Journal of Steroid Biochemistry & Molecular Biology 111 (2008) 37–40

Contents lists available at ScienceDirect

Journal of Steroid Biochemistry and Molecular Biology

journa l homepage: www.e lsev ier .com/ locate / j sbmb

Lithocholic acid down-regulation of NF-�B activity throughvitamin D receptor in colonic cancer cells

Jun Suna,∗, Reba Mustafib, Sonia Cerdab, Anusara Chumsangsrib, Yinglin Rick Xiaa,

Yan Chun Lib, Marc Bissonnetteb

chestity of C

condmonathesizumanway.eatmWe agicalIL-1�e effeave indatio

cated

a Department of Medicine, Gastroenterology and Hepatology, University of Rochester, Rob Department of Medicine, Martin Boyer Laboratories and IBD Research Center, Univers

a r t i c l e i n f o

Article history:Received 6 August 2007Accepted 14 January 2008

Keywords:Vitamin D receptorIL-1�NF-�B activityLithocholic acidIntestineInflammation

a b s t r a c t

Lithocholic acid (LCA), a seD3 (1,25(OH)2D3), the horVDR. Therefore, we hypopresent study, we used hinflammatory NF-�B pathof 1,25(OH)2D3. LCA pretr�B p65 phosphorylation.as a read-out of the bioloIL-8 secretion induced byThus, LCA recapitulated th(MEF) cells lacking VDR hTNF�-induced I�B� degraCollectively, our data indi

1. Introduction

The primary bile acids in human bile are chenodeoxycholic acidand cholic acid. They are secreted into the intestine to aid theabsorption of lipid-soluble nutrients. Within the intestines, the pri-mary bile acids are converted to the secondary bile acids. Bothprimary and secondary bile acids are reabsorbed by the intestinesand delivered back to the liver via the portal circulation. Bile acidsregulate gene expression by acting as ligands for the nuclear recep-tor or by activating kinase signaling pathways [1–3].

Recently, lithocholic acid, a potent tumor-promoting sec-ondary bile acid, was demonstrated to be a vitamin D receptor(VDR) ligand [4]. LCA binding to VDR induced the expression ofcolonic cytochrome P4503A (CYP3A) and dehydroepiandrosteronesulfotransferase, enzymes that detoxify LCA [4–7]. LCA activa-tion of the VDR also induced the expression of the multi-drugresistance-associated protein 3 (MP3) [8]. This basolateral mem-brane transporter exports LCA from the colonocyte to the blood for

∗ Corresponding author at: Department of Medicine, University of Rochester, 601Elmwood Avenue, Box 646, Rochester, NY 14642, USA. Tel.: +1 585 276 3798;fax: +1 585 275 8118.

E-mail address: jun sun@umrc.rochester.edu (J. Sun).

0960-0760/$ – see front matter © 2008 Elsevier Ltd. All rights reserved.doi:10.1016/j.jsbmb.2008.01.003

er, NY 14642, USAhicago, Chicago, IL 60637, USA

ary bile acid, is a vitamin D receptor (VDR) ligand. 1,25-Dihydroxyvitaminl form of vitamin D, is involved in the anti-inflammatory action throughe that LCA acts like 1,25(OH)2D3 to drive anti-inflammatory signals. Incolonic cancer cells to assess the role of LCA in regulation of the pro-

We found that LCA treatment increased VDR levels, mimicking the effectent inhibited the IL-1�-induced I�B� degradation and decreased the NF-lso measured the production of IL-8, a well-known NF-�B target gene,effect of LCA expression on NF-�B pathway. LCA significantly decreased. These LCA-induced effects were very similar to those of 1,25(OH)2D3.

cts of 1,25(OH)2D3 on IL-1� stimulated cells. Mouse embryonic fibroblasttrinsically high NF-�B activity. LCA pretreatment was not able to preventn in MEF VDR (−/−), whereas LCA stabilized I�B� in MEF VDR (+/−) cells.that LCA activated the VDR to block inflammatory signals in colon cells.

© 2008 Elsevier Ltd. All rights reserved.

excretion by the liver or kidney. Thus, LCA activation of the VDRaccelerates detoxification of the major secondary bile acids.

Moreover, recent study demonstrated that LCA can substitute

for vitamin D in vivo [9]. LCA elevates the serum calcium in vita-min D-deficient rats. LCA in the diet also replaces vitamin Din the mobilization of calcium from bone. Further, LCA inducesCYP24-hydroxylase mRNA gene expression in the kidney of vita-min D-deficient rats. It is clear, therefore, that LCA can be absorbedinto the circulation to bind to the VDR [9].

1,25(OH)2D3, the hormonal form of vitamin D, is involved in theanti-inflammatory action through VDR [10,11]. Our pervious studydemonstrated that VDR negatively regulated the pro-inflammatoryNF-�B pathway [12]. Mouse embryonic fibroblast (MEF) cells lack-ing VDR have intrinsically high NF-�B activity. VDR mutation ledto a marked increase in nuclear p65 DNA binding and NF-�B tran-scriptional activity; consistently, induction of IL-6 by TNF� or IL-1�was much more robust in VDR (−/−) than in VDR (+/−) cells [12].Collectively, cells lacking VDR are in a pro-inflammatory or pre-inflammatory state with high NF-�B activity.

LCA, like 1,25(OH)2D3, is expected to drive anti-inflammatorysignals that could protect the colon against VDR-independent bileacid-induced pro-inflammatory signals. In present study, we usehuman colonic epithelial cells to assess the role of LCA in regula-tion of NF-�B pathway. We also use MEF lacking VDR to determine

istry & Molecular Biology 111 (2008) 37–40

V

38 J. Sun et al. / Journal of Steroid Biochem

whether the LCA regulation of NF-�B is dependent on VDR. Our dataindicate that LCA activates the VDR to block inflammatory signalsin colon cells.

2. Materials and methods

2.1. Cell culture

Human colonic epithelial Caco-2 cells and HT29C19A (clone19A) cells were maintained in DMEM supplemented with 10% fetalbovine serum (FBS), penicillin–streptomycin and l-glutamine asprevious described [13]. MEFs were isolated from E13.5 embryosgenerated from VDR (+/−) × VDR (+/−) mouse breeding [14]. Thecells were cultured in DMEM containing 10% FBS. Cells from eachembryo were genotyped by PCR. VDR (+/−) and VDR (−/−) MEFswere used in experiments after more than 15 passages when theyhad been immortalized as shown previously [15].

2.2. Immunoblotting

Cultured cells were rinsed twice in ice-cold HBSS, lysed inprotein loading buffer (50 mM Tris, pH 6.8, 100 mM dithiothre-itol, 2% SDS, 0.1% bromphenol blue, 10% glycerol) and sonicated.Equal amount of proteins or equal volumes of total cultured celllysates were separated by SDS-polyacrylamide gel electrophore-sis, transferred to nitrocellulose, and immunoblotted with primaryantibodies (1:500 to 1000 dilution): anti-phospho-p65 on serine536 (Cell Signaling Technology, Beverly, MA, USA), anti-I�B�, anti-

DR (Santa Cruz Biotechnology, Santa Cruz, CA, USA), or �-actin(Sigma–Aldrich, St. Louis, MO, USA) antibodies and visualized byECL as previous described [13].

2.3. IL-8 secretion

Confluent Caco-2 cells were grown in 12-well plates. After treat-ment, the supernatant was collected and assayed for IL-8 using theR&D Systems human IL-8 ELISA kit (R&D, Inc., Minneapolis, MN,USA) according to the manufacturer’s instructions.

2.4. Statistical analysis

Data are expressed as mean ± S.D. Differences were analyzed byStudent’s t-test. p-Values <0.05 were considered significance.

3. Results

3.1. LCA treatment increases VDR expression

VDR is a nuclear transcription factor responsible for mediat-ing the biological activities of 1,25(OH)2D3. As a ligand-dependentreceptor, VDR expression is highly regulated by the vitamin D sta-tus [16,17]. Therefore, pretreatment of 1,25(OH)2D3 can increase theVDR protein expression. To confirm that LCA acts like 1,25(OH)2D3,we treated the human colonic epithelial Caco-2 cells with LCA andassessed VDR protein levels. Incubation of Caco-2 cells for 24 hwith non-toxic concentrations of LCA (20 �M) increased VDR levels,mimicking the effect of 1,25(OH)2D3 (Fig. 1A). In addition, the effectof LCA on VDR expression was also tested in the human colonicepithelial HT29C19A cells and mouse embryonic fibroblast cells. InHT29C19A and MEF cells, LCA treatment for 3 h up-regulated VDRexpression (Fig. 1B and C), the similar trend as the Caco-2 cells did.It reveals that our results are not restricted to one cell line.

Fig. 1. LCA induces VDR expression. (A) Caco-2 cells were incubated with1,25(OH)2D3 (20 nM), LCA (20 �M) or ethanol (–) for 3 h. Cells were lysed and pro-teins (35 �g) probed for VDR expression. Note: VDR expression is increased by LCAand 1,25(OH)2D3. (B) LCA treatment increased VDR expression in HT29C19A cells.Cells were incubated with 1,25(OH)2D3 (20 nM), LCA (20 �M) or ethanol (–) for 3 h.Cells were lysed for VDR expression. (C) LCA treatment increased VDR expression inMEF cells. Cells were incubated with LCA (20 �M) or ethanol (–) for 3 h. Cells werelysed for VDR expression. Beta-actin is the loading control. Images shown are froma single experiment and are representative of three separate repeats.

3.2. LCA pretreatment inhibits the IL-1ˇ-induced I�B˛degradation

To assess the effects of LCA on inflammatory signaling, we pre-treated Caco-2 cells for 24 h with LCA, or 1,25(OH)2D3 as a positivecontrol. Cells were then stimulated with IL-1�. In unstimulatedcells, NF-�B is bound to the inhibitor I�B� and restricted to thecytoplasm. With inflammatory stimuli, I�B� is phosphorylated and

subsequently degraded, releasing NF-�B to traffic to the nucleus[18]. We examined the expression of I�B� in the intestinal epithe-lial cells by immunoblotting. As expected, IL-1� treatment reducedI�B� level; LCA pretreatment was able to stabilize I�B� as the posi-tive control 1,25(OH)2D3 did (Fig. 2A I�B�). We further test whetherLCA has the similar effect in HT29C19A cells. Cells treated with IL-1� or TNF� induced I�B� degradation, whereas LCA pretreatmentwas able to stabilize I�B� (Fig. 2B I�B�).

3.3. LCA decreases the NF-�B p65 phosphorylation

It should be noted that NF-�B activity is regulated by phospho-rylation. Increased level of phosphorylated-p65 (p-p65) indicatesthe high activity of NF-�B pathway [19]. Interestingly, LCA treat-ment was able to inhibit the phosphorylation of the NF-�B subunit,p65 (Fig. 2A p-p65). Without stimulation, the p-p65 level was verylow in the control cells. With IL-1� stimulation, the level of p-p65increased dramatically. In contrast, the p-p65 remained low in theLCA pretreated cells after stimulation with IL-1�. Collectively, LCApretreatment decreased the NF-�B activity by stabilizing I�B� anddecreasing p-p65.

J. Sun et al. / Journal of Steroid Biochemistry &

Fig. 2. Effects of LCA and 1,25(OH)2D3 on IL-1� stimulated colon cancer cells. (A)Caco-2 cells were incubated for 24 h with 20 �M lithocholic acid (LCA, L) or 20 nM1,25(OH)2D3 (D3). Cells were then treated with 10 ng/ml IL-1� (I) or vehicle (Ctl).After 30 min cells were lysed and whole cell lysates probed for I�B� and p-p65expression. (B) HT29C19A cells were incubated for 24 h with 20 �M lithocholic acid(LCA, L). Cells were then treated with 10 ng/ml IL-1� (I), 10 ng/ml TNF� (T), or vehicle(Ctl). After 30 min cells were lysed and whole cell lysates probed for I�B� and �-actin expression. Images shown are from a single experiment and are representativeof two to three separate repeats.

3.4. IL-8 secretion regulated by LCA

Because NF-�B is a key regulator involved in the synthesis ofinflammatory cytokines, we also measured the production of IL-8, a well-known NF-�B target gene, as a read-out of the biologicaleffect of LCA expression on NF-�B pathway [20]. As shown here,LCA significantly decreased IL-8 secretion induced by IL-1� (Fig. 3,

Fig. 3. LCA inhibits IL-1� stimulated IL-8 secretion in colon cancer cells. Caco-2 cellswere incubated for 24 h with 20 �M lithocholic acid (LCA, L) or 20 nM 1,25(OH)2D3

(D3). Cells were then treated with 10 ng/ml IL-1� (I) or vehicle (Ctl) for 24 h. Thesupernatant was collected and assayed for IL-8 using the R&D Systems human IL-8ELISA kit (R&D, Inc., Minneapolis, MN, USA) according to the manufacturer’s instruc-tions. LCA and 1,25(OH)2D3 inhibit IL-1�-induced IL-8 secretion in Caco-2 cells,*p < 0.05 compared with IL-1� alone. Data are the mean ± S.D. of a single experimentassayed in triplicate and are representative of three separate experiments.

Molecular Biology 111 (2008) 37–40 39

Fig. 4. Effects of LCA is abolished in cells lacking VDR. MEF cells were incubatedfor 24 h with 20 �M lithocholic acid (LCA, L). Cells were then treated with 10 ng/mlTNF� (T) or vehicle (Ctl). After 30 min cells were lysed and whole cell lysates wereprobed for VDR and I�B� expression. Images shown are from a single experimentand are representative of two separate repeats.

p < 0.05). These LCA-induced effects were very similar to those of1,25(OH)2D3. Thus, LCA recapitulated the effects of 1,25(OH)2D3 onIL-1� stimulated Caco-2 cells.

3.5. VDR knockout cells abolish the LCA effect

We further investigated the effect of VDR ablation on LCA reg-ulation of NF-�B using MEFs derived from VDR-null mice, becausefibroblasts play an important role in inflammatory reactions andhave readily inducible NF-�B activity. In the present study, we usedVDR+/− cells, as we wanted to compare the difference betweenVDR-null and one allele of the Vdr gene. Similar results were seenwhen VDR+/+ MEFs were used. As shown in Fig. 4, VDR expressionwas enhanced with LCA pretreatment in VDR+/− cells, whereasno VDR signal was detected in VDR−/− cells. In VDR+/−, LCA pre-treatment stabilized I�B� level. And without LCA incubation inVDR+/−, TNF� stimulation decreased I�B� level. However, LCA can-not protect from TNF�-induced I�B� degradation in VDR−/− (Fig. 4).Additionally, I�B� baseline was lower in VDR−/− compared to thatin the VDR+/−, which is consistent with our previous results thatVDR expression negatively regulates NF-�B pathway and the base-line of I�B�, the inhibitor of NF-�B activity, in VDR−/− is lower thanthe wild-type cells (VDR+/−) [12].

4. Discussion

Our studies demonstrated that the VDR-dependent effects of

LCA on the pro-inflammatory NF-�B signals in colon cancer cells.LCA is able to inhibit the inflammation by enhancing the VDRexpression, stabilizing I�B�, decreasing NF-�B phosphorylation,and inhibiting IL-8 secretion. LCA pretreatment will not change theI�B� level in VDR−/− cells without the receptor binding with LCA.These results indicate that LCA–VDR delivers an anti-inflammatorysignal in colon cancer cells.

To our knowledge, it is the first report on the LCA activationof VDR, thus inhibiting TNF�- and IL-1�-induced NF-�B activa-tion in colonic epithelial cells. 1,25(OH)2D3 is involved in theanti-inflammatory action through VDR [10,11]. VDR expression ishighly regulated by the vitamin D status [16,17]. Therefore, pre-treatment of 1,25(OH)2D3 can increase the VDR protein expression.LCA, like 1,25(OH)2D3, also increases VDR expression and drivesanti-inflammatory signals that could protect the colon against VDR-independent bile acid-induced pro-inflammatory signals. Basedon our observation and previous studies [12], VDR expressionnegatively regulates NF-�B pathway. VDR is enhanced with LCApretreatment, thus decreasing the NF-�B activity. In cells lack-ing VDR, the baseline of the NF-�B activity is higher than thewild-type cells (VDR+/−). LCA pretreatment will not change the

istry &

[

[

[

[

[

[

[17] K.D. Healy, J.B. Zella, J.M. Prahl, H.F. DeLuca, Regulation of the murine renal

40 J. Sun et al. / Journal of Steroid Biochem

NF-�B activity in VDR−/− without the receptor binding with LCA.These data indicate that LCA regulation of NF-�B activity is VDRdependent.

It should be noted that high fecal LCA concentrations in themicromolar range compensate for the lower affinity of LCA forthe VDR compared to 1,25(OH)2D3 [4]. Although LCA binds to VDRwith low affinity, LCA can induce its own catabolism through theVDR. The molecular and functional comparison of 1,25(OH)2D3 andlithocholic acid has been reported [7]. Recent in vivo study [9]demonstrated that LCA can substitute for vitamin D in the eleva-tion of serum calcium in vitamin D-deficient rats. Therefore, it isclear that LCA can bind to VDR specifically and replace vitamin D.Thus, we believe LCA plays a quantitatively important role in VDRactivation in the colon [21].

Epidemiological studies indicate that Western style high fatdiets enhance colon cancer risk [22–24]. Since dietary factors arebelieved to contribute up to 70% of this tumor burden, increasedunderstanding of the pathogenic mechanisms involving the dietcould advance efforts to prevent this disease. Several mechanismsof inflammation and tumor promotion by dietary fat or bile acidshave been identified in experimental models of colon cancer. Inthese models the effects of fat on tumorigenesis occur during thepromotional phase. Bile acids can activate NF-�B, AP-1 and �-catenin in colon cancer cells [25–27]. These transcription factorsplay key roles in controlling inflammation, proliferation and colonictumorigenesis.

Studies of bile acid activated VDR pathways that can suppressinflammation will elucidate an unexplored area of diet-regulatedtumor suppressor effects. In the future, studies in VDR−/− andVDR+/+ mice will help to understand whether LCA inhibits inflam-mation by a VDR-dependent mechanism. New understanding ofpathways to block VDR-independent pro-inflammatory effects,while enhancing anti-inflammatory LCA–VDR signals will help

develop potentially new chemopreventive strategies for colonmalignancy.

Acknowledgements

We thank Sumalatha Kuppireddi for her technical support forIL-8 secretion. This study was supported by NIH grants DK075386(J.S.), University of Rochester Start-up fund (J.S.), NIH grantsCA36745 (M.B.), and P30DK42086 (Digestive Disease Research CoreCenter).

References

[1] D. Qiao, W. Chen, E.D. Stratagoules, J.D. Martinez, Bile acid-induced activationof activator protein-1 requires both extracellular signal-regulated kinase andprotein kinase C signaling, J. Biol. Chem. 275 (20) (2000) 15090–15098.

[2] P. Qin, L.A. Borges-Marcucci, M.J. Evans, D.C. Harnish, Bile acid signaling throughFXR induces intracellular adhesion molecule-1 expression in mouse liver andhuman hepatocytes, Am. J. Physiol. Gastrointest. Liver Physiol. 289 (2005)G267–G273.

[3] R.T. Stravitz, Y.P. Rao, Z.R. Vlahcevic, E.C. Gurley, W.D. Jarvis, P.B. Hylemon, Hep-atocellular protein kinase C activation by bile acids: implications for regulationof cholesterol 7 alpha-hydroxylase, Am. J. Physiol. Gastrointest. Liver Physiol.271 (1996) G293–G303.

[

[

[

[

[[

[

[

[

[

Molecular Biology 111 (2008) 37–40

[4] M. Makishima, T.T. Lu, W. Xie, G.K. Whitfield, H. Domoto, R.M. Evans, M.R.Haussler, D.J. Mangelsdorf, Vitamin D receptor as an intestinal bile acid sensor,Science 296 (5571) (2002) 1313–1316.

[5] B. Chatterjee, I. Echchgadda, C.S. Song, Vitamin D receptor regulation of thesteroid/bile acid sulfotransferase SULT2A1, Methods Enzymol. 400 (2005)165–191.

[6] I. Echchgadda, C.S. Song, A.K. Roy, B. Chatterjee, Dehydroepiandrosterone sul-fotransferase is a target for transcriptional induction by the vitamin D receptor,Mol. Pharmacol. 65 (2004) 720–729.

[7] P.W. Jurutka, P.D. Thompson, G.K. Whitfield, K.R. Eichhorst, N. Hall, C.E.Dominguez, J.C. Hsieh, C.A. Haussler, M.R. Haussler, Molecular and functionalcomparison of 1,25-dihydroxyvitamin D(3) and the novel vitamin D receptorligand, lithocholic acid, in activating transcription of cytochrome P450 3A4, J.Cell Biochem. 94 (5) (2005) 917–943.

[8] T.C. McCarthy, X. Li, C.J. Sinal, Vitamin D receptor-dependent regulation of colonmultidrug resistance-associated protein 3 gene expression by bile acids, J. Biol.Chem. 280 (2005) 23232–23242.

[9] J.A. Nehring, C. Zierold, H.F. DeLuca, Lithocholic acid can carry out in vivo func-tions of vitamin D, Proc. Natl. Acad. Sci. U.S.A. 104 (2007) 10006–10009.

10] A. Gurlek, M.R. Pittelkow, R. Kumar, Modulation of growth factor/cytokine syn-thesis and signaling by 1alpha, 25-dihydroxyvitamin D2: implications in cellgrowth and differentiation, Endocr. Rev. 23 (2002) 763–786.

[11] X.P. Yu, T. Bellido, S.C. Manolagas, Down-regulation of NF-kappa B protein levelsin activated human lymphocytes by 1,25-dihydroxyvitamin D3, Proc. Natl. Acad.Sci. U.S.A. 92 (1995) 10990–10994 (Published erratum appears in Proc. Natl.Acad. Sci. U.S.A. 93 (1) (1996) 524).

12] J. Sun, J. Kong, Y. Duan, F. Szeto, A.P. Liao, J.L. Madara, Y.C. Li, Increased NF-kappaBactivity in fibroblasts lacking the vitamin D receptor, Am. J. Physiol. Endocrinol.Metab. 291 (2) (2006) E315–E322.

13] R. Mustafi, S. Cerda, A. Chumsangsri, A. Fichera, M. Bissonnette, Protein Kinase-zeta inhibits collagen I-dependent and anchorage-independent growth andenhances apoptosis of human Caco-2 cells, Mol. Cancer Res. 4 (9) (2006)683–694.

14] Y.C. Li, A.E. Pirro, M. Amling, G. Delling, R. Baron, R. Bronson, M.B. Demay,Targeted ablation of the vitamin D receptor: an animal model of vitamin D-dependent rickets type II with alopecia, Proc. Natl. Acad. Sci. U.S.A. 94 (1997)9831–9835.

15] G.J. Todaro, H. Green, Quantitative studies of the growth of mouse embryo cellsin culture and their development into established lines, J. Cell Biol. 17 (1963)299–313.

16] M.E. Sandgren, H.F. DeLuca, Serum calcium and vitamin D regulate 1,25-dihydroxyvitamin D3 receptor concentration in rat kidney in vivo, Proc. Natl.Acad. Sci. U.S.A. 87 (1990) 4312–4314.

vitamin D receptor by 1,25-dihydroxyvitamin D3 and calcium, Proc. Natl. Acad.Sci. U.S.A. 100 (2003) 9733–9737.

18] G. Bonizzi, M. Karin, The two NF-kappaB activation pathways and their role ininnate and adaptive immunity, Trends Immunol. 25 (2004) 280–288.

19] H. Sakurai, H. Chiba, H. Miyoshi, T. Sugita, W. Toriumi, IkappaB kinases phos-phorylate NF-kappaB p65 subunit on serine 536 in the transactivation domain,J. Biol. Chem. 274 (1999) 30353–30356.

20] C. Kunsch, C.A. Rosen, NF-kappa B subunit-specific regulation of theinterleukin-8 promoter, Mol. Cell Biol. 13 (10) (1993) 6137–6146.

21] D.P. Burkitt, Epidemiology of cancer of the colon and rectum, Cancer 28 (1971)3–13.

22] J.R. Hecht, Dietary fat and colon cancer, Adv. Exp. Med. Biol. 399 (1996) 157–163.23] M. Lipkin, B. Reddy, H. Newmark, S.A. Lamprecht, Dietary factors in human

colorectal cancer, Annu. Rev. Nutr. 19 (1999) 545–586.24] B.S. Reddy, Y. Maeura, Tumor promotion by dietary fat in azoxymethane-

induced colon carcinogenesis in female F344 rats: influence of amount andsource of dietary fat, J. Natl. Cancer Inst. 72 (1984) 745–750.

25] A.W. Bull, B.K. Soullier, P.S. Wilson, M.T. Hayden, N.D. Nigro, Promotion ofazoxymethane-induced intestinal cancer by high-fat diet in rats, Cancer Res.39 (1979) 4956–4959.

26] F. Hirano, H. Tanada, Y. Makino, K. Okamoto, M. Hiramoto, H. Handa, I. Makino,Induction of the transcription factor AP-1 in cultured human colon adenocarci-noma cells following exposure to bile acids, Carcinogenesis 17 (1996) 427–433.

27] S.A. Shah, Y. Volkov, Q. Arfin, M.M. Abdel-Latif, D. Kelleher, Ursodeoxycholic acidinhibits interleukin 1 beta [corrected] and deoxycholic acid-induced activationof NF-kappaB and AP-1 in human colon cancer cells, Int. J. Cancer 118 (2006)532–539.