SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

The increasing incidence of hepatocellular carcinoma (HCC) is of great concern not only in the United States but throughout the world. Although sorafenib, a multikinase inhibitor with antiangiogenic and antiproliferative effects, currently sets the new standard for advanced HCC, tumor response rates are usually quite low. An understanding of the underlying mechanisms for sorafenib resistance is critical if outcomes are to be improved. In this study we tested the hypothesis that hypoxia caused by the antiangiogenic effects of sustained sorafenib therapy could induce sorafenib resistance as a cytoprotective adaptive response, thereby limiting sorafenib efficiency. We found that HCCs, clinically resistant to sorafenib, exhibit increased intratumor hypoxia compared with HCCs before treatment or HCCs sensitive to sorafenib. Hypoxia protected HCC cells against sorafenib and hypoxia-inducible factor 1 (HIF-1α) was required for the process. HCC cells acquired increased P-gp expression, enhanced glycolytic metabolism, and increased nuclear factor kappa B (NF-κB) activity under hypoxia. EF24, a molecule having structural similarity to curcumin, could synergistically enhance the antitumor effects of sorafenib and overcome sorafenib resistance through inhibiting HIF-1α by sequestering it in cytoplasm and promoting degradation by way of up-regulating Von Hippel-Lindau tumor suppressor (VHL). Furthermore, we found that sustained sorafenib therapy led to increased intratumor hypoxia, which was associated with sorafenib sensitivity in HCC subcutaneous mice tumor models. The combination of EF24 and sorafenib showed synergistically effects against metastasis both in vivo and in vitro. Synergistic tumor growth inhibition effects were also observed in subcutaneous and orthotopic hepatic tumors. Conclusion: Hypoxia induced by sustained sorafenib treatment confers sorafenib resistance to HCC through HIF-1α and NF-κB activation. EF24 overcomes sorafenib resistance through VHL-dependent HIF-1α degradation and NF-κB inactivation. EF24 in combination with sorafenib represents a promising strategy for HCC. (HEPATOLOGY 2013)

Hepatocellular carcinoma (HCC) is one of the most common cancer types worldwide.1 Most HCC patients are diagnosed at advanced stages, and therefore there is an urgent need for new systemic therapies.1 Sorafenib, a multikinase inhibitor, is the first and only drug that is clinically approved for patients with advanced HCC.2 Sorafenib not only blocks raf kinase, but also the kinase activity of vascular endothelial growth factor (VEGF) and platelet-derived growth factor (PDGF) receptors.3 Although sorafenib showed survival benefits in large randomized Phase III studies, the response rate is actually quite low.4 In terms of absolute numbers of months gain, the prolongation of survival is modest.5 Therefore, it is important to identify the molecular mechanisms of sorafenib resistance and to improve the response of sorafenib in HCC.

It is widely accepted that hypoxia in solid tumors is associated with chemotherapy failure, selection of more invasive and resistant clones, and poor patient prognosis.6, 7 Because of the failure of conventional DNA damaging chemotherapy, antiangiogenic therapy has been widely investigated in many cancer clinical trails. However, the efficacy is often transient, with acquired resistance, which may be due, at least in part, to increased tumor hypoxia and activation of compensatory survival pathways that negatively impact the therapeutic outcome.8-10 Indeed, a previous study has indicated that administration of sorafenib resulted in a decrease of vasculature, leading to an elevated level of tumor hypoxia within short-term treatment.8 In addition, recent studies in mouse models of different cancers showed an unwanted side effect of sorafenib: after initial antitumor activity, enhanced tumor progression and increased metastasis occurred.9, 10 In this study we hypothesized that tumor hypoxia induced by the antiangiogenic effects of sorafenib and selection of highly resistant cells adapted to depletion of oxygen and nutrition may be potential mechanisms of escape from sustained sorafenib treatment in HCC. If so, combination of sorafenib with a “sensitizer” that can suppress hypoxia-mediated effects may be necessary for complete elimination of HCC through sorafenib therapy.

EF24, a molecule having structural similarity to curcumin, has been reported to inhibit hypoxia-inducible factor 1 (HIF-1α) accumulation under hypoxia, but the mechanisms are still unclear.11 Previously, we have found that EF24 could potently inhibit HCC cell proliferation and induce apoptosis. We also provided evidence that the molecular mechanism is possibly inhibiting nuclear factor kappa B (NF-κB) activity.12 Our purpose in this study was to investigate whether hypoxia induced by sustained sorafenib treatment could lead to sorafenib resistance and to examine whether EF24 can reverse the resistance by inhibiting HIF-1α and NF-κB using HCC cell and mice models.

Patients and Methods

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Patients, Cell Lines, and Animals.

After selection, 36 HCC patients were enrolled in this study as described in the Supporting Information. Five HCC cell lines and nude mice were used in this study and are described in the Supporting Material.

Hypoxic Treatment.

Cells were placed in a sealed hypoxia chamber (Billups-Rothenberg) equilibrated with certified gas containing 1% O2, 5% CO2, and 94% N2.

Apoptosis Assay.

In brief, following treatment cells were washed twice with cold phosphate-buffered saline (PBS), resuspended in binding buffer at a concentration of 1 × 106 cells/mL. Then 100 μL of the cell suspension (1 × 105) was transferred to a 5-mL culture tube. Five μL of Annexin V-FITC and 5 μL of propidium iodide were added to each tube. Cells were gently vortexed and incubated for 15 minutes at room temperature (25°C) in a dark environment. Next, 400 μL of binding buffer (1×) was added to each tube and the cell suspension was subsequently analyzed by flow cytometry.

Immunofluorescence Assay.

Briefly, HCC cells were washed and fixed with 3% paraformaldehyde (PFA) or 100% methanol and permeabilized in 0.1% Triton X-100. Incubation with monoclonal mouse anti-HIF-1α antibody (Novus Biologicals) for 30 minutes was followed by incubation with fluorescein isothiocyanate (FITC)-labeled antimouse IgG secondary antibody (Santa Cruz Biotechnology, Santa Cruz, CA). Immunostained cells were photographed under an inverted fluorescence microscope (×400).

HIF-1α Silenced by Lentiviral Vector-Mediated Short Hairpin RNA (shRNA).

Silencing of HIF-1α was achieved by way of lentiviral transduction of the following specific shRNA vectors, obtained from Santa Cruz Biotechnology: HIF-1α specific shRNA sc-35561, and scramble shRNA control sc-108080. After transduction, stable cell lines expressing the shRNA were isolated by way of selection with puromycin.

Electrophoretic Mobility Shift Assay (EMSA).

The detailed methodology has been described.13 Briefly, nuclear extract (10 μg) was incubated with 1 μg of poly (deoxyinosinic-deoxycytidylic acid) in binding buffer for 30 minutes at 4°C. DNA binding activity was confirmed with a biotin-labeled oligonucleotide bio-NF-κB probe using an EMSA kit.

In Vivo Metastasis Analysis.

HepG2 cells (1 × 106/0.2 mL) were injected into nude mice by way of tail vein to imitate tumor metastasis. Experimental animals (n = 15/group) received vehicle, EF24 (10 mg/kg/d), sorafenib (10 mg/kg/d), or EF24 (10 mg/kg/d) + sorafenib (10 mg/kg/d) treatment five times per week beginning on the day of implantation. The mice were killed 5 weeks after the inoculation and lungs were removed and fixed in formaldehyde.

Subcutaneous HCC Experiments.

Male BALB/c (5-6 weeks old) mice (n = 10/group) were inoculated subcutaneously in the flank with 2 × 106 Huh-7 cells or 3 × 106 Hep3B cells suspended in PBS. Sorafenib tosylate was used in in vivo experiments. EF24 solution was prepared in 20% dimethyl sulfoxide (DMSO) and 15% Tween 80 in 0.9% saline. When tumors of ∼100 mm3 were detected (around 6 days), mice were administered once daily with sorafenib (orally, 10 mg/kg on days 1-5 of each week) and/or with EF24 (intraperitoneally, 10 mg/kg on days 1-5 of each week).

Orthotopic Huh-7 Hepatic Tumor Experiments.

Huh-7-luciferase-transfected (Luci-Huh-7) cells were harvested from subconfluent cultures and washed once in serum-free medium and resuspended in PBS. Only suspensions consisting of single cells, with >90% viability, were used for the injections. Mice were anesthetized with pentobarbital sodium, a small left abdominal flank incision was made, and Luci-Huh-7 cells 5 × 106 in 200 μL PBS were injected into the spleen parenchyma. To avoid intrasplenic tumor growth, the spleen was removed after 10 minutes. The incision was closed in two layers using vicryl 5/0 for the abdominal wall and vicryl 4/0 for the skin. After 1 week of implantation, mice were randomized into the following treatment groups (n = 8/group) based on the bioluminescence measured after IVIS imaging (Xenogen, Alameda, CA): (1) Control (treated with vehicles); (2) Sorafenib (orally, 10 mg/kg/d); (3) EF24 (intraperitoneally, 10 mg/kg/d); (4) EF24 (intraperitoneally, 10 mg/kg/d) + Sorafenib (orally, 10 mg/kg/d). Mice were imaged by the bioluminescence IVIS Imaging System weekly and then mice were sacrificed.

Immunohistochemical Analysis.

Expression of Ki-67, cleaved-caspase 3, HIF-1α, CD31, and Von Hippel-Lindau tumor suppressor (VHL) was evaluated using an immunohistochemical method described previously.12

Statistical Analysis.

Statistical analyses were performed with mean ± standard deviation (SD) values using Student's t-test and two-way analysis of variance (ANOVA) with the Bonferroni's correction. Statistical significance was concluded at P < 0.05.

Details for cell culture and treatments, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) and clonogenic assay, western blot and Taqman real-time polymerase chain reaction (PCR), measurement of glucose uptake, the rate of glycolysis and lactate production, Luciferase reporter assay for NF-κB transcriptional activity, HIF-1α reporter gene assay, lenti-shRNA-mediated VHL silence, migration and invasion assay, subcutaneous and orthotopic hepatic tumor experiments are given in the Supporting Material.

Results

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

HCCs from Patients Resistant to Sorafenib Exhibit Increased Intratumor Hypoxia Compared with HCCs Without Sorafenib Treatment or HCCs Sensitive to Sorafenib Treatment.

Review of an institutional database of 1,645 HCC patients revealed 36 cases meeting our criteria as described (Fig. 1A; Supporting Table S1). Tumor vessel density, assessed by CD31 immunostaining, decreased 77.2% and 43.1%, respectively, in sorafenib-resistant HCC specimens or HCCs sensitive to sorafenib compared with HCCs without treatment (Fig. 1B,C); meanwhile, immunostaining of HIF-1α increased 62.2% and 17.6%, respectively. The cell apoptotic index, assessed by cleaved caspase-3 immunostaining, decreased 78.8% in sorafenib-resistant specimens compared with sorafenib-sensitive specimens. HIF-1α transcriptional activity was assessed through measuring the messenger RNA (mRNA) levels of HIF-1α-dependent genes (VEGF, glucose transporter 1 [GLUT-1], CA9, CXCR-4, and MDR1). As shown in Fig. 1D, all examined target genes were expressed at the highest levels in sorafenib-resistant HCCs.

thumbnail image

Figure 1. HCCs resistant to sorafenib exhibit decreased vessel density and increased hypoxia compared with HCCs without sorafenib treatment or HCCs sensitive to sorafenib. (A) The criteria of HCC patient selection and outcome reporting (PR, partial response; SD, stable disease). (B) Representative immunostaining for CD31 (top row), HIF-1α (middle row), and cleaved caspase-3 (bottom row). (C) Vessel density decreased, HIF-1α staining increased in sorafenib-treated HCCs specimens. Apoptosis decreased in sorafenib-resistant HCCs compared with sorafenib sensitive HCCs. (D) Expression of HIF-1α dependent genes in tumor lysates from corresponding patients (**P < 0.01, ***P < 0.001, two-way ANOVA with Bonferroni post-test).

Download figure to PowerPoint

EF24 Overcomes Hypoxia-Mediated Sorafenib Resistance by Synergistically Enhancing the Cell Viability Inhibition and Apoptosis.

As shown in Fig. 2A, briefly culturing HCC cells under hypoxia induced significant resistance of HCC cells to sorafenib. Consistent with the cell viability results, apoptosis induced by sorafenib was also significantly attenuated by hypoxia (Fig. 2B; Supporting Fig. S1D). Hypoxia could also lead to a significant increase in the clonogenic survival of HCC cells after sorafenib treatment relative to cells maintained under normoxia (Fig. S2). EF24 enhanced the growth inhibition effects of sorafenib on HCC, which were more significant under hypoxia compared with under normoxia (Fig. 2A; Fig. S2). The coefficient of drug interaction (CDI) values, calculated as described previously,13 for different concentrations of sorafenib were all less than 1 (Fig. S1B), indicating the synergistic effects of EF24 in combination with sorafenib. The concentration of EF24 (0.5 μM) used in the combination experiment was determined according to the maximal synergistic effects (Fig. S1C). EF24 could reverse hypoxia-mediated sorafenib resistance and induced more significant apoptosis than sorafenib alone under hypoxia (Fig. 2B; Fig. S1D).

thumbnail image

Figure 2. Hypoxia-mediated sorafenib resistance can be overcome by EF24 in vitro. (A) Cells were incubated with sorafenib at various concentrations (36 hours) in the presence or absence of EF24 (0.5 μM), and then the cell viability was determined by MTT assay. ***P < 0.001, compared with cells treated with sorafenib under normoxia; #P < 0.001, compared with cells treated with sorafenib alone under hypoxia; *P < 0.05, compared with cells treated with sorafenib alone under normoxia (two-way ANOVA with Bonferroni post-test). (B) Apoptosis were determined under the same conditions as (A). *P < 0.05 (Student's t test), compared with cells treated with sorafenib under normoxia; **P < 0.05 (Student's t test), compared with cells treated with sorafenib under normoxia; #P < 0.05 (Student's t test), compared with cells treated with sorafenib alone under hypoxia.

Download figure to PowerPoint

Hypoxia Increases the Expression of Multidrug Resistance 1 (MDR1) and Enhances the Glucose Metabolism in HCC Cells.

HIF-1α protein was significantly stabilized in HCC cells under hypoxia, accompanied by the up-regulation of its target gene, VEGF (Fig. 3A). One of the most important mechanisms explaining the contribution of HIF-1α to chemoresistance is that HIF-1α is able to activate the MDR1 gene.14 MDR1 encodes for P-gp, which can decrease the intracellular concentration of chemotherapeutic drugs, including sorafenib.15 We found here that P-gp was significantly induced by hypoxia in the presence of sorafenib (Fig. 3A). As reported, hypoxia can exert its antiapoptotic functions by alterations in energy metabolism in tumors.16 Interestingly, we found that the expression of GLUT-1 (an HIF-1α dependent gene), a main glucose transporter protein, was obviously increased under hypoxia (Fig. 3A). Hypoxia could significantly enhance the glucose uptake, increase the rate of glycolysis, and lactate levels of HCC cells (Fig. 3B; Fig. S4).

thumbnail image

Figure 3. Hypoxia stabilizes HIF-1α, increases P-gp expression and glucose uptake. (A) Cells were exposed to hypoxia (24 hours), and protein expression was examined by western blot. (B) Exposure to hypoxia (24 hours) increased the glucose uptake. A significant difference from cells under normoxia is denoted by “*,” P < 0.05 (Student's t test).

Download figure to PowerPoint

HIF-1α Is Required for Hypoxia-Mediated Sorafenib Resistance.

To investigate the exact role of HIF-1α in hypoxia-mediated sorafenib resistance, we introduced lentivirus-mediated shRNA (lenti-shRNA) targeting HIF-1α into HCC cells. Hypoxia resulted in a substantial increase in HIF-1α protein that was unaffected in parental or lenti-shRNA control cells. In contrast, HIF-1α protein was barely detected in HIF-1α silenced cells (Fig. S5A). HIF-1α silence reversed the hypoxia-mediated sorafenib resistance. However, HIF-1α silence had almost no effect on cells under normoxia (Fig. 4A,B). Knockdown of HIF-1α decreased the glucose uptake (Fig. 4C) and partially reduced the NF-κB activity induced by hypoxia (Fig. S5B). Expression of HIF-1α target genes and other cell apoptosis-related genes were also examined (Fig. 4D).

thumbnail image

Figure 4. HIF-1α is required for hypoxia-mediated sorafenib resistance. (A) Cell viability was determined in HIF-1α silencing or control cells after sorafenib treatment (36 hours). *P < 0.001, compared with shRNAcon cells treated with sorafenib under hypoxia. (B) Apoptosis was analyzed using Annexin V assay after treatment (36 hours). A significant difference from shRNAcon cells under hypoxia is denoted by “*,” a significant difference from shRNAcon cells treated with sorafenib under hypoxia by “#,” P < 0.05 (Student's t test). (C) Glucose uptake was examined under normoxia or hypoxia after sorafenib treatment (24 hours). A significant difference from shRNAcon cells under hypoxia is denoted by “#,” a significant difference from shRNAcon cells treated with sorafenib under hypoxia by “*,” P < 0.05 (Student's t test). (D) Expression of target genes was detected by western blot, lane 1 (shRNA con), lane 2 (shRNA HIF-1α).

Download figure to PowerPoint

EF24 Inhibits HIF-1α Protein Accumulation in a VHL-Dependent Manner in HCC Cells.

EF24 treatment resulted in a down-regulation of HIF-1α protein and its targets in a dose-dependent manner under hypoxia (Fig. 5A). To examine whether EF24 could decrease the transcriptional activity of HIF-1α, a reporter plasmid was introduced. As expected, hypoxia could induce HIF-1α's transcriptional activity, as shown in Fig. 5B. EF24 treatment resulted in a dose-dependent reduction of the transcriptional activity of HIF-1α. However, HIF-1α mRNA levels remained unchanged, signifying that EF24 is a posttranscriptional regulator of HIF-1α (Fig. 5C).

thumbnail image

Figure 5. EF24 inhibits HIF-1α and its target genes in a VHL-dependent manner. (A) HCC cells preincubated under hypoxia for 12 hours were treated with EF24 for another 12 hours. Effects of EF24 on the expression of HIF-1α and its target genes were determined by western blot. (B) HCC cells transiently transfected with the hypoxia response element (HRE)-luc construct pBI-GL V6L and treated to the indicated drug concentrations for 24 hours under hypoxia. Luciferase values were normalized to total cell protein content. (C) The mRNA level of HIF-1α was examined by real-time PCR. (D) VHL silenced or control cells were treated with or without EF24 (0.5 μ μ) for 24 hours. Protein expression was determined by western blot. (E) Cells were treated with 0.5 μM EF24 with or without MG-132 (10 μM) under hypoxia for 24 hours. (F) EF24 sequesters HIF-1α in the cytoplasm and decreases the levels of HIF-1α under hypoxia as detected by IF staining.

Download figure to PowerPoint

The level of VHL, a negative regulator of HIF-1α, was also investigated.17 EF24 treatment caused a dose-dependent increase of VHL both in protein (Fig. 5A) and mRNA levels (Fig. S7A), indicating that EF24 regulates VHL expression at the transcriptional level. VHL silence by lenti-shRNA was used to address whether EF24 inhibits HIF-1α accumulation by way of VHL up-regulation. The results proved that VHL silence could prevent EF24-mediated HIF-1α reduction (Fig. 5D). We conclude, therefore, that EF24 destabilized HIF-1α by increasing VHL expression under hypoxia.

EF24 Sequesters HIF-1α Protein in the Cytoplasm and Promotes the Proteasomal Degradation of HIF-1α.

To further understand the mechanism through which EF24 decreases HIF-1α, we examined its ability to target HIF-1α for proteosomal degradation. MG132 (proteasome inhibitor) treatment in the presence of EF24 prevents degradation of HIF-1α under hypoxia (Fig. 5E), suggesting that EF24 could promote proteasomal degradation of HIF-1α. As reported, protection of HIF-1α against degradation by VHL was a multistep mechanism, including hypoxia-induced nuclear translocation of HIF-1α and other intracellular signals.18 The immunofluorescence (IF) staining results revealed that HIF-1α is accumulated and imported into the nuclei in control and sorafenib-treated cells under hypoxia compared with cells under normoxia. EF24 treatment changed the nuclear localization of HIF-1α whose nuclear accumulation was hindered and the overall staining is significantly reduced under hypoxia in the presence of sorafenib (Fig. 5F).

Synergistic Inhibitory Effects of EF24 and Sorafenib on HCC Cell Migration, Invasion, and Metastasis to Lung.

Results of the transwell assays demonstrated that the migration and invasive ability of HCC cells was hindered by EF24 or sorafenib under hypoxia (Fig. 6A; Fig. S8A). Moreover, administration of EF24 and sorafenib together inhibited the migration and invasion more significantly. To examine the therapeutic efficacy of EF24 and sorafenib against metastasis in vivo, HepG2 cells were injected into nude mice by way of tail vein to imitate tumor metastasis. When EF24 or sorafenib was administrated, the average number of foci per mouse was reduced by 23.6% and 43.1%, respectively (Fig. 6B). Importantly, the average number was reduced by 69.1% in the combination group. Furthermore, no gross signs of toxicity were observed (Fig. S8B).

thumbnail image

Figure 6. EF24 and sorafenib synergistically inhibit migration, metastasis, and subcutaneous tumor growth. (A) Representative images of the migration (i) and invasion (ii) assay in HepG2 cells under hypoxia; the inhibitory effect of sorafenib (10 μM) on migration and invasion was significantly enhanced by (0.5 μM) EF24. (***P < 0.001, two-way ANOVA with Bonferroni post-test). (B) The representative hematoxylin and eosin (H&E) staining of pulmonary metastasis (arrows) at 5 weeks (left panel) and the average number of foci per mouse were calculated (right panel); (**P < 0.01, ***P < 0.001, two-way ANOVA with Bonferroni post-test). (C) Huh-7 subcutaneous tumor volumes were measured and transformed to relative tumor volume (left panel); (**P < 0.01, ***P < 0.001, two-way ANOVA with Bonferroni post-test) and representative images of Huh-7 xenograft tumors at the end of treatment (right panel). (D) Representative images of sections stained with an anti-HIF-1α antibody, or with an anticleaved caspase-3 antibody (left panel); cleaved caspase-3-positive cells were counted to give the apoptosis index; cells expressing HIF-1α were counted to determine the hypoxia level (right panel) (**P < 0.01, ***P < 0.001, two-way ANOVA with Bonferroni post-test).

Download figure to PowerPoint

Combination Treatment of Sorafenib and EF24 Exerts a More Potent Antitumor Effect in Subcutaneous Xenograft Models.

Although Huh-7 tumors treated with sorafenib or EF24 were smaller than control tumors at the end of the treatment (Fig. 6C), we observed that mice in the sorafenib group gained eventually accelerated tumor growth after the 23rd day (black arrow in Fig. 6C), indicating tumors become less sensitive to sorafenib treatment. A combination of EF24 and sorafenib inhibited tumor growth in a prolonged and significant manner versus either agent alone (Fig. 6C). Similar sustained tumor growth inhibition in combined treated tumors versus eventually accelerated tumor growth in sorafenib-treated tumors was also noted in Hep3B tumor models (Fig. S8C). There was moderate weight loss in control and treated groups (Fig. S8D). Tumors from control mice showed a low number of apoptotic cells. Sorafenib alone could not significantly increase the number of apoptotic cells. However, the combination showed an obvious increase in apoptotic cells (Fig. 6D). The control and sorafenib-treated tumors had high NF-κB DNA-binding activity. However, in combination or EF24 groups, the activity was significantly reduced (Fig. S8E). Sustained sorafenib treatment could increase the protein level of HIF-1α compared with control (Fig. 6D; Fig. S9A). However, combination therapy or EF24 treatment led to up-regulation of VHL and down-regulation of HIF-1α (Fig. S9A). Expression of the key tumor survival genes, examined by western blot, was similar to that in vitro (Fig. S9A).

EF24 Can Potentiate the Antitumor Effects of Sorafenib in Huh-7 Orthotopic Tumor Models by Inhibiting HIF-1α by Way of VHL.

The results illustrated a gradual increase in orthotopic Huh-7 hepatic tumor growth in the control group compared with the treated groups. Tumors treated with combination therapy were significantly smaller than tumors treated with sorafenib or EF24 (Fig. 7A-C; Fig. S10A,B). There were more apoptotic cells, fewer Ki-67-positive cells, and CD31-stained vessels in tumors treated with either sorafenib or EF24 compared with control tumors. The combination therapy resulted in even much fewer Ki-67-positive, more apoptotic cells, and fewer vessels (Fig. 7D; Fig. S11A-C). Sorafenib treatment increased the DNA-binding activity of NF-κB, whereas concurrent treatment with EF24 decreased it (Fig. 7E). EF24 could up-regulate VHL protein level and decrease HIF-1α protein level in tumors (Fig. 7F; Fig. S11D). Expression of key tumor survival genes, examined by western blot of tumor homogenates, was similar to that in vitro (Fig. 7F).

thumbnail image

Figure 7. EF24 potentiates the antitumor effects of sorafenib on orthotopic hepatic tumors. (A) Representative bioluminescence images corresponding to Huh-7 orthotopic hepatic tumors. (B) Volume of Huh-7 orthotopic tumors was determined at different timepoints. Data points represent the mean ± SD (*P < 0.05, ***P < 0.001, two-way ANOVA with Bonferroni post-test). (C) The ratio of liver weight/body weight (*P < 0.05, **P < 0.01, two-way ANOVA with Bonferroni post-test). (D) Representative images of sections stained with anti-Ki-67 antibody, or with anticleaved caspase-3 antibody, or with anti-CD31 antibody. (E) The nuclear fraction of the tumor tissue was subjected to EMSA to measure the NF-κB DNA-binding activity. (F) Tumor tissue lysates were subjected to western blot to measure the expression of the target proteins.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

The present study has for the first time demonstrated that intratumor hypoxia could be generated by sustained sorafenib treatment both in human HCC patients and mice models. Tumor hypoxia, microvessel density, HIF-1α, and its target gene levels were investigated to better understand the effects of sorafenib on the tumor microenvironment. The results indicated that sustained sorafenib treatment resulted in decreased microvessel density, increased HIF-1α protein levels, and its transcriptional activity. Importantly, a moderate tumor growth inhibition at the early stage of low-dose sorafenib treatment and an eventual accelerated tumor growth accompanied with increased tumor hypoxia in subcutaneous and orthotopic tumor models reflect that the levels of intratumor hypoxia might be predictive of HCC response to sorafenib.

The in vitro studies suggest that hypoxia is able to induce sorafenib resistance by decreasing the growth inhibition and apoptosis mediated by sorafenib. Hypoxia can stabilize HIF-1α protein in the presence of sorafenib, increase the expression and function of HIF-1α targets, such as MDR1, GLUT-1, and VEGF. MDR1 is an important contribution to HIF-1α-mediated P-gp expression to hypoxia-induced drug resistance, which has been observed in a plethora of tumor cells. In addition, we also provide evidence that changes of glucose metabolism, including enhanced glucose uptake, increased rate of glycolysis, and lactate levels could be other important adaptation processes for cell survival after sorafenib treatment under hypoxia. Hypoxia can increase the activity of NF-κB and subsequently increase the expression of its antiapoptotic target genes such as Bcl-2, survivin, and decrease the apoptotic target genes such as Bax. HIF-1α silence using lenti-shRNA abolished these effects and strikingly elevated the sensitivity of HCC to sorafenib both in vivo and in vitro. All these data strongly suggest that hypoxia-induced HIF-1α might play an important role in conferring sorafenib resistance in HCC and provide a novel strategy through which sorafenib efficiency may be improved. The effects of sorafenib on tumor hypoxia also suggest that combination with an HIF-1α inhibitor could be a rational strategy for HCC treatment.

VHL protein, expressed in most tissues and cell types, can directly mediate the ubiquitylation and proteasomal degradation of HIF-1α by way of physical interaction with the core of the oxygen-dependent degradation domain.18 However, it is now clear that in hypoxic cells prolyl hydroxylase 2 (PHD2) activity is still present, acting to degrade HIF-1α. VHL can engage HIF-1α in hypoxia and promote VHL-mediated HIF-1α ubiquitylation and degradation.18 Our previous study has suggested that down-regulation of VHL may be partly responsible for doxorubicin resistance and adenovirus-expressing VHL enhances the efficacy of doxorubicin by targeting HIF and NF-κB.13 Therefore, regulation of VHL expression during hypoxia may provide an effective therapy for HCC. We demonstrated here that low-dose EF24 could abolish the accumulation of HIF-1α protein and lead to suppression of its target genes by way of up-regulating VHL protein.

In light of the fact that EF24 could inhibit HIF-1α and reduce DNA-binding activity of NF-κB,12 we hypothesized that EF24 can overcome hypoxia-mediated sorafenib resistance. This hypothesis was supported by the cell viability and apoptosis results, which showed that EF24 synergically inhibited the cell viability and increased the apoptosis induced by sorafenib under hypoxia. We also found that EF24 in combination with sorafenib displayed synergistic tumor growth inhibition both in subcutaneous and orthotopic HCC mice models. Importantly, in this study a lower daily dose of EF24, compared with our previous study,12 was able to inhibit HIF-1α protein accumulation through up-regulating VHL. We also found that EF24 combined with sorafenib has synergistic effects against metastasis both in vivo and in vitro. One important possibility we thought might explain the increased inhibition of tumor growth exerted by the combination is that the HIF-1α degradation and NF-κB inhibition by EF24 was sufficient to counteract the hypoxia-mediated drug resistance induced by the antiangiogenic effects of sustained sorafenib treatment. Indeed, we observed that administration of EF24 could significantly decrease the expression of HIF-1α-dependent genes, which could be induced by sustained sorafenib treatment alone. The combination of EF24 with sorafenib also markedly reduced angiogenesis relative to either agent alone.

In summary, our results have provided direct evidence that HIF-1α and NF-κB pathways controlled by sorafenib-induced hypoxia are the driving force for HCC cells to maintain survival under sustained sorafenib treatment. We further provide evidence that inhibition of HIF-1α and NF-κB by EF24 overcomes sorafenib resistance in HCC (Fig. 8). These results indicate that EF24 in combination with sorafenib represents a promising strategy for the treatment of HCC and needs further clinical investigation.

thumbnail image

Figure 8. Schematic model depicting the possible mechanisms of sorafenib resistance induced by hypoxia and the underlying molecular mechanisms through which EF24 overcomes the resistance.

Download figure to PowerPoint

References

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information
  • 1
    Villanueva A, Llovet JM. Targeted therapies for hepatocellular carcinoma. Gastroenterology 2011; 140: 1410-1426.
  • 2
    El-Serag HB, Marrero JA, Rudolph L, Reddy KR. Diagnosis and treatment of hepatocellular carcinoma. Gastroenterology 2008; 134: 1752-1763.
  • 3
    Colombo M. Sorafenib in advanced hepatocellular carcinoma: a further step toward personalized therapy of liver cancer. Gastroenterology 2009; 136: 1832-1835.
  • 4
    Llovet JM, Ricci S, Mazzaferro V, Hilgard P, Gane E, Blanc JF, et al. Sorafenib in advanced hepatocellular carcinoma. N Engl J Med 2008; 359: 378-390.
  • 5
    Dufour JF. The evasive promise of antiangiogenic therapy. J Hepatol 2009; 51: 970-972.
  • 6
    Trédan O, Galmarini CM, Patel K, Tannock IF. Drug resistance and the solid tumor microenvironment. J Natl Cancer Inst 2007; 99: 1441-1454.
  • 7
    Tak E, Lee S, Lee J, Rashid MA, Kim YW, Park JH, et al. Human carbonyl reductase 1 upregulated by hypoxia renders resistance to apoptosis in hepatocellular carcinoma cells. J Hepatol 2011; 54: 328-339.
  • 8
    Chang YS, Adnane J, Trail PA, Levy J, Henderson A, Xue D, et al. Sorafenib (BAY 43-9006) inhibits tumor growth and vascularization and induces tumor apoptosis and hypoxia in RCC xenograft models. Cancer Chemother Pharmacol 2007; 59: 561-574.
  • 9
    Ebos JM, Lee CR, Cruz-Munoz W, Bjarnason GA, Christensen JG, Kerbel RS. Accelerated metastasis after short-term treatment with a potent inhibitor of tumor angiogenesis. Cancer Cell 2009; 15: 232-239.
  • 10
    Pàez-Ribes M, Allen E, Hudock J, Takeda T, Okuyama H, Viñals F, et al. Antiangiogenic therapy elicits malignant progression of tumors to increased local invasion and distant metastasis. Cancer Cell 2009; 15: 220-231.
  • 11
    Thomas SL, Zhong D, Zhou W, Malik S, Liotta D, Snyder JP, et al. EF24, a novel curcumin analog, disrupts the microtubule cytoskeleton and inhibits HIF-1. Cell Cycle 2008; 7: 2409-2417.
  • 12
    Liang Y, Yin D, Hou L, Zheng T, Wang J, Meng X, et al. Diphenyl difluoroketone: a potent chemotherapy candidate for human hepatocellular carcinoma. PLoS One 2011; 6: e23908.
  • 13
    Wang J, Ma Y, Jiang H, Zhu H, Liu L, Sun B, et al. Overexpression of von Hippel-Lindau protein synergizes with doxorubicin to suppress hepatocellular carcinoma in mice. J Hepatol 2011; 55: 359-368.
  • 14
    Comerford KM, Wallace TJ, Karhausen J, Louis NA, Montalto MC, Colgan SP. Hypoxia-inducible factor-1-dependent regulation of the multidrug resistance (MDR1) gene. Cancer Res 2002; 62: 3387-3394.
  • 15
    Poller B, Wagenaar E, Tang SC, Schinkel AH. Double-transduced MDCKII cells to study human P-glycoprotein (ABCB1) and breast cancer resistance protein (ABCG2) interplay in drug transport across the blood-brain barrier. Mol Pharm 2011; 8: 571-582.
  • 16
    Rohwer N, Cramer T. Hypoxia-mediated drug resistance: novel insights on the functional interaction of HIFs and cell death pathways. Drug Resist Updat 2011; 14: 191-201.
  • 17
    Foxler DE, Bridge KS, James V, Webb TM, Mee M, Wong SC, et al. The LIMD1 protein bridges an association between the prolyl hydroxylases and VHL to repress HIF-1 activity. Nat Cell Biol 2012; 14: 201-208.
  • 18
    Tanimoto K, Makino Y, Pereira T, Poellinger L. Mechanism of regulation of the hypoxia-inducible factor-1 alpha by the von Hippel-Lindau tumor suppressor protein. EMBO J 2000; 19: 4298-4309.

Supporting Information

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
HEP_26224_sm_SuppFig1.tif2219KSupporting Information
HEP_26224_sm_SuppFig2.tif6922KSupporting Information
HEP_26224_sm_SuppFig3.tif5080KSupporting Information
HEP_26224_sm_SuppFig4.tif1916KSupporting Information
HEP_26224_sm_SuppFig5.tif4265KSupporting Information
HEP_26224_sm_SuppFig6.tif4254KSupporting Information
HEP_26224_sm_SuppFig7.tif3395KSupporting Information
HEP_26224_sm_SuppFig8.tif2542KSupporting Information
HEP_26224_sm_SuppFig9.tif2457KSupporting Information
HEP_26224_sm_SuppFig10.tif2421KSupporting Information
HEP_26224_sm_SuppFig11.tif2654KSupporting Information
HEP_26224_sm_SuppTab1.tif1758KSupporting Information
HEP_26224_sm_SuppInfo.doc136KSupporting Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.