Heterogeneity of Jagged1 expression in human and mouse intestinal tumors: implications for targeting Notch signaling.

Inhibition of Notch signaling is effective in inhibiting colon tumorigenesis, but targeting specific components of the pathway may provide more effective strategies. Here we show that the expression of Jagged1, a ligand for canonical Notch signaling, was restricted to enteroendocrine cells or undetectable in the mucosa of the human small and large intestine, respectively. In contrast, increased expression characterized half of human colon tumors, although not all tumors with elevated Wnt signaling displayed elevated Jagged1. Increased Jagged1 was also present in intestinal tumors of Apc(1638N/+) and Apc(Min/+) mice, but to a higher level and more frequently in the former, and in 90% of mouse tumors Notch signaling was elevated when Jagged1 was elevated. In the human HT29Cl16E colonic carcinoma cell line, induction of goblet cell differentiation by contact inhibition of growth depended on the loss of Jagged1-mediated Notch activation, with signaling through Notch1 and Notch2 acting redundantly. Therefore, targeting of Jagged1 could be effective in downregulating Notch signaling in a subset of tumors, but may avoid the limiting gastrointestinal toxicity caused by pharmacological inhibition of Notch signaling.

KLF4 regulation in intestinal epithelial cell maturation.

The Krüppel-like factor 4 (KLF4) transcription factor suppresses tumorigenesis in gastrointestinal epithelium. Thus, its expression is decreased in gastric and colon cancers. Moreover, KLF4 regulates both differentiation and growth that is likely fundamental to its tumor suppressor activity. We dissected the expression of Klf4 in the normal mouse intestinal epithelium along the crypt-villus and cephalo-caudal axes. Klf4 reached its highest level in differentiated cells of the villus, with levels in the duodenum>jejunum>ileum, in inverse relation to the representation of goblet cells in these regions, the lineage previously linked to KLF4. In parallel, in vitro studies using HT29cl.16E and Caco2 colon cancer cell lines clarified that KLF4 increased coincident with differentiation along both the goblet and absorptive cell lineages, respectively, and that KLF4 levels also increased during differentiation induced by the short chain fatty acid butyrate, independently of cell fate. Moreover, we determined that lower levels of KLF4 expression in the proliferative compartment of the intestinal epithelium are regulated by the transcription factors TCF4 and SOX9, an effector and a target, respectively, of beta-catenin/Tcf signaling, and independently of CDX2. Thus, reduced levels of KLF4 tumor suppressor activity in colon tumors may be driven by elevated beta-catenin/Tcf signaling.

Molecular mechanisms of action and prediction of response to oxaliplatin in colorectal cancer cells.

The platinum compound oxaliplatin has been shown to be an effective chemotherapeutic agent for the treatment of colorectal cancer. In this study, we investigate the molecular mechanisms of action of oxaliplatin to identify means of predicting response to this agent. Exposure of colon cancer cells to oxaliplatin resulted in G2/M arrest and apoptosis. Immunofluorescent staining demonstrated that the apoptotic cascade initiated by oxaliplatin is characterised by translocation of Bax to the mitochondria and cytochrome c release into the cytosol. Oxaliplatin treatment resulted in caspase 3 activation and oxaliplatin-induced apoptosis was abrogated by inhibition of caspase activity with z-VAD-fmk, but was independent of Fas/FasL association. Targeted inactivation of Bax or p53 in HCT116 cells resulted in significantly increased resistance to oxaliplatin. However, the mutational status of p53 was unable to predict response to oxaliplatin in a panel of 30 different colorectal cancer cell lines. In contrast, the expression profile of these 30 cell lines, assessed using a 9216-sequence cDNA microarray, successfully predicted the apoptotic response to oxaliplatin. A leave-one-out cross-validation approach was used to demonstrate a significant correlation between experimentally observed and expression profile predicted apoptosis in response to clinically achievable doses of oxaliplatin (R=0.53; P=0.002). In addition, these microarray experiments identified several genes involved in control of apoptosis and DNA damage repair that were significantly correlated with response to oxaliplatin.

c-Myc overexpression sensitises colon cancer cells to camptothecin-induced apoptosis.

The proto-oncogene c-Myc is overexpressed in 70% of colorectal tumours and can modulate proliferation and apoptosis after cytotoxic insult. Using an isogenic cell system, we demonstrate that c-Myc overexpression in colon carcinoma LoVo cells resulted in sensitisation to camptothecin-induced apoptosis, thus identifying c-Myc as a potential marker predicting response of colorectal tumour cells to camptothecin. Both camptothecin exposure and c-Myc overexpression in LoVo cells resulted in elevation of p53 protein levels, suggesting a role of p53 in the c-Myc-imposed sensitisation to the apoptotic effects of camptothecin. This was confirmed by the ability of PFT-alpha, a specific inhibitor of p53, to attenuate camptothecin-induced apoptosis. p53 can induce the expression of p21(Waf1/Cip1), an antiproliferative protein that can facilitate DNA repair and drug resistance. Importantly, although camptothecin treatment markedly increased p21(Waf1/Cip1) levels in parental LoVo cells, this effect was abrogated in c-Myc-overexpressing derivatives. Targeted inactivation of p21(Waf1/Cip1) in HCT116 colon cancer cells resulted in significantly increased levels of apoptosis following treatment with camptothecin, demonstrating the importance of p21(Waf1/Cip1) in the response to this agent. Finally, cDNA microarray analysis was used to identify genes that are modulated in expression by c-Myc upregulation that could serve as additional markers predicting response to camptothecin. Thirty-four sequences were altered in expression over four-fold in two isogenic c-Myc-overexpressing clones compared to parental LoVo cells. Moreover, the expression of 10 of these genes was confirmed to be significantly correlated with response to camptothecin in a panel of 30 colorectal cancer cell lines.

p21(WAF1/cip1) is an important determinant of intestinal cell response to sulindac in vitro and in vivo.

Sulindac, a nonsteroidal anti-inflammatory drug, inhibits intestinal tumorigenesis in humans and rodents. Sulindac induced complex alterations in gene expression, but only 0.1% of 8063 sequences assayed were altered similarly by the drug in rectal biopsies of patients treated for 1 month and during response of colonic cells in culture. Among these changes was induction of the cyclin-dependent kinase inhibitor, p21(WAF1/cip1). In Apc1638(+/-) mice, targeted inactivation of p21 increased intestinal tumor formation in a gene-dose-dependent manner, but inactivation of p21 completely eliminated the ability of sulindac to both inhibit mitotic activity in the duodenal mucosa and to inhibit Apc-initiated tumor formation. Thus, p21 is essential for tumor inhibition by this drug. The array data can be accessed on the Internet at http://sequence.aecom.yu.edu/genome/.

c-myc/p53 interaction determines sensitivity of human colon carcinoma cells to 5-fluorouracil in vitro and in vivo.

Colon carcinoma cells overexpress c-myc due to defective Wnt signaling, but only patients whose tumors have an amplified c-myc gene show improved disease-free and overall survival in response to 5-fluoruracil (5FU). Here we show that in two colon carcinoma cell lines that do not have an amplified c-myc gene but differ in their p53 status, high c-myc levels can be further elevated by introducing a c-myc expression vector. Whereas sensitivity to low serum-induced apoptosis was imposed on the parental lines independent of p53 status and was unaffected by further elevation of c-myc, sensitivity to 5FU-induced apoptosis was dependent on both the higher c-myc levels due to the expression vector and wild-type p53 function. The elevated c-myc levels led to higher c-myc transactivation activity in the p53 wild-type cell line, but not in the mutant p53 cell line. The requirement for both elevated c-myc and p53 for 5FU sensitivity was confirmed using antisense c-myc and pifithrin-alpha, a specific inhibitor of p53. Finally, the in vitro data predicted that only patients with both amplified c-myc and wild-type p53 in their primary tumors would be responsive to 5FU-based therapy, which was borne out by analysis of tumors from 135 patients entered into a Phase III clinical trial of 5FU-based adjuvant therapy. The data provide significant insight into mechanisms that establish colon tumor cell sensitivity to 5FU, clearly demonstrate the necessity of exercising caution in considering combining novel strategies that target elevated c-myc with standard 5FU-based therapy, and suggest alternative therapeutic strategies that target c-myc and/or p53 mutations in the treatment of colon cancer.

Down-regulation of beta-catenin TCF signaling is linked to colonic epithelial cell differentiation.

The beta-catenin TCF pathway is implicated in the regulation of colonic epithelial cell proliferation, but its role in the regulation of cell differentiation is unknown. The colon carcinoma cell line, Caco-2, spontaneously undergoes G(0)/G(1) cell cycle arrest and differentiates along the absorptive cell lineage over 21 days in culture. In parallel, we show that beta-catenin-TCF activity and complex formation are significantly down-regulated. The down-regulation of beta-catenin-TCF signaling was independent of APC, which we characterized as having a nonsense mutation in codon 1367 in Caco-2 cells, but was associated with a decrease in TCF-4 protein levels. Total beta-catenin levels increased during Caco-2 cell differentiation, although this was attributable to an increase in the membrane, E-cadherin-associated, fraction of beta-catenin. Importantly, down-regulation of beta-catenin-TCF signaling in undifferentiated Caco-2 cells by three different mechanisms, ectopic expression of E-cadherin, wild-type APC, or dominant negative TCF-4, resulted in an increase in the promoter activities of two genes that are well-established markers of cell differentiation, alkaline phosphatase and intestinal fatty acid binding protein. These studies demonstrate, therefore, that in addition to its established role in the regulation of cell proliferation, down-regulation of the beta-catenin-TCF pathway is associated with the promotion of a more-differentiated phenotype in colonic epithelial cells.

Resistance to butyrate-induced cell differentiation and apoptosis during spontaneous Caco-2 cell differentiation.

BACKGROUND & AIMS: The short-chain fatty acid butyrate induces cell cycle arrest, differentiation, and apoptosis in colon cancer cells, but often induces opposite effects in normal colonic epithelial cells. We determined whether response to butyrate is dependent on the basal differentiation status of colonic epithelial cells. METHODS: Caco-2 cells at progressive stages of differentiation were treated with butyrate, and endpoints were measured. RESULTS: Response of Caco-2 cells to butyrate was dependent on their differentiation status. Butyrate maximally stimulated cell cycle arrest, apoptosis, alkaline phosphatase activity, transepithelial resistance, cell migration, urokinase receptor expression, and interleukin 8 secretion in undifferentiated Caco-2 cells, whereas differentiated Caco-2 cells were essentially resistant to these effects. Consistently, butyrate selectively induced histone hyperacetylation in undifferentiated Caco-2 cells. This resistance was also observed during HT29cl.19A cell differentiation, but not in the nondifferentiating SW620 cell line. Finally, the rate of butyrate use significantly increased as Caco-2 cells underwent spontaneous differentiation. CONCLUSIONS: Colonic epithelial cells become progressively more refractory to the effects of butyrate during absorptive cell differentiation. We postulate that this resistance is caused by the rapid use of butyrate by differentiated Caco-2 cells, which likely results in low intracellular concentrations and subsequently in its inability to inhibit histone deacetylase.

Targeted inactivation of the p21(WAF1/cip1) gene enhances Apc-initiated tumor formation and the tumor-promoting activity of a Western-style high-risk diet by altering cell maturation in the intestinal mucosal.

Elimination of both alleles of the gene that encodes the cyclin kinase inhibitor p21(WAF1/cip1) increases the frequency and size of intestinal tumors in Apc1638+/- mice that inherit a mutant allele of the Apc gene, and intermediate effects are seen if a single p21 allele is inactivated. The increased tumor formation is associated with altered cell maturation in the intestinal mucosa of the p21-deficient mice--increased cell proliferation, and decreased apoptosis, and goblet cell differentiation--that is also a function of p21 gene dosage. Moreover, a Western-style diet that mimics principal risk factors for colon cancer (high fat and phosphate, low calcium and vitamin D) accelerates tumor formation in Apc1638+/- mice, and the loss of a single or both p21 alleles is additive with the tumor-promoting effects of this diet, resulting in more and larger tumors, and a highly significant decrease in survival time. Thus, p21 normally suppresses Apc-initiated tumor formation and is haplo-insufficient in this regard. This is consistent with recent reports that Apc initiates tumor formation by up-regulating c-myc expression through altered beta-catenin-Tcf signaling and that c-myc then up-regulates cdk4, whose activity is inhibited by p21. Decreased expression of p21 is also a marker of poor prognosis in patients, and the data presented suggest that dietary alterations in patients undergoing treatment for colon cancer might be highly effective in improving outcome.

The Sp family of transcription factors in the regulation of the human and mouse MUC2 gene promoters.

Modulation of mucin gene expression is an important component both in the early steps of colon cancer development and in later tumor progression. Previous work from our laboratory and others has suggested that the Sp family of transcription factors may play an important role in the regulation of the human MUC2 gene. To determine whether this was an essential element, we extended our work to the cloning and analysis of 3.5 kb of the 5'-flanking region of the mouse Muc2 (mMuc2) gene. Comparative analysis between the mouse and human MUC2 promoter regions has identified a strong sequence homology between the mouse and human genes, including the presence of GC-rich boxes, the location and composition of which are maintained in the mouse and human genes. We show that these GC boxes are binding sites for Sp-family transcription factors and are functionally important since mithramycin, an inhibitor of Sp1/Sp3 binding, blocks MUC2 gene expression in HT29 cells. Furthermore, by a combination of gel shift analysis and site-directed mutagenesis, we have identified the relative contribution of individual GC boxes, and of the factors they bind, to the regulation of the mouse Muc2 promoter, which appears to be different in the mouse and human genes. Finally, we demonstrate by overexpressing Sp1 and Sp3 that the functional difference between the proximal promoter region of the MUC2 gene in the two species is not attributable to differential ability of this region to bind members of the Sp family of transcription factors, but rather to the different anatomy of the individual GC boxes in the mouse and human proximal promoters.

Dissociation of staurosporine-induced apoptosis from G2-M arrest in SW620 human colonic carcinoma cells: initiation of the apoptotic cascade is associated with elevation of the mitochondrial membrane potential (deltapsim).

We have identified an alternative apoptotic cascade induced in SW620 human colonic carcinoma cells by the protein kinase antagonist staurosporine (stsp). Consistent with its effect in other colonic epithelial cells, stsp induced G2-M arrest and apoptosis of SW620 cells. However, despite the paradigm that growth arrest triggers apoptotic cascades, apoptosis was detected before G2-M arrest. Reports have linked dissipation of the mitochondrial membrane potential (deltapsim) to the initiation of apoptosis and have linked elevation of the deltapsim to the escape from apoptosis However, neither apoptosis nor cell cycle arrest were altered by the collapse of the deltapsim, and increased deltapsim enhanced the initiation of apoptosis but blocked G2-M arrest. Although reactive oxygen species (ROS) have been implicated in some colonic epithelial cell and stsp-induced cascades, neither antioxidants nor the inhibition of RNA or protein synthesis altered apoptosis of SW620 cells. Finally, cytosolic cytochrome c has been linked to activation of caspase-3 and dissipation of the deltapsim. However, caspase-3 activation preceded the accumulation of cytochrome c in the cytosol and was accompanied by transient elevations in both the deltapsim and mitochondria-associated cytochrome c. Therefore, we have identified a distinct apoptotic cascade in SW620 cells that was induced independently of growth arrest, dissipation of the deltapsim, ROS production, or synthesis of de novo RNA or protein, and we have linked its efficient initiation to early elevation of the deltapsim.

Genetic reprogramming in pathways of colonic cell maturation induced by short chain fatty acids: comparison with trichostatin A, sulindac, and curcumin and implications for chemoprevention of colon cancer.

The short-chain fatty acid butyrate, produced by microbial fermentation of dietary fiber in the large intestine, is a physiological regulator of major pathways of colonic epithelial cell maturation: cell cycle arrest, lineage-specific differentiation, and apoptosis. Microarray analysis of 8,063 sequences demonstrated a complex cascade of reprogramming of SW620 colonic epithelial cells upon treatment with butyrate characterized by the progressive recruitment of gene sets as a function of time. Comparison with the effects of trichostatin A, in conjunction with differences in the kinetics of alteration of histone acetylation induced by butyrate and trichostatin A, identified subsets of induced and repressed genes likely coordinately regulated by altered histone acetylation. The butyrate response was also compared in detail with that of sulindac, a nonsteroidal anti-inflammatory drug with significant chemopreventive activity for colon cancer, and curcumin, a component of mustard and curry structurally and functionally related to sulindac that also has chemopreventive activity. Although gene clusters were identified that showed similar responses to butyrate and sulindac, the data were characterized by the extensive differences in the effects of the two agents. This was striking for functional classes of genes involved in signaling pathways and in cell cycle progression, although butyrate and sulindac induce a similar G0-G1 arrest, elevation of beta-catenin-Tcf signaling, and apoptotic cascade. As regards cell cycle arrest, the underlying mechanism in response to butyrate was most similar to that of the Caco-2 cell line that had spontaneously undergone a G0-G1 arrest and least similar to the G2-M arrest stimulated by curcumin. Thus, high-throughput microarray analysis of gene expression profiles can be used to characterize and distinguish the mechanisms of response of colonic epithelial cells to physiological and pharmacological inducers of cell maturation. This has important implications for characterization of chemopreventive agents and recognition of potential toxicity and synergies. The data bases, gene clusters, and analyses are available at http:// sequence.aecom.yu.edu/genome/.

The integrin-linked kinase regulates the cyclin D1 gene through glycogen synthase kinase 3beta and cAMP-responsive element-binding protein-dependent pathways.

The cyclin D1 gene encodes the regulatory subunit of a holoenzyme that phosphorylates and inactivates the pRB tumor suppressor protein. Cyclin D1 is overexpressed in 20-30% of human breast tumors and is induced both by oncogenes including those for Ras, Neu, and Src, and by the beta-catenin/lymphoid enhancer factor (LEF)/T cell factor (TCF) pathway. The ankyrin repeat containing serine-threonine protein kinase, integrin-linked kinase (ILK), binds to the cytoplasmic domain of beta(1) and beta(3) integrin subunits and promotes anchorage-independent growth. We show here that ILK overexpression elevates cyclin D1 protein levels and directly induces the cyclin D1 gene in mammary epithelial cells. ILK activation of the cyclin D1 promoter was abolished by point mutation of a cAMP-responsive element-binding protein (CREB)/ATF-2 binding site at nucleotide -54 in the cyclin D1 promoter, and by overexpression of either glycogen synthase kinase-3beta (GSK-3beta) or dominant negative mutants of CREB or ATF-2. Inhibition of the PI 3-kinase and AKT/protein kinase B, but not of the p38, ERK, or JNK signaling pathways, reduced ILK induction of cyclin D1 expression. ILK induced CREB transactivation and CREB binding to the cyclin D1 promoter CRE. Wnt-1 overexpression in mammary epithelial cells induced cyclin D1 mRNA and targeted overexpression of Wnt-1 in the mammary gland of transgenic mice increased both ILK activity and cyclin D1 levels. We conclude that the cyclin D1 gene is regulated by the Wnt-1 and ILK signaling pathways and that ILK induction of cyclin D1 involves the CREB signaling pathway in mammary epithelial cells.

Divergent phenotypic patterns and commitment to apoptosis of Caco-2 cells during spontaneous and butyrate-induced differentiation.

Caco-2 cells differentiate spontaneously when cultured in confluence and on exposure to the physiologically relevant short-chain fatty acid, butyrate. This study aimed to compare the phenotype induced by these pathways and their relations to cell turnover. Caco-2 cells were treated with butyrate at a nontoxic concentration of 2 mM for 3 days, or allowed to spontaneously differentiate for 0-21 days. Brush border hydrolase activities and carcinoembryonic antigen (CEA) expression, transepithelial resistance and dome formation, expression of components of the urokinase system, and cell turnover by flow cytometry, and the degree of DNA fragmentation were quantified. Butyrate induced increases in alkaline phosphatase activity and CEA expression but not the activities of other hydrolases, while culture alone induced progressive increases in the activities/expression of all markers. Butyrate induced a significantly greater increase in transepithelial resistance (TER) than occurred during culture alone but the densities of domes were similar. Butyrate induced a ninefold increase in urokinase receptor expression and twofold increase in urokinase activity, while culture alone induced a significantly smaller increase in receptor expression, an increase in plasminogen activator inhibitor-1 but no change in activity. While both stimuli induced cell cycle arrest, only butyrate increased the proportion of cells undergoing apoptosis. In conclusion, differentiation of Caco-2 cells can proceed along multiple pathways but does not necessarily lead to apoptosis. The phenotypic changes during spontaneous differentiation mimic those that occur in normal colonic epithelial cells in vivo during their migration from the crypt base to neck, while butyrate-induced effects more closely follow those occurring when normal colonic epithelial cells migrate from crypt neck to the surface compartment.

Short-chain fatty acid metabolism, apoptosis, and Apc-initiated tumorigenesis in the mouse gastrointestinal mucosa.

Short-chain fatty acids (SCFAs) are physiological regulators of growth and differentiation in the gastrointestinal tract, and we have previously shown that apoptosis induced in colonic cell lines by these compounds is dependent on their metabolism by B-oxidation in the mitochondria (B. G. Heerdt et al., J. Biol. Chem., 266: 19120-19126, 1991; Cancer Res., 54: 3288-3293, 1994). Because tumors initiated by an inherited Apc mutation have been reported to be linked to decreases in apoptosis in the flat mucosa of the gastrointestinal tract, the aims were to determine whether elimination of efficient metabolism of SCFAs affected apoptosis in the gastrointestinal mucosa of the mouse, and whether this altered tumorigenesis initiated by an inherited Apc mutation. We, therefore, generated mice that have a chain-terminating mutation in the Apc gene and that were either wild-type for SCFA metabolism, or deficient, due to homozygous deletion of the gene (Scad) that encodes the enzyme short-chain acyl dehydrogenase, which catalyzes the first step in SCFA B-oxidation. Scad+/+ mice maintained on a wheat bran-fiber-supplemented diet gained significantly more weight than mice maintained on AIN76A, but this was eliminated by the Scad mutation, demonstrating that uptake and metabolism of SCFAs in the gastrointestinal tract can be a significant energy source. As predicted, on either AIN76A or wheat bran diet, the Scad mutation almost completely eliminated apoptosis in the flat mucosa of the proximal colon and reduced apoptosis by 50% in the distal colon compared with littermates that were wild-type for Scad. The mutation also reduced apoptosis by approximately 50% in the duodenum in AIN76A-fed mice. These reductions in apoptosis had no effect on incidence, frequency, or site specificity of tumors initiated by the Apc mutation. Therefore, the metabolism of SCFAs by the gastrointestinal mucosa plays a role in modulating apoptosis, but a general decrease in apoptosis in the mucosa of the gastrointestinal tract is not linked to gastrointestinal tumorigenesis initiated by an inherited Apc mutation.

Butyrate-induced apoptotic cascade in colonic carcinoma cells: modulation of the beta-catenin-Tcf pathway and concordance with effects of sulindac and trichostatin A but not curcumin.

Short-chain fatty acids play a critical role in colonic homeostasis because they stimulate pathways of growth arrest, differentiation, and apoptosis. These effects have been well characterized in colonic cell lines in vitro. We investigated the role of beta-catenin-Tcf signaling in these responses to butyrate and other well-characterized inducers of apoptosis of colonic epithelial cells. Unlike wild-type APC, which down-regulates Tcf activity, butyrate, as well as sulindac and trichostatin A, all inducers of G0-G1 cell cycle arrest and apoptosis in the SW620 colonic carcinoma cell line, up-regulate Tcf activity. In contrast, structural analogues of butyrate that do not induce cell cycle arrest or apoptosis and curcumin, which stimulates G2-M arrest without inducing apoptosis, do not alter Tcf activity. Similar to the cell cycle arrest and apoptotic cascade induced by butyrate, the up-regulation of Tcf activity is dependent upon the presence of a mitochondrial membrane potential, unlike the APC-induced down-regulation, which is insensitive to collapse of the mitochondrial membrane potential. Moreover, the butyrate-induced increase in Tcf activity, which is reflected in an increase in beta-catenin-Tcf complex formation, is independent of the down-regulation caused by expression of wild-type APC. Thus, butyrate and wild-type APC have different and independent effects on beta-catenin-Tcf signaling. These data are consistent with other reports that suggest that the absence of wild-type APC, associated with the up-regulation of this signaling pathway, is linked to the probability of a colonic epithelial cell entering an apoptotic cascade.