Pifithrin-α Enhances Chemosensitivity by a p38 Mitogen-Activated Protein Kinase-Dependent Modulation of the Eukaryotic Initiation Factor 4E in Malignant Cholangiocytes

Pifithrin-α is a small molecule that was originally identified as a p53 inhibitor on chemical screens, and it represents a promising new class of therapeutic agents that are being developed and evaluated for use in cancer and neurodegenerative diseases (Komarov et al., 1999; Gudkov and Komarova, 2005). The utility of pifithrin-α in anticancer therapy remains undefined, partly due to a lack of knowledge regarding the mechanisms and targets by which it may exert cellular effects. Although originally identified as an inhibitor of p53, recent studies have shown that pifithrin-α can modulate intracellular signaling independent of p53 by acting as a potent agonist of the aryl hydrocarbon receptor (AhR) (Hoagland et al., 2005). Moreover, there is limited information about the potential interactions with other chemotherapeutic agents.

We examined the utility of pifithrin-α as an adjunct to therapy for the treatment of human cholangiocarcinomas, malignancies arising from the biliary tract. These tumors are highly refractory to conventional treatments; consequently, they are associated with a poor prognosis. Recent reports indicate that global incidence and mortality of this aggressive cancer are increasing (Patel, 2006). Due to a limited ability to detect the tumor early, cholangiocarcinoma is often diagnosed at an advanced disease stage when curative resection is not possible. Thus, there is an urgent need for newer, more effective anticancer approaches for this cancer. An understanding of signaling mechanisms regulating chemotherapy resistance in cholangiocarcinoma may eventually result in the development of more effective therapies and enhance patient long-term survival.

Dysregulation of protein synthesis has recently been recognized as a contributing event in tumorigenesis. The eukaryotic translation initiation factor-4E (eIF-4E) is an mRNA cap binding protein that can regulate the initiation of translation and is becoming increasingly recognized as a critical mediator of tumor cell growth and responses to environmental stresses such as chemotherapy. eIF-4E binds to the m⁷-G cap on mRNA, and it is a rate-limiting determinant of cap-dependent translation. Overexpression of eIF-4E has been shown to contribute to tumorigenesis by enhancing the growth and survival of transformed cells in many cancers (for review, see De and Graff, 2004). Increased eIF-4E expression has been correlated with the increased rate of translation of several genes, including many known malignancy-associated proteins such as ornithine decarboxylase, cyclin D1, c-myc, vascular endothelial growth factor, and fibroblast growth factor (for review, see Mamane et al., 2004). eIF-4E may be an important contributor of resistance to chemotherapeutic agents by activating antiapoptotic or cell survival pathways. Indeed, modulation of eIF-4E by antisense approaches to decrease expression of eIF-4E or by altered eIF-4E binding to the mRNA cap have been shown to modulate cell survival as well as to decrease tumor cell growth. Translational regulation of survival proteins can result in resistance to therapy, and increased eIF-4E can modulate apoptotic responses to hypoxia in human malignant biliary tract epithelia (cholangiocytes) (Marienfeld et al., 2004).

The activity of eIF-4E can be modulated by eIF-4E inhibitory proteins. The activity of these proteins can be modulated by phosphorylation with a decrease in phosphorylation enhancing binding to eIF-4E and an increase in phosphorylation resulting in their detachment from eIF-4E, thereby enabling eIF-4E to bind to mRNAs. Phosphorylation of eIF-4E at serine 209 occurs once eIF-4E is bound to the scaffolding protein eIF-4G by the mitogen-activating protein kinase signal-integrating kinases Mnk-1 and Mnk-2, which can be activated by extracellular signal-regulated kinase or p38 MAPK kinase. We have recently shown that p38 MAPK signaling is aberrantly activated in malignant cholangiocytes, and moreover, that it can directly regulate protein translation and eIF-4E (Tadlock and Patel, 2001; Yamagiwa et al., 2003). Thus, aberrant p38 MAPK signaling can stimulate eIF-4E activity and promote either tumor proliferation or survival. Thus, targeting eIF-4E may be an effective strategy in developing successful treatments for human cholangiocarcinoma. Developing such strategies necessitates understanding molecular mechanisms of regulation of eIF-4E in cancer cells.

Recent studies have demonstrated that eIF-4E can be regulated by the p53 transcription factor (Horton et al., 2002; Constantinou and Clemens, 2005; Tilleray et al., 2006). To elucidate potential mechanisms involved in eIF-4E-mediated chemoresistance, we evaluated the effect of aberrant eIF-4E expression to pifithrin-α exposure. We asked the following questions: Is eIF-4E overexpressed in human cholangiocarcinoma? What is the cellular effect of pifithrin-α on responses to chemotherapy? Is eIF-4E involved in the pifithrin-α response? If so, does this involve p38 MAPK signaling? Our study shows that eIF-4E is a target of pifithrin-α, and it supports the potential utility of pifithrin-α to improve the cellular response to chemotherapeutic agents.

Materials and Methods

Cell Lines and Culture. H69 nonmalignant and KMCH malignant human cholangiocytes were obtained as described previously (Park et al., 1999a,b). KMCH cells were stably transfected with dominant-negative MKK3 as described previously, and the stable transfectants were designated as KM-MKK3dn (Yamagiwa et al., 2003). H69, KMCH, and KM-MKK3dn cells were cultured in Dulbecco's modified Eagle's medium with necessary supplements as described previously, and unless otherwise noted, culture media were supplemented with 10% fetal bovine serum.

Tissue Microarray and Analysis. Tissue microarray slides containing samples from normal livers and human cholangiocarcinoma were obtained from Abxis (Seoul, Korea). Each unique tumor sample was represented in duplicate. Immunohistochemistry was performed following the manufacturer's protocol and using primary monoclonal antibodies for eIF-4E at a dilution of 1:500, and a biotinylated secondary antibody with detection using streptavidin-horseradish peroxidase (Zymed Laboratories, San Francisco, CA). Each section was scored using a numerical value of 1 to 4 for none, mild, moderate, or intense staining intensity. The number of positive cells in each section was determined and assigned numerical values of 1 to 5 for 0, <25, 25 to 50, 50 to 75, or >75% positive cells, respectively. The values for the staining intensity and positive cells were multiplied, and values of 1 to 5, 6 to 10, 11 to 15, and 16 to 20 were assigned an expression score of 1, 2, 3, and 4, respectively.

RNA Interference. The design and validation of siRNA to eIF-4E was described previously (Yamagiwa et al., 2003). Validated siRNA to p53 and scrambled nucleotide control siRNA were obtained from Ambion (Austin, TX). Transfection of siRNA was performed using the Nucleofector system (Amaxa Biosystems, Koln, Germany). One million cells were resuspended in 100 μl of Nucleofector solution T at room temperature followed by addition of 1.5 μg of siRNA eIF-4E, p53, or scrambled nucleotide control, and the cells were transfected by electroporation with Nucleofector. Cells were then resuspended in DMEM containing 10% serum before use for subsequent studies.

Cytotoxicity Assay. Cells were transfected with siRNA to either eIF-4E or p53 or scrambled nucleotide control, and then they were seeded into 96-well plates at a cell density of 10,000 cells/well in 200 μl of DMEM with 10% fetal bovine serum. After 24 h, the cells were washed with 1× phosphate-buffered saline and pretreated for 24 h with varying concentrations of pifithrin-α in serum-free DMEM. The media were then changed and replaced with media containing 0 (diluent control) or 100 μM gemcitabine. Cell viability was assessed after 24 h using a colorimetric assay (Cell Titer 96 Aqueous; Promega, Madison, WI). Cytotoxicity is expressed as a percentage of control. For studies to evaluate the involvement of aryl hydrocarbon receptor signaling, cells were pretreated with 0.1 and 1 μM α-naphthoflavone, before incubation with pifithrin-α and gemcitabine as noted above.

Clonogenic Growth Assay. Cells were seeded in 96-well plates (10,000 cells/well) in an agar suspension containing media with 20% fetal bovine serum. Then, cells were incubated at 37°C with varying concentrations of pifithrin-α and gemcitabine. Growth in soft agar was assessed after 7 days, following addition of Alamar Blue (Bio-Source International, Camarillo, CA), and measurement of fluorescence was accomplished using a CytoFluor Multiwell Plate Reader with excitation at 530 nm and emission at 580 nm.

Immunoblot Analysis. KMCH and KM-MKK3dn cells were grown to ∼70% confluence in 100-mm dishes. Cells were washed twice with phosphate-buffered (PBS) before incubation with varying concentrations of pifithrin-α in serum-free DMEM for 48 h. Subsequently, cells were washed with ice-cold PBS and then lysed with 0.5 ml of lysis buffer containing protease inhibitors for total protein (Tadlock and Patel, 2001). Equivalent amounts of protein samples were mixed with 4× sample buffer, separated on 4 to 12% gradient polyacrylamide gels (Novex; San Diego, CA), and then transferred to nitrocellulose membrane (Millipore Corporation, Billerica, MA). The membranes were blocked with 5% nonfat dry milk in Tris-buffered saline, pH 7.4, containing 0.05% Tween 20 (TBST) for 1 h and then incubated overnight at 4°C with the respective anti-human primary antibody (1:1000). The membrane was washed three times for 5 min with TBST and then incubated with IRDye700 (Invitrogen, Carlsbad, CA) and IRDye800-labeled (Rockland, Gilbertsville, PA) secondary antibodies (1:2000) for 30 min. Blots were stripped and reprobed with β-actin (1:2000 primary antibody; 1:4000 secondary antibody). Protein expression was visualized and measured using a LI-COR Odyssey infrared imaging system (LI-COR Bioscience, Lincoln, NE).

MAPK Assay. KMCH cells (1.5 × 10⁶) in 100-mm dishes were incubated with 25 μM SB203580 or diluent control for 24 h before the addition of 25 μM pifithrin-α or diluent control for 48 h. After 48 h, total protein was obtained, and total and active site phosphorylation-specific p38 MAPK activity was measured using a commercially available assay kit from BioSource International (Invitrogen), following the manufacturer's instructions.

Cell Cycle Analysis. KMCH cells (1.5 × 10⁶) in 100-mm dishes were serum-starved for 24 h. Subsequently, cells were incubated with 0 or 25 μM pifithrin-α in 10% serum-containing medium. After 24 h, cells were harvested and resuspended (0.5 to 1 × 10⁶ cells/ml) in propidium iodide solution containing 0.1 mM propidium iodide, 0.1% Triton X-100, and 0.25 mg/ml RNase A in PBS at 4°C for 1 h in the dark. Cell cycle analysis was performed using an FACSAria flow cytometer (BD Biosciences, San Jose, CA). Ten thousand events were recorded, and the percentage of cells in the cell cycle phases was analyzed using FACSDiva software version 4.1 (BD Biosciences).

Reverse Transcription-Polymerase Chain Reaction. Total RNA was isolated using the ToTALLY RNA kit (Ambion). cDNA was prepared from 2 μg RNA using the Moloney murine leukemia virus reverse transcriptase kit (Invitrogen) and random primers. The PCR primer sequences were CYP1A1, 5′-TCT TTC TCT TCC TGG CTA TC-3′ (sense) and 5′-CTG TCT CTT CCC TTC ACT CT-3′ (antisense); and β-actin, 5′-CAG AGC AAG AGA GGC ATC CT-3′ (sense) and 5′-TTG AAG GTC TCA AAC ATG AT-3′ (antisense). The PCR mixture contained 1 μl of cDNA, 1× PCR SuperMix high-fidelity buffer (Invitrogen) containing a mixture of Taq and Pyrococus GB-D DNA polymerase, Mg²⁺, and dNTPs, 200 nM each primers in a final volume of 50 μl. Amplification was performed on a Genius thermal cycler (Techne, Princeton, NJ). Reactions were hot-started at 95°C for 5 min and then amplified for 35 cycles (30 s at 95°C, 30 s at 55°C, and 30 s at 72°C), followed by a final 10-min extension at 72°C. Controls with total RNA were performed for each set of primers. Then, 1 μl of each PCR reaction was resolved on a 500 DNA chip and analyzed on an Agilent 2100 Bioanalyzer (Agilent Technologies, Waldbronn, Germany).

DNA Sequencing. KMCH cells were grown in 100-mm dishes in 10% serum containing DMEM until confluence. Cells were washed twice with ice-cold PBS, and genomic DNA was isolated from the cells using a commercially available DNA extraction kit (Chemicon International, Temecula, CA). PCR amplification was performed as described above using published sequencing primers for p53 exons 4 to 9 (Rassidakis et al., 2005). Reactions were hot-started at 95°C for 5 min and then amplified for 35 cycles (30 s at 95°C, 30 s at 56°C, and 1 min at 72°C), followed by a final 5-min extension at 72°C. PCR reaction products were confirmed by on a 500 DNA chip with Agilent 2100 Bioanalyzer. p53 sequencing was performed at the TGen DNA Sequencing Facility (Phoenix, AZ), and sequence data analyzed using Mutation Surveyor (SoftGenetics, LLC, State College, PA).

Materials. Cell culture media, supplements, and PCR primers were obtained from Invitrogen. Pifithrin-α and SB203580 were obtained from Calbiochem (San Diego, CA). α-Naphthoflavone was from Sigma-Aldrich (St. Louis, MO). Antibodies to eIF-4E, CYP1A1, and β-actin were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Antibodies to p53, Ser209-eIF-4E, and phospho-Mnk-1 were obtained from Cell Signaling Technology Inc. (Beverly, MA). All other reagents were obtained from Sigma-Aldrich.

Statistical Analysis. Data are expressed as the mean ± 95% confidence limits from at least three separate experiments. The difference between groups was analyzed using a double-sided Student's t test. Statistical significance was considered as P < 0.05.

Results

Is eIF-4E Expression Increased in Human Cholangiocarcinoma? We have recently shown that the expression of eIF-4E is increased in malignant human cholangiocarcinoma cell lines compared with that in nonmalignant cholangiocytes (Marienfeld et al., 2004). However, it is unknown whether eIF-4E is overexpressed in human cholangiocarcinoma. Therefore, we assessed the expression of eIF-4E in 46 human cholangiocarcinomas using a tissue microarray. Compared with expression in normal liver, there was an increase in eIF-4E expression in cholangiocarcinoma samples (Fig. 1A). A semiquantitative analysis indicated low expression in all benign tissue samples. However increased expression (score of ≥2) was observed in 85.9% cholangiocarcinomas and was markedly increased (score ≥3) in 51.1% cholangiocarcinomas (Fig. 1B). Thus, eIF-4E expression is increased in human cholangiocarcinoma.

Can Pifithrin-α Alter the Sensitivity of Malignant Cholangiocytes to Gemcitabine? Pifithrin-α is currently under investigation as an adjunct to chemotherapy to minimize the side effects of anticancer therapies or as the basis for the development of new drugs (Zhu et al., 2002). Although pifithrin-α may be potentially useful for the treatment of chemoresistant tumors such as cholangiocarcinoma, there is a lack of information about the effect of pifithrin-α in cholangiocytes, and its utility in enhancing chemotherapeutic response is unknown. To assess the cellular responses of pifithrin-α, we first assessed the cytotoxicity of pifithrin-α. Indeed, pifithrin-α did not alter KMCH cell viability over a concentration range from 0 to 100 μM (Fig. 2A). Gemcitabine is the most effective single-agent currently available for cholangiocarcinoma. However, treatment responses are poor. Thus, we next evaluated the potential of pifithrin-α as an adjunct to gemcitabine. Incubation of KMCH cells with 50 μM pifithrin-α for 24 h enhanced sensitivity to gemcitabine compared with controls (Fig. 2B). Likewise, an increase in anchorage-independent growth in soft agar was also observed in cells incubated with pifithrin-α (Fig. 2C). These observations suggest that pifithrin-α is a potentially useful adjunct to chemotherapeutic agents such as gemcitabine for the treatment of cholangiocarcinoma. Chemosensitization by pifithrin-α was not observed in H69 nonmalignant human cholangiocytes (Fig. 2D), suggesting that the observed effects were specific to malignant cells and supporting the potential use of pifithrin-α as an adjunct to chemotherapeutic agents.

Is eIF-4E Involved in the Chemosensitization Effects of Pifithrin-α? Having established that pifithrin-α can enhance the effects of gemcitabine, we next evaluated the involvement of eIF-4E in mediating this effect. Pifithrin-α enhanced the response to gemcitabine in KMCH cells transfected with scrambled nucleotide control siRNA (Fig. 3A). However, in cells transfected with siRNA to eIF-4E, the effect of pifithrin-α on enhancing gemcitabine sensitivity was decreased. Thus, eIF-4E is involved in pifithrin-α chemosensitization. These observations indicated a previously unrecognized mechanism by which pifithrin-α can sensitize cancer cells to chemotherapeutic agents. To further investigate the mechanisms involved, we next assessed the effect of pifithrin-α on eIF-4E phosphorylation (Fig. 3B). An increase in eIF-4E phosphorylation at serine 209 was observed during incubation with pifithrin-α, suggesting that pifithrin-α chemosensitization may involve modulation of eIF-4E phosphorylation by pifithrin-α.

Does Pifithrin-α Modulate eIF-4E Phosphorylation via a p38 MAPK-Dependent Mechanism? Phosphorylation of eIF-4E can occur via a signaling cascade involving modulation of phosphorylation of 4E-binding proteins by the Mnk-1 kinase, a downstream substrate of p38 MAPK activation (Pyronnet, 2000). We assessed the contribution of this p38 MAPK-dependent kinase-signaling pathway in the modulation of eIF-4E phosphorylation by pifithrin-α. KMCH cells were stably transfected with dominant-negative MKK3, an upstream activator of p38 MAPK, which decreases p38 MAPK activation. Compared with control cells with functionally active p38 MAPK activation, pifithrin-α did not increase eIF-4E phosphorylation in KM-MKK3dn cells, which are stably transfected with dominant-negative MKK3 (Fig. 4A). Incubation with pifithrin-α increased phospho-p38 MAPK activity by ∼5-fold; moreover, pifithrin-α-induced phospho-p38 MAPK activity was reduced in the presence of SB203580, a pharmacological inhibitor of p38 MAPK activation (Fig. 4B). Furthermore, we found an increase in phospho-Mnk-1 (data not shown) consistent with the activation of p38 MAPK and eIF-4E phosphorylation. In addition, the pifithrin-α dependent increase in eIF-4E phosphorylation was inhibited by preincubation of cells with SB203580 (Fig. 4C). These findings indicate that activation of p38 MAPK is required for pifithrin-α-dependent phosphorylation of eIF-4E.

Does Pifithrin-α Mediate Chemosensitization by a Mechanism Involving Aryl Hydrocarbon Receptor Activation? Pifithrin-α has recently been recognized as an agonist of the AhR that can modulate cell cycle progression. Therefore, we next evaluated the effect of AhR blockade using α-naphthoflavone (αNF) on pifithrin-α chemosensitization. Cells pretreated with 0.1 or 1 μM αNF for 24 h before pifithrin-α treatment had decreased sensitivity to gemcitabine consistent with the postulate that the effects of pifithrin-α are mediated through the AhR (Fig. 5A). To further confirm this, we evaluated the effect of pifithrin-α on the activation of CYP1A1, a well-characterized downstream target gene of AhR activation. By reverse-transcription PCR, we observed a significant increase in CYP1A1 mRNA and protein levels in cells incubated with pifithrin-α (Fig. 5, B and C). However, pretreatment of KMCH cells with αNF increased CYP1A1 mRNA and protein levels compared with untreated controls (Fig. 5, B and C). We speculate that αNF may enhance the binding of pifithrin-α to AhR and enhance receptor activation. Thus, further studies on the biochemical interactions between pifithrin-α and exogenous ligands for AhR will be required to clarify the precise mechanism by which pifithrin-α can modulate the expression of downstream targets such as CYP1A1. Aryl hydrocarbon receptor-mediated signaling has been shown to involve p38 MAP kinase (Chen et al., 2003). Indeed, CYP1A1 protein levels were decreased in KM-MKK3dn cells (Fig. 5D), suggesting that p38 MAPK signaling mediates AhR signaling during pifithrin-α chemosensitization. We then evaluated whether pifithrin-α exposure altered cell cycle progression. The percentage of KMCH cells in the G₀/G₁ phase increased from 64.4 ± 1.6% of untreated cells to 70.3 ± 1.0% of pifithrin-α-treated cells (n = 4; P = 0.0004), indicating that pifithrin-α abrogates cell cycle progression into S phase.

Does Pifithrin-α Chemosensitization Involve p53-Dependent Mechanisms? Although our results show that pifithrin-α chemosensitivity could result from the activation of aryl hydrocarbon receptor signaling, pifithrin-α is an inhibitor of p53, raising the potential that inhibition of p53-dependent responses to chemotherapy may contribute to the observed effects. Wild-type p53 is a transcription factor that can regulate the expression of proteins involved in cell cycle regulation. To examine the potential contribution of p53-dependent mechanisms in the observed effects of pifithrin-α, we examined the expression of p53, analyzed for mutations by direct sequencing, and evaluated the effects of modulation of p53. First, we noted that p53 expression was decreased in KMCH cells compared with nonmalignant H69 cells (Fig. 6A) by immunoblot analysis. Next, we sequenced p53 exons 4 to 9 and analyzed for the presence of any mutations. In KMCH malignant cholangiocytes, a single missense mutation TAT > TGT: Y > C was identified in exon 6 at codon 220. This mutation results in a total loss of p53 transactivation activity (Mitsumoto et al., 2004). This functionally inactive mutation has been noted in several other cancers, such as colorectal, lung, liver, ovarian, and pancreatic (Kalthoff et al., 1993; Kohler et al., 1993; Kubicka et al., 1995; el-Mahdani et al., 1997; Andriani et al., 2004). No p53 mutations were identified in the nonmalignant H69 cells. In addition, the effect of p53-dependent responses to chemotherapy response was assessed in KMCH cells. Transfection of KMCH cells with siRNA to p53 did not alter cellular resistance to gemcitabine (Fig. 6B). These findings that p53 is mutated, functionally inactive, and does not contribute to chemosensitivity to gemcitabine in KMCH cells exclude the possibility that the observed effects of pifithrin-α are mediated by p53-dependent mechanisms.

Discussion

In the present study, we have identified a novel function for pifithrin-α in enhancing the response of human cholangiocarcinoma cells to gemcitabine. The mechanism involves activation of AhR signaling and modulation of eIF-4E phosphorylation involving p38 MAPK signaling pathway. Enhanced expression of eIF-4E is common in human cholangiocarcinoma, and these studies identify eIF-4E as a potential target that can be modulated by pifithrin-α. This study has several important clinical considerations that warrant further study given their relevance to treatments of solid tumors. Pifithrin-α has been reported to sensitize refractory tumors to chemotherapy (Xu et al., 2005; Yin et al., 2006). This effect has been attributed to inhibition of p53, and it has been postulated that chemosensitization arises from failed cell cycle arrest with insufficient time for DNA repair. These studies have not specifically evaluated the possibility that pifithrin-α may act in a p53-independent mechanism. Thus, these observations need to be reevaluated, and the potential involvement of p53-independent mechanisms such as AhR-mediated signaling needs to be considered.

eIF-4E may be characterized as an oncoprotein, and it is overexpressed in many different cancers. Overexpression of eIF-4E has been associated with enhanced translation of several factors promoting tumor survival, angiogenesis, and metastasis. Thus, manipulation of eIF-4E expression or function is a valid therapeutic strategy for tumors in which this protein is overexpressed. However, the regulation of eIF-4E expression and function are not completely understood, and eIF-4E has received scant attention as a potential therapeutic target. As a rate-limiting factor for protein translation, modulation of eIF-4E may globally influence protein translation as well as the translation of downstream targets characterized by extensively structured 5′-untranslated region that are selectively translated by overexpressed eIF-4E and may contribute to tumor behavior. Small molecule drugs, inhibitory peptides and proteins, antisense and gene therapy are all potential approaches by which eIF-4E expression could be manipulated or eIF-4E function could be inhibited to abrogate eIF-4E-mediated effects on tumor growth.

Although we have shown that pifithrin-α enhances eIF-4E phosphorylation at serine 209, via a p38 MAPK-Mnk-1 kinase pathway and that pifithrin-α can modulate chemosensitivity, the mechanisms by which eIF-4E phosphorylation contribute to chemosensitivity and the mediators involved are unknown. The mechanistic and functional significance of eIF-4E phosphorylation remains enigmatic (Raught and Gingras, 1999; Scheper and Proud, 2002). Whereas early reports showed that eIF-4E phosphorylation resulted in an increased affinity of eIF-4E for capped mRNA, more recent studies have reported decreased affinity of eIF-4E to the mRNA cap structure following phosphorylation of eIF-4E at serine 209 (Minich et al., 1994; Scheper et al., 2002). Furthermore, the role of eIF-4E phosphorylation in the regulation of protein translation is not completely understood, and it varies between different conditions. We speculate that eIF-4E phosphorylation results in the separation of the eIF-4E-mRNA cap complex, thereby decreasing the efficiency of translation of mRNA that enhances cell cycle progression and that is selectively translated by overexpression of eIF-4E. Future studies to clarify the protein translation machinery in response to pifithrin-α may be useful to clarify the functional role of eIF-4E phosphorylation.

Our observations showing activation of AhR and p38 MAPK signaling suggest the possibility that p38 MAPK signaling can be directly activated by AhR signaling, consistent with a recent study by Weiss et al. (2005) showing AhR-dependent activation of p38 MAPK in response to the polyhalogenated aromatic hydrocarbon dioxin. In this context, it is noteworthy that exposure to dioxins predisposes to cholangiocarcinoma formation in rats and that the p38 MAPK signaling pathway is aberrantly activated and contributes to tumor growth in human cholangiocarcinoma (Tadlock and Patel, 2001; Walker et al., 2005). Targeting downstream mediators of p38 MAPK such as eIF-4E is therefore an appropriate strategy for intervention in cholangiocarcinoma.

The clinical utility of pifithrin-α has been signified by its potential for use as an adjunct to chemotherapy to reduce adverse effects by reversibly inhibiting p53-dependent responses to chemotherapy-induced genotoxic stress in normal cells and tissues (Komarov et al., 1999). Our study indicates that pifithrin-α may have additional benefits by enhancing chemosensitivity in tumor cells with functionally inactive mutated p53 responses. The role of p53 in tumor cell responses to chemotherapy is complex and variable, and the loss of p53-dependent responses can enhance as well as reduce resistance to chemotherapy in a tumor-specific context (Gudkov and Komarova, 2005). Although we have focused on studies in malignant cholangiocytes, we believe that studies in other tumor cell types are also justified based on our observations. More importantly, pifithrin-α analogs that have the ability to modulate p53-independent responses without inhibition of p53 should be developed. Such analogs would be extremely useful therapeutics, because they may decrease the potential risk of allowing survival of genetically altered cells that would otherwise be eliminated via p53-dependent apoptosis. Moreover, tumor cells that have functional p53 responses may respond differently to pifithrin-α. In this context, it is worth noting the observations of Hoagland et al. (2005) that pifithrin-α is structurally similar to many ligands of the aryl hydrocarbon receptor. Thus, the development of pifithrin-α analogs that can selectively target aryl hydrocarbon receptor signaling independent of effects on p53 are likely to be valuable as biological tools and for clinical use in cancer treatment.