Background
Bile acids are polar derivatives of cholesterol which are synthesized in the liver and stored in the gall bladder [
1]. During digestion bile is excreted into the intestinal tract where bile acids aid in the absorption of dietary fats. Although the majority of the bile acids is reabsorbed and reused a small fraction (1–4%) is not reabsorbed and passes into the colon [
2]. Here the primary bile acids, those bile acids that are produced in the liver, are modified by enteric bacteria dehydroxylating the cholesterol core and removing the conjugated amino acid to produce unconjugated secondary bile acids. These secondary bile acids, principally deoxycholic acid, have been associated with increased risk for colon cancer
Epidemiological and animal model studies support the concept that bile acids may play a role in the development of colon cancer. Studies of populations that eat high fat diets which promote more bile acid production show increased risk for colon cancer [
3,
4] and patients diagnosed with colon cancer have elevated levels of serum bile acids, especially deoxycholic acid (DCA) [
3,
5]. In studies using animal models DCA was found to act synergistically with carcinogens to increase colon tumorigenesis [
6,
7] and could cause transformation of cells in vitro [
8]. Collectively these observations suggest that DCA may be a tumor promoter. However it should be noted that not all bile acids act to promote colon tumor development. Ursodeoxycholic acid (UDCA) suppresses the development of colon tumors in AOM-treated rats [
9,
10] and two studies in human subjects suggest that UDCA can reduce the risk of developing colorectal cancer [
11‐
13]. Hence, in spite of having very similar chemical structures, these two bile acids have very distinct biological activities both at the organismal level as well as in vitro [
14]. To date the mechanism that accounts for this difference in function is not clear.
The mechanism through which bile acids bring about there biological effects is not well understood, however, there is a growing body of evidence indicating that bile acids can regulate gene expression [
15‐
18]. DCA has been shown to activate a number of mitogenic and apoptosis associated signaling pathways which is consonant with its proposed tumor promoting abilities including the epidermal growth factor receptor and the raf/mek/erk pathway [
19‐
21], protein kinase C [
22‐
24], the AP-1 transcription factor [
25‐
27], and Cox2 [
17] all of which are known to be dysregulated during colon tumorigenesis. Much less is known about the signaling mechanisms activated by UDCA. However, in general UDCA displays activities that are in opposition to those exhibited by DCA. For instance UDCA can suppress activation of ras, EGFR-raf/mek/erk pathway and AP-1 [
19] and is cytoprotective as opposed to cytotoxic DCA [
28,
29]. Similarly, while DCA interferes with functioning of the p53 tumor suppressor, UDCA does not [
21]. Interestingly, we found that bile acids are not readily taken up by colonic cells [
30], but instead initiate intracellular signaling by their action at cell membrane [
22] in ligand independent manner possibly through specialized domains like caveolae [
31]. Given the mode of action of DCA and UDCA on cell membranes, it is likely that these two bile acids can act on some of the same signaling pathways.
Understanding if DCA and UDCA utilize same pathways to bring about diametrically opposed biological outcome is very important in characterizing the role bile acids play in tumorigenesis. In addition, before therapeutically targeting DCA activated signaling to overcome the DCA mediated colon carcinogenesis, it is important to address whether UDCA employs similar signaling pathways as DCA. In order to gain insight into overlap in signaling pathways activated by DCA and UDCA, we isolated cell lines resistant to UDCA induced growth arrest and then tested these for their response to DCA induced apoptosis, since, the most significant biological effects for DCA and UDCA have been shown to be apoptosis and growth arrest, respectively. Characterization of these resistant cells demonstrated that some were also cross resistance to the effects of DCA suggesting that DCA and UDCA signaling activity may overlap. Importantly, we found evidence that resistance to some DCA-activated signaling lead to a more neoplastic phenotype. The relevance of these finding to colon cancer are discussed.
Methods
Reagents
DCA, cholic acid, and hyoDCA were obtained from Sigma Chemical (St. Louis, MO) and UDCA from Calbiochem (La Jolla, CA). All were maintained as 100 mM stock solutions in water. Upon addition of bile acids to media, no change in pH was observed. Etoposide, cisplatin, and adriomyosin were all obtained from Sigma Chemical Co. (St. Louis, MO)
Cell culture
The HCT116 colon cancer cell line was used as the parental cell line in all experiments and was purchased from the American Type Culture Collection (Rockville, Maryland). All cell lines were propagated at 37°C and 5% CO2 in a humidified atmosphere in Dulbecco's modified Eagle's medium (DMEM) (Gibco/BRL, Gaithersburg/MD) supplemented with 10% fetal bovine serum (Gibco/BRL), 100 units of penicillin, 100 mg of streptomycin, 2 mM L-glutamine, 4 mM sodium pyruvate and 100 μM non-essential amino acids.
Derivation of UDCA resistant cell lines
Parental HCT116 cells were plated onto a 162 cm2 flask with 50 milliliters of fresh DMEM and allowed to attach and grow for 24 hours. This produced a cell monolayer that was approximately 40% confluent. These cells were mutagenized by incubation with ethyl methane sulfonate (Sigma) at a final concentration of 500 μg/ml for 12 hours. Cells were then rinsed three times with DMEM, re-fed with fresh DMEM before returning to the incubator for 24 hours. Cells were then split into 20 ten centimeter dishes and allowed to grow for 24 hours prior to the addition of UDCA to a final concentration of 500 μM in each dish. Cells were refed with fresh DMEM supplemented with 500 μM UDCA once a week for four weeks at which time colonies of UDCA resistant cells appeared. From this treatment 47 UDCA resistant colonies emerged, 41 of which were successfully expanded into peremanent cell lines. Once the lines were expanded into 10 cm dishes, cells were maintained in DMEM supplemented with 250 μM UDCA. These lines were designated HOMUR cells for HCT116 Odd Morphology UDCA resistant.
Screening HOMUR lines for cross resistance to DCA and hyoDCA
HOMUR cells were plated at 100,000 cells per 35 mm dish and then incubated for 24 hours prior to the addition of either DCA or hyoDCA to a final concentration of 500 μM. Cells exposed to DCA were incubated for 24 hours and then harvested and the fraction of cells undergoing apoptosis determined as described below. HOMUR cells exposed to hyoDCA were incubated for 48 hours with this bile acid and then the fraction of apoptotic cells determined.
Apoptosis assay
For apoptosis assays 100,000 HCT116 cells were plated onto 60 mm tissue culture plates and allowed to attach for 24 hours. This procedure produced a cell monolayer that was 30–40% confluent at the time bile acids were added. The cells were treated with 500 μM bile acids for the times indicated. The media were removed and saved and the remaining attached cells rinsed in PBS and harvested by trypsinization. The cell pellet was re-suspended in the saved media. The number of apoptotic cells was then quantitated by staining with acridine orange and ethidium bromide as described previously [
14].
Anchorage independent growth
To test for anchorage independent growth cells were grown in 0.6% agarose as follows. A stock of 1.2% LMP agarose (Gibco) was autoclaved and then the solution equilibrated at 37°C for 30 minutes. The LMP agarose was diluted 1:1 with DMEM and one milliliter of the mixture poured into each well of a 6 well plate to form a basal layer. This basal layer was allowed to solidify for 10 minutes at 4°C prior to reequilibrating at room temperature for 30 minutes. The top layer was similar to the basal layer, but contained 5,000 cell per well. The top layer was allowed to solidify at room temperature for approximately 15 minutes and the plates were then transferred to a 37°C incubator with 5% CO2. The following day, one milliliter of medium was added to each well, and the cells refed every 3–4 days for 2.5 weeks. Three sets of experiments were performed in triplicate. The total number of colonies was counted and the percent colony formation determined.
Statistical analysis
Statistical analysis of data was performed using Sigmastat statistical analysis software. In all cases a p value of <0.05 was considered the threshhold for significance.
Discussion
In the present study we derived a set of cells that were resistant to UDCA and then tested these cells for cross resistance to the effects of two other bile acids to ascertain whether there was overlap in the signaling pathways that mediate bile acid-induced cell death. We were able to demonstrate that there is an overlap in the signaling mechanisms activated by UDCA which lead to growth arrest and those activated by DCA which bring about apoptosis. Careful examination of the number of UDCA-induced growth arrest resistant cells revealed that the majority of these lines also exhibited resistance to DCA-induced apoptosis a finding that is consistent with the concept that the signaling activities of these two bile acids may overlap. Most of the HOMUR lines exhibited some degree of resistance suggesting that the extent of overlap in signaling activities may be extensive. Hence, it seems likely that the signaling activities induced by bile acids and which are responsible for growth arrest and for apoptosis may have much in common.
The likely extensive overlap in signaling activities between DCA and UDCA raises the question of how these two bile acids can exhibit such distinctly different biological activities. Considering that all bile acids have such similar chemical structures it is not unexpected that they can also activate many of the same intracellular signaling mechanisms. However, there is also slight evidence that some bile acids interact with intracellular signaling in unique ways. For example DCA can stimulate the EGFR/ras/mek/erk signal transduction pathway, yet UDCA has been shown to suppress this same pathway [
19,
21]. Hence, although the same pathways may be targeted for modulation by the different bile acids the effect that they have on these pathways, activation or inhibition, may determine the ultimate biological outcome of exposure to these agents. This suggests that the distinction between tumor promotion and prevention may be very subtle
These notions emphasize the importance of elucidating the identity and the nature of the unique signaling mechanisms activated by DCA and UDCA. Insight into the characteristics of these pathways can be gleaned from our characterization of the HOMUR cells. Our observation that only the most profoundly DCA resistant HOMUR 7 line exhibits extensive growth in soft agar supports the notion that resistance to DCA-induced apoptosis favors a more tumorogenic phenotype. Our finding that HOMUR 7 cells are also markedly resistant to three commonly used anti-cancer agents suggests that DCA-induced apoptosis may utilize pathways that are also employed by cancer therapeutics. Hence, profound resistance to DCA correlates with acquired resistance to multiple other drugs each of which is known to cause cell death through very different mechanisms. Collectively these results suggest that resistance to bile acid-induced apoptosis is tumorigeneic and is consistent with findings made using natural human tumors [
33].
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
JDM was responsible for data analysis, drafting the manuscript and the overall direction of this study. AAP generated and partially characterized HOMUR clones. SA, SJL and RAF carried out growth curves, apoptosis assays and agarose anchorage independent growth assays. WQ carried out PARP western blots. PH assisted in characterizing HOMUR clones. All authors read and approved the manuscript.