AKR1B10 (aldo-keto reductase 1B10) is overexpressed in liver and lung cancer, and plays a critical role in tumour development and progression through promoting lipogenesis and eliminating cytotoxic carbonyls. AKR1B10 is a secretory protein and potential tumour marker; however, little is known about the regulatory mechanism of AKR1B10 expression. The present study showed that AKR1B10 is induced by mitogen EGF (epidermal growth factor) and insulin through the AP-1 (activator protein-1) signalling pathway. In human HCC (hepatocellular carcinoma) cells (HepG2 and Hep3B), EGF (50 ng/ml) and insulin (10 nM) stimulated endogenous AKR1B10 expression and promoter activity. In the AKR1B10 promoter, a putative AP-1 element was found at bp −222 to −212. Deletion or mutation of this AP-1 element abrogated the basal promoter activity and response to EGF and AP-1 proteins. This AP-1 element bound to nuclear proteins extracted from HepG2 cells, and this binding was stimulated by EGF and insulin in a dose-dependent manner. Chromatin immunoprecipitation showed that the AP-1 proteins c-Fos and c-Jun were the predominant factors bound to the AP-1 consensus sequence, followed by JunD and then JunB. The same order was followed in the stimulation of endogenous AKR1B10 expression by AP-1 proteins. Furthermore, c-Fos shRNA (short hairpin RNA) and AP-1 inhibitors/antagonists (U0126 and Tanshinone IIA) inhibited endogenous AKR1B10 expression and promoter activity in HepG2 cells cultured in vitro or inoculated subcutaneously in nude mice. U0126 also inhibited AKR1B10 expression induced by EGF. Taken together, these results suggest that AKR1B10 is up-regulated by EGF and insulin through AP-1 mitogenic signalling and may be implicated in hepatocarcinogenesis.

INTRODUCTION

AKR1B10 (aldo-keto reductase 1B10), also named ARL-1 (aldose reductase-like-1), is overexpressed in primary liver and lung carcinomas [1,2]. AKR1B10 promotes cancer cell growth and survival through increasing lipogenesis and eliminating cytotoxic carbonyls, and is thus considered as a tumour-promoting protein [3,4]. Recently, AKR1B10 was found to be a secretory protein and potential tumour marker [5]. As an enzyme, AKR1B10 activity is regulated by oxidative stress and S-thiolation at the protein level [6]. However, little is known about the expression regulation of AKR1B10.

AKR1B10 was first identified from human HCC (hepatocellular carcinoma) [1]. This protein belongs to the AKR (aldo-keto reductase) superfamily, a group of proteins implicated in xenobiotic detoxification, cell growth and proliferation, carcinogenesis and cancer therapeutics [710]. AKR1B10 is a monomeric enzyme that efficiently reduces carbonyl groups in lipid peroxides, which protects host cells from carbonyl lesions [1115]. Previous studies from our and other laboratories have shown that AKR1B10 can also reduce the C13 ketonic group in daunorubicin and idarubicin, leading to chemoresistance of cancer cells to these cytostatic agents [13,16,17]. In human mammary epithelial cells, AKR1B10 is up-regulated with tumorigenic transformation and blocks the ubiquitin-dependent degradation of ACCA (acetyl-CoA carboxylase-α), a rate-limiting enzyme in de novo fatty acid synthesis, promoting fatty acid/lipid synthesis, and cell growth and survival [4,18]. In human colon carcinoma cells (HCT-8) and lung carcinoma cells (NCI-H460), siRNA (small-interfering RNA)-induced AKR1B10 silencing results in carbonyl sensitivity and apoptotic cell death secondary to lipid depletion and mitochondrial dysfunction [3,4]. In the airway epithelium, AKR1B10 is up-regulated by cigarette smoke and may activate procarcinogens in smoke, such as polycyclic aromatic hydrocarbons [1921]. Therefore AKR1B10 plays a critical role in cell transformation, growth and survival. It is important to elucidate the expression regulation of AKR1B10.

Recently, Nishinaka and colleagues [22] reported that AKR1B10 expression is induced by Nrf2 (NF-E2-related factor 2) or in response to ethoxyquin, an antioxidant. However, this may not explain the up-regulation of this protein in tumours, such as HCC, where mitogenic signalling plays a critical role in cancer development and progression. EGF (epidermal growth factor), which signals through binding to a transmembrane tyrosine kinase receptor, is a key regulator of cell survival and proliferation [23,24]. EGF can induce cell transformation and enhances in vitro growth of human epithelial- and mesenchymal-derived tumours [25,26]. EGF is a mitogen of hepatocytes and is up-regulated in liver regeneration and HCC. In animals, targeted expression of EGF leads to hepatocellular lesions and malignancies [2729]. Therefore EGF is a carcinogenic factor implicated in the development and progression of HCC. Bioenergetic and lipogenic changes are early events in hepatocarcinogenesis [30]. Insulin is a hormone central to the regulation of energy metabolism and lipid synthesis in the liver [31]. Deregulation of insulin and insulin-like growth factor pathways is implicated in hepatocarcinogenesis [32,33]. In particular, insulin receptor substrate-2 is induced in liver preneoplastic lesions and up-regulated with malignant transformation [32]. The present study found for the first time that AKR1B10 is up-regulated by the mitogens EGF and insulin and may play a potential role in hepatocarcinogenesis.

MATERIALS AND METHODS

Tissue procurement

Surgically resected specimens (frozen) were collected through the NCI (National Cancer Institute)-sponsored CHTN (Cooperative Human Tissue Network). All specimens were quality controlled by pathologists. Tissue microarrays were purchased from Cybrdi™ Tissue Array and US Biomax. Pathological diagnosis is available for all HCC tissues in the tissue microarrays.

Cell culture

The human HCC cell lines HepG2 and HEP3B were purchased from the A.T.C.C. (Manassas, VA, U.S.A.) and maintained in DMEM (Dulbecco's modified Eagle's medium) containing 10% fetal bovine serum, 2 mM glutamine, 100 units/ml penicillin and 100 μg/ml streptomycin at 37°C in 5% CO2.

AKR1B10 activity assay

Cells were lysed on ice in a buffer containing 20 mM NaH2PO4 (pH 7.0), 2 mM 2-mercaptoethanol, 5 μM leupeptin and 20 μM PMSF for 30 min, followed by centrifugation at 10000 g at 4°C for 10 min. Soluble proteins (50 μg) were used for the AKR1B10 activity assay in a reaction mixture consisting of 125 mM sodium phosphate (pH 7.0), 0.2 mM NADPH, 50 mM KCl and 20 mM DL-glyceraldehyde at 35°C for 10 min. Oxidized NADPH was monitored at 340 nm to indicate enzyme activity.

RNA preparation and RT (reverse transcription)–PCR

Total RNA was extracted using TRIzol® reagent (Invitrogen) and quantified by measuring at A260. For semi-quantitative RT–PCR, RNA was treated with RNase-free DNase 1, and the first-strand cDNA was synthesized from 1 μg of total RNA with oligo-dT primers and Superscript II® retrotranscriptase, following the manufacturer's protocol (Invitrogen). PCR reactions were performed using standard procedures. β-Actin mRNA in each sample was run in parallel for an internal control. The primers for AKR1B10 were 5′-CTGGATCCGGCAAGATTAAGGAGAT-3′ (forward) and 5′-GACTGCGGCCGCGATATCCACCAGG-3′ (reverse) and the primers for β-actin were 5′-ATCATTGCTCCTCCTGAGCGC-3′ (forward) and 5′-TGAACTTTGGGGGATGCTCGC-3′ (reverse).

Expression vectors and luciferase activity assay

The −2067bp/AP-1 (activator protein-1) wild-type, −2067bp/AP-1 mutant, −255bp/AP-1 wild-type, −255bp/AP-1 mutant and −211bp/AP-1 negative AKR1B10 promoter-luciferase vectors were constructed with a PCR-based method [34]. c-Fos and c-Jun expression vectors were provided by Dr Dennis K. Watson (Hollings Cancer Center, Medical University of South Carolina, Charleston, SC, U.S.A.) [35]; JunB and JunD were a gift from Dr Michael Karin (Department of Pharmacology, University of California at San Diego, San Diego, CA, U.S.A.) [36]. c-Fos shRNA (short hairpin RNA) expression vector was obtained from Dr Serpil C. Erzurum (Department of Pathobiology, Lerner Research Institute, Cleveland Clinic, Cleveland, OH, U.S.A.) [37]. Transient transfection and luciferase activity assays were conducted as described previously [34].

Western blot analysis

Cells were lysed in cell lysis buffer (Roche) containing a protease inhibitor cocktail. For frozen tissues, the tissues were homogenized in the same cell lysis buffer, followed by centrifugation at 14000 g at 4°C for 15 min to collect soluble proteins. Protein separation and antibody probing were conducted as described previously [4]. Loading amounts of proteins were corrected by re-probing membranes with an anti-β-actin monoclonal antibody (1:40000 dilution; Sigma). AKR1B10 and β-actin bands in each specimen were quantified by densitometry, and AKR1B10 intensity was corrected by the intensity of β-actin in the same sample. AKR1B10 expression in each tumour was expressed as the fold increase in its density over the normal cells.

EMSA (electrophoretic mobility-shift assay)

Nuclear proteins were extracted and interacted with double-stranded DNA probes containing the putative (5′-TGTTGAATTTGTTGACTCATCCTGAA-3′) or mutant (5′-TGTTGAATTTGTTtttTtATCCTGAA-3′) AP-1 element (underlined and lower cases respectively), as described previously [38]. In competitive experiments, unlabelled wild-type or mutant AP-1 oligonucleotides (100-fold excess) were added to the labelled probes before binding. Binding complexes were separated in 0.5× TBE (Tris/borate/EDTA) at 150 V for 2 h on a 6% non-denaturing polyacrylamide gel, followed by exposure to Kodak X-ray films.

ChIP (chromatin immunoprecipitation) assay

HepG2 cells were fixed in 1% formaldehyde for 10 min and cell lysates were prepared for immunoprecipitation as described previously [39]. Immune complexes were collected with 45 μl of Protein G–agarose beads and DNA was extracted using proteinase K digestion and phenol/chloroform extraction. PCR amplification was performed with forward (5′-CCTACCTTCCAACTTTTGGCTG-3′) and reverse (5′-CTCTTAGGTGACAGTATCATTGG-3′) primers at −254 and −171 bp of the AKR1B10 promoter.

In situ hybridization

In situ hybridization was performed as described previously with modifications [40]. Briefly, 10 μm frozen sections of HCC were fixed in 4% paraformaldehyde in PBS for 5 min and soaked in 0.1 M triethanolamine (pH 8.0) and 0.25% acetic anhydride for 10 min with stirring, followed by dehydration in 50, 70, 95 and 100% ethanol. After being dried, sections were hybridized with 35S-labelled AKR1B10 antisense or sense (blank control) riboprobes (5×104 c.p.m./1 μl) at 50°C overnight. After being dehydrated in ethanol as above, slides were dried under vacuum, embedded with emulsion (Kodak) and exposed at 4°C for 10–14 days, followed by developing and haematoxylin-counter-staining.

Immunohistochemistry

Tissue microarrays or tumour sections (5 μm) were dewaxed and hydrated. Antigen retrievals were conducted by immersing into preheated citric acid buffer (pH 6.5) at 90–95°C for 20 min with microwaving. After being blocked with 5% horse serum for 30 min, the slides were incubated with a specific rabbit anti-AKR1B10 antibody (1:50 dilution) [18], as indicated, at 4°C in a humidified box overnight, followed by incubation with HRP (horseradish peroxidase)-conjugated anti-rabbit IgG (1:800 dilution; Pierce) at room temperature (20°C) for 1 h. Enhanced DAB (diaminobenzidine) solution (Pierce) was used to visualize the signals. The staining intensity was evaluated blindly by at least one researcher and a pathologist. AKR1B10 expression was designated as negative, positive and strongly positive based on the staining intensity.

HepG2 cell inoculation and treatment

HepG2 cells (107) were suspended in 100 μl of DMEM and matrigel mix [1:1 (v/v)] (BD Bioscience), and injected into the back flank of female nude mice to form a palpable mass. Two days later, treatments were conducted as indicated. After 24 h, the masses were removed and split into three parts. One was used for total RNA extraction and RT–PCR, one was utilized for protein preparation and Western blotting, and the rest was fixed in 10% saline-buffered formalin and processed for sectioning (5 μm) and immunohistochemistry. Animal experiments were performed in accordance and with approval of the Southern Illinois University School of Medicine Laboratory Animal Care and Use Commitee (LACUC).

Statistical analysis

Student's t tests or χ2 tests of independence, as appropriate, were used to determine statistical significance with P<0.05.

RESULTS

AKR1B10 is overexpressed in HCC tissues

AKR1B10 was reported to be overexpressed in HCC [41]. The present study semi-quantitatively verified the expression of AKR1B10 in HCC tissues using Western blot analysis (Figure 1A). In 34 frozen tumours matched with adjacent normal tissues, AKR1B10 was highly increased in 10 cases (29.4%), in which the amount of AKR1B10 in tumours was over 100-fold higher than in the matching normal tissue; and in 12 tumours (35.3%), AKR1B10 was increased by 2–100-fold compared with the matching normal tissue. Figure 1(A) (lower panel) shows AKR1B10 protein levels in 34 HCC tissues. AKR1B10 mRNA in tumour tissues was confirmed using in situ hybridization (Figure 1B). Tissue microarrays were further used to verify and compare AKR1B10 expression in different HCC tissues (Figure 1C). Results showed that of the 116 HCC cases, AKR1B10 was positive in 37 (31.9%) and strongly positive in 40 (34.5%), consistent with the results of the Western blot analysis of frozen tissues. In these cases, AKR1B10 expression in the HCC was inversely correlated with tumour grades [Spearman's rank correlation=−0.3435; 95% confidence interval=−0.4989 to −0.1666; and P (two-tailed)=0.0002] (Table 1). It is noteworthy that AKR1B10 displayed slightly differential expression in the normal liver, consistent with a previous report [13].

AKR1B10 expression in HCCs

Figure 1
AKR1B10 expression in HCCs

(A) Upper panel, AKR1B10 protein in representative HCC (T) and normal (N) tissues. Lower panel, AKR1B10 protein in 34 individual HCC cases, which is presented as fold increase over normal after being corrected by β-actin. (B) In situ hybridization. Adjacent frozen sections of human HCCs were probed by 35S-labelled AKR1B10 antisense RNA (left-hand panel) with 35S-labelled sense RNA as a control (right-hand panel). The arrows indicate AKR1B10 mRNA. (C) Immunohistochemistry. Tissue microarrays of HCCs were assessed by immunohistochemical staining for AKR1B10 protein. Images demonstrate AKR1B10 expression in HCC, but not in the normal liver tissue.

Figure 1
AKR1B10 expression in HCCs

(A) Upper panel, AKR1B10 protein in representative HCC (T) and normal (N) tissues. Lower panel, AKR1B10 protein in 34 individual HCC cases, which is presented as fold increase over normal after being corrected by β-actin. (B) In situ hybridization. Adjacent frozen sections of human HCCs were probed by 35S-labelled AKR1B10 antisense RNA (left-hand panel) with 35S-labelled sense RNA as a control (right-hand panel). The arrows indicate AKR1B10 mRNA. (C) Immunohistochemistry. Tissue microarrays of HCCs were assessed by immunohistochemical staining for AKR1B10 protein. Images demonstrate AKR1B10 expression in HCC, but not in the normal liver tissue.

Table 1
AKR1B10 expression in HCCs

AKR1B10 expression reversely correlated with tumour grade (Spearman's rank correlation): Spearman r=−0.3435; 95% confidence interval=−0.4989 to −0.1666; and P (two-tailed)=0.0002. M, male; F, female.

  AKR1B10 (n=116) 
Grade Sex (M/F) Negative Positive Strongly positive 
30/9 12 
30/7 26 28 23 
34/6 12 
Subtotal (%) 94/22 39 (33.6) 37 (31.9) 40 (34.5) 
  AKR1B10 (n=116) 
Grade Sex (M/F) Negative Positive Strongly positive 
30/9 12 
30/7 26 28 23 
34/6 12 
Subtotal (%) 94/22 39 (33.6) 37 (31.9) 40 (34.5) 

EGF and insulin stimulate AKR1B10 expression and promoter activity

EGF and insulin are important mitogens involved in hepatocarcinogenesis [27,29,32]. The induction of AKR1B10 in HCC prompted us to evaluate the effect of EGF and insulin on AKR1B10 expression in the HCC cell lines HepG2 and Hep3B. The results showed that both EGF (50 ng/ml) and insulin (10 nM) stimulated endogenous AKR1B10 expression, leading to an increase in AKR1B10 mRNA, protein and enzyme activity by 2–3 times (Figure 2A). More sensitively, EGF (50 ng/ml) and insulin (10nM) greatly enhanced the activity of the −255bp AKR1B10 promoter fragment in driving luciferase reporter expression in HepG2 and Hep3B cells (Figure 2B). These data suggest that EGF and insulin stimulate AKR1B10 expression in HCC cells.

Induction of AKR1B10 expression by EGF and insulin

Figure 2
Induction of AKR1B10 expression by EGF and insulin

(A) Induction of endogenous AKR1B10 by EGF and insulin. HepG2 cells (106) were seeded on to 10 cm cell culture dishes and fed the next day with fresh medium containing 50 ng/ml EGF or 10 nM insulin. The cells were collected at indicated time for Western blotting (panel i), enzyme activity (panel ii) and RT–PCR (panel iii), as described in the Materials and methods section. (B) Stimulation of the AKR1B10 promoter by EGF and insulin. HepG2 and Hep3B cells (104/well) were applied in 24-well plates and transfected with 1 μg of the −255bp AKR1B10 promoter–luciferase construct and 0.5 μg of β-galactosidase expression vectors as described in the Materials and methods section. After incubation for 24 h in medium containing 50 ng/ml EGF or 10nM insulin, luciferase activity was measured with β-galactosidase activity as an internal control. Results are means±S.D. from three independent experiments. * P<0.05; ** P<0.01, compared with the control.

Figure 2
Induction of AKR1B10 expression by EGF and insulin

(A) Induction of endogenous AKR1B10 by EGF and insulin. HepG2 cells (106) were seeded on to 10 cm cell culture dishes and fed the next day with fresh medium containing 50 ng/ml EGF or 10 nM insulin. The cells were collected at indicated time for Western blotting (panel i), enzyme activity (panel ii) and RT–PCR (panel iii), as described in the Materials and methods section. (B) Stimulation of the AKR1B10 promoter by EGF and insulin. HepG2 and Hep3B cells (104/well) were applied in 24-well plates and transfected with 1 μg of the −255bp AKR1B10 promoter–luciferase construct and 0.5 μg of β-galactosidase expression vectors as described in the Materials and methods section. After incubation for 24 h in medium containing 50 ng/ml EGF or 10nM insulin, luciferase activity was measured with β-galactosidase activity as an internal control. Results are means±S.D. from three independent experiments. * P<0.05; ** P<0.01, compared with the control.

EGF and insulin stimulate AKR1B10 expression through AP-1 signalling

Stimulation of the −255 bp AKR1B10 promoter activity by EGF and insulin indicates a regulatory mechanism at the promoter level. A putative AP-1 element was identified at −222 to −212 bp of the AKR1B10 promoter [34]. AP-1 acts in the EGF and insulin signalling cascade, regulating target gene expression [25,42,43]. Therefore we characterized this putative AP-1 consensus sequence. Results showed that deletion (e.g. −211bp/luciferase vector) or targeted mutations (in both −255bp/luciferase and −2067bp/luciferase constructs) of this AP-1 site abolished AKR1B10 promoter activity in driving luciferase reporter expression and its response to EGF in HepG2 cells (Figure 3A). In contrast, this AP-1 consensus sequence enhanced the activity of the conventional promoter SV40 (simian virus 40; Figure 3B). These data suggest that this AP-1 consensus sequence is essential for the basal activity and that of the AKR1B10 promoter and its response to EGF.

Regulatory role of AP-1 element in the AKR1B10 promoter

Figure 3
Regulatory role of AP-1 element in the AKR1B10 promoter

(A) AP-1 activity in the AKR1B10 promoter. AP-1 deleted or mutant AKR1B10 promoter–luciferase (Luc) vectors indicated were transiently introduced into HepG2 cells as described in the Materials and methods section. EGF was used at 50 ng/ml for 24 h. Luciferase activity was measured with β-galactosidase as an internal control. Upper panel, sequences of wild-type (WT) and mutated AP-1 elements; lower panel, luciferase activity. Results are means±S.D. from three independent experiments. ** P<0.01 compared with wild-type AP-1; # P<0.01 compared with the non-EGF control (Ctrl). (B) Trans-activity. AP-1 consensus was inserted upstream of SV40 promoter to drive luciferase expression. Upper panel, AP-1/SV40 promoter; lower panel, luciferase activity. Results are means±S.D. from three independent experiments. *P<0.05 compared with the vector control. Del, deletion; Mut, mutant.

Figure 3
Regulatory role of AP-1 element in the AKR1B10 promoter

(A) AP-1 activity in the AKR1B10 promoter. AP-1 deleted or mutant AKR1B10 promoter–luciferase (Luc) vectors indicated were transiently introduced into HepG2 cells as described in the Materials and methods section. EGF was used at 50 ng/ml for 24 h. Luciferase activity was measured with β-galactosidase as an internal control. Upper panel, sequences of wild-type (WT) and mutated AP-1 elements; lower panel, luciferase activity. Results are means±S.D. from three independent experiments. ** P<0.01 compared with wild-type AP-1; # P<0.01 compared with the non-EGF control (Ctrl). (B) Trans-activity. AP-1 consensus was inserted upstream of SV40 promoter to drive luciferase expression. Upper panel, AP-1/SV40 promoter; lower panel, luciferase activity. Results are means±S.D. from three independent experiments. *P<0.05 compared with the vector control. Del, deletion; Mut, mutant.

The regulatory activity of this AP-1 site was confirmed by an EMSA assay (Figure 4A). Nuclear proteins from HepG2 cells were incubated with isotope-labelled AP-1 oligonucleotides and a single specific-binding complex was detected. The binding specificity was verified by two competitive binding assays using either a 100-fold excess of non-radioactive wild-type or mutant AP-1 oligonucleotides. The results showed that the wild-type AP-1 oligonucleotides specifically competed with the labelled probes and significantly reduced the formation of the radioactive complex, whereas the mutant AP-1 oligonucleotides did not. Furthermore, exposing the HepG2 cells to EGF and insulin induced a dose-dependent increase in the AP-1-binding complex (Figure 4B). These data suggest that EGF and insulin stimulate AKR1B10 expression through this AP-1 consensus sequence.

Specific binding to the AP-1 element in the AKR1B10 promoter

Figure 4
Specific binding to the AP-1 element in the AKR1B10 promoter

(A and B) EMSA. Nuclear proteins were extracted from HepG2 cells and EMSA were conducted as described in the Materials and methods section. (A) AP-1 probe alone (lane 1), AP-1 with nuclear extracts (lane 2), AP-1 with a 100-fold excess of unlabelled wild-type AP-1 (lane 3) and AP-1 with a 100-fold excess of unlabelled mutant AP-1 (lane 4). (B) Labelled AP-1 oligonucleotides were interacted with nuclear proteins of HepG2 cells exposed to EGF or insulin at the indicated concentrations. (C) ChIP assay. Immunoprecipitation and PCR amplification were conducted as described in the Materials and methods section. Rabbit IgG was used as a negative control and DNA input (10%) was utilized as a positive control.

Figure 4
Specific binding to the AP-1 element in the AKR1B10 promoter

(A and B) EMSA. Nuclear proteins were extracted from HepG2 cells and EMSA were conducted as described in the Materials and methods section. (A) AP-1 probe alone (lane 1), AP-1 with nuclear extracts (lane 2), AP-1 with a 100-fold excess of unlabelled wild-type AP-1 (lane 3) and AP-1 with a 100-fold excess of unlabelled mutant AP-1 (lane 4). (B) Labelled AP-1 oligonucleotides were interacted with nuclear proteins of HepG2 cells exposed to EGF or insulin at the indicated concentrations. (C) ChIP assay. Immunoprecipitation and PCR amplification were conducted as described in the Materials and methods section. Rabbit IgG was used as a negative control and DNA input (10%) was utilized as a positive control.

A ChIP assay further confirmed the specific binding of AP-1 proteins to this element in AKR1B10 promoter. As shown in Figure 4(C), immunoprecipitated chromosomal DNA was amplified by PCR using primers specific to the region of the AKR1B10 promoter that harbours the AP-1 consensus. The results showed that c-Fos and c-Jun were the predominant proteins bound to this AP-1 site, followed by JunD and then JunB.

Transient expression of AP-1 proteins stimulates AKR1B10 expression and promoter activity

The role of the AP-1 consensus sequence in AKR1B10 expression was further confirmed by transfection or co-transfection of HepG2 cells with c-Fos, c-Jun, JunB or JunD. As shown in Figure 5(A), all of the tested AP-1 proteins stimulated endogenous AKR1B10 expression, leading to a significant increase in protein levels and enzyme activity.

Up-regulation of AKR1B10 expression by AP-1 proteins

Figure 5
Up-regulation of AKR1B10 expression by AP-1 proteins

(A) Induction of endogenous AKR1B10 by AP-1 proteins. HepG2 cells (106) were spread in 10 cm cell-culture dishes and transfected with c-Fos, c-Jun, JunB or JunD. After incubation for 36 h, the cells were harvested for Western blotting (panel i) and enzyme activity assay (panel ii) as described in the Materials and methods section. (B) Stimulation of AKR1B10 promoter by AP-1 proteins. HepG2 cells (104/well) were spread in 12-well plates and co-transfected with the −255bp or −2067bp AKR1B10 promoter–luciferase construct and c-Fos, c-Jun, JunB or JunD. After incubation for 36 h, the cells were harvested for luciferase activity assay with β-galactosidase as an internal control. Results are means±S.D. from three independent experiments. *P<0.05 and **P<0.01, compared with the control (Ctrl). AP-1 (+), promoter with wild-type AP-1 element; AP-1 (−), promoter without AP-1 element; AP-1 WT, wild-type AP-1; AP-1 Mut, mutant AP-1.

Figure 5
Up-regulation of AKR1B10 expression by AP-1 proteins

(A) Induction of endogenous AKR1B10 by AP-1 proteins. HepG2 cells (106) were spread in 10 cm cell-culture dishes and transfected with c-Fos, c-Jun, JunB or JunD. After incubation for 36 h, the cells were harvested for Western blotting (panel i) and enzyme activity assay (panel ii) as described in the Materials and methods section. (B) Stimulation of AKR1B10 promoter by AP-1 proteins. HepG2 cells (104/well) were spread in 12-well plates and co-transfected with the −255bp or −2067bp AKR1B10 promoter–luciferase construct and c-Fos, c-Jun, JunB or JunD. After incubation for 36 h, the cells were harvested for luciferase activity assay with β-galactosidase as an internal control. Results are means±S.D. from three independent experiments. *P<0.05 and **P<0.01, compared with the control (Ctrl). AP-1 (+), promoter with wild-type AP-1 element; AP-1 (−), promoter without AP-1 element; AP-1 WT, wild-type AP-1; AP-1 Mut, mutant AP-1.

AP-1 transcription factors also stimulated AKR1B10 promoter activity when co-transfected. As demonstrated by luci-ferase activity assay in HepG2 cells, c-Fos and c-Jun enhanced AKR1B10 promoter (−255 bp) activity by more than 20 times each, but JunB and Jun D were much less effective, consistent with the data from the ChIP assay. Deletion of the AP-1 element (e.g. the −211bp promoter/luciferase construct) completely abrogated the stimulatory role of c-Jun, JunB and JunD, but the stimulation by c-Fos remained, although noticeably decreased (Figure 5B, panel i), suggesting that c-Fos may also regulate AKR1B10 expression through a mechanism independent of the AP-1 consensus sequence. Similar results were observed in studies with the −2067bp/luciferase constructs with a wild-type or mutant AP-1 site (Figure 5B, panel ii).

Inhibition of AKR1B10 expression by AP-1 inhibitors and c-Fos shRNA in vitro and in vivo

The regulatory role of AP-1 signalling in AKR1B10 expression was further confirmed both in vitro and in vivo by AP-1 inhibitors and RNA interference-mediated c-Fos knockdown. As shown in Figure 6(A) (panel i), exposing HepG2 cells to AP-1 inhibitors U0126 (10 or 20 μM) and Tanshinone IIA (20 μM) significantly reduced endogenous AKR1B10 expression. AP-1 inhibitor U0126 (10 μM) also blocked AKR1B10 expression induced by EGF (Figure 6A, panel ii). Consistently, these AP-1 inhibitors significantly inhibited the −2067bp AKR1B10 promoter activity (Figure 6B). Interestingly, c-Fos shRNA alone reduced AKR1B10 promoter activity by more than 80%, indicating the essential role of c-Fos in AKR1B10 promoter activity.

Inhibition of AKR1B10 expression by AP-1 inhibitors

Figure 6
Inhibition of AKR1B10 expression by AP-1 inhibitors

(A) Inhibition of endogenous AKR1B10 expression by AP-1 inhibitors. HepG2 cells (106) were seeded on to 6 cm cell culture dishes and fed the next day with fresh medium containing U0126, Tanshinone IIA or EGF as indicated for 24 h. The cells were collected at the indicated time for RT–PCR (panel i) and Western blotting (panel ii) as described in the Materials and methods section. [In lane 3 in (panel ii), U0126 was used at 10 μM and EGF was at 50 ng/ml.] (B) Inhibition of the AKR1B10 promoter by AP-1 inhibitors and c-Fos shRNA. HepG2 cells (104/well) were applied in 24-well plates and transfected with 1 μg of −2076bp AKR1B10 promoter–luciferase construct and 0.5 μg of β-galactosidase expression vectors as described in the Materials and methods section. The next day the cells were treated with AP-1 inhibitors U0126 or Tanshinone IIA at 10 or 20 μM each for 24 h, and subjected to luciferase activity assay. For c-Fos silencing, c-Fos shRNA was co-transfected into HepG2 cells for 72 h. Results are means±S.D. from three independent experiments. *P<0.05 and **P<0.01, compared with the control (−2076bp AKR1B10 promoter–luciferase construct with wild-type AP-1).

Figure 6
Inhibition of AKR1B10 expression by AP-1 inhibitors

(A) Inhibition of endogenous AKR1B10 expression by AP-1 inhibitors. HepG2 cells (106) were seeded on to 6 cm cell culture dishes and fed the next day with fresh medium containing U0126, Tanshinone IIA or EGF as indicated for 24 h. The cells were collected at the indicated time for RT–PCR (panel i) and Western blotting (panel ii) as described in the Materials and methods section. [In lane 3 in (panel ii), U0126 was used at 10 μM and EGF was at 50 ng/ml.] (B) Inhibition of the AKR1B10 promoter by AP-1 inhibitors and c-Fos shRNA. HepG2 cells (104/well) were applied in 24-well plates and transfected with 1 μg of −2076bp AKR1B10 promoter–luciferase construct and 0.5 μg of β-galactosidase expression vectors as described in the Materials and methods section. The next day the cells were treated with AP-1 inhibitors U0126 or Tanshinone IIA at 10 or 20 μM each for 24 h, and subjected to luciferase activity assay. For c-Fos silencing, c-Fos shRNA was co-transfected into HepG2 cells for 72 h. Results are means±S.D. from three independent experiments. *P<0.05 and **P<0.01, compared with the control (−2076bp AKR1B10 promoter–luciferase construct with wild-type AP-1).

We further verified in vivo the role for AP-1 signalling in AKR1B10 expression. In the present study, we inoculated HepG2 cells into female nude mice to mimic an in vivo environment. After a palpable mass was formed, EGF, U0126 or c-Fos shRNA was administered intraperitoneally or intratumorally. Palpable masses were collected 24 h later and processed for RT–PCR, Western blot analysis and immunohistochemistry. As shown in Figure 7, AKR1B10 expression in the tumours was significantly induced by EGF, but inhibited by the AP-1 antagonist U0126 and c-Fos shRNA. Taken together, these in vitro and in vivo data suggest that AP-1 signalling participates in the regulation of AKR1B10 expression in HepG2 cells. In addition, it appears that oncogenic c-Fos silencing induced tumour cell death (Figure 7D).

AKR1B10 expression regulation by AP-1 in tumours in nude mice

Figure 7
AKR1B10 expression regulation by AP-1 in tumours in nude mice

HepG2 cells (1×107) were inoculated into nude mice. After palpable masses were formed, animals were exposed to EGF, U0126 or c-Fos shRNA, administered by intraperitoneal (i.p.) or intratumoral (i.t.) injection. Doses were as follows: EGF at 100 μl (10 ng/ml) intratumorally or 10 ng/kg intraperitoneally, U0126 at 100 μl (10 μM) intratumorally or 20 μmol/kg intraperitoneally, and c-Fos at 1 μg intratumorally. After 24 h, the tumours were collected for RT–PCR, Western blotting and immunohistochemistry studies as described in the Materials and methods section. (A) A representative tumour (indicated by the arrow). (B) RT–PCR for AKR1B10 mRNA. (C) Western blotting for AKR1B10 protein. (D) AKR1B10 immunohistochemistry. Data are representatives of at least two tumours in each treatment.

Figure 7
AKR1B10 expression regulation by AP-1 in tumours in nude mice

HepG2 cells (1×107) were inoculated into nude mice. After palpable masses were formed, animals were exposed to EGF, U0126 or c-Fos shRNA, administered by intraperitoneal (i.p.) or intratumoral (i.t.) injection. Doses were as follows: EGF at 100 μl (10 ng/ml) intratumorally or 10 ng/kg intraperitoneally, U0126 at 100 μl (10 μM) intratumorally or 20 μmol/kg intraperitoneally, and c-Fos at 1 μg intratumorally. After 24 h, the tumours were collected for RT–PCR, Western blotting and immunohistochemistry studies as described in the Materials and methods section. (A) A representative tumour (indicated by the arrow). (B) RT–PCR for AKR1B10 mRNA. (C) Western blotting for AKR1B10 protein. (D) AKR1B10 immunohistochemistry. Data are representatives of at least two tumours in each treatment.

DISCUSSION

AKR1B10 is induced in human HCC, but the regulatory mechanism remains to be fully understood. The present study demonstrated that AKR1B10 is up-regulated by the mitogens EGF and insulin through the mitogenic AP-1 signalling pathway, and may function as a downstream effector of EGF and insulin promoting hepatocarcinogenesis and tumour progression.

The proto-oncogenic proteins c-Fos and c-Jun are downstream effectors of EGF and insulin signalling [25,42,43], and are found to be overexpressed in HCC tissues [44]. The present study characterized an AP-1 consensus sequence at −222 to −212 bp in the AKR1B10 promoter. We investigated nuclear protein binding to the AP-1 element and the effect of AP-1 site mutation or deletion, AP-1 gene co-transfection or silencing, and AP-1 inhibitors on endogenous AKR1B10 expression and its promoter activity. Two different lengths of AKR1B10 prompter fragments (−255bp and −2067bp) were tested. All results indicate that this AP-1 element is essential for AKR1B10 promoter activity and its response to EGF and insulin. AP-1 protein is a heterodimeric transcription factor. Fos and Jun nuclear oncogenic proteins are found in mammals. c-Fos belongs to the Fos family, consisting of c-Fos, FosB and Fra-1/2 (Fos-related antigen 1/2). Fos proteins contain a leucine-zipper motif and often dimerize with Jun proteins (c-Jun, JunB and JunD) to form the heterodimeric protein AP-1. The adjacent basic domain in AP-1 can bind to DNA at the specific AP-1 element. Hence AP-1 protein is a DNA- sequence-specific-binding transcription factor that activates AP-1-dependent gene transcription [45]. The expression of Fos and Jun proteins is rapidly and transiently induced by various mitogenic factors, such as growth factors, cytokines, polypeptide hormones and stress [46]. In the present study, we tested the regulatory activity of different members from the Fos and Jun families on endogenous AKR1B10 expression and on cloned AKR1B10 promoter fragments (−255bp and −2067bp). EMSA and ChIP assays both supported the hypothesis that AP-1 binds directly to the AP-1 consensus in the AKR1B10 promoter; and c-Fos and c-Jun are predominant trans-activating factors that bind to the AP-1 element and regulate AKR1B10 expression. Silencing of c-Fos by shRNA led to more than 80% reduction in AKR1B10 promoter activity. These data suggest that c-Fos and c-Jun may form a heterodimer playing a major role in regulating AKR1B10 expression. Interestingly c-Fos displayed substantial stimulating activity towards the −211bp AKR1B10 promoter fragment that lacks the AP-1 element, suggesting that c-Fos may also act as a stimulator via a mechanism independent of the AP-1 site. Further study is merited to elucidate this mechanism. It is noteworthy that various AP-1 proteins displayed differential stimulatory activity to endogenous AKR1B10 expression and the promoter luciferase reporters (Figure 5), which is consistent with the fact that various Fos/Jun heterodimers differ in their ability to transactivate AP-1-dependent genes [46]. Taken together, the results of the present study indicate that AKR1B10 is a novel target gene of AP-1 in its oncogenesis.

AKR1B10 was overexpressed in some HCC tissues. Therefore we investigated AP-1-mediated AKR1B10 expression in HepG2 cells inoculated in nude mice. This observation mimicked, at least in part, the in vivo environment in tumorigenesis, thus having greater pathophysiological significance. Either intraperitoneal or intratumoral delivery of EGF induced AKR1B10 expression in the tumours, but, in contrast, the AP-1 antagonist U0126 inhibited the tumoral expression of AKR1B10. Similarly, c-Fos shRNA, administered intratumorally to enhance its bioavailability, also significantly reduced AKR1B10 expression in the tumours. Taken together, these in vivo data strengthen the notion that EGF regulates AKR1B10 expression through AP-1 signalling in tumorigenesis.

In human HCC tissues, AKR1B10 expression inversely correlated with tumour grade (Spearman's rank correlation coefficient=−0.3435; 95% confidence interval=−0.4989 to −0.1666; and P=0.0002). This is consistent with that reported in lung cancer, in which AKR1B10 up-regulation was significantly more frequent in well-differentiated squamous carcinoma cells and precancerous squamous metaplasia than in poorly differentiated carcinomas [2]. A clear explanation on this inverse correlation is currently lacking, but it may be ascribed to its complicated expression regulation at the promoter level. As previously reported, there exist in the AKR1B10 promoter multiple putative oncogenic and tumour-suppressor protein-binding sites, such as c-Ets-1, C/EBP (CCAAT/enhancer-binding protein) and p53 [34]. Deregulation or non-specific switching on/off of certain transcriptional mechanisms, which occurs in more malignant HCC cells, may down-regulate the expression of AKR1B10. Extensive studies on AKR1B10 expression regulation are ongoing in our laboratory.

EGF and insulin are critical mitogens involved in energy metabolism, cell growth and proliferation. Enhanced EGF and insulin signalling has been reported in hepatocarcinogenesis, but the underlying molecular mechanisms need to be fully elucidated [27,29,31,42]. Enhanced bioenergetics and lipogenesis are important early events in cancer development, including HCC [30,47]. In tumour cells, newly synthesized lipids are mainly phospholipids, meeting the need for rapid cell division. More importantly, newly synthesized lipids are enriched with saturated or mono-unsaturated fatty acids. The saturated fatty acids tend to partition into detergent-resistant membrane microdomains or rafts, mediating cell migration, signal transduction and intracellular trafficking [48]. Our previous studies have shown that AKR1B10 associates with the rate-limiting enzyme ACCA in de novo fatty acid synthesis and prevents its degradation through the ubiquitination–proteasome pathway, promoting fatty acid/lipid synthesis [18]. This finding of AKR1B10 as an effector of EGF suggests a novel mechanism of regulating lipogenesis in cancer cells, i.e. an EGF–AKR1B10–lipogenesis axis. Taken together with the well-documented role of AKR1B10 in cytotoxic carbonyl elimination, this EGF–AKR1B10 axis may exert an important effect in hepatocarcinogenesis, representing a novel carcinogenic mechanism of this mitogen.

In summary, mitogenic EGF stimulates AKR1B10 expression in HCC cells via AP-1 signalling, which may promote HCC development and progression via eliminating cytotoxic carbonyls and promoting lipogenesis. The present study identified a novel EGF–AKR1B10 axis in HCC cells, which may be a potential target for therapeutic intervention in cancer.

Abbreviations

     
  • ACCA

    acetyl-CoA carboxylase-α

  •  
  • AKR1B10

    aldo-keto reductase 1B10

  •  
  • AP-1

    activator protein-1

  •  
  • ChIP

    chromatin immunoprecipitation

  •  
  • DMEM

    Dulbecco's modified Eagle's medium

  •  
  • EGF

    epidermal growth factor

  •  
  • EMSA

    electrophoretic mobility-shift assay

  •  
  • HCC

    hepatocellular carcinoma

  •  
  • RT

    reverse transcription

  •  
  • shRNA

    short hairpin RNA

  •  
  • SV40

    simian virus 40

AUTHOR CONTRIBUTION

Deliang Cao designed the research, obtained grant support and wrote the paper. Ziwen Liu performed the majority of the experiments except those described below. Ruilan Yan performed the AKR1B10 enzyme activity assays and Western blotting. Ahmed Al-salman conducted some luciferase activity assays and the AP-1 inhibitor tests. Yi Shen helped with animal studies and performed the immunohistochemistry of the tumour sections. Yiwen Bu performed the RT–PCR and Chenfei Huang tested the effect of AP-1 inhibitor on EGF stimulation. Jun Ma performed the immunohistochemistry of HCC and Di-Xian Luo performed the Western blot analysis of AKR1B10 in HCC. Yuyang Jiang, Andrew Wilber, Yin-Yuan Mo, Mei Chris Huang and Yupei Zhao were involved in the design of the study, discussion of the results, and paper preparation and revision.

ACKOWLEDGEMENTS

We thank Mr Navneet for his help with making Figure 1(A) (lower panel).

FUNDING

This work was supported in part by the National Cancer Institute [grant number CA122327].

References

References
1
Cao
D.
Fan
S. T.
Chung
S. S.
Identification and characterization of a novel human aldose reductase-like gene
J. Biol. Chem.
1998
, vol. 
273
 (pg. 
11429
-
11435
)
2
Fukumoto
S.
Yamauchi
N.
Moriguchi
H.
Hippo
Y.
Watanabe
A.
Shibahara
J.
Taniguchi
H.
Ishikawa
S.
Ito
H.
Yamamoto
S.
, et al. 
Overexpression of the aldo-keto reductase family protein AKR1B10 is highly correlated with smokers' non-small cell lung carcinomas
Clin. Cancer Res.
2005
, vol. 
11
 (pg. 
1776
-
1785
)
3
Yan
R.
Zu
X.
Ma
J.
Liu
Z.
Adeyanju
M.
Cao
D.
Aldo-keto reductase family 1 B10 gene silencing results in growth inhibition of colorectal cancer cells: implication for cancer intervention
Int. J. Cancer
2007
, vol. 
121
 (pg. 
2301
-
2306
)
4
Wang
C.
Yan
R.
Luo
D.
Watabe
K.
Liao
D. F.
Cao
D.
Aldo-keto reductase family 1 member B10 promotes cell survival by regulating lipid synthesis and eliminating carbonyls
J. Biol. Chem.
2009
, vol. 
284
 (pg. 
26742
-
26748
)
5
Luo
D. X.
Huang
M. C.
Ma
J.
Gao
Z.
Liao
D. F.
Cao
D.
Aldo-keto reductase family 1, member B10 is secreted through a lysosome-mediated non-classical pathway
Biochem. J.
2011
, vol. 
438
 (pg. 
71
-
80
)
6
Shen
Y.
Zhong
L.
Markwell
S.
Cao
D.
Thiol-disulfide exchanges modulate aldo-keto reductase family 1 member B10 activity and sensitivity to inhibitors
Biochimie
2010
, vol. 
92
 (pg. 
530
-
537
)
7
Hyndman
D.
Bauman
D. R.
Heredia
V. V.
Penning
T. M.
The aldo-keto reductase superfamily homepage
Chem. Biol. Interact.
2003
, vol. 
143/144
 (pg. 
621
-
631
)
8
Lee
K. W.
Ko
B. C.
Jiang
Z.
Cao
D.
Chung
S. S.
Overexpression of aldose reductase in liver cancers may contribute to drug resistance
Anticancer Drugs
2001
, vol. 
12
 (pg. 
129
-
132
)
9
Barski
O. A.
Tipparaju
S. M.
Bhatnagar
A.
The aldo-keto reductase superfamily and its role in drug metabolism and detoxification
Drug Metab. Rev.
2008
, vol. 
40
 (pg. 
553
-
624
)
10
Jin
Y.
Penning
T. M.
Aldo-keto reductases and bioactivation/detoxication
Annu. Rev. Pharmacol. Toxicol.
2007
, vol. 
47
 (pg. 
263
-
292
)
11
Shen
Y.
Zhong
L.
Johnson
S.
Cao
D.
Human aldo-keto reductases 1B1 and 1B10: a comparative study on their enzyme activity toward electrophilic carbonyl compounds
Chem. Biol. Interact.
2011
, vol. 
191
 (pg. 
192
-
198
)
12
Spite
M.
Baba
S. P.
Ahmed
Y.
Barski
O. A.
Nijhawan
K.
Petrash
J. M.
Bhatnagar
A.
Srivastava
S.
Substrate specificity and catalytic efficiency of aldo-keto reductases with phospholipid aldehydes
Biochem. J.
2007
, vol. 
405
 (pg. 
95
-
105
)
13
Martin
H. J.
Maser
E.
Role of human aldo-keto-reductase AKR1B10 in the protection against toxic aldehydes
Chem. Biol. Interact.
2009
, vol. 
178
 (pg. 
145
-
150
)
14
Zhong
L.
Liu
Z.
Yan
R.
Johnson
S.
Zhao
Y.
Fang
X.
Cao
D.
Aldo-keto reductase family 1 B10 protein detoxifies dietary and lipid-derived α, β-unsaturated carbonyls at physiological levels
Biochem. Biophys. Res. Commun.
2009
, vol. 
387
 (pg. 
245
-
250
)
15
Zu
X.
Yan
R.
Robbins
S.
Krishack
P. A.
Liao
D. F.
Cao
D.
Reduced 293T cell susceptibility to acrolein due to aldose reductase-like-1 protein expression
Toxicol. Sci.
2007
, vol. 
97
 (pg. 
562
-
568
)
16
Balendiran
G. K.
Fibrates in the chemical action of daunorubicin
Curr. Cancer Drug Targ.
2009
, vol. 
9
 (pg. 
366
-
369
)
17
Zhong
L.
Shen
H.
Huang
C.
Jing
H.
Cao
D.
AKR1B10 Induces cell resistance to daunorubicin and idarubicin by reducing C13 ketonic group
Toxicol. Appl. Pharmacol.
2011
, vol. 
255
 (pg. 
40
-
47
)
18
Ma
J.
Yan
R.
Zu
X.
Cheng
J. M.
Rao
K.
Liao
D. F.
Cao
D.
Aldo-keto reductase family 1 B10 affects fatty acid synthesis by regulating the stability of acetyl-CoA carboxylase-α in breast cancer cells
J. Biol. Chem.
2008
, vol. 
283
 (pg. 
3418
-
3423
)
19
Wang
R.
Wang
G.
Ricard
M. J.
Ferris
B.
Strulovici-Barel
Y.
Salit
J.
Hackett
N. R.
Gudas
L. J.
Crystal
R. G.
(2010) Smoking-induced upregulation of AKR1B10 expression in the airway epithelium of healthy individuals
Chest
, vol. 
138
 (pg. 
1402
-
1410
)
20
Quinn
A. M.
Harvey
R. G.
Penning
T. M.
Oxidation of PAH trans-dihydrodiols by human aldo-keto reductase AKR1B10
Chem. Res. Toxicol.
2008
, vol. 
21
 (pg. 
2207
-
2215
)
21
Li
C. P.
Goto
A.
Watanabe
A.
Murata
K.
Ota
S.
Niki
T.
Aburatani
H.
Fukayama
M.
AKR1B10 in usual interstitial pneumonia: expression in squamous metaplasia in association with smoking and lung cancer
Pathol. Res. Pract.
2008
, vol. 
204
 (pg. 
295
-
304
)
22
Nishinaka
T.
Miura
T.
Okumura
M.
Nakao
F.
Nakamura
H.
Terada
T.
Regulation of aldo-keto reductase AKR1B10 gene expression: involvement of transcription factor Nrf2
Chem. Biol. Interact.
2011
, vol. 
191
 (pg. 
185
-
191
)
23
Pessin
J. E.
Okada
S.
Insulin and EGF receptors integrate the Ras and Rap signaling pathways
Endocr. J.
1999
, vol. 
46
 
Suppl.
(pg. 
S11
-
S16
)
24
Bublil
E. M.
Yarden
Y.
The EGF receptor family: spearheading a merger of signaling and therapeutics
Curr. Opin. Cell Biol.
2007
, vol. 
19
 (pg. 
124
-
134
)
25
Stoscheck
C. M.
King
L. E.
Jr
Role of epidermal growth factor in carcinogenesis
Cancer Res.
1986
, vol. 
46
 (pg. 
1030
-
1037
)
26
Singletary
S. E.
Baker
F. L.
Spitzer
G.
Tucker
S. L.
Tomasovic
B.
Brock
W. A.
Ajani
J. A.
Kelly
A. M.
Biological effect of epidermal growth factor on the in vitro growth of human tumors
Cancer Res.
1987
, vol. 
47
 (pg. 
403
-
406
)
27
Tanabe
K. K.
Lemoine
A.
Finkelstein
D. M.
Kawasaki
H.
Fujii
T.
Chung
R. T.
Lauwers
G. Y.
Kulu
Y.
Muzikansky
A.
Kuruppu
D.
, et al. 
Epidermal growth factor gene functional polymorphism and the risk of hepatocellular carcinoma in patients with cirrhosis
JAMA, J. Am. Med. Assoc.
2008
, vol. 
299
 (pg. 
53
-
60
)
28
Mullhaupt
B.
Feren
A.
Fodor
E.
Jones
A.
Liver expression of epidermal growth factor RNARapid increases in immediate–early phase of liver regeneration
J. Biol. Chem.
1994
, vol. 
269
 (pg. 
19667
-
19670
)
29
Buckley
A. F.
Burgart
L. J.
Kakar
S.
Epidermal growth factor receptor expression and gene copy number in fibrolamellar hepatocellular carcinoma
Hum. Pathol.
2006
, vol. 
37
 (pg. 
410
-
414
)
30
Bannasch
P.
Klimek
F.
Mayer
D.
Early bioenergetic changes in hepatocarcinogenesis: preneoplastic phenotypes mimic responses to insulin and thyroid hormone
J. Bioenerg. Biomembr.
1997
, vol. 
29
 (pg. 
303
-
313
)
31
Fritsche
L.
Weigert
C.
Haring
H. U.
Lehmann
R.
How insulin receptor substrate proteins regulate the metabolic capacity of the liver: implications for health and disease
Curr. Med. Chem.
2008
, vol. 
15
 (pg. 
1316
-
1329
)
32
Boissan
M.
Beurel
E.
Wendum
D.
Rey
C.
Lecluse
Y.
Housset
C.
Lacombe
M. L.
Desbois-Mouthon
C.
Overexpression of insulin receptor substrate-2 in human and murine hepatocellular carcinoma
Am. J. Pathol.
2005
, vol. 
167
 (pg. 
869
-
877
)
33
Tanaka
S.
Sugimachi
K.
Maehara
S.
Harimoto
N.
Shirabe
K.
Wands
J. R.
Sugimachi
K.
Oncogenic signal transduction and therapeutic strategy for hepatocellular carcinoma
Surgery
2002
, vol. 
131
 (pg. 
S142
-
S147
)
34
Liu
Z.
Zhong
L.
Krishack
P. A.
Robbins
S.
Cao
J. X.
Zhao
Y.
Chung
S.
Cao
D.
Structure and promoter characterization of aldo-keto reductase family 1 B10 gene
Gene
2009
, vol. 
437
 (pg. 
39
-
44
)
35
Pei
H.
Yordy
J. S.
Leng
Q.
Zhao
Q.
Watson
D. K.
Li
R.
EAPII interacts with ETS1 and modulates its transcriptional function
Oncogene
2003
, vol. 
22
 (pg. 
2699
-
2709
)
36
Claret
F. X.
Hibi
M.
Dhut
S.
Toda
T.
Karin
M.
A new group of conserved coactivators that increase the specificity of AP-1 transcription factors
Nature
1996
, vol. 
383
 (pg. 
453
-
457
)
37
Chambellan
A.
Leahy
R.
Xu
W.
Cruickshank
P. J.
Janocha
A.
Szabo
K.
Cannady
S. B.
Comhair
S. A.
Erzurum
S. C.
Pivotal role of c-Fos in nitric oxide synthase 2 expression in airway epithelial cells
Nitric Oxide
2009
, vol. 
20
 (pg. 
143
-
149
)
38
Zhang
D.
Cao
D.
Russell
R.
Pizzorno
G.
p53-dependent suppression of uridine phosphorylase gene expression through direct promoter interaction
Cancer Res.
2001
, vol. 
61
 (pg. 
6899
-
6905
)
39
Chung
S. W.
Chen
Y. H.
Yet
S. F.
Layne
M. D.
Perrella
M. A.
Endotoxin-induced down-regulation of Elk-3 facilitates heme oxygenase-1 induction in macrophages
J. Immunol.
2006
, vol. 
176
 (pg. 
2414
-
2420
)
40
Tsang
M. C.
Lo
A. C.
Cheung
P. T.
Chung
S. S.
Chung
S. K.
Perinatal hypoxia-/ischemia-induced endothelin-1 mRNA in astrocyte-like and endothelial cells
NeuroReport
2001
, vol. 
12
 (pg. 
2265
-
2270
)
41
Heringlake
S.
Hofdmann
M.
Fiebeler
A.
Manns
M. P.
Schmiegel
W.
Tannapfel
A.
Identification and expression analysis of the aldo-ketoreductase1-B10 gene in primary malignant liver tumours
J. Hepatol.
2010
, vol. 
52
 (pg. 
220
-
227
)
42
Ish-Shalom
D.
Christoffersen
C. T.
Vorwerk
P.
Sacerdoti-Sierra
N.
Shymko
R. M.
Naor
D.
De Meyts
P.
Mitogenic properties of insulin and insulin analogues mediated by the insulin receptor
Diabetologia.
1997
, vol. 
40
 (pg. 
S25
-
S31
)
43
Cui
X.
Kim
H. J.
Kuiatse
I.
Kim
H.
Brown
P. H.
Lee
A. V.
Epidermal growth factor induces insulin receptor substrate-2 in breast cancer cells via c-Jun NH2-terminal kinase/activator protein-1 signaling to regulate cell migration
Cancer Res.
2006
, vol. 
66
 (pg. 
5304
-
5313
)
44
Yuen
M. F.
Wu
P. C.
Lai
V. C.
Lau
J. Y.
Lai
C. L.
Expression of c-Myc, c-Fos, and c-Jun in hepatocellular carcinoma
Cancer
2001
, vol. 
91
 (pg. 
106
-
112
)
45
Gao
S. Y.
Li
E. M.
Cui
L.
Lu
X. F.
Meng
L. Y.
Yuan
H. M.
Xie
J. J.
Du
Z. P.
Pang
J. X.
Xu
L. Y.
Sp1 and AP-1 regulate expression of the human gene VIL2 in esophageal carcinoma cells
J. Biol. Chem.
2009
, vol. 
284
 (pg. 
7995
-
8004
)
46
Tulchinsky
E.
Fos family members: regulation, structure and role in oncogenic transformation
Histol. Histopathol.
2000
, vol. 
15
 (pg. 
921
-
928
)
47
Yahagi
N.
Shimano
H.
Hasegawa
K.
Ohashi
K.
Matsuzaka
T.
Najima
Y.
Sekiya
M.
Tomita
S.
Okazaki
H.
Tamura
Y.
, et al. 
Co-ordinate activation of lipogenic enzymes in hepatocellular carcinoma
Eur. J. Cancer.
2005
, vol. 
41
 (pg. 
1316
-
1322
)
48
Simons
K.
Toomre
D.
Lipid rafts and signal transduction
Nat. Rev. Mol. Cell Biol.
2000
, vol. 
1
 (pg. 
31
-
39
)