This article is mentioned in:

Introduction

Cholangiocarcinoma (CCA) is one of the highly aggressive malignant tumors, arising from varying locations within ductular epithelium of biliary tree (1,2). As the second most common primary hepatobiliary malignancy, CCA has very low postoperative 5-year survival rate, and treatment with radiotherapy and chemotherapy also carries poor overall survival rates (1,2). Disappointingly, epidemiologic report showed that the overall incidence and mortality of CCA seem to be increasing (3). Hence, defining the molecular mechanisms of CCA is urgent and required for identifying early diagnosis and effective chemotherapy markers for CCA patients.

FBW7 (F-box and WD repeat domain-containing 7), also known as CDC4, AGO and SEL10, is the substrate recognition component of an evolutionary conserved SCF (complex of SKP1, CUL1 and F-box protein)-type E3 ubiquitin ligase (4,5). Human FBW7 gene encodes three transcripts that are produced by alternative splicing and each mRNA consists of an isoform-specific first exon linked to ten shared exons, resulting in three protein isoforms (FBW7α, β and γ) that only vary at the N-terminus (6). Three FBW7 isoforms have different subcellular location and tissue expression profile. FBW7α, β and γ locate in nucleoplasm, cytoplasm, and nucleolus, respectively (7). FBW7α mRNA is expressed at much higher levels than either FBW7β or FBW7γ in most human tissues and varieties of exponentially growing human cells (6,8). Consistently, adult mouse tissues express Fbw7α mRNA ubiquitously, while the FBW7β displays higher levels in the brain and FBW7γ isoform exhibits an increased expression in muscle tissue (9). Grim et al utilized gene targeting to create isoform-specific Fbw7-null mutations in human colon cancer cell line HCT116 and found that FBW7α was the major isoform that mediates the stability of cyclin E, c-Myc, and SREBP1 (sterol regulatory element binding protein 1) (8). These studies suggest that FBW7α may play more important roles in most physiological and pathological process than other isoforms.

FBW7 targets multiple well-known oncoproteins including c-Myc (10), cyclin E (11), mTOR (12), Mcl-1 (13), Notch-1 (14,15) and AIB1 (16) for ubiquitination-mediated destruction, thus, it is a recognized tumor suppressor (4,5). Consistent with the notion that FBW7 is a tumor suppressor in various human malignancies, FBW7 mutations are frequently identified in a variety of human malignancies. The frequency of FBW7 mutations in all primary human cancers analyzed was approximately 6% (87/1556). However, the mutation frequency in cholangiocarcinomas reached up to 35% (7/20) and the report showed that FBW7 hotspot mutants (4/7) in cholangiocarcinoma not only reveal defects in localization and substrate binding but also can abrogate wild-type FBW7 function through a dominant negative mechanism (17). It has been reported that the expression of FBW7 was downregulated in tumor tissues compared with adjacent non-tumorous tissues in intrahepatic (IHCC; 43 tumor specimens vs. 10 paired intrahepatic bile duct tissues) and perihilar (PHCC; 64 tumor specimens vs. 10 paired perihilar bile duct tissues) CCA (18). These results indicate that FBW7 is a general tumor suppressor in various cancers and it may play more important inhibitory roles in CCA progression. However, the function of FBW7 in the proliferation of CCA remains unknown. In the present study, we report that overexpression of FBW7α inhibited CCA cell proliferatin in vitro and in vivo. Further study indicated that FBW7α exerts antitumor activity in CCA at least in part through downregulating c-Myc and cyclin E.

Materials and methods

Cell culture

CCA cell lines QBC-939 and MZ-cha1 were cultured in RPMI-1640 (HyClone) supplemented with 10% fetal bovine serum (Gibco) and 100 U/ml penicillin and 100 mg/ml streptomycin and were maintained in the humidified incubator with 95% air and 5% CO2 at 37°C. QBC-939 cells were obtained from Shuguang Wang (The Third Military Medical University, China); Mz-Cha1 cells were kindly provided by Dr Yabing Chen (University of Alabama at Birmingham, Birmingham, AL, USA).

Small interfering RNA and cell transfection

p21CIP/WAF1 siRNA (CUUCGACUUUGUCACCGAGdTdT) (19), FBW7α siRNA-1 (GGGCAACAACGACGCCGAAdTdT) (20), FBW7α siRNA-2: (GUGAAGUUGUUGGAGUAGAdTdT) (21) and nonspecific siRNA were purchased from Invitrogen. siRNAs were transfected with X-tremeGENE transfection reagent to knock down p21CIP/WAF1 or FBW7α expression in QBC-933 cells following the manufacturer's instructions.

Generation of lentiviruses and infection

The cDNA encoding FBW7α, c-Myc or cyclin E were cloned in the lentiviral vector pLVCS2.0. Oligonucleotide encoding shRNA targeting c-Myc or cyclin E or control oligonucleotide was cloned in the lentiviral vector GV112. The generation of lentivirus vectors was performed by co-transfecting pLV-CS2.0, GV112 or pLV-CS2.0/GV112 carrying the expression cassette with helper plasmids pVSV-G and pHR into 293T cells. The viral supernatant was collected 48 h after transfection and CCA cells were infected with viral supernatants containing 10 µg/ml Polybrene for 24 h and then fresh medium was added to the infected cells. The shRNA target sequences are listed as follows: the target sequence for c-myc shRNA-1: CAAGGTA GTTATCCTTAAA; the target sequence for c-Myc shRNA-2: GTTGAAACACAAACTTGAA; the target sequence for cyclin E shRNA-1: ACATAGAGAACTGTGTCAA; the target sequence for cyclin E shRNA-2: AATTCTTCTGGATTGG TTA; the target sequence for control shRNA: TTCTCCGA ACGTGTCACGT.

Western blot analysis

Briefly, Equal amounts of protein lysates were separated by SDS-PAGE and transferred onto PVDF membranes. After incubation with the primary antibody, the membranes were extensive washed by TBST buffer, then were incubated with horseradish peroxidase conjugated secondary antibody and visualized by chemiluminescence. Antibodies for FBW7 and c-Myc were purchased from Abcam; Antibodies for p21CIP/WAF1, mTOR, AIB1 and Notch1 were purchased from Cell Signaling Technology; Antibodies for cyclin E and Mcl-1 were purchased from Santa Cruz Biotechnology; Anti-β-actin antibody was purchased from Sigma.

Real-time RT-PCR

Briefly, total RNA was isolated with TRIzol reagent (Invitrogen) according to the manufacturer's instructions. Reverse transcription was performed using Revertra Ace qCR RT Master mix (Toyobo) with random primer and olig dT primer. Real-time PCR reactions were performed using FastStart Universal SYBR Green Master (Roche). Relative quantification was achieved by normalization to the amount of GAPDH. The primers used for real-time PCR are listed as follow: FBW7α forward (22): GGAGATGG ACCAGGAGAGTG; FBW7α reverse: GTTGGTGTTGC TGAACATGG; c-Myc forward: GCTGCTTAGACGCTG GATTT; c-Myc reverse: CACCGAGTCGTAGTCGAGGT; cyclin E forward: CTCCAGGAAGAGGAAGGCAA; cyclin E reverse: TCGATTTTGGCCATTTCTTCA; p21CIP/WAF1 forward: CAGGGGAGCAGGCTGAAG; p21CIP/WAF1 reverse: GGATTAGGGCTTCCTCTTGG; GAPDH forward: CAC TCCTCCACCTTTGACGC; GAPDH reverse: TGCTGTAGC CAAATTCGTTGT.

CHX (cycloheximide) treatment

CCA cells were treated with CHX (60 µg/ml) for indicated times in the absence or presence of FBW7α overexpression. Total protein was isolated at different times after CHX treatment, and the levels of c-Myc and cyclin E protein were determined by western blot analysis. The band density of c-Myc and cyclin E protein was quantified by using Scion Imaging software and normalized to β-actin levels.

MTT assay

CCA cell proliferation was analyzed by MTT assay. A total of 3×103 QBC-939 or MZ-cha1 cells were seeded in 96-well plates and MTT was added to each well every 24 h. The plates were incubated for 4 h before addition of solubilization solution (10% SDS in 0.01 M HCl). The absorbance was measured at 560 nm using a microplate reader.

Focus formation assay

Five hundred cells were cultured in six-well plates in RPMI-1640 with 10% FBS. Cells were grown for 2–3 weeks, then colonies were stained with 0.05% crystal violet for 30 min and counted.

Cell cycle analysis

For cell cycle analysis, 4×105 CCA cells were synchronized by serum starvation for 24 h and induced to re-enter the cell cycle by an exchange of 10% fetal bovine serum for 9 h. Cells were harvested and fixed in 75% ethanol at 4°C overnight. Cells were incubated with RNase A at 37°C for 30 min, and then stained with propidium iodide (PI) at 37°C for 30 min. Cell cycle was measured by flow cytometry.

Tumor xenograft experiments

After infection by lentivirus, QBC-939 cells were selected with puromycin for 1 week, and then used to inject into nude mice. The protocols for the in vivo studies were approved by Institutional Animal Care and Use Committee of Laboratory Animal Center of Xiamen University. Male nude mice (4–6-week-old) were obtained from Laboratory Animal Center of Xiamen University. Nude mice were injected subcutaneously in both flanks with 4×106 QBC-939-FBW7α and control cells, respectively. Ten days after cell injection, the volume of the tumor was monitored and calculated following the formula: Volume = length × Width2 ×0.52. After 28 days, tumors were harvested and weighed, and then were dissected and fixed in 10% formalin and embedded in paraffin for Ki67 staining. Remaining tumors were homogenized in RIPA for western blot analysis.

Immunohistochemistry

Slides were soaked in preheated citrate buffer (pH 6.0) and heated in a microwave for 20 min to retrieve antigen. After cooling, slides were washed with PBS three times, and then incubated with Ki67 antibody (1:200; Invitrogen) overnight at 4°C. On the second day, slides were washed with PBS three times, and then incubated with a horseradish peroxidase-conjugated secondary antibody for 1 h at room temperature. After washing, DAB reagent was added to visualize the labeled protein.

Statistical analysis

All data are shown as the mean ± SD from the number of replicates described in results. The statistically significant effects between mean values (p<0.05) were assessed with the two-tailed Student's t-test in SPSS.

Results

Overexpression of FBW7α inhibits CCA cell proliferation

To investigate the role of FBW7α in CCA cell proliferation, the FBW7α-encoding lentivirus was used to express FBW7α in two CCA cell lines, QBC939 and MZ-Cha1, and then cell proliferation was measured by MTT assay. As shown in Fig. 1A and B, FBW7α-encoding lenivirus, but not control lentivirus, efficiently increased the levels of the FBW7α protein in these cells. Significantly, upregulation of FBW7α inhibited proliferation of QBC939 and MZ-Cha1 cells (Fig. 1A and B). Furthermore, upregulation of FBW7α decreased focus formations in QBC939 and MZ-Cha1 cells (Fig. 1C and D). Together, these results indicate that FBW7α is important for CCA cell proliferation.

Figure 1. Overexpression of FBW7α inhibits CCA cell proliferation. CCA cells were infected with control or FBW7α-encoding lentivirus for 48 h and then these cells were used to perform western blot analysis (A and B), MTT assay (A and B), focus formation assay (C and D) and cell cycle analysis (G and H), respectively. QBC-939 cells were transfected with FBW7α-specific siRNA or control RNA for 48 h, then these cells were used to perform Real-time analysis (E) and MTT assay (F). All data are mean ± SD, n=3, *p<0.05, **p<0.01.

We also observed that the decrease in growth of FBW7α-overexpression MZ-Cha1 cells was less than FBW7α-overexpression QBC-939 cells at 96 h (Fig. 1A and B), the reason may be that the growth of MZ-Cha1 cells is quite nutrition-dependent in culture medium and nutrition reduction in culture medium after 96-h cell culture may result in a reduced growth of control cells. Importantly, our focus forming assay demonstrated that upregulation of FBW7α significantly decreased focus formations in MZ-Cha1 cells as QBC-939 (Fig. 1C and D), suggesting that FBW7α has the same tumor suppressor potential in the MZ-Cha1 cells and QBC-939 cells.

To investigate the effect of knockdown of FBW7α on CCA cell proliferation, RNA interference was used to knock down the expression of FBW7α in QBC-939 cells and the cell proliferation was measured by MTT assay. Since it is very difficult to detect endogenous FBW7α protein expression, we use real-time PCR to detect the level of FBW7α mRNA. As shown in Fig. 1E, FBW7α-specific siRNA, but not control RNA, efficiently reduced the levels of endogenous FBW7α mRNA. Although downregulation of FBW7α slightly increased the mean of OD560 in MTT assay at 96 h, they were not statistically significant (Fig. 1F). Since the basal levels of FBW7α protein were too low in CCA cells, knockdown of FBW7α could not cause significant changes in cell proliferation.

As upregulation of FBW7α resulted in a decrease in CCA cell proliferation, cell cycle analysis was performed to detect whether overexpression of FBW7α induces cell arrest in a specific phase of the cell cycle. The results showed that the percentage of FBW7α-overexpression cells at G1 phase was significantly increased as compared with control cells, and this was associated with a concomitant decrease of cells at the S and G2/M of the cell cycle (Fig. 1G and H). These results indicate that upregulation of FBW7α induces CCA cell arrest in the G1 phase of the cell cycle, which is at least in part responsible for the suppression of cell proliferation by FBW7α.

FBW7α promotes the protein degradation of c-Myc and cyclin E

To explore the molecular mechanism underlying the crucial role of FBW7α in control of CCA cell proliferation, we examined the effect of FBW7α overexpression on the expression of several cell proliferation/survival-related proteins which are identified as the substrates of FBW7 (10–16). As shown in Fig. 2A, overexpression of FBW7α significantly reduced the protein levels of c-Myc and cyclin E, but had no significant effect on the protein levels of Notch, mTOR, AIB1 and Mcl-1 in either QBC939 or MZ-Cha1 cells. Overexpression of FBW7α did not affect the mRNA levels of c-Myc and cyclin E (Fig. 2B), indicating that FBW7α decreases the expression of c-Myc and cyclin E at the post-transcriptional level in CCA cells. It has been reported that c-Myc can inhibit cell cycle inhibitor p21CIP/WAF1 expression at transcriptional level (23,24). Our results showed that overexpression of FBW7α increased p21CIP/WAF1 expression at both protein and mRNA levels (Fig. 2A and 2B), and downregulation of c-Myc increased p21CIP/WAF1 expression at both protein and mRNA levels (Fig. 3A) in CCA cells, indicating that FBW7α increases cell cycle inhibitor p21CIP/WAF1 expression at least in part through downregulating c-Myc expression.

Figure 2. FBW7α promotes protein degradation of c-Myc and cyclin E. (A and B) CCA cells were infected with control or FBW7α-encoding lentivirus for 72 h, then western blotting and real-time PCR were performed to detect the expression of FBW7α target genes. (C) Overexpression of FBW7α promoted protein degradation of c-Myc and cyclin E when cycloheximide (CHX) blocked protein synthesis. All data are mean ± SD, n=3. FBW7α promotes the protein degradation of c-Myc and cyclin E. (D) Overexpression of FBW7α promoted the protein degradation of c-Myc and cyclin E when cycloheximide (CHX) blocked protein synthesis. (E) Proteasome inhibitor MG132 inhibited FBW7α-mediated downregulation of the protein levels of c-Myc and cyclin E. All data are mean ± SD, n=3.

Figure 3. Downregulation of c-Myc or cyclin E inhibits CCA cell proliferation. CCA cells were infected by sh-ctrl, sh-c-Myc-encoding or shCyclin E-encoding lentivirus for 48 h and these cells were seeded to perform western blotting (A), MTT assay (A), focus formation assay (B) and cell cycle analysis (C), respectively. All data are mean ± SD, n=3, *p<0.05, **p<0.01.

To determine whether FBW7α accelerates the degradation of c-Myc and cyclin E proteins, we infected QBC939 and MZ-Cha1 cells with control lentivirus or FBW7α-encoding lentivirus, and then used cycloheximide (CHX) to block protein synthesis. Overexpression of FBW7α significantly accelerated the degradation of c-Myc and cyclin E proteins when CHX blocked protein synthesis (Fig. 2C and D), but the effect of FBW7α on the degradation of c-Myc and cyclin E proteins was suppressed by the proteasome inhibitor MG132 (Fig. 2E). These results indicate that FBW7α inhibits c-Myc and cyclin E protein expression dependent on ubiquitin-proteasome degradation pathway in CCA cells. This notion is consistent with previous reports in other cells (10,11).

Downregulation of either c-Myc or cyclin E inhibits CCA cell proliferation

To determine whether FBW7α inhibits CCA proliferation through promoting c-Myc or cyclin E protein degradation, the effect of downregulation of c-Myc and cyclin E on CCA proliferation were investigated. We respectively used two different shRNA to knock down c-Myc and cyclin E in QBC-939 cells, and then measured cell proliferation, focus formation and cell cycle progression. As shown in Fig. 3A, sh-c-Myc and shCyclin E efficiently knocked down the expression of c-Myc and cyclin E, and downregulation of c-Myc and cyclin E significantly inhibited CCA cells proliferation, respectively. Downregulation of c-Myc and cyclin E also significantly inhibited focus formation (Fig. 3B), and induced cell cycle arrest in G1 phase (Fig. 3C). These results indicate that downregulation of c-Myc and cyclin E indeed contribute to the inhibitory effect of FBW7α on CCA proliferation.

Restoration of the expression of c-Myc, but not cyclin E, partially rescues proliferation of FBW7α-overexpressing CCA cells

To further determine whether FBW7α inhibits CCA proliferation through promoting c-Myc or cyclin E protein degradation, FBW7α-overexpressing cells were infected with c-Myc-encoding lentivirus and cyclin E-encoding lentivirus to restore the expression of c-Myc and cyclin E, and then cell proliferation was measured by MTT assay, respectively. Consistent with the notion of Fig. 2 that FBW7α could promote c-Myc and cyclin E protein degradation in CCA cells, overexpression of FBW7α promoted degradation of exogenous c-Myc (compare lanes 2 with lanes 5) or cyclin E (compare lanes 3 with lanes 6) protein (Fig. 4A). Although, FBW7α could degrade exogenous c-Myc and cyclin E proteins, as shown in Fig. 4B, infection with c-Myc-encoding lentivirus and cyclin E-encoding lentivirus efficiently restored the expression of c-Myc and cyclin E in FBW7α-overexpression cells. Restoration of c-Myc expression, but not cyclin E expression, suppressed FBW7α-medicated upregulation of the cell cycle inhibitor p21CIP/WAF1 in FBW7α-overexpression cells (Fig. 4B). Restoration of c-Myc expression partially rescued the proliferation of FBW7α-overexpression cells as expected (Fig. 4C). Surprisingly, restoration of cyclin E expression failed to rescue the proliferation of FBW7α-overexpression cells (Fig. 4C). It may be due to relatively high levels of p21CIP/WAF1 in FBW7α-overexpression cells (Fig. 4B), which counteracts the promoting effect of cyclin E on cell proliferation. This notion was supported by the observation that knockdown of p21CIP/WAF1 partially restored the proliferation of FBW7α-overexpression cells (Fig. 4D and E).

Figure 4. Restoration of the expression of c-Myc, but not cyclin E, partially rescues the proliferation of FBW7α-overexpressing CCA cells. (A) QBC-939 cells were infected with control or FBW7α-encoding lentivirus for 48 h, then control cells and FBW7α-overexpressing cells were infected with control, c-Myc-encoding or cyclin E-encoding lentivirus for 24 h, respectively. Then these cells were seeded to perform western blot analysis after 24 h. (B and C) After QBC-939 cells were infected with control or FBW7α-encoding lentivirus for 48 h, the cells were infected with control, c-Myc-encoding or cyclin E-encoding lentivirus for 24 h. Then, QBC-939 cells were seeded to perform western blot analysis (B) and MTT assay (C). (D and E) Knockdown of p21CIP/WAF1 partially restored the proliferation of FBW7α-overexpressing cells. After QBC-939 cells were infected with control or FBW7α-encoding lentivirus for 48 h, these cells were transfected with control siRNA or p21CIP/WAF1-specific siRNA for 24 h. Then, QBC-939 cells were seeded to perform western blot analysis (D) and MTT assay (E). Ctrl represents the infection with pLV-CS 2.0-encoding lentivirus; FBW7α represents the infection with pLV-CS2.0-FBW7α-encoding lentivirus; NC represents negtive control for si-p21; si-p21 represents p21-specific siRNA. All data are mean ± SD, n=3, *p <0.05, **p<0.01.

Overexpression of FBW7α inhibits CCA tumorigenesis in nude mice

To invistigate the role of FBW7α in CCA progression in vivo, we examined the effect of FBW7α overexpression on the growth of CCA xenograft tumors in nude mice injected with FBW7α-overexpression or control QBC939 cells. As shown in Fig. 5A, FBW7α-overexpression CCA tumors grew much slower than control tumors. At the end of study (day 28), tumor weight of FBW7α-overexpression group (0.0223±0.0185 g) was only 15% of the control group (0.151±0.0728 g) (Fig. 5B). Western blot results showed that the protein levels of c-Myc and cyclin E in FBW7α-overexpression CCA tumors were much lower than that of control tumors, whereas p21CIP/WAF1 expression was much higher (Fig. 5C). Furthermore, immunohistochemical results showed that FBW7α-overexpression CCA tumors had much less Ki67-positive cells (Ki-67 is known as a marker of cell proliferation) than control tumors (Fig. 5D), indicating that overexpression of FBW7α inhibits CCA tumor cell proliferation in vivo. Collectively, these results demonstrate that FBW7α is able to inhibit CCA tumor growth in nude mice.

Figure 5. Overexpression of FBW7α inhibits CCA tumorigenesis in nude mice. (A) Overexpression of FBW7α inhibited CCA tumor growth. n=9; **p<0.01. (B) Tumor weight was quantitated at the end of the experiment. n=9; **p<0.01. (C) Related protein levels in tumors were detected by western blot analysis at the end of the experiment. (D) Overexpression of FBW7α inhibited tumor cell proliferation. Tumor cell proliferation was detected by Ki-67 staining. n=4; **p<0.01.

Discussion

The mutation rates of FBW7 are relatively tumor type-dependent. Cholangiocarcinoma and T-cell acute lymphocytic leukemia harbor 35 and 31% frequency of mutations for FBW7, respectively, while more than ten other types of tumors (such as stomach, colon, breast, bladder, endometrium) only harbor 0–9% frequency of mutations (17). Immunohistochemistry staining showed that the expression of FBW7 was downregulated in tumor tissues compared with adjacent non-tumorous tissues in intrahepatic (IHCC) and perihilar (PHCC) CCA (18). The relatively high frequency of impaired FBW7 function in CCA implicate that FBW7 may play a specific role in CCA progression. In the present study, we defined the role of FBW7 in CCA cell proliferation. Our results showed that overexpression of FBW7α significantly inhibited CCA cell proliferation and CCA xenograft tumor growth, suggesting that FBW7α plays a tumor suppressor role in CCA progression.

Several oncogenic proteins, such as c-Myc, c-Jun, cyclin E, mTOR, Notch-1, Notch-3, and AIB1, have been identified as the substrates of FBW7 (4,5). However, FBW7 targets different substrates for degradation in a cell type-specific manner and the role of FBW7 in cell proliferation may depend on cell and signaling context. For example, conditional deletion of FBW7 in mouse keratinocytes significantly increased the protein levels of Notch1 and c-Myc, but not c-Jun, Notch3 and cyclin E (25). FBW7 deletion in mouse thymocytes significantly increased the expression of Notch1, Notch3 and c-Myc, but not cyclin E (26). Although conditional deletion of FBW7 in hematopoietic stem cells, keratinocytes or T cells promoted cell proliferation (25–27), FBW7 deletion inhibited cell proliferation in mouse embryonic fibroblasts (28,29). In the present study, we found that the decrease in the amounts of c-Myc and cyclin E (rather than mTOR, Notch, AIB1 and Mcl-1) in FBW7α-overexpression CCA cells were the most pronounced and reproducible in different CCA cell lines, suggesting that downregulation of c-Myc and cyclin E may mediate the inhibitory effect of FBW7α in CCA cell proliferation. This notion is supported by the observation that knockdown of c-Myc or cyclin E significantly inhibited CCA cell proliferation. Noteworthy, while restoration of c-Myc expression in FBW7α-overexpressing CCA cells partially rescued the cell proliferation, restoration of cyclin E expression in FBW7α-overexpressing CCA cells failed to rescue the cell proliferation. Restoration of c-Myc expression in FBW7α-overexpression CCA cells could significantly downregulate p21CIP/WAF1 expression, but restoration of cyclin E expression in FBW7α-overexpression CCA cells could not downregulate p21CIP/WAF1 expression. These results suggest that high level of p21CIP/WAF1 could abolish the promotive effect of cyclin E on proliferation in cyclin E/FBW7α-overexpression CCA cells. Downregulation of p21CIP/WAF1 in FBW7α-overexpressing CCA cells could partially rescue the cell proliferation.

Our study demonstrated that FBW7α plays an essential inhibitory role in CCA progression, suggesting that targeting FBW7α pathway becomes an attracting strategy in CCA treatment. Our results showed that restoration of the expression of c-Myc partially rescues proliferation of FBW7α-overexpressing CCA cells, suggesting that downregulation of c-Myc is partially responsible for FBW7α-induced inhibition of CCA cell proliferation. Consistent with our results, cre-mediated ablation of c-Myc (but not inactivation of notch signaling) prevented the stimulatory effect of FBW7 ablation on keratinocyte or T cell proliferation in mice (25,26), suggesting that FBW7-c-Myc axis is important for controlling cell proliferation. The expression of FBW7α substrate c-Myc was induced in different animal models of CCA and promoted CCA progression (30–32), suggesting that c-Myc plays an important role in CCA progression.

It has been reported that FBW7 hotspot mutants in human CCA not only reveal defects in localization and substrate binding, but also can abrogate wild-type FBW7 function through a dominant negative mechanism (17). Since mutated endogenous FBW7 proteins are not only functionally inactivated, but also impair wild-type FBW7 function, it is not a good strategy to target FBW7 directly for CCA treatment. Given that our results demonstrate that downregulation of c-Myc expression plays an important role in inhibition of proliferation of FBW7α-overexpression CCA cells, therefore, targeting FBW7α substrate c-Myc with a pharmacological c-Myc inhibitor may be a viable strategy for CCA treatment. It has been reported that a small molecular c-Myc inhibitor 10058-F4 could induce p21CIP/WAF1 expression, inhibit tumor cell proliferation and enhance chemosensitivity of human hepatocellular carcinoma cells and acute myeloid leukemia (33,34). Therefore, c-Myc inhibitor may represent a promising therapeutic approach for CCA with FBW7 mutated or downregulated.

Acknowledgements

This work was supported by grants from the Natural Science Foundation of China (no. 81272246 to W.L.), the Xiamen Science and Technology Plan (no. 3502Z20154037 to M.L.) and Natural Science Foundation of Fujian Province (no. 2015J01502 to Z.Z.).

References