β-Catenin is involved in alterations in mitochondrial activity in non-transformed intestinal epithelial and colon cancer cells

Background: Alteration in respiratory activity and mitochondrial DNA (mtDNA) transcription seems to be an important feature of cancer cells. Leukotriene D4 (LTD4) is a proinflammatory mediator implicated in the pathology of chronic inflammation and cancer. We have shown earlier that LTD4 causes translocation of β-catenin both to the mitochondria, in which it associates with the survival protein Bcl-2 identifying a novel role for β-catenin in cell survival, and to the nucleus in which it activates the TCF/LEF transcription machinery. Methods: Here we have used non-transformed intestinal epithelial Int 407 cells and Caco-2 colon cancer cells, transfected or not with wild type and mutated (S33Y) β-catenin to analyse its effect on mitochondria activity. We have measured the ATP/ADP ratio, and transcription of the mtDNA genes ND2, ND6 and 16 s in these cells stimulated or not with LTD4. Results: We have shown for the first time that LTD4 triggers a cellular increase in NADPH dehydrogenase activity and ATP/ADP ratio. In addition, LTD4 significantly increased the transcription of mtDNA genes. Overexpression of wild-type β-catenin or a constitutively active β-catenin mutant mimicked the effect of LTD4 on ATP/ADP ratio and mtDNA transcription. These elevations in mitochondrial activity resulted in increased reactive oxygen species levels and subsequent activations of the p65 subunit of NF-κB. Conclusions: The present novel data show that LTD4, presumably through β-catenin accumulation in the mitochondria, affects mitochondrial activity, lending further credence to the idea that inflammatory signalling pathways are intrinsically linked with potential oncogenic signals.

Leukotriene D 4 (LTD 4 ) is a powerful proinflammatory mediator derived from arachidonic acid through the 5-lipoxygenase pathway. Arachidonic acid is the common precursor of a group of mediators collectively called eicosanoids (Funk, 2001). Leukotriene D 4 is the most potent of the cysteinyl leukotrienes (LTC 4 , LTD 4 , and LTE 4 ). Leukotriene D 4 is known to mediate its effects through specific cell surface receptors belonging to the G protein-coupled receptor family. The CysLT 1 and CysLT 2 receptors has been most studied (Heise et al, 2000;Lynch et al, 1999;Sarau et al, 1999) of which, the CysLT 1 receptor has been proven to have a much higher affinity for LTD 4 (Heise et al, 2000). Leukotriene D 4 signalling has been implicated in chronic inflammatory conditions such as asthma and inflammatory bowel diseases (IBDs) (Drazen, 2002).
Inflammatory bowel diseases are associated with an increased incidence of neoplastic transformation (Ekbom et al, 1990), suggesting that there is a link between inflammation and cancer (Sheng et al, 1997). Furthermore, earlier studies have shown that colon cancer is underrepresented in a population of patients with ulcerative colitis, who where treated with non-steroidal anti-inflammatory drugs (Smalley and DuBois, 1997). This suggests that inflammatory mediators, such as LTD 4, could be essential factors in mediating the coupling between IBD and colon cancer.
In 1930, Warburg showed that some types of cancer cells are capable of constitutive upregulation of glucose metabolism, even in the presence of the abundant oxygen. This phenomenon is known as the Warburg effect (Warburg, 1956). Cancer cells synthesise ATP mainly through 'anaerobic glycolysis', a metabolic state that is linked to high glucose uptake and local acidification due to lactate production, even in the presence of oxygen. Cancer cells often upregulate glycolysis enzymes as a result of constitutive signalling through the Akt pathway, or because of the expression of oncogenes such as Ras or Src (Elstrom et al, 2004;Pelicano et al, 2006). Earlier we have shown that LTD 4 induces Akt phosphorylation in Int 407 cells (Paruchuri et al, 2005).
A major finding in understanding increases in respiration is based on the well-established fact that there is an increase in the average of ATP/ADP ratio (Matsunaga et al, 1996). The stimulation of both respiration (oxidative phosphorylation) and glycolysis is presumably responsible for the rise in the ATP/ADP ratio, which begins almost simultaneously with the increase in respiration.
Certain types of chronic inflammation such as ulcerative colitis have long been associated with a high risk of cancer development (Coussens and Werb, 2002;Hussain et al, 2003). It is believed that increased free radical generation is one important mechanism that promotes the progression of chronic inflammation into malignant transformation (Hussain et al, 2003). Cancer patients commonly have decreased glucose clearance capacity, high glycolytic activity and raised lactate production. Therefore, it has been suggested that the observed pro-oxidative shift is mediated by an increased availability of mitochondrial energy substrate. The 'inflammatory oxidative conditions' are typically associated with an excessive stimulation of NAD(P)H oxidase by cytokines and other inflammatory mediators (Kohli et al, 2007). Increased reactive oxygen species (ROS) production or changes in intracellular glutathione levels are often involved with pathological changes, both of which are indicative of signal cascade, or gene expression dysregulation (Waris and Ahsan, 2006).
b-Catenin is a multifunctional protein, its function is altered in most colon cancers. In non-transformed cells it is present at the cell membrane, where together with E-cadherin forms part of the adherent-type junction (Conacci-Sorrell et al, 2002;Cullen et al, 2004). In the presence of a Wnt signal, b-catenin translocates to the nucleus in which it activates transcription factors of the TCF/LEF family (Korinek et al, 1997). b-Catenin has an essential role in normal cell physiology, which makes it an unsuitable target for antagonising with specific inhibitors or siRNA. We have shown earlier that b-catenin translocates to the nuclei of Int 407 cells after exposure to the pro-inflammatory mediator LTD 4 . We also found that after LTD 4 stimulation, b-catenin was present in mitochondria where it was associated with the anti-apoptotic protein, Bcl-2 . Here we further examine the effects of b-catenin on mitochondrial activity.

Cell culture
The non-transformed human intestinal epithelial cell line (Int 407) which shows typical epithelial morphology and growth, was isolated from jejunum and ileum of a human embryo of 2 months gestation (Henle and Deinhardt, 1957). Int 407 cells were cultured as a monolayer in Eagle's basal medium supplemented with 15% newborn calf serum. The colon cancer cell line Caco-2 was grown in Dulbecco's modified Eagle medium (Sigma Chemicals Co, St Louis, MO, USA) with 10% fetal bovine serum (Wikström et al, 2003). All media was supplemented with 2 mM L-glutamine, 55 IU ml -1 penicillin and 55 mg ml -1 streptomycin. The cell lines were cultured at 371C in a humidified atmosphere with 5% CO 2 . The cells were regularly tested to ensure the absence of mycoplasma contamination.

Transfection
Transfection was performed in complete medium for 24 h using PolyFect transfection reagent (Qiagen GmbH, Hilden, Germany), according to the manufacturer's instructions. In all transfection experiments it was routinely confirmed that empty vector had no effect.

Reverse transcription -PCR
Messenger RNA (mRNA) for the p65 NF-kB subunit was synthesised using the forward primer: GCGAGAGGAGCACAGAT ACCACCAA and the reverse primer: GGCAGATCTTGAGCTCGG CAGTGTT. mRNA isolation cDNA synthesis and quantitative real-time PCR Messenger RNA was isolated using the RNeasy Plus Mini Kit from Qiagen following the manufacturer's instructions. Complementary DNA (cDNA) was synthesised using the Fermentas Revert Aid H Minus First Strand cDNA Synthesis kit following the manufacturer's instructions, or with random hexamer primers and superscript II reverse transcriptase (Invitrogen, Carlsbad, CA, USA), according to the standard procedures. Complementary DNA was used as a template in quantitative real-time PCR reactions with TaqMan Gene Expression Assay (Applied Biosystems, Cambridge, UK) and Maxima Probe qPCR Mastermix (Fermentas Life Sciences, Vilnius, Lithuania). For relative quantification of expression levels, the comparative C t method was used. The point at which the fluorescence crosses the threshold was taken as the [C t ] value. Expression levels of the different genes of interest were normalised to the expression level of the housekeeping gene, b-actin or GAPDH.

ATP/ADP ratio determination
The transfected and untransfected cells were grown to 60% confluency, and thereafter stimulated or not with 40 nM LTD 4 . Cells were collected in 20% TCA and frozen on dry ice for ATP determination. ATP was determined using the ATP Kit SL (luminescent assays) from BioThema (Handen, Sweden). Untreated samples were diluted 40-fold and measured bioluminometrically. Thereafter, ATP was irreversibly converted to AMP with ATP sulfurylase in the presence of molybdate. In all, 20 ml of sample were mixed with 180 ml of Buffer I (50 mM/5 mM Tris-HCl/ MgCl 2 , pH 8.0, 10 mM Na 2 MoO 4 , 2.5 mM GMP, and 0.5 U ATP sulfurylase), incubated for 20 min at 301C, boiled for 2.5 min and chilled on ice. Samples were diluted with 800 ml milliQ-water and ATP was measured bioluminometrically. ADP was converted to ATP by pyruvate kinase. In total, 100 ml of the diluted sample from the previous step was mixed with 100 ml of the Buffer II (50 mM/ 5 mM Tris-HCl/MgCl 2 , pH 8.0, 38 mM KCl, 0.5 mM phosphopyruvate, and 50 U pyruvate kinase), samples were incubated at room temperature for 30 min. To terminate the reaction 300 ml of H 2 O was added and ATP was measured bioluminometricaly (Lundin, 2000). ADP was calculated as the difference between the ATP measured after incubation with pyruvate kinase, and the residual ATP measured after treatment with sulfurylase alone.

Transient transfection and luciferase assays
Luciferase assays were carried out using the Dual Luciferase Reporter Assay System from Promega. Each plasmid was used at a final concentration of 1 mg ml -1 , except for the control Renilla luciferase vector, which was always used at 0.2 mg ml -1 to standardise for transfection efficiency. Vector DNA was allowed to form complexes with PolyFect. Cells in 12-well plates were washed once in serum-free medium, and the DNA-Polyfect mixture was added. Cells were transfected at 371C for 24 h in complete medium, after which the medium was changed to normal growth medium, and the cells allowed to recover for 24 h. Before any further stimulation, the cells were left in serum-free medium for 1 h. The cells were then pre-treated or not with 30 mM NAC (N-acetyl-L-cysteine, Sigma Chemicals Co, an antioxidant capable of neutralising ROS) for 15 min and then incubated in the absence or presence of 40 nM LTD 4 for the indicated period of time. After stimulation with LTD 4 , cells were washed in PBS and lysed using 250 ml well -1 of the DLR passive lysis buffer provided in the assay. Lysed samples were collected and briefly centrifuged to precipitate the debris. A 20 ml volume of each lysate was used to measure luciferase activity with 50 ml luciferase assay substrate using a MiniLumat LB 9506 (Berthold Technologies, Hannover, Germany) luminometer. The control Renilla luciferase signal was recorded after the subsequent addition of 50 ml of Stop&Glow buffer, and the level of expression given as a ratio. In every experiment, triplicate samples were prepared and analysed for each condition.

Dihydroethidium staining
The oxidative fluorescent dihydroethidium (DHE) dye was used to evaluate the intracellular production of O 2 À K (Kitada et al, 2003). Dihydroethidium is cell permeable and reacts with O 2 À K to form ethidium, which in turn intercalates with DNA, providing nuclear fluorescence at an excitation wavelength of 520 nm and emission wavelength of 610 nm. The cells were grown to 60% confluency, and were transfected with HA-S33Y-Kb-catenin or HA-wt-Kbcatenin. After 24 h, the cells were serum starved overnight and the following day incubated with 50 mM DHE at 371C for 30 min. Then the cells were washed with PBS and serum-free culture media was added before fluorescent microscopy (Kohli et al, 2007). Exogenous addition of H 2 O 2 was used as a positive control. All images shown are of living cells maintained under physiological conditions by fluorescent microscopy using an Olympus microscope IX81 (Hamburg, Germany), objective Â 20.

Cell fractionation
Treatments were terminated by the addition of ice-cold lysis buffer (20 mM NaHepes pH 8.0, 2 mM MgCl 2 , 1mM EDTA, 5 mM orthovandate, 60 mg ml -1 PMSF and 4 mg ml -1 leupeptin) and the cells were placed on ice. After N 2 -decompression at 1000 psi for 10 min using a cell disruption bomb (Parr Instrument Company, Moline, IL, USA) and centrifugation at 200 g, the supernatant was centrifuged at 10 000 g for 10 min, and further centrifugation at 200 000 g for 1 h, to obtain an isolated cytosolic fraction.

Leukotriene D 4 increases metabolic activity in mitochondria
We have shown earlier that LTD 4 increases cell survival and causes translocation of b-catenin to mitochondria. Therefore, we analysed whether LTD 4 could affect mitochondrial activity. We observed a significant increase in the NADPH dehydrogenase activity in both the non-transformed epithelial cell line Int 407 and the cancer cell line Caco-2 ( Figures 1A and B) in response to the LTD 4 stimulation (40 nM). The maximum effect was reached after 15 min of stimulation with a significant increase after 1 h. The activity declined again after 2 h of stimulation. As a consequence of increased metabolism in the cell, an increase in the ATP/ADP ratio occurs, which contributes to plasma membrane depolarisation followed by activation of calcium channels leading to an increase in intracellular calcium (Fridlyand et al, 2005). Leukotriene D 4 increased intracellular calcium levels in both Int 407 and Caco-2 cells (Nielsen et al, 2005). We also found an increase in the ATP/ ADP ratio that was delayed compared with NADPH dehydrogenase activity, with a maximum effect after 1 h. The change was statistically significant at time points between 15 min and 1 h of LTD 4 stimulation in both cell lines ( Figures 1C and D).

Leukotriene D 4 increases mitochondrial gene transcription
Next, we examined whether LTD 4 had any effect on mitochondrial gene transcription. The mitochondrial DNA (mtDNA) is present at a high copy number per cell. It contains genes encoding 13 polypeptides, essential for protein synthesis in mitochondria, and a non-coding region called the displacement loop (D-loop). The D-loop is involved in the control of replication and transcription of mtDNA. The two mtDNA strands differ in their G þ T content and can therefore be separated into a heavy strand (H-strand), and a light strand (L-strand). In human cells each strand contains one single promoter for transcriptional initiation (Clayton, 1991;Ojala et al, 1981), and an additional initiation site for heavy strand transcription (Montoya et al, 1983). We chose to measure mRNA of one gene from each transcript of the two strands of the D-loop that control replication and transcription of mtDNA, in order to detect changes in mitochondrial-encoded gene activity. We observed a significant increase in mRNA for ND2 (NADH dehydrogenase subunit 2) both in Int 407 and Caco-2 cells (Figures 2A and B). This increase in transcription level occurred after 15 min of LTD 4 stimulation, with the maximum three-fold increase observed after 2 h in Int 407 cells and after 1 h in Caco-2 cells. The same was seen in both cell lines for the other gene transcripts, ND6 (NADH dehydrogenase subunit 6) and 16 s, with the maximum effect after 2 h stimulation with 40 nM LTD 4 in Int 407 cells and 1 h for the Caco-2 cells ( Figures 2C-E and F). The maximum increase (three-fold) was found for 16 s mRNA in Int 407 cells after 2 h of LTD 4 treatment ( Figure 2E).

b-Catenin affects mitochondrial activity
Int 407 is a non-transformed cell line showing epithelial cell morphology, in which b-catenin is mainly present at the sites of cell -cell junctions (i.e., at the plasma membrane) and follows a normal degradation pattern (Mezhybovska et al, 2005). The Caco-2 cell line, however, has a mutation in the APC gene. APC is a part of the b-catenin degradation complex. Caco-2 cells have a nonfunctional b-catenin degradation pathway, leading to accumulation of b-catenin in the cells. As seen in Figure 3A, the endogenous level of b-catenin in the Caco-2 cells is significantly higher than in Int 407 cells (12 mg of total protein from Int 407 cells and 4 mg from Caco-2 cells were loaded on the gel). Next we examined whether b-catenin affects mitochondrial activity. We have shown that b-Catenin is involved in mitochondria function M Mezhybovska et al b-catenin translocates to the mitochondria and interacts with the survival protein Bcl-2 . We overexpressed the following b-catenin constructs: b-cat wt and b-cat S33Y, and measured ATP/ADP ratio in the cells. It is worth mentioning that both b-catenin constructs are expressed equally in the cells (data not shown). Overexpression of both wild-type b-cat wt and mutant b-cat S33Y gave at least a 1.5-fold increase in ATP/ ADP ratio in Int 407 cells ( Figure 3B) and a 2.5-fold increase in Caco-2 cells ( Figure 3C). This effect was almost totally dependent on the b-catenin-mediated stimulation of the mitochondrial activity and not on the increase of number of the mitochondria, as seen from Figure 1A. The increase in viable cell number after wt or S33Y b-catenin overexpression were o0.2-fold after 24 h (118±11% (wt) and 122±10% (S33Y) of that seen after transfection with empty vector), which is o15% of the total increase in ATP/ADP ratio. Interestingly, LTD 4 stimulation of cells overexpressing the b-cat wt construct caused a further increase in ATP/ADP ratio compared with unstimulated cells. However, LTD 4 stimulation in cells overexpressing the b-cat S33Y construct did not cause any significant changes in mitochondrial activity ( Figures 3B and C). The empty vector control had no effect. The basal expression levels of mitochondrial-encoded genes were different between Int 407 and Caco-2 cells. The basal levels of mitochondrial gene expression were significantly higher in Caco-2 cells (Table 1A). The difference in b-catenin levels in the two cell lines obviously correlated with the basal expression profiles of their mitochondrial-encoded genes.
In Int 407 cells, there was a dramatic increase in the expression levels of all three genes ND2, ND6 and 16 s after transfection with b-cat wt or b-cat S33Y (Table 1B). These stimulated levels in Int 407 cells were compatible to basal expression levels of Caco-2 cells. After the same treatment in Caco-2 cells, the effect of LTD 4 on gene activity was less prominent. Activity of the ND2 and ND6 genes was not significantly different from cells transfected with empty vector, but 16 s transcription was increased 1.5-fold (Table 1C). In the Int 407 cells that had been transfected with wild-type or mutant b-catenin, LTD 4 stimulation induced a significant increase in ND2, ND6 and 16 s transcription. Stimulation of Caco-2 cells overexpressing wild-type b-catenin was not significantly different from the unstimulated cells for ND2, ND6 and 16 s expression. Stimulation of Caco-2 cells that overexpressed constitutively active b-catenin, resulted in a significant increase compared with unstimulated cells (the effect was 1.5-fold for all analysed genes). The difference in the response between Int 407 and Caco-2 cells to LTD 4 stimulation might be explained by the substantial difference in basal levels of gene expression between the two cell lines (Table 1A).

b-Catenin-mediated increase in mitochondria activity leads to NF-jB activation through ROS
Oxidative phosphorylation is the main source of ROS production. Increased mitochondrial activity is expected to induce ROS production. As we observed earlier that LTD 4 stimulation leads to an accumulation of b-catenin in mitochondria, we next examined whether LTD 4 stimulation and b-catenin overexpression resulted in elevated levels of ROS. The mean fluorescence intensity value per cells for unstimulated empty vector was 11.2 relative units (RU), calculated by the image program ImageJ (W Burger & MJ Burge, NIH, USA). Extracellular addition of H 2 O 2 was used as a positive control (mean fluorescent intensity per cell was 26.5 RU). Leukotriene D 4 increased ROS production in Int 407 cells, although the effect was less prominent than in the positive control (mean fluorescence per cell 16.1 RU; Figure 4). Overexpression of b-catenin constructs also caused an increase in ROS production. The downstream target for ROS in many cell types is the NF-kB family of transcription factors (Bubici et al, 2006). We found that overexpression of b-catenin resulted in the activation of the p65 subunit of NF-kB ( Figures 5A and B). Furthermore, the p65 subunit mRNA of NF-kB is increased in LTD 4 -stimulated cells. A similar increase was observed after overexpression of the b-cat wt or b-cat S33Y constructs. Leukotriene D 4 stimulation of cells overexpressing b-cat wt or b-cat S33Y construct did not significantly change the level of the p65 mRNA ( Figures 5A and B). We next analysed whether LTD 4 or b-catenin could affect the activity of NF-kB. After LTD 4 stimulation, we detected a three-fold increase in the activity of NF-kB responsive elements in Int 407 cells and a 1.5-fold increase in Caco-2 cells (Figures 5C and D). Overexpression of b-cat wt or b-cat S33Y significantly increased activity of NF-kB responsive elements in Int 407 (two-fold), and Caco-2 cells (1.4-fold). To analyse whether the increase in NF-kB activity was ROS mediated, we used an antioxidant to block ROS formation. Preincubation with N-acetylcysteine (NAC) significantly reduced the effect of LTD 4 and b-catenin overexpression in both cell lines (Figures 5C and D). We also analysed how NF-kB activation affected its downstream target gene Bcl-2. Leukotriene D 4 induced a significant increase in Bcl-2 expression that was further increased in cells transfected with either wt b-catenin or   were totally dependent on the LTD 4 -induced ROS production as shown by the use of the ROS inhibitor NAC that totally blocked the LTD 4 -induced increase in Bcl-2 expression ( Figure 6A). To show the specificity of ROS/NF-kB activation we also analysed the wellknown target gene of b-catenin/TCF activity Cyclin D1. We found that both LTD 4 and b-catenin overexpression significantly increased Cyclin D1 expression ( Figure 6B). In contrast to the  Figure 6B).

DISCUSSION
We have earlier observed that exposure of Int 407 cells to LTD 4 causes translocation of b-catenin to the mitochondria, and association with the anti-apoptotic protein Bcl-2, thereby enhancing cell survival Paruchuri et al, 2006). Here we suggest a possible mechanism linking b-catenin to survival signalling in the inflammatory environment. We observed an increase in complex I activity and an increase in ATP/ADP ratio in Int 407 and Caco-2 cells. It is worth mentioning that both Lef-1 and mitochondria transcription factor TFAM1 bears the similar HMG domain that is not only responsible for DNA binding but for protein -protein interaction as well. It has earlier been shown that ATP/ADP ratio is crucial for the decision of the cell's fate (Leist et al, 1997;Richter et al, 1996). Furthermore, Vrbacky et al (2003), showed that a decrease in ATP/ADP ratio in BSC-40 cells is a hallmark of apoptosis. It has been shown in tumour cells (HeLa, MCF-7, and HL-60) that mitochondrial ATP supported increased cellular proliferation rates (Guppy et al, 2002;Sweet and Singh, 1995). In agreement with earlier findings our data strongly indicate that an increase in ATP/ADP ratio favours survival signalling in the cell. Leukotriene D 4 increased mitochondrial gene activity in both Int 407 and Caco-2 cells as the activity of the mtDNA encoding for complex I was also increased transiently in both Int 407 cells and Caco-2 cells.
Changes in mitochondria gene activity as well as mutations in the mitochondrial genome are reported for many cancers (Pelicano et al, 2004;Polyak et al, 1998;Waris and Ahsan, 2006). ND2 is overexpressed in human acute myeloid leukaemia cells (Matsunaga et al, 1996). An increased level of 16 s RNA was shown in polyps of familial polyposis coli patients (Yamamoto et al, 1989). Mutations in 16 s RNA has been reported for colorectal tumours as well (Polyak et al, 1998). b-Catenin overexpression in the nontransformed cell line, Int 407, had a dramatic effect on mitochondrial gene activity, but in cancer cells b-catenin overexpression did not show any significant differences. Caco-2 cells have higher levels of b-catenin compared with Int 407 cells and as a result, higher activity of the mitochondrial genome. Therefore, the effect of b-catenin overexpression on mitochondrial gene activity in Caco-2 cells was not as prominent as in Int 407 cells.
Reactive oxygen species are formed predominantly through oxidative phosphorylation. Formation of ROS is often increased in cancer in order to maintain cell growth and proliferation (Chen et al, 2007;Chen et al, 1998;Irani et al, 1997). Therefore, it was expected that ROS production would be elevated in this system. We observed increased ROS production in Int 407 cells (but not to the extent of the positive control, H 2 O 2 ), upon LTD 4 stimulation and after b-catenin overexpression. The cell fate decision for suicide or survival is dependent on signalling pathways activated upon oxidative stress. High doses lead to cell death while a moderate increase favours cell survival signalling (Martindale and Holbrook, 2002). The observed change in ROS was not as significant as in H 2 O 2 stimulation, leading to the conclusion that a moderate increase in ROS levels in this system mediates survival signalling.
Earlier we showed that LTD 4 facilitates survival signalling through both PI-3 kinase/Akt/GSK-3b-and PKC/Erk1/2-dependent mechanisms Paruchuri et al, 2002). Erk activation as a result of oxidant injury was reported in 1996 (Guyton et al, 1996), resulting in cell survival and tumour growth. Akt activation in response to oxidant injury contributes to survival signalling in peroxide-induced apoptosis in a human glioblastoma cell line (Sonoda et al, 1999). Both Erk and Akt signalling has been shown to be upregulated upon LTD 4 stimulation in the nontransformed intestinal cell line Int 407 Paruchuri et al, 2002). In the Caco-2 cancer cell line, we observed ROS formation in untreated cells, in which the treatment with LTD 4 did not have any additional effect (data not shown). This is not surprising since the elevated levels of ROS have been connected to cancer development for several decades (Pelicano et al, 2004). It was shown that extracellular addition of O 2 resulted in activation of NF-kB, and that this effect was ROS mediated (Cazals et al, 1999;Suzuki et al, 2000). We found an increase in activity of the p65 subunit of NF-kB after LTD 4 stimulation for 1 h in Int 407 and Caco-2 cells. Earlier reports from our group did not find any significant effect on NF-kB activation in Int 407 cells after 8 h or longer stimulation (Bengtsson et al, 2008). Overexpression of wild-type b-cateninand bÀcatenin-S33Y constructs also activated the p65 subunit of NF-kB. We also checked whether the activation of p65 subunit resulted in its increased activity. We used a luciferase assay to test activity of the p65 response element. The p65 responsive element activity was elevated after LTD 4 stimulation as well as after b-catenin overexpression in both cell lines. There is still some confusion in the literature regarding possible cross-talk of NF-kB and b-catenin signalling. b-Catenin has been reported to regulate NF-kB in tumour cells (Amit and Ben-Neriah, 2003). Other reports suggest that NF-kB is a regulator of the b-catenin pathway (Wang et al, 2007), whereas some studies reported that ROS activity is subject to negative feedback regulation by NF-kB (Bubici et al, 2006). We used the antioxidant (NAC) to validate whether ROS is upstream or downstream of NF-kB. p65 activation seems to be ROS mediated as the LTD 4 mediated increase in ROS was reduced after antioxidant treatment. We also b-Catenin is involved in mitochondria function found that the well-known downstream target gene of NF-kB activity Bcl-2 was induced by LTD 4 stimulation or b-catenin overexpression. These effects were mediated through ROS production, which is in contrast to the effects of LTD 4 stimulation and b-catenin overexpression on b-catenin/Tcf-mediated Cyclin D1 expression.
In conclusion, we have shown earlier that LTD 4 induces b-catenin translocation to the mitochondria and increases survival by interacting with the cell survival protein Bcl-2. Here we describe for the first time that b-catenin is involved in signal transduction leading to increased activity of the respiratory chain, which in turn leads to an increased ROS production and as a consequence, NF-kB activation. NF-kB is one of the key players in inflammation, linking chronic inflammation and cancer development. Consequently, the present data add further support to the idea that NF-kB activation might be crucial for LTD 4 -induced cell survival.