levels in the nucleus. The LIM2099 cell line carries a unique mutation which results in deletion of serine 45 (Table II): further studies will be needed to determine whether this mutation has different functional outcomes from the more frequent amino acid substitution of Ser.45

Constitutive activation of the Ras/Raf signaling pathways and sensitivity to signal inhibitors

Activation of K-Ras or of B-Raf leads to MEK-dependent phosphorylation and activation of the downstream kinases Erk-1 and Erk-2. We used this parameter to assess the constitutive activation of the Ras/Raf pathway in LIM cell lines. Cells were harvested in log-growth under normal culture conditions, and specific Erk phosphorylation was determined on cell lysates by the ratios of phospho-Erk to total Erk proteins (Fig. 4, panel a). Some degree of constitutive Erk activation was reproducibly detected in 7 of the cell lines. Surprisingly, there was no clear correlation between K-Ras mutations and Erk activation: for example, both LIM 2463B and LIM 2550 have a heterozygous Q61H mutation in K-Ras, but have a 40-fold difference in specific activation of Erk; LIM 1899, which carries a heterozygous G12 mutation of K-Ras, has no significant Erk activation. In contrast, all the B-Raf mutants clearly showed constitutive Erk phosphorylation. To address the physiological significance of Erk activation, we utilized the specific inhibitor UO126,47 which blocks the interaction between MEK and Erk and thus prevents phosphorylation and activation of the latter. Cells in log-growth were exposed to increasing concentrations of the inhibitor, and their proliferation rate determined by the MTT assay. Inhibition of growth relative to untreated cells is shown in Figure 4, panel b. The LIM cell lines differ significantly in their sensitivity to UO126: 5 of the cell lines displayed extreme sensitivity to the inhibitor (IC50 < 0.5 μM), whereas the others were either less sensitive (IC50 > 2 μM) or resistant (LIM 2099 and LIM2550) to UO126. All B-Raf mutants, but none of the K-Ras mutants, fell in the sensitive group; the 5th sensitive cell line, LIM 1863, has no mutations in either K-Ras or B-Raf but has slightly elevated Erk phosphorylation: it is likely that this cell line either carries a mutation in other components of the Ras/MAPK pathway or that the MAPK pathway is activated through autocrine secretion of growth factors.

thumbnail image

Figure 4. Ras/Raf/PI3K activation and inhibition. (a) Constitutive activation of the MAPK pathway downstream of Ras and Raf was monitored by immunoblotting with phospho-specific antibodies to Erk-1 and Erk-2. Lysates from exponentially growing cells (10 μg/lane) were separated by SDS-PAGE on 4–12% Bis/Tris gels and transferred to nitrocellulose membranes. The membrane was probed with antibody to phospho-MAPK, stripped, and reprobed with antibodies to Erk-1 and Erk-2 total protein (left panels). Cells carrying a Ras mutation are indicated by an asterisk, and cells carrying B-Raf mutation by a triangle. Reactive bands were quantitated by wide-line integration on ImageQuant and the ratios of phospho-Erk to total Erk plotted (right panel). (b) Specific inhibition of MAPK activation by the inhibitor UO126 was determined by the MTT assay as described in the legend to Figure 3c. (c) PTEN protein levels in LIM cell lines were assessed by immunoblotting as described in Panel A. Immunoreactivity with PTEN antibody (top) and with tubulin antibodies (loading control; bottom) are shown in the left panel. The ratios of PTEN to tubulin were determined as described in Panel A and are shown in the right panel. (d) The effects on cell growth of inhibition of the PI-3-K pathway by LY294002 was determined by the MTT assay as described in the legend to Figure 3c.

Download figure to PowerPoint

PI-3K activation and inhibition

The most accepted readout for PI3K activation is phosphorylation of the serine/threonine kinase Akt. Phosphorylated Akt mediates the prosurvival properties of PI3K; however, PIP3-dependent phosphorylation of Akt is prevented by PTEN, a phosphatase that removes the 3′ phosphate from PIP3 and thus counteracts the effects of PI3K activation. We attempted to monitor Akt phosphorylation in LIM cell lines by immunoblotting, but failed to detect any constitutive phosphorylation of Akt (data not shown). Akt phosphorylation is acutely induced by stimulation of serum-starved cells, but may be difficult to detect in cells in log-growth. We also determined the expression of PTEN in the cell lines, as PTEN inactivation by either mutation or epigenetic silencing has been reported in CRC. PTEN appears to be mutated in only one LIM cell line (LIM 2099). The LIM 2099 PTEN protein migrates with higher molecular mass (Fig. 4, panel c); we have not identified this mutation at the genomic DNA level and we cannot say whether it has any physiological significance. PTEN expression was undetectable in 3 LIM cell lines (LIM 2405, 2408 and 2537); notably, LIM 2537 also harbors an activating mutation in the p85 subunit of PI3K and should therefore have greatly enhanced signaling through this pathway. We measured the sensitivity of the cells to a specific inhibitor of PI3K, LY 294002,48 using the MTT proliferation assay (Fig. 4d). All cell lines were inhibited by LY294002 within its reported range of activity (1–10 μM); the most sensitive to PI3K inhibition (IC50 < 2 μM) included the two PI3K mutant cell lines LIM 2537 and LIM 2550; however, there was no clear correlation between mutation status, including K-Ras mutations, and sensitivity to LY294002. Similar results were obtained using a second PI3K inhibitor, Wortmannin (data not shown).

Discussion

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Mutational spectrum in LIM cell lines

Our series of CRC cell lines shows an abnormally high frequency of microsatellite instability (Table I). Although 2 of the cell lines were derived from patients with Lynch syndrome (HNPCC), the age at diagnosis, sex and tumor location of other patients do not explain the higher MSI frequency. We have no longer have access to the primary tumors from which the cell lines were derived, thus cannot confirm the MSI status or presence of mutations in the original tumor; however, we have used cells from very early passage, making it unlikely that mutations have accumulated during cell culture. All the cell lines examined carried mutations or alterations in the principal pathways activated in CRC: APC, β-catenin, K-Ras, B-Raf, PI3K and p53.

Mutations in APC or β-catenin are mutually exclusive, as previously reported,49 and cumulatively occur in 9 of the 10 LIM cell lines. Furthermore, LIM 2463B, which has wt alleles at both the APC and β-catenin loci, shows loss of APC protein and mRNA because of promoter hypermethylation. Activating mutations in the Ras/MAPK pathway were also present with high frequency; in particular, B-Raf mutations were more prevalent than reported in sporadic CRC, but never coincident with K-Ras mutations. Given the high frequency of mutation in individual genes, it is not surprising that the vast majority of LIM cell lines carry mutations in multiple pathways (Fig. 2), thus mirroring the accumulation of mutations reported during the adenoma-carcinoma progression. However, the frequency of individual mutations was somewhat different from that reported in primary CRC: TP53 mutation is higher than expected, at the upper limit of the reported range42 and the prevalence of β-cat mutations is very high (4/10) but does not correlate with the MSI status. The combination of genetic events in the LIM cell lines is also divergent from that observed in primary tumors, including the coincidence of APC and B-Raf mutations, or of β-catenin and K-Ras mutations. This divergence may be apparent, because of the very limited sample number, or real, reflecting a possible selection of these combinations for continued growth in vitro; thus, permanent cell lines may represent a subset of the primary tumors with growth advantage in culture. Nonetheless, the LIM series provides a useful range of cell lines for the investigation of the relevance of each activated pathway to self-renewal, survival and invasion, thus providing an experimental tool for the assessment of signaling therapies.

Wnt pathway inhibition

Our results show a clear difference in β-catenin distribution between cells with APC mutations, and cells with β-catenin mutations: in the former, β-catenin maintains at least in part its association with the inner plasma membrane, whereas in the latter β-catenin is localized exclusively to the cytosol and nucleus. More importantly, cells with APC mutations retain responsiveness to stimulation and to inhibition of extracellular wnt signaling. Our results strongly suggest the existence of autocrine wnt signaling in APC mutant cells, which is reversed by the extracellular wnt inhibitor DKK leading to relocalization of β-catenin to the plasma membrane and to an increase in cell-to-cell adhesion (Fig. 3 and data not shown). The existence of autocrine wnt signaling, and its relevance to cell growth, is supported by the profound inhibition of cell proliferation caused by DKK-1 in LIM cell lines carrying a homozygous APC mutation (Fig. 3c). These results are consistent with the observation that soluble Frizzled-related proteins (sFRPs), which neutralize the wnt ligands, are often downregulated in colorectal cancer and cause apoptosis in CRC cell lines when reexpressed.50 However, in contrast to the results of Suzuki et al.,50 we failed to detect DKK-dependent inhibition of cell growth in the majority of cells carrying β-catenin mutations, consistent with the lack of redistribution of β-catenin and of changes in morphology: thus, in our experiments, extracellular signal inhibition appears specific for the APC mutant cell lines. One possible explanation for the conflicting results is that all our cell lines are homozygous for the β-catenin mutation, whereas the cells used by Suzuki et al. (HCT-116) are heterozygous, and thus may maintain some functional regulation of wnt signaling.

Inhibition of the MAPK and PI3K pathways

Activation of Ras leads directly to activation of the PI3K pathway51 and, through the Raf kinase, to sequential phosphorylation and activation of the Mek and MAPK kinases. Thus, mutational activation of K-Ras and B-Raf should be functionally equivalent in activating Mek/MAPK, but not in activating PI3K. We have used selective inhibitors of MEK or PI3K activity to dissect the relative contribution of these pathways to the growth of cells expressing mutated K-Ras or B-Raf. All cells with activating B-Raf mutations were exquisitely sensitive to inhibition of the MAPK pathway, while there was no correlation between K-Ras mutational status and sensitivity to UO126. These results, although counterintuitive, mirror the observation by Haigis et al.52 in a mouse model of colorectal cancer which lead the authors to conclude that “Mek is not an important mediator of K-Ras signaling during colon carcinogenesis.”52 In contrast, the sensitivity to MEK inhibitors of many cancers expressing activated B-Raf has been confirmed by numerous preclinical and clinical studies (reviewed in Ref.53). It appears that, although Mek is a direct target of B-Raf activation, other pathways activated by K-Ras are responsible for its cellular effects in carcinomas. However, PI3K activation either by mutation (Table II) or by silencing of PTEN (Fig. 4c) did not confer enhanced sensitivity to the PI3K inhibitor LY294002: the growth of all the cell lines was inhibited to some extent by this compound, but there was no correlation with mutational status.

In conclusion, our study shows clear-cut differences in responsiveness to signaling inhibitors between individual mutations within an activated signaling pathway: APC, but not β-catenin, mutant cells respond to inhibition of extracellular wnt signaling, and B-Raf, but not K-Ras, mutants respond to MeK inhibition. Obviously these results need to be further tested in animal models, but contribute to an understanding of the activated signaling pathways and provide a solid base for the assessment of potential signal therapy.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

The authors are grateful to Stephen Cody (Ludwig Institute, Melbourne) for his assistance with confocal microscopy, and Michael Krypuy for assistance with TP53 mutation testing.

References

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information
  • 1
    Kinzler KW,Vogelstein B. Lessons from hereditary colorectal cancer. Cell 1996; 87: 15970.
  • 2
    Hanahan D,Weinberg RA. The hallmarks of cancer. Cell 2000; 100: 5770.
  • 3
    Behrens J. The role of the Wnt signalling pathway in colorectal tumorigenesis. Biochem Soc Trans 2005; 33: 6725.
  • 4
    Morin PJ,Sparks AB,Korinek V,Barker N,Clevers H,Vogelstein B,Kinzler KW. Activation of β-catenin-Tcf signaling in colon cancer by mutations in β-catenin or APC. Science 1997; 275: 178790.
  • 5
    van den Brink GR,Offerhaus GJ. The morphogenetic code and colon cancer development. Cancer Cell 2007; 11: 10917.
  • 6
    Johnson V,Volikos E,Halford SE,Eftekhar Sadat ET,Popat S,Talbot I,Truninger K,Martin J,Jass J,Houlston R,Atkin W,Tomlinson IP, et al. Exon 3 β-catenin mutations are specifically associated with colorectal carcinomas in hereditary non-polyposis colorectal cancer syndrome. Gut 2005; 54: 2647.
  • 7
    Tan LP,Ng BK,Balraj P,Lim PK,Peh SC. No difference in the occurrence of mismatch repair defects and APC and CTNNB1 genes mutation in a multi-racial colorectal carcinoma patient cohort. Pathology 2007; 39: 22834.
  • 8
    Suraweera N,Robinson J,Volikos E,Guenther T,Talbot I,Tomlinson I,Silver A. Mutations within Wnt pathway genes in sporadic colorectal cancers and cell lines. Int J Cancer 2006; 119: 183742.
  • 9
    Thorstensen L,Lind GE,Lovig T,Diep CB,Meling GI,Rognum TO,Lothe RA. Genetic and epigenetic changes of components affecting the WNT pathway in colorectal carcinomas stratified by microsatellite instability. Neoplasia 2005; 7: 99108.
  • 10
    Akagi K,Uchibori R,Yamaguchi K,Kurosawa K,Tanaka Y,Kozu T. Characterization of a novel oncogenic K-ras mutation in colon cancer. Biochem Biophys Res Commun 2007; 352: 72832.
  • 11
    Noda H,Kato Y,Yoshikawa H,Arai M,Togashi K,Nagai H,Konishi F,Miki Y. Frequent involvement of ras-signalling pathways in both polypoid-type and flat-type early-stage colorectal cancers. J Exp Clin Cancer Res 2006; 25: 23542.
  • 12
    Oliveira C,Westra JL,Arango D,Ollikainen M,Domingo E,Ferreira A,Velho S,Niessen R,Lagerstedt K,Alhopuro P,Laiho P,Veiga I, et al. Distinct patterns of KRAS mutations in colorectal carcinomas according to germline mismatch repair defects and hMLH1 methylation status. Hum Mol Genet 2004; 13: 230311.
  • 13
    Fang JY,Richardson BC. The MAPK signalling pathways and colorectal cancer. Lancet Oncol 2005; 6: 3227.
  • 14
    Nagasaka T,Sasamoto H,Notohara K,Cullings HM,Takeda M,Kimura K,Kambara T,MacPhee DG,Young J,Leggett BA,Jass JR,Tanaka N, et al. Colorectal cancer with mutation in BRAF. KRAS, and wild-type with respect to both oncogenes showing different patterns of DNA methylation. J Clin Oncol 2004; 22: 458494.
  • 15
    Davies H,Bignell GR,Cox C,Stephens P,Edkins S,Clegg S,Teague J,Woffendin H,Garnett MJ,Bottomley W,Davis N,Dicks E, et al. Mutations of the BRAF gene in human cancer. Nature 2002; 417: 94954.
  • 16
    Carpenter CL,Cantley LC. Phosphoinositide kinases. Biochemistry 1990; 29: 1114756.
  • 17
    Philp AJ,Campbell IG,Leet C,Vincan E,Rockman SP,Whitehead RH,Thomas RJ,Phillips WA. The phosphatidylinositol 3′-kinase p85α gene is an oncogene in human ovarian and colon tumors. Cancer Res 2001; 61: 74269.
  • 18
    Campbell IG,Russell SE,Choong DY,Montgomery KG,Ciavarella ML,Hooi CS,Cristiano BE,Pearson RB,Phillips WA. Mutation of the PIK3CA gene in ovarian and breast cancer. Cancer Res 2004; 64: 767881.
  • 19
    Nosho K,Yamamoto H,Takahashi T,Mikami M,Taniguchi H,Miyamoto N,Adachi Y,Arimura Y,Itoh F,Imai K,Shinomura Y. Genetic and epigenetic profiling in early colorectal tumors and prediction of invasive potential in pT1 (early invasive) colorectal cancers. Carcinogenesis 2007; 28: 136470.
  • 20
    Samuels Y,Wang Z,Bardelli A,Silliman N,Ptak J,Szabo S,Yan H,Gazdar A,Powell SM,Riggins GJ,Willson JK,Markowitz S, et al. High frequency of mutations of the PIK3CA gene in human cancers. Science 2004; 304: 554.
  • 21
    Ollikainen M,Gylling A,Puputti M,Nupponen NN,Abdel-Rahman WM,Butzow R,Peltomaki P. Patterns of PIK3CA alterations in familial colorectal and endometrial carcinoma. Int J Cancer 2007; 121: 91520.
  • 22
    Samuels Y,Diaz LA,Jr,Schmidt-Kittler O,Cummins JM,Delong L,Cheong I,Rago C,Huso DL,Lengauer C,Kinzler KW,Vogelstein B,Velculescu VE. Mutant PIK3CA promotes cell growth and invasion of human cancer cells. Cancer Cell 2005; 7: 56173.
  • 23
    Wang J,Kuropatwinski K,Hauser J,Rossi MR,Zhou Y,Conway A,Kan JL,Gibson NW,Willson JK,Cowell JK,Brattain MG. Colon carcinoma cells harboring PIK3CA mutations display resistance to growth factor deprivation induced apoptosis. Mol Cancer Ther 2007; 6: 114350.
  • 24
    Samowitz WS,Slattery ML,Sweeney C,Herrick J,Wolff RK,Albertsen H. APC mutations and other genetic and epigenetic changes in colon cancer. Mol Cancer Res 2007; 5: 16570.
  • 25
    De Galitiis F,Cannita K,Tessitore A,Martella F,Di Rocco ZC,Russo A,Adamo V,Iacobelli S,Martinotti S,Marchetti P,Ficorella C,Ricevuto E. Novel P53 mutations detected by FAMA in colorectal cancers. Ann Oncol 2006; 17( Suppl 7): vii78vii83.
  • 26
    Halazonetis TD,Gorgoulis VG,Bartek J. An oncogene-induced DNA damage model for cancer development. Science 2008; 319: 13525.
  • 27
    Besse B,Ropert S,Soria JC. Targeted therapies in lung cancer. Ann Oncol 2007; 18( Suppl 9): ix135ix142.
  • 28
    Saal LH,Johansson P,Holm K,Gruvberger-Saal SK,She QB,Maurer M,Koujak S,Ferrando AA,Malmstrom P,Memeo L,Isola J,Bendahl PO, et al. Poor prognosis in carcinoma is associated with a gene expression signature of aberrant PTEN tumor suppressor pathway activity. Proc Natl Acad Sci USA 2007; 104: 75649.
  • 29
    Whitehead RH,Macrae FA,St John DJ,Ma J. A colon cancer cell line (LIM1215) derived from a patient with inherited nonpolyposis colorectal cancer. J Natl Cancer Inst 1985; 74: 75965.
  • 30
    Whitehead RH,Jones JK,Gabriel A,Lukies RE. A new colon carcinoma cell line (LIM1863) that grows as organoids with spontaneous differentiation into crypt-like structures in vitro. Cancer Res 1987; 47: 26839.
  • 31
    Whitehead RH,Zhang HH,Hayward IP. Retention of tissue-specific phenotype in a panel of colon carcinoma cell lines: relationship to clinical correlates. Immunol Cell Biol 1992; 70(Part 4): 22736.
  • 32
    Sizeland AM,Burgess AW. The proliferative and morphologic responses of a colon carcinoma cell line (LIM 1215) require the production of two autocrine factors. Mol Cell Biol 1991; 11: 400514.
  • 33
    Vincan E,Darcy PK,Farrelly CA,Faux MC,Brabletz T,Ramsay RG. Frizzled-7 dictates three-dimensional organization of colorectal cancer cell carcinoids. Oncogene 2007; 26: 234052.
  • 34
    Lachal S,Ford J,Shulkes A,Baldwin GS. PPARα agonists stimulate progastrin production in human colorectal carcinoma cells. Regul Pept 2004; 120: 24351.
  • 35
    Andrew SM,Teh JG,Johnstone RW,Russell SM,Whitehead RH,McKenzie IF,Pietersz GA. Tumor localization by combinations of monoclonal antibodies in a new human colon carcinoma cell line (LIM1899). Cancer Res 1990; 50: 522530.
  • 36
    Whitehead RH,van Eeden P,Lukeis RE. A cell line (LIM 2463) derived from a tubulovillous adenoma of the rectum. Int J Cancer 1991; 48: 6936.
  • 37
    Usadel H,Brabender J,Danenberg KD,Jeronimo C,Harden S,Engles J,Danenberg PV,Yang S,Sidransky D. Quantitative adenomatous polyposis coli promoter methylation analysis in tumor tissue, serum, and plasma DNA of patients with lung cancer. Cancer Res 2002; 62: 3715.
  • 38
    Krypuy M,Ahmed AA,Etemadmoghadam D,Hyland SJ,DeFazio A,Fox SB,Brenton JD,Bowtell DD,Dobrovic A. High resolution melting for mutation scanning of TP53 exons 5-8. BMC Cancer 2007; 7: 168.
  • 39
    Faragher IG,Whitehead RH. Rapid diagnostic test for hereditary nonpolyposis colon cancer kindred using polymerase chain reaction. Dis Colon Rectum 1998; 41: 93840.
  • 40
    Vecsey-Semjen B,Becker KF,Sinski A,Blennow E,Vietor I,Zatloukal K,Beug H,Wagner E,Huber LA. Novel colon cancer cell lines leading to better understanding of the diversity of respective primary cancers. Oncogene 2002; 21: 464662.
  • 41
    Yang AD,Fan F,Camp ER,van Buren G,Liu W,Somcio R,Gray MJ,Cheng H,Hoff PM,Ellis LM. Chronic oxaliplatin resistance induces epithelial-to-mesenchymal transition in colorectal cancer cell lines. Clin Cancer Res 2006; 12: 414753.
  • 42
    Russo A,Bazan V,Iacopetta B,Kerr D,Soussi T,Gebbia N. The TP53 colorectal cancer international collaborative study on the prognostic and predictive significance of p53 mutation: influence of tumor site, type of mutation, and adjuvant treatment. J Clin Oncol 2005; 23: 751828.
  • 43
    Zhao L,Vogt PK. Helical domain and kinase domain mutations in p110α of phosphatidylinositol 3-kinase induce gain of function by different mechanisms. Proc Natl Acad Sci USA 2008; 105: 26527.
  • 44
    Yokoyama N,Yin D,Malbon CC. Abundance, complexation, and trafficking of Wnt/β-catenin signaling elements in response to Wnt3a. J Mol Signal 2007; 2:11.
  • 45
    Bafico A,Liu G,Yaniv A,Gazit A,Aaronson SA. Novel mechanism of Wnt signalling inhibition mediated by Dickkopf-1 interaction with LRP6/Arrow. Nat Cell Biol 2001; 3: 6836.
  • 46
    Mao B,Wu W,Davidson G,Marhold J,Li M,Mechler BM,Delius H,Hoppe D,Stannek P,Walter C,Glinka A,Niehrs C. Kremen proteins are Dickkopf receptors that regulate Wnt/β-catenin signalling. Nature 2002; 417: 6647.
  • 47
    Favata MF,Horiuchi KY,Manos EJ,Daulerio AJ,Stradley DA,Feeser WS,Van Dyk DE,Pitts WJ,Earl RA,Hobbs F,Copeland RA,Magolda RL, et al. Identification of a novel inhibitor of mitogen-activated protein kinase kinase. J Biol Chem 1998; 273: 1862332.
  • 48
    Vlahos CJ,Matter WF,Hui KY,Brown RF. A specific inhibitor of phosphatidylinositol 3-kinase, 2-(4-morpholinyl)-8-phenyl-4H-1-benzopyran-4-one (LY294002). J Biol Chem 1994; 269: 52418.
  • 49
    Sparks AB,Morin PJ,Vogelstein B,Kinzler KW. Mutational analysis of the APC/β-catenin/Tcf pathway in colorectal cancer. Cancer Res 1998; 58: 11304.
  • 50
    Suzuki H,Watkins DN,Jair KW,Schuebel KE,Markowitz SD,Chen WD,Pretlow TP,Yang B,Akiyama Y,Van Engeland M,Toyota M,Tokino T, et al. Epigenetic inactivation of SFRP genes allows constitutive WNT signaling in colorectal cancer. Nat Genet 2004; 36: 41722.
  • 51
    Rodriguez-Viciana P,Warne PH,Dhand R,Vanhaesebroeck B,Gout I,Fry MJ,Waterfield MD,Downward J. Phosphatidylinositol-3-OH kinase as a direct target of Ras. Nature 1994; 370: 52732.
  • 52
    Haigis KM,Kendall KR,Wang Y,Cheung A,Haigis MC,Glickman JN,Niwa-Kawakita M,Sweet-Cordero A,Sebolt-Leopold J,Shannon KM,Settleman J,Giovannini M, et al. Differential effects of oncogenic K-Ras and N-Ras on proliferation, differentiation and tumor progression in the colon. Nat Genet 2008; 40: 6008.
  • 53
    Wang D,Boerner SA,Winkler JD,LoRusso PM. Clinical experience of MEK inhibitors in cancer therapy. Biochim Biophys Acta 2007; 1773: 124855.

Supporting Information

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
IJC_24289_sm_SuppFig1.tif2177KSupplementary Figure 1: Morphology of LIM 2463B. LIM 2463B cells grow as spindly, adherent cells which at high density form large and tightly-packed cell clusters (A). These ‘organoids’ are initially anchored to the adherent monolayer by a stalk, but eventually detach from the plate and float freely in the medium (B). Cells within the ‘organoids’ are of uniform size and appearance. Microscope: Nikon TE-2000, 10x lens (A) and 20 x lens (B).
IJC_24289_sm_SuppFig2.tif3256KSupplementary Figure 2: Expression of E-cadherin and Vimentin in LIM cell lines Exponentially growing cells were lysed in DOC buffer as described in Methods. 10μg of total cellular protein were loaded in each lane of 4-12% Bis/Tris gels. Samples were transferred to nitrocellulose membrane and probed with antibodies to the epithelial marker E-cadherin (panel A) or to the mesenchymal marker Vimentin (panel B), followed by Odyssey anti-mouse IRDye 680. The nitrocellulose membranes were scanned on an Odyssey Infrared Imaging instrument.
IJC_24289_sm_SuppTabs.doc124KSupplementary Table 1: Primer sequence and product size for APC ex16. F: forward primer; R: reverse primer Supplementary Table 2: Primers sequence and product size of B-Raf.and K-Ras PCR. F: forward primer: R: reverse primer. Supplementary Table 3: Primer sequence and product size for β-Catenin. F: forward primer; R: reverse primer Supplementary Table 4: Primer sequence for PIK3CA and PIK3R1. F: forward primer; R: reverse primer Supplementary Table 5: Colony forming efficiency of LIM cell lines. Cells were plated at high (5×104cells/ml) or low (5×103 cells/ml) density in semi-solid agar. After 12 days, plates were stained with crystal violet, and colonies containing >50 cells were scored. Colony forming efficiency is defined as the number of colonies per 100 cells plated. The data are average and standard deviation of three replicate plates. Results are representative of two independent experiments.
IJC_24289_sm_SuppInfo.doc29KSupplementary methods

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.