The role of COX-2/PGE2 in gossypol-induced apoptosis of colorectal carcinoma cells

Our previous study showed that gossypol (GOS) exhibits potent cytotoxic effects via apoptosis induction against human colorectal carcinoma cells; however, the role of cyclooxygenase (COX)-2/prostaglandin (PG)E(2) on GOS-induced apoptosis is still unknown. In the present study, 12-O-tetradecanoylphorbol-13-acetate (TPA) addition significantly inhibited GOS-induced apoptosis in human colorectal carcinoma HT-29 cells in accordance with inducing COX-2 protein/PGE(2) production. TPA inhibition of GOS-induced apoptosis was blocked by adding protein kinase (PK)C inhibitors including staurosporine (ST), GF109203X (GF), and H7, characterized by the occurrence of cleaved caspase 3 proteins and a decrease in COX-2 protein/PGE(2) production in HT-29 cells. The addition of COX activity inhibitors, including NS398 (NS), aspirin (AS), diclofenac (DI), and indomethacin (IN), suppressed TPA protection of GOS-induced apoptosis with decreased PGE(2) production in HT-29 cells. Application of PGE(2), but not it analogs PGD(2), PGJ(2), or PGF(2α), protected HT-29 cells from GOS-induced DNA ladders, and the E-prostanoid (EP(1)) receptor agonist, 17PT-PGE(2), mimicked the protection induced by PGE(2), whereas the selective EP(2) receptor agonist, butaprostol (BUT), the EP(3) receptor agonist, sulprostol (SUL), and the EP(4) receptor agonist, PGE(1) alcohol (PGE(1)), showed no significant effects on GOS-induced apoptosis in HT-29 cells. PGE(2) 's protection against GOS-induced apoptosis was reversed by adding the selective EP(1) receptor antagonist, SC-19220. Furthermore, GOS had an effective apoptotic effect on COLO205 colorectal carcinoma cells which expressed undetectable level of endogenous COX-2 protein than HT-29 cells, and the decreased COX-2 protein level via COX-2 siRNA or addition of COX-2 activity inhibitor NS significantly elevated GOS-induced cell death in HT-29 cells. COLO205-T cells were established through sustained TPA incubation of COLO205 cells, and COLO205-T cells showed a lower sensitivity to GOS-induced cell death with increased COX-2 (not Bcl-2 and Mcl-1) protein than parental COLO-205 cells. A decrease in COX-2 protein expression in COLO205-T cells by COX-2 siRNA transfection or enhanced GOS-induced cell death according to MTT assay and DNA integrity assay. The notion of COX-2/PGE(2) activation against GOS-induced apoptosis in colon carcinoma cells was demonstrated, and the combination of GOS and COX-2 inhibitors to treat colon carcinoma possesses clinical potential worthy of further investigation.     

HT-29 colorectal cancer cells undergoing apoptosis overexpress COX-2 to delay ursolic acid-induced cell death

Colorectal cancer is one of the most common cancer types and the third leading cause of cancer-related death in the western world. Generally, colorectal cancers are resistant to anticancer drugs. Several lines of evidence support a critical role for cyclooxygenase-2 (COX-2) during colorectal tumorigenesis and its role in chemoresistance. In this study, we focused our interest on the role played by COX-2 in apoptosis induced in HT-29 human colorectal cancer cells by ursolic acid (UA), a triterpenoid found in a large variety of plants. We showed that UA-induced apoptosis and that COX-2 was overexpressed only in apoptotic cells. We demonstrated that this overexpression was mediated by the p38 MAP kinase pathway as inhibiting its activation using a p38-specific inhibitor, SB 203580, abrogated COX-2 expression. Inhibiting COX-2 expression either by using a p38-specific inhibitor or COX-2-specific siRNA increased apoptosis. These results demonstrated that COX-2 was involved in a resistance mechanism to UA-induced apoptosis in HT-29 cells. Cells undergoing apoptosis were able to trigger a resistance mechanism by overexpressing a protein such as COX-2 to delay their death. Furthermore, we demonstrated that this resistance mechanism was independent of PGE₂ production as the addition of the specific COX-2 activity inhibitor, NS-398, did not affect apoptosis in UA-treated cells.     

Effect of PPAR activators on cytokine-stimulated cyclooxygenase-2 expression in human colorectal carcinoma cells.

Cyclooxygenase-2 (COX-2) expression is up-regulated in colorectal cancer tissue. Peroxisome proliferator-activated receptors (PPARs) are expressed in human colorectal tissue and activation of PPARs can alter COX-2 expression. In macrophages, activation of PPARs down-regulates COX-2 expression. We examined the effect of PPARalpha and PPARgamma ligands on untreated and TNF-alpha-induced COX-2 expression in the human colorectal epithelial cell line HT-29. The expression of PPARalpha and PPARgamma was confirmed in these cells. TNF-alpha, an inflammatory cytokine, increased COX-2 expression via the NFkappaB pathway. In the absence of TNF-alpha, WY14643 (PPARalpha activator) caused an increase, while BRL49653 (PPARgamma activator) did not alter COX-2 expression. When HT-29 cells were incubated with TNF-alpha and WY14643, a further increase in COX-2 expression was detected. Incubation with TNF-alpha and BRL49653 caused an additional twofold increase in COX-2 expression. Our results suggest that both PPARalpha signaling and TNF-alpha signaling increase COX-2 expression by independent pathways, while PPARgamma stimulates COX-2 expression by up-regulation of the TNF-alpha pathway.     

Mitochondrial localization of cyclooxygenase-2 and calcium-independent phospholipase A2 in human cancer cells: implication in apoptosis resistance

Cyclooxygenase-2 (COX-2) is inducible by myriad stimuli. The inducible COX-2 in primary cultured human cells has been reported to localize to nuclear envelope, endoplasmic reticulum, nucleus and caveolae. As COX-2 plays an important role in tumor growth, we were interested in its subcellular location in cancer cells. We examined COX-2 localization in several cancer cell lines by confocal microscopy. A majority of COX-2 was colocalized with heat shock protein 60, a mitochondrial protein, in colon cancer (HT-29, HCT-15 and DLD-1), breast cancer (MCF7), hepatocellular cancer (HepG2) and lung cancer cells (A549) with a similar distribution pattern. By contrast, COX-2 was not localized to mitochondria in human foreskin fibroblasts or endothelial cells. Immunoblot analysis of COX-2 in mitochondrial and cytosolic fractions confirmed localization of COX-2 to mitochondria in HT-29 and DLD-1 cells but not in fibroblasts. Calcium-independent phospholipase A2 was colocalized with heat shock protein 60 to mitochondria not only in cancer cells (HT-29 and DLD-1) but also in fibroblasts. HT-29 which expressed more abundant mitochondrial COX-2 than DLD-1 was highly resistant to arachidonic acid and H2O2-induced apoptosis whereas DLD-1 was less resistant and human fibroblasts were highly susceptible. Treatment of HT-29 cells with sulindac or SC-236, a selective COX-2 inhibitor, resulted in loss of resistance to apoptosis. These results suggest that mitochondrial COX-2 in cancer cells confer resistance to apoptosis by reducing the proapoptotic arachidonic acid.     

Autophagy delays sulindac sulfide-induced apoptosis in the human intestinal colon cancer cell line HT-29

Autophagy is a major catabolic process allowing the renewal of intracellular organelles by which cells maintain their homeostasis. We have previously shown that autophagy is controlled by two transduction pathways mediated by a heterotrimeric Gi3 protein and phosphatidylinositol 3-kinase activities in the human colon cancer cell line HT-29. Here, we show that 3-methyladenine, an inhibitor of autophagy, increases the sensitivity of HT-29 cells to apoptosis induced by sulindac sulfide, a nonsteroidal anti-inflammatory drug which inhibits the cyclooxygenases. Similarly, HT-29 cells overexpressing a GTPase-deficient mutant of the G(alpha i3) protein (Q204L), which have a low rate of autophagy, were more sensitive to sulindac sulfide-induced apoptosis than parental HT-29 cells. In both cell populations we did not observe differences in the expression patterns of COX-2, Bcl-2, Bcl(XL), Bax, and Akt/PKB activity. However, the rate of cytochrome c release was higher in Q204L-overexpressing cells than in HT-29 cells. These results suggest that autophagy could retard apoptosis in colon cancer cells by sequestering mitochondrial death-promoting factors such as cytochrome c.     

Oroxylin A improves the sensitivity of HT-29 human colon cancer cells to 5-FU through modulation of the COX-2 signaling pathway

5-Fluorouracil (5-FU) is a principal drug for the treatment of colorectal cancer. Due to its low response and high toxicity, synergistic effects of 5-FU in combination with other drugs have been widely researched. This study investigated whether oroxylin A improved the sensitivity of HT-29 human colon cancer cells to 5-FU. A correlation between COX-2 inhibition by oroxylin A and a synergistic effect of 5-FU on the growth of HT-29 cells was observed, and a COX-2 pathway for this effect was recognized; oroxylin A evidently elevated the level of reactive oxygen species in HT-29 cells, which subsequently inhibited COX-2 expression and enhanced the susceptibility of HT-29 cells to 5-FU. Likely also related to COX-2 inhibition, oroxylin A decreased PGE(2) levels in HT-29 cells. The synergistic effect of 5-FU induced by oroxylin A was also found in the suppression of Bcl-2 and in the activation of P53, Bax, PARP, and procaspase-3 proteins in HT-29 cells. Ultimately, a combination of 5-FU with oroxylin A significantly reduced the growth of HT-29 tumors in nude mice compared with treatment with 5-FU or oroxylin A alone. In conclusion, a combination of 5-FU and oroxylin A has a significant synergistic effect in the inhibition of HT-29 cell proliferation in vitro and controls HT-29 tumor growth in vivo. This synergistic effect may be mainly related to COX-2 inhibition by oroxylin A in HT-29 cells.     

Celecoxib and tauro-ursodeoxycholic acid co-treatment inhibits cell growth in familial adenomatous polyposis derived LT97 colon adenoma cells

Chemoprevention would be a desirable strategy to avoid duodenectomy in patients with familial adenomatous polyposis (FAP) suffering from duodenal adenomatosis. We investigated the in vitro effects on cell proliferation, apoptosis, and COX-2 expression of the potential chemopreventives celecoxib and tauro-ursodeoxycholic acid (UDCA). HT-29 colon cancer cells and LT97 colorectal micro-adenoma cells derived from a patient with FAP, were exposed to low dose celecoxib and UDCA alone or in combination with tauro-cholic acid (CA) and tauro-chenodeoxycholic acid (CDCA), mimicking bile of FAP patients treated with UDCA. In HT-29 cells, co-treatment with low dose celecoxib and UDCA resulted in a decreased cell growth (14-17%, p<0.01). A more pronounced decrease (23-27%, p<0.01) was observed in LT97 cells. Cell growth of HT-29 cells exposed to 'artificial bile' enriched with UDCA, was decreased (p<0.001), either in the absence or presence of celecoxib. In LT97 cells incubated with 'artificial bile' enriched with UDCA, cell growth was decreased only in the presence of celecoxib (p<0.05). No clear evidence was found for involvement of proliferating cell nuclear antigen, caspase-3, or COX-2 in the cellular processes leading to the observed changes in cell growth. In conclusion, co-treatment with low dose celecoxib and UDCA has growth inhibitory effects on colorectal adenoma cells derived from a patient with FAP, and further research on this combination as promising chemopreventive strategy is desired.     

Application of non-steroidal anti-inflammatory drugs to enhance 5-fluorouracil efficacy in experimental systems

The elevated cyclooxygenase-2 (COX-2) expression has been shown to affect the carcinogenesis and tumor progression processes, including cell proliferation, motility and angiogenesis. COX-2 is overexpressed in approximately 80% of sporadic colorectal carcinomas and COX-2 enzyme is the best defined target of non-steroidal anti-inflammatory drugs (NSAIDs). In the chemotherapy of colorectal carcinomas 5-fluorouracil (5-FU) has been the most important of the basic drugs for more than 40 years. In order to improve the effectiveness of 5-FU therapy different biological modifiers i.e. inhibitors of its catabolism or activators of anabolism have been studied recently. The rate-limiting enzyme of 5-FU catabolism is dihydropyrimidine dehydrogenase (DPD) since more than 80% of the administered 5-FU is catabolized by DPD. Tumoral DPD has become of clinical interest because elevated intratumoral DPD can decrease the tumor response to 5-FU therapy. The main purpose of our experiments was to investigate the effect of COX inhibitors on the efficacy of 5-FU on high and low COX-2 expressing HCA-7 and HT-29 human colon adenocarcinoma cell lines, respectively, and also on xenografts derived from HT-29 cells. The cytotoxic and antitumor effects of 5-FU in the presence of low doses of indomethacin (non-selective COX-2 inhibitor) and that of NS-398 (highly selective COX-2 inhibitor) on HT-29 and HCA-7 cells and also on the HT-29 xenograft were investigated. In addition, our intention was to understand the mechanism(s) by which NSAIDs could enhance the cytotoxic effect of 5-FU. Our data indicated that the elevated COX-2 expression of HCA-7, the collagen-induced HT-29-C cells and of the HT-29 xenograft were associated with reduced 5-FU sensitivity. Based on the fact that at the same time DPD activity was also increased it might be conceivable that a possible explanation for the decrease of 5-FU sensitivity is the co-existence of high COX-2 and DPD activity. Indomethacin or NS-398 enhanced in a simultaneous and significant manner the sensitivity and cytotoxic effect of 5-FU on high COX-2 expressing cells and xenografts through the modulation of DPD - decrease of its mRNA expression and/or enzyme activity. Based on our results it could be presumable that 5-FU efficacy is limited by the COX-2 associated high DPD expression and activity in patients with colorectal cancer as well, therefore further clinical studies are warranted to decide if NSAIDs in the therapeutic protocol might improve the antitumor potency of 5-FU. Réti A. Application of non-steroidal anti-inflammatory drugs to enhance 5-fluorouracil efficacy in experimental systems.     

Selective cyclooxygenase-2 inhibitors show a differential ability to inhibit proliferation and induce apoptosis of colon adenocarcinoma cells

Although the influence of selective cyclooxygenase (COX)-2 inhibitors on the proliferation of colon adenocarcinoma cells have been the subject of much investigation, relatively little research has compared the effects of different COX-2 inhibitors. Celecoxib strongly suppressed the proliferation of COX-2 expressing HT-29 cells at 10-40 microM. NS-398 and nimesulide also inhibited cell proliferation, whereas rofecoxib, meloxicam, and etodolac did not. Only celecoxib induced apoptosis of HT-29 cells, as detected on the basis of DNA fragmentation, TUNEL positivity, and caspase-3/7 activation. DNA fragmentation was also increasd in COX-2 non-expressing cell lines (SW-480 and HCT-116) by exposure to celecoxib for 6-24 h. All six COX-2 inhibitors suppressed the production of prostaglandin E(2) by HT-29 cells, suggesting that the pro-apoptotic effect of celecoxib was unrelated to inhibition of COX-2. Inactivation of Akt might explain the differential pro-apoptotic effect of these selective COX-2 inhibitors on colon adenocarcinoma cells.     

Prostaglandin-independent effects of aspirin on cell cycle and putrescine synthesis in human colon carcinoma cells

Aspirin consumption has been reported to be able to reduce colorectal cancer risk in humans and in animal models of colon carcinogenesis. Although the mechanism involved in such an effect is not yet clear, both prostaglandin-dependent and -independent effects have been proposed. Using HT-29 Glc(-/+)cells, which originate from a human colon adenocarcinoma, we demonstrated in this study a dose-dependent effect of millimolar concentration of aspirin on cell growth that was concomitant with a rapid accumulation of the cells in the G0/G1 phase, followed by an accumulation in the G2/M phase and by a minor increase in the proportion of cells undergoing nuclear condensation. Cell membrane integrity and cell release into the culture medium were not affected by this treatment. The aspirin effects were apparently unrelated to prostaglandin biosynthesis inhibition, since although these cells were found to express high levels of cyclooxygenase 1 (COX-1) and low levels of COX-2 proteins, they did not produce any measurable net amounts of prostaglandins, based on both utilization of radiolabelled arachidonic acid and the radioimmunoassay of prostaglandins E2 and F2 alpha. In contrast, we identified polyamine biosynthesis as a cellular target of aspirin, since the treatment of HT-29 Glc(-/+) cells with aspirin reduced the flux of L-ornithine through ornithine decarboxylase, an effect that could not be explained by an acute action of the drug on the ornithine decarboxylase catalytic activity. Since polyamine biosynthesis is strictly necessary for HT-29 cell growth, our data suggest that reduced flux through ornithine decarboxylase may participate in the antiproliferative activity of aspirin towards colonic tumoral cells. It is concluded that in HT-29 Glc(-/+) cells that are not functional for prostaglandin production, aspirin can affect cell growth, cell cycle, and polyamine biosynthesis without affecting cell membrane integrity.     

Differential mucin expression in colon carcinoma HT-29 clones with variable resistance to 5-fluorouracil and methotrexate

A current challenge is to define the biological characteristics of colon tumor cells resistant to chemotherapy. Distinct sub-populations of mucus-secreting cells were previously obtained from the colon cancer cell line HT-29 after long-term treatment with the anti-cancer drugs, 5-fluorouracil (5-FU) and methotrexate (MTX). Since mucins are increasingly implicated as playing a role in carcinogenesis, we studied the pattern of mucin expression in two HT-29 clones of mucus-secreting and two clones of enterocyte-like phenotype which differ in their capacity to resist to 5-FU and/or MTX. The expression of both transmembrane (MUC1, MUC3, MUC4) and secreted gel-forming (MUC2, MUC5AC, MUC5B, MUC6) mucins in clones was studied by northern and/or western blotting. The four HT-29 clones showed three cellular phenotypes: (1) The mucus-secreting clone HT29-5F12 consists of unpolarized cells with mucus secretions that have anti-colonic mucin immunoreactivity, and mainly expresses MUC2 and is resistant to 5-FU and sensitive to MTX; (2) The mucus-secreting clone HT29-5M21 forms a monolayer of polarized cells with strong anti-gastric mucin immunoreactivity and mainly expresses MUC5AC and MUC5B and is resistant to MTX and sensitive to 5-FU; (3) The two enterocyte-like clones, HT29-5F7 and HT29-5M12 are resistant to both MTX and 5-FU and express mainly MUC1 and MUC5B, respectively. These clones which originate from a same colorectal tumour and display different patterns of mucin expression as well as differing resistance to MTX and 5-FU will make useful in vitro models for studying the potential role of mucins or other biological markers in drug resistance pathways.     

A novel nitro-oxy substituted analogue of rofecoxib reduces human colon cancer cell growth

Rofecoxib is a specific COX-2 inhibitor able to exert antiproliferative activity against colorectal cancer cells. It was withdrawn from the market after the demonstration of an increased risk of cardiovascular complications after prolonged use. Nevertheless, it remains an interesting compound for laboratory research as an experimental COX-2 inhibitor. In this study, the antiproliferative activity of a novel dinitro-oxy-substituted analogue of rofecoxib (NO-rofe), potentially less cardiotoxic, has been investigated in vitro on human colon cancer cells and compared with the action of the parent drug. Due to the fact that COX-2 inhibition is the main characteristic of coxibs, we performed all experiments in COX-2-overexpressing (HT-29) and COX-2-negative (SW-480) human colon cancer cells, to elucidate whether the observed effects were dependent on COX-2 inhibition. Moreover, experiments were performed in order to evaluate whether COX-2 pharmacological inhibition may affect beta-catenin/E-cadherin signaling pathway. NO-rofe exerted a significant antiproliferative activity on COX-2 positive HT-29 human colon cancer cells, being less effective on the COX-2 negative SW-480 human colon cancer cell line. In particular, the rofecoxib analogue retained similar potencies with respect to COX-2 inhibition but was much more active than rofecoxib in inhibiting the growth of human colon cancer cells in vitro. In addition, this novel compound resulted in the induction of membrane β-catenin/E-cadherin expression, a feature that may significantly contribute to its antiproliferative activity.     

Necrosis predominates in the cell death of human colon adenocarcinoma HT-29 cells treated under variable conditions of photodynamic therapy with hypericin.

Photodynamic therapy (PDT) represents a new rapidly-developing anticancer approach based on administration of a non- or weakly-toxic photosensitizer and its activation with light of appropriate wavelength. Hypericin, one of the promising photosensitizers, is known to induce apoptosis with high efficiency in various cell line models. However, here we report the prevalence of necrosis accompanied by suppression of caspase-3 activation in colon adenocarcinoma HT-29 cells exposed to an extensive range of PDT doses evoked by variations in two variables -- hypericin concentration and light dose. Necrosis was the principal mode of cell death despite different PDT doses and the absence of anti-apoptotic Bcl-2 expression, even if the same condition induced caspase-3 activity at similar toxicity in HeLa cells. Introduction of Bcl-2 into HT-29 cells invoked caspase-3 activation, changed the Bcl-X(L) expression pattern, increased the apoptosis ratio with no effect on overall toxicity, and supported arrest in the G(2)/M-phase of cell cycle. Since it is known that Bcl-2 suppression in HT-29 is reversible and linked to the over-expression of mutated p53 and also considering our data, we suggest that the mutation in p53 and events linked to this feature may play a role in cell death signalling in HT-29 colon cancer cells.     

5-Fluorouracil mediated anti-cancer activity in colon cancer cells is through the induction of Adenomatous Polyposis Coli: Implication of the long-patch base excision repair pathway

Colorectal cancer (CRC) patients with APC mutations do not benefit from 5-FU therapy. It was reported that APC physically interacts with POLβ and FEN1, thus blocking LP-BER via APC's DNA repair inhibitory (DRI) domain in vitro. The aim of this study was to elucidate how APC status affects BER and the response of CRC to 5-FU. HCT-116, HT-29, and LOVO cells varying in APC status were treated with 5-FU to evaluate expression, repair, and survival responses. HCT-116 expresses wild-type APC; HT-29 expresses an APC mutant that contains DRI domain; LOVO expresses an APC mutant lacking DRI domain. 5-FU increased the expression of APC and decreased the expression of FEN1 in HCT-116 and HT-29 cells, which were sensitized to 5-FU when compared to LOVO cells. Knockdown of APC in HCT-116 rendered cells resistant to 5-FU, and FEN1 levels remained unchanged. Re-expression of full-length APC in LOVO cells caused sensitivity to 5-FU, and decreased expression of FEN1. These knockdown and addback studies confirmed that the DRI domain is necessary for the APC-mediated reduction in LP-BER and 5-FU. Modelling studies showed that 5-FU can interact with the DRI domain of APC via hydrogen bonding and hydrophobic interactions. 5-FU resistance in CRC occurs with mutations in APC that disrupt or eliminate the DRI domain's interaction with LP-BER. Understanding the type of APC mutation should better predict 5-FU resistance in CRC than simply characterizing APC status as wild-type or mutant.     

TGF-β acts as a dual regulator of COX-2/PGE2 tumor promotion depending of its cross-interaction with H-Ras and Wnt/β-catenin pathways in colorectal cancer cells

Transforming growth factor-β (TGF-β) plays a dual role acting as tumor promoter or suppressor. Along with cyclooxygenase-2 (COX-2) and oncogenic Ras, this multifunctional cytokine is deregulated in colorectal cancer. Despite their individual abilities to promote tumor growth and invasion, the mechanisms of cross regulation between these pathways is still unclear. Here, we investigate the effects of TGF-β, Ras oncogene and COX-2 in the colorectal cancer context. We used colon adenocarcinoma cell line HT-29 and Ras-transformed IEC-6 cells, both treated with prostaglandin E₂ (PGE₂), TGF-β or a combined treatment with these agents. We demonstrated that PGE₂ alters the subcellular localization of E-cadherin and β-catenin and enhanced the tumorigenic potential in HT-29 cells. This effect was inhibited by TGF-β, indicating a tumor suppressor role. Conversely, in Ras-transformed IEC-6 cells, TGF-β induced COX-2 expression and increased invasiveness, acting as a tumor promoter. In IEC-6 Ras-transformed cells, TGF-β increased nuclear β-catenin and Wnt/β-catenin activation, opposite to what was seen in the PGE₂ and TGF-β joint treatment in HT-29 cells. Together, our findings show that TGF-β increases COX-2 levels and induces invasiveness cooperating with Ras in a Wnt/β-catenin activation-dependent manner. This shows TGF-β dual regulation over COX-2/PGE₂ tumor promotion depending on the H-Ras and Wnt/β-catenin pathways activation status in intestinal cancer cells.     

Conjugated linoleic acid inhibits cell proliferation and ErbB3 signaling in HT-29 human colon cell line

Conjugated linoleic acid (CLA) has chemoprotective properties in experimental cancer models, and in vitro studies have shown that CLA inhibits HT-29 colon cancer cell growth. ErbB2 and ErbB3 have been implicated in the development of colon cancer, and both proteins are expressed at high levels in the HT-29 cell line. Activation of ErbB2/ErbB3 heterodimers is regulated by the ErbB3 ligand heregulin. To examine CLA regulation of HT-29 cell proliferation and apoptosis and the influence of CLA on the ErbB3 signaling pathway, HT-29 cells were cultured in the presence of CLA and/or heregulin. CLA inhibited DNA synthesis and induced apoptosis of HT-29 cells. Although the addition of heregulin-alpha led to an increase in cell number, it was not able to counteract the negative growth regulatory effect of CLA. Immunoprecipitation/Western blot studies revealed that CLA inhibited heregulin-alpha-stimulated phosphorylation of ErbB2 and ErbB3, recruitment of the p85 subunit of phosphoinositide 3-kinase (PI3-kinase) to the ErbB3 receptor, ErbB3-associated PI3-kinase activities, and phosphorylation of Akt. CLA decreased ErbB2 and ErbB3 mRNA and protein levels in a dose-dependent manner. In conclusion, we demonstrate that CLA inhibits cell proliferation and stimulates apoptosis in HT-29 cells and that this may be mediated by its ability to downregulate ErbB3 signaling and the PI3-kinase/Akt pathway.