SEARCH

SEARCH BY CITATION

Abstract

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

Cyclophilin B (CypB) performs diverse roles in living cells, but its role in hepatocellular carcinoma (HCC) is largely unclear. To reveal its role in HCC, we investigated the induction of CypB under hypoxia and its functions in tumor cells in vitro and in vivo. Here, we demonstrated that hypoxia-inducible factor 1α (HIF-1α) induces CypB under hypoxia. Interestingly, CypB protected tumor cells, even p53-defective HCC cells, against hypoxia- and cisplatin-induced apoptosis. Furthermore, it regulated the effects of HIF-1α, including those in angiogenesis and glucose metabolism, via a positive feedback loop with HIF-1α. The tumorigenic and chemoresistant effects of CypB were confirmed in vivo using a xenograft model. Finally, we showed that CypB is overexpressed in 78% and 91% of the human HCC and colon cancer tissues, respectively, and its overexpression in these cancers reduced patient survival. Conclusions: These results indicate that CypB induced by hypoxia stimulates the survival of HCC via a positive feedback loop with HIF-1α, indicating that CypB is a novel candidate target for developing chemotherapeutic agents against HCC and colon cancer. (HEPATOLOGY 2011;).

Cyclophilins (Cyps) were discovered as cellular binding proteins for the immunosuppressive drug, cyclosporin A (CsA).1 They help nascent proteins fold properly via peptidyl-prolyl cis-trans isomerase (PPIase) activity. CypB, a Cyp family member, mainly localizes to the endoplasmic reticulum (ER) lumen2 and attenuates ER stress via its PPIase activity.3 Furthermore, CypB is a functional regulator of the hepatitis C virus replication machinery through its interaction with NS5A and NS5B.4-7

Hepatocellular carcinoma (HCC) is a major health problem worldwide and is the most frequently occurring primary cancer of the liver.8 Although its pathophysiology remains to be clearly understood, fundamental liver dysfunction, particularly cirrhosis, is a predisposing factor for the development of HCC.9 Because fibrogenesis during the development of cirrhosis ultimately destroys the normal blood supply of the liver, HCC with cirrhosis has limited blood supply, which, ultimately, leads to local hypoxia. The insufficient blood supply of the rapidly growing tumor tissues also induces hypoxia in the central region of the tumor. Hypoxia within HCC, in turn, activates hypoxia-inducible factor 1 alpha (HIF-1α), which acts as a transcriptional factor for the expression of a variety of essential genes, including those encoding vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (BFGF) in the hypoxic microenvironment.10, 11

HIF-1, which is composed of alpha (HIF-1α) and beta (HIF-1β) subunits, is a master regulator in tumor angiogenesis, growth, resistance to anticancer drugs, and metastasis.12, 13 Although proteasome pathways rapidly degrade HIF-1α under normoxia, this protein is stable under hypoxia, translocates to the nucleus, and binds to hypoxia response elements (HREs) within the promoter of its target genes.14 Reportedly, the activation of many signal pathways, such as the PI3 kinase, Akt, and Ras pathways, enhances HIF-1α synthesis14; however, the mechanism for the transcriptional regulation of HIF-1α messenger RNA (mRNA) remains largely unknown.

Here, we first show that HIF-1α upregulates cyclophilin B (CypB) expression at the transcriptional level and this CypB expression, in turn, up-regulates not only HIF-1α expression at the transcriptional level, but also its transactivity in a positive feedback loop in HCC. Furthermore, we demonstrate that CypB regulates angiogenesis via HIF-1α-mediated VEGF production and protects HCC cells against stresses, including those induced by hypoxia and cisplatin treatment, by using in vitro and in vivo models. We also show that CypB is overexpressed in 78% and 91% of HCC and colon cancer tissues, respectively, by using human tissue microarrays.

Materials and Methods

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

Cell Culture and Hypoxic Condition.

Human HCC cell lines PLC/PRF/5 and Hep3B as well as hepatoblastoma-derived HCC cell line HepG2 were grown in Roswell Parm Memorial Institute medium or Dulbecco's modified Eagle's medium (DMEM), supplemented with 10% fetal bovine serum (FBS), 100 units/mL of penicillin, and 100 μg/mL of streptomycin. Human HCC cell line Huh7 cells were grown in DMEM supplemented with 10% FBS, 100 units/mL penicillin, 100 μg/mL of streptomycin, 1% L-glutamine, and 1× nonessential amino acid (NEAA) miture. Human umbilical vein endothelial cells (HUVECs), isolated from human umbilical cord veins by collagenase treatment, were cultured in M199 supplemented with 20% FBS, 100 units/mL of penicillin, 100 μg/mL of streptomycin, 3 ng/mL of BFGF, and 5 units/mL of heparin. For hypoxic exposure, cells were placed in a hypoxia chamber in an atmosphere consisting of 94.9% N2, 5% CO2, and 0.1% O2.

Real-Time Quantitative Reverse-Transcription Polymerase Chain Reaction Analysis.

Real-time polymerase chain reaction (PCR) amplification was performed by using the SYBR Green PCR Master Mix (Invitrogen/Applied Biosystems, Carlsbad, CA) and the ABI Prism 7300 real-time PCR system (Applied Biosystems), according to the manufacturer's instructions. Calculations, based on the 2math image method,15 were performed by using the following equation: R (ratio) = 2math image. The integrity of the amplified DNA was confirmed by determining the melting temperature. Data are expressed as fold change of the treatment groups in relation to the controls and were normalized to the levels of glyceraldehyde 3-phosphate dehydrogenase (GAPDH). The primer sequences, designed by Primer3 and UCSC In-Silico PCR, were as follows: CypB forward, 5′-AATTCC ATCGTGTAATCAAGGACTT-3′; CypB reverse, 5′-TCTTGACTGTCGTGATGAAGAACT-3′; HIF-1α forward, 5′-TGATGACCAGCAACTTGAGG-3′; HIF-1α reverse, 5′-TGGGGCATGGTAAAAGAAAG-3′; GAPDH forward, 5′-TGACCACAGTCCATGCCAT-3′; GAPDH reverse, 5′-TTCTAGACGGCAGGTCA GGT-3′.

Reactive Oxygen Species Analysis.

Reactive oxygen species (ROS) were measured by using 2′,7′-dichlorfluorescein-diacetate (DCF-DA). Cells were loaded with 10 μM of DCF-DA at 37°C for 30 minutes and resuspended in 1 mL of phosphate-buffered saline. Fluorescence was measured by a flow cytometer. Mean dichlorodihydrofluorescein (DCF) fluorescence intensity was measured with excitation at 488 nm and emission at 525 nm.

Comet Assay.

The assay was conducted as described previously,16 with slight modifications.17, 18 In brief, slides were electrophoresed in alkaline electrophoresis buffer containing 300 mM of NaOH and 1 mM of Na-EDTA (ethylenedimainetetraacetic acid). Slides were then stained with SYBR Gold (Invitrogen) and observed under fluorescence microscopy.

Endothelial Tube Formation Assay.

Cells were transfected with pcDNA3-CypB/WT (wild type) or CypB small interfering RNA (siRNA) and incubated under normoxic or hypoxic conditions in serum-free medium, and conditioned medium were collected for 24 hours. For the tube formation assay, HUVECs were cultured on Matrigel (BD Biosciences, Franklin Lakes, NJ)-coated 12-well plates with M199 medium. Cells were incubated in 250 μL of M199 (without growth factors) containing 100 μL of conditioned medium from HepG2 cells cultured under normoxic or hypoxic conditions. After incubation for 24 hours, morphogenic changes in cells were examined under microscopy and photographed at ×40 magnification. One tube was designated as a three-branch point, and the tube numbers were counted at ×40 magnification. Three independent experiments with triplicate samples were performed.

Animals and HCC Xenografts.

Female athymic BALB nu/nu mice (5-6 weeks old) were purchased from Orient Bio, Inc. (Sungnam, Korea). Animals were placed in a pathogen-free environment and allowed to acclimate for 1 week before being used in the study. The experimental protocol (KHUASP[SE]-10-017) was approved by the Institutional Animal Care and Use Committee of Kyung Hee University (Seoul, Korea).

Huh7 and HepG2 cells (1 × 107) stably transfected with Mock or pcDNA3-CypB/WT were injected subcutaneously into mice (n = 10 mice/group). Mice were then injected intraperitoneally with or without cisplatin at a concentration of 4 mg/kg daily for 6 days. Tumor weights were calculated with the formula (L × l2)/2, where L is the tumor length and l is the tumor width, both of which were measured with a set of calipers.

In Situ Apoptosis Assay.

Tumor tissue samples from mice subjected to different treatments were sectioned by using a cryostat and mounted on silane-coated slides. In situ apoptosis assay was performed by using the DeadEnd colorimetric terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) system (Promega, Madison WI). Positive apoptotic nuclei were stained dark brown.

Tissue Samples.

Formalin-fixed and paraffin-embedded samples of HCC (n = 78) and colon cancer (n = 123) were obtained from patients. No patient had received any form of treatment before surgery. Informed consent was obtained from all patients. The study was approved by the Institutional Review Board of the Catholic University of Korea, College of Medicine (Seoul, Korea).

To construct the tissue microarray block, two pathologists screened the histologic sections and selected areas representative of the tumor cells. Two and one core samples from cancerous and noncancerous areas of each specimen, respectively, were obtained and placed in a new recipient paraffin block by using a commercially available microarray instrument (Beecher Instruments, Sun Prairie, WI).

Immunohistochemistry Analysis.

Detailed methods for immunohistochemistry (IHC) analysis are described in the Supporting Information.

Statistical Analysis.

The results are expressed as the mean ± standard error of the mean (SEM) of at least three independent experiments. The difference between two means was analyzed with the Student's t-test and considered statistically significant when P < 0.05. P values for 5-year survival curves were evaluated by the Kaplan-Meier survival curves using the log-rank test.

Results

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

CypB Upregulation Under Hypoxia in HCC.

To To determine whether CypB would be induced by hypoxia, we subjected human HCC cell lines, such as Huh7, PLC/PRF/5, and Hep3B, as well as hepatoblastoma derived HCC cell line HepG2, to hypoxic conditions for up to 12 hours. HIF-1α protein levels increased rapidly in both cell lines (Fig. 1A). As expected, CypB protein levels also increased after 3 hours of hypoxic exposure and increased until 12 hours (Fig. 1A). We conducted reverse-transcription (RT)-PCR by using Huh7 and HepG2 cells to corroborate these findings. CypB mRNA levels increased substantially in cells exposed to hypoxia for the indicated periods, but the levels did not change under normoxia (Fig. 1B).

thumbnail image

Figure 1. CypB is up-regulated during hypoxia in HCC cells. (A) Western blotting analysis of CypB and HIF-1α in the indicated HCC cell lines. Actin was analyzed as the loading control. (B) RT-PCR analysis of total RNA from Huh7 and HepG2 cells exposed to normoxic or hypoxic conditions for the indicated periods. GAPDH was employed as the loading control. (C) CypB mRNA decay in Huh7 and HepG2 cells. Cells were exposed to hypoxia for 9 hours, then incubated with 5 μg/mL of actinomycin D (Act. D) under normoxic or hypoxic conditions. (D) Real-time qRT-PCR analysis of CypB mRNA expression in Huh7 and HepG2 cells cultured under hypoxic conditions for 0 and 12 hours (left) and exposed to hypoxia and actinomycin D as described in (C) (right). Data are the means ± SEM from six independent experiments. *P < 0.05, compared with normoxia; **P < 0.05 compared with 0 hours in normoxia; ##P < 0.05, compared with 0 hours in hypoxia.

Download figure to PowerPoint

To determine whether CypB mRNA would be induced by mRNA stabilization or transcriptionally by hypoxia, Huh7 and HepG2 cells grown under hypoxia for 9 hours were treated with 5 μg/mL of actinomycin D and transferred to normoxic or hypoxic conditions for another 12 hours. The resultant rate of CypB mRNA decay was similar under both conditions (Fig. 1C). We also observed similar CypB mRNA levels by real-time quantitative RT-PCR (qRT-PCR) (Fig. 1D). Taken together, these results indicated that CypB mRNA induction by hypoxia reflects mRNA synthesis, rather than mRNA stabilization.

Activation of CypB Promoter by HIF-1α.

Because CypB was transcriptionally upregulated under hypoxic conditions, we tested whether HIF-1α would be involved in hypoxia-mediated CypB upregulation. CypB and HIF-1α levels were increased in a dose-dependent manner by HIF-1α inducers CoCl2 and deferoxamine (DFO) in both Huh7 and HepG2 cells (Fig. 2A, left). In addition, they were transiently transfected with pcDNA3-HIF-1α expression vector, which harbored mutant HIF-1α that was not degraded, even under normoxic conditions.19 CypB was induced under normoxic conditions in the transfected cells, compared with the cells transfected with the pcDNA3 empty vector (Mock) (Fig. 2A, right). Furthermore, HIF-1α inhibitors, such as 17-allylamino-17-demethoxygeldanamycin (17-AAG) and deguelin, abrogated hypoxia-mediated CypB induction (Fig. 2B, left). Knockdown of endogenous HIF-1α expression by using HIF-1α siRNA also showed the same results as the inhibitors (Fig. 2B, right).

thumbnail image

Figure 2. CypB up-regulation during hypoxia was mediated by HIF-1α in HCC cells. (A) Western blotting analysis of HIF-1α and CypB in Huh7 and HepG2 cells treated with HIF-1α inducers, such as CoCl2 and DFO (left) and Huh7 and HepG2 cells, transiently transfected with pcDNA3 or pcDNA3-HIF-1α expression vector (right). Actin was analyzed as the loading control. (B) Western blotting analysis of HIF-1α and CypB in Huh7 and HepG2 cells treated with 3 μM of 17-AAG, 1 μM of deguelin, and HIF-1α siRNA for 12 hours. Actin was analyzed as the loading control. (C) Luciferase reporter constructs of the CypB promoter. The four putative HRE sites and their mutated sites are indicated by gray boxes and empty boxes, respectively. (D) Luciferase reporter assay of Huh7 and HepG2 cells transfected with the luciferase reporter constructs shown in (C) and exposed to hypoxia or cotransfected with pcDNA3-HIF-1α. Relative luciferase activities are expressed in comparison with the activity of the pGL3-Basic. Data are the mean ± SEM from six independent experiments. *P < 0.05, compared with pGL3-CypB-150; #P < 0.05, compared with pGL3-CypB-350. (E) EMSA of 10-μg nuclear extracts obtained from Huh7 cells exposed to hypoxia for 12 hours and incubated with 32P-labeled WT or mutated oligonucleotide probes. A 100-fold molar excess of unlabeled WT probe (cold) was used for the competition study. (F) ChIP assay of CypB in Huh7 and HepG2 cells exposed to hypoxia for 12 hours (H) or unexposed (N). Input, amplified CypB from a 1:100 dilution of total input chromatin as the positive control; Immunoglobulin G (IgG), immunoprecipitation with nonspecific IgG as the negative control.

Download figure to PowerPoint

Next, we attempted to ascertain whether the CypB promoter would contain HRE sites. First, bioinformatic analysis of the CypB promoter revealed the presence of four putative consensus HRE sites, located at −43 (HRE1), −135 (HRE2), −266 (HRE3), and −701 base pairs (bp) (HRE4) upstream of the transcriptional initiation site (Fig. 2C). As the HRE sequences include an HIF-1α-binding site (HBS) and HIF1 ancillary sequence (HAS),20 we screened the CypB promoter for both HBS and HAS. We confirmed both HBS and HAS at sites HRE1 and HRE3. We also found a typical consensus HRE site in the CypB promoters of the mouse, rat, and monkey. To further confirm the responses of the four putative consensus HRE sites to hypoxic conditions, we designed several luciferase reporter constructs and conducted luciferase assays. The luciferase assay demonstrated that only the pGL3-CypB-350 construct had a significant increase in luciferase activity under hypoxic conditions, compared with normoxic conditions; pGL3-CypB-350M and pGL3-CypB-150M/350M constructs, containing 5′-AAAG-3′, rather than 5′-CGTG-3′, at the HRE site, exhibited reduced luciferase activity. The empty pGL3 basic plasmid did not alter luciferase activity under either normoxic or hypoxic conditions (Fig. 2D). To confirm the activity of HIF-1α transactivation on the CypB promoter, we overexpressed HIF-1α via transfection of the pcDNA3-HIF-1α expression vector under normoxic conditions. Results were the same as those observed under hypoxic conditions, in terms of luciferase activity (Fig. 2D). We then evaluated the physical interaction between HIF-1α and HRE3 of the CypB promoter via an electrophoretic mobility shift assay (EMSA). As indicated in Fig. 2E, only the WT oligonucleotide incubated with the nuclear extracts from Huh7 cells under hypoxic conditions exhibited a strong, mobility-shifted band, whereas the mutated oligonucleotide and 100-fold excess of cold oligonucleotide yielded noticeably attenuated bands. The nuclear extracts from HepG2 cells also showed similar results (data not shown). To confirm these results, we conducted chromatin immunoprecipitation (ChIP) assays. HIF-1α directly bound to the CypB promoter located at −266 bp (HRE3) (Fig. 2F). These results indicate that CypB is a hypoxia-inducible gene and its expression can be induced transcriptionally by HIF-1α in HCC.

Cell Protection Against Hypoxia-, Cisplatin-, and H2O2-Induced Apoptosis by CypB.

HIF-1α activates the transcription of diverse genes related to growth and survival in solid tumor cells.21 We demonstrated that CypB is directly induced by HIF-1α. Therefore, we hypothesized that CypB is involved in HCC cell growth and survival, particularly in response to cellular stresses induced by hypoxia, cisplatin treatment, and H2O2 treatment. Initially, to examine the direct role of CypB in cell death under these stress conditions, we prepared a stably transfected cell line by transfecting Huh7, PLC/PRF/5, HepG2, and Hep3B cells with Mock and pcDNA-CypB/WT or a CypB knockdown cell line with scrambled siRNA and CypB siRNA. After the hypoxia, cisplatin, and H2O2 treatments, the pcDNA-CypB/WT–transfected cell line evidenced better survival rates than the Mock-transfected cell line, but the CypB siRNA-transfected cell line evidenced markedly reduced cell survival, compared with the scrambled siRNA-transfected cell line (Fig. 3A). Interestingly, the same results were observed in the p53-defective HCC cells, such as Hep3B cells (Fig. 3A). This observation is important, because p53, the key tumor-suppressor gene, is reportedly inactivated by mutation in many cancer cells, and p53 inactivation is one of the main causes of resistance to chemotherapeutic agents in HCC.

thumbnail image

Figure 3. Effects of CypB on hypoxia-, cisplatin-, and H2O2-induced cell death. (A) A (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay of the indicated HCC cell lines transfected with Mock, pcDNA3-CypB/WT, or CypB siRNA. Each transfectant was exposed to hypoxic conditions for 3 days, 60 μM of cisplatin for 24 hours, or 200 μM of H2O2 for 24 hours. Data are the means ± SEM from six independent experiments. *P < 0.05, compared with Mock; #P < 0.05, compared with scrambled siRNA transfection. (B) Flow cytometric analysis for ROS measurement in transfected Huh7 cells after exposure to hypoxic conditions for 48 hours. (C) Western blotting analysis of cleaved PARP and cleaved caspase 3 in Huh7 cells transfected with pcDNA3-CypB/WT (left) or CypB siRNA (right) and exposed to the different treatments. (D) Comet assay of transfected Huh7 cells after cisplatin treatment. Arrowheads represent chromosomal DNA fragmentation as a sign of cisplatin-induced apoptosis. Scale bar, 100 μm. (E) TUNEL assay of transfected Huh7 cells under hypoxia. Arrows indicate TUNEL-positive cells. Scale bar, 20 μm.

Download figure to PowerPoint

Furthermore, recent studies have shown that ROS are induced by hypoxic conditions and stimulate cell death in tumor cells.22, 23 Cisplatin is also known to induce apoptosis via ROS generation.24 Therefore, we measured ROS levels in Mock-, pcDNA3-CypB/WT-, scrambled siRNA-, and CypB siRNA-transfected Huh7 cells after 48 hours of exposure to hypoxia. As anticipated, the highest and lowest levels of ROS were detected in the CypB siRNA-transfected and pcDNA3-CypB/WT-transfected cells, respectively (Fig. 3B). Treatment with cisplatin generated ROS in the CypB siRNA-transfected HepG2 cells, whereas the pcDNA3-CypB/WT-transfected cells did not have significantly increased ROS (data not shown). The same results were observed in the HepG2 cells (Supporting Fig. 1A) Furthermore, assessment of apoptotic markers, such as cleaved poly(ADP-ribose) polymerase (PARP) and cleaved caspase-3, yielded similar results as those shown in Fig. 3A (Fig. 3C; Supporting Fig. 1B). Comet and TUNEL assays showed similar results after cisplatin (Fig. 3D; Supporting Fig. 1C) and hypoxic treatments (Fig. 3E; Supporting Fig. 1D), respectively. Taken together, these findings indicated that CypB protects cells against apoptosis induced by various stresses and renders HCC cells chemoresistant to cisplatin.

CypB Regulation of Transcriptional and Transactivational Activity of HIF-1α via Interaction with STAT3.

CypB contributes to signal transducer and activator of transcription 3 (STAT3) signaling,25 and STAT3 regulates the transcription of HIF-1α.26 Therefore, we tested whether CypB would up-regulate not only the expression level of HIF-1α at the transcriptional level, but also its transactivity. To determine whether CypB would be associated with HIF-1α expression, we used CypB siRNA and CsA as CypB inhibitors. Interestingly, CypB siRNA and CsA treatments under hypoxic conditions reduced HIF-1α expression levels (Fig. 4A). Down-regulation of HIF-1α mRNA expression was verified by real-time qRT-PCR analysis (Fig. 4B). To confirm whether reduced HIF-1α expression would be associated with the interaction between CypB and STAT3, we conducted coimmunoprecipitation experiments with CypB and STAT3. Results revealed that CypB interacted specifically with STAT3 (Fig. 4C; Supporting Fig. 2A).

thumbnail image

Figure 4. CypB-regulated transcription and transactivation of HIF-1α. (A) Western blotting analysis of HIF-1α in Huh7and HepG2 cells cultured for 12 hours under normoxic or hypoxic conditions with CypB-siRNA and 10 μM of CsA. (B) Real-time qRT-PCR analysis of HIF-1α mRNA expression in Huh7 and HepG2 cells cultured for 12 hours under normoxic or hypoxic conditions with CypB-siRNA. Data are the means ± SEM from six independent experiments. *P < 0.05, compared with normoxia; #P < 0.05, compared with scrambled siRNA in hypoxia. (C) Coimmunoprecipitation of CypB with STAT3 in Huh7 cells treated without or with hypoxia for 12 hours, then analyzed with CypB or STAT3 antibody. Western blotting analysis was conducted with the antibodies by using cell lysates (input). (D) Redistribution of CypB in Huh7 cells exposed to normoxic or hypoxic conditions for 12 hours, followed by immunostaining with CypB antibody and fluorescein isothiocyanate secondary antibody. Scale bar, 20 μm. (E) ChIP assay of the STAT3-binding site within the −209-bp region in Huh7 cells exposed to hypoxia for 12 hours (H) or unexposed (N), then analyzed with CypB or STAT3 antibody. Input, amplified HIF-1α from a 1:100 dilution of total input chromatin as the positive control; IgG, immunoprecipitation with nonspecific IgG as the negative control. (F) Luciferase reporter assay of Huh7 cells transiently transfected with CypB siRNA and luciferase reporter gene plasmids pGL3-VEGF-HRE-Luc, pGL3-EPO-HRE-Luc, and pGL3-GLUT1-HRE-Luc or β-gal expression vector pCMV, then treated for an additional 12 hours under hypoxic conditions. Data are the mean ± SEM from six independent experiments. *P < 0.05, compared with scrambled siRNA in hypoxia.

Download figure to PowerPoint

As reported earlier, intracellular CypB is detected principally within the ER lumen.21 To investigate the cellular location of CypB under hypoxic conditions, Huh7 cells with or without exposure to hypoxia were monitored via confocal microscopy. As shown in Fig. 4D, intracellular CypB was detected principally in the ER under normoxic conditions. Interestingly, after incubation under hypoxic conditions, CypB was detected in the nucleus as well as in the ER. The same results were observed in HepG2 cells (Supporting Fig. 2B).

To confirm that CypB participates in the regulation of HIF-1α mRNA, together with STAT3, we performed ChIP assays. We found three putative consensus STAT3-binding sites on the HIF-1α promoter, located at −209, −629, and −726 bp upstream of the transcriptional initiation site, and confirmed that CypB and STAT3 bind to the HIF-1α promoter at −209 bp (Fig. 4E; Supporting Fig. 2C). We found that CypB and STAT3 did not bind to the HIF-1α promoter at −629 and −726 bp (data not shown). Taken together, the results indicated that CypB binds to the HIF-1α promoter via interaction with STAT3.

Next, to determine whether the CypB would control the transactivational activity of HIF-1α, we assessed the effects of CypB on the expression of HIF-1α-specific genes, including VEGF, erythropoietin (EPO), and glucose transporter 1 (GLUT1), via luciferase assays. The hypoxia-dependent induction of the VEGF, EPO, and GLUT1 promoters were suppressed by CypB siRNA (Fig. 4F, Supporting Fig. 2D), compared with that by scrambled siRNA. These observations indicated that CypB regulates not only the expression level of HIF-1α transcriptionally, but also its transactivity via interaction with STAT3.

CypB Regulates Angiogenesis via VEGF Production.

To determine the effect of CypB on tumor progression in HCC, we performed enzyme-linked immunosorbent assay (ELISA) assays to assess VEGF expression and endothelial tube formation in various conditioned media. Overexpression of CypB increased the secretion of VEGF in hypoxia (Fig. 5A; Supporting Fig. 3A) and enhanced endothelial tube formation in the hypoxia-conditioned medium (Fig. 5B; Supporting Fig. 3B). On the other hand, knockdown of CypB decreased the secretion of VEGF in hypoxia (Fig. 5A; Supporting Fig. 3A) and blocked endothelial tube formation in the hypoxia-conditioned medium (Fig. 5B; Supporting Fig. 3B). Taken together, these results indicated that CypB is involved in angiogenesis in HCC.

thumbnail image

Figure 5. Role of CypB in angiogenesis. (A) Hypoxia-induced VEGF production regulated by CypB. VEGF was analyzed in media collected from Huh7 cells using an ELISA kit, and its concentration was calculated according to the curve plotted by standard VEGF. Data are the mean ± SEM from six independent experiments. *P < 0.05, compared with Mock in hypoxia; #P < 0.05, compared with scrambled siRNA in hypoxia. (B) Endothelial tube formation. Huh7 cells were transfected with CypB/WT or CypB specific-siRNA and incubated under normoxia or hypoxia in serum-free medium for 24 hours to collect conditioned medium. HUVECs were incubated in medium containing conditioned medium on Matrigel-coated plates. Data are the means ± SEM. *P < 0.05, compared with Mock in hypoxia; #P < 0.05, compared with scrambled siRNA in hypoxia. Scale bar, 200 μm.

Download figure to PowerPoint

Effect of CypB on Tumorigenesis and Cisplatin Resistance in a Xenograft HCC Model.

To determine the effects of CypB on tumorigenesis and cisplatin resistance in vivo, we injected 1 × 107 Huh7 and HepG2 cells stably transfected with Mock or pcDNA3-CypB/WT in 10 athymic nude mice per group for xenoplantation. Interestingly, mice injected with Huh7 and HepG2 cells transfected with pcDNA3-CypB/WT showed significantly increased tumor growth, compared with those injected with Huh7 and HepG2 cells transfected with Mock (Fig. 6A). Furthermore, after cisplatin treatment, mice injected with Huh7 and HepG2 cells transfected with pcDNA3-CypB/WT showed slightly decreased tumor growth, compared with the untreated group, whereas the mice injected with Huh7 and HepG2 cells transfected with Mock had significantly inhibited tumor growth (Fig. 6A). These results were confirmed by measuring tumor weight (Fig. 6B).

thumbnail image

Figure 6. Effect of CypB on tumor growth and cisplatin resistance in the xenograft HCC model. (A) Tumor volume in athymic mice (n = 10) injected with Huh7 and HepG2 cells transfected with Mock and pGL3-CypB/WT. Huh7 and HepG2 tumor-bearing mice were treated with or without cisplatin (4 mg/kg). (B) Tumor weight in the indicated groups. The data are the mean ± standard deviation. *P < 0.05, compared with Mock; #P < 0.05, compared with Mock+cisplatin. (C) Western blotting analysis of HA-Tag and HIF-1α expressions in the xenografted tumor derived from Huh7 and HepG2 cells. (D) In situ TUNEL assay. TUNEL staining was performed to detect apoptotic cells in the xenografted tumor derived from Huh7 and HepG2 cells. TUNEL-positive cells were counted. Data are the mean ± SEM. *P < 0.05, compared with Mock+cisplatin.

Download figure to PowerPoint

To confirm the overexpression of CypB in the tumor specimens, we performed western blotting analysis. Strong expression of hemagglutinin (HA)-tagged CypB and HIF-1α was observed in the tumor specimens from mice injected with Huh7 and HepG2 cells transfected with pcDNA3-CypB/WT, compared with those injected with Huh7 and HepG2 cells transfected with Mock with or without cisplatin treatment (Fig. 6C). The same specimens were subjected to an in situ apoptosis TUNEL assay. Fewer apoptotic nuclei were noted in the tumor specimens from mice injected with Huh7 and HepG2 cells transfected with pcDNA3-CypB/WT than in those from mice injected with Huh7 and HepG2 cells transfected with Mock after cisplatin treatment (Fig. 6D). Collectively, these data indicate that CypB has a crucial role in HCC cell survival and chemoresistance to cisplatin in vivo.

CypB Overexpression in HCC and Colon Cancer Tissues.

To explore the clinical relevance of CypB, we evaluated its expression levels in human HCC and colon cancer tissues by using IHC analysis. Pathologically confirmed HCC, colon cancer, and corresponding noncancerous tissues were also obtained. HCC and colon cancer tissues showed intense CypB staining, compared with the corresponding noncancerous normal tissues (Fig. 7A,B). We also confirmed CypB upregulation in 7 and 9 of 10 HCC and colon cancer samples, respectively, by western blotting analysis (Fig. 7C,D). Furthermore, in 61 (78%) of the 78 HCC samples and 112 (91%) of the 123 colon cancer samples, strong immunopositivity of CypB was clearly observed (Table 1). The specimens exhibiting ++ immunoreactivity were considered positive. Interestingly, the level of CypB expression was not associated with tumor grade or developmental stage.

thumbnail image

Figure 7. CypB expression in human HCC and colon cancer tissues. (A, B) Representative H&E staining and IHC analysis of CypB expression from normal liver and colon tissues, HCC, and colon cancer tissues. Scale bar, 100 μm. (C, D) Western blotting analysis of CypB expression in normal (N) and HCC and colon cancer (T) tissues. (E, F) Kaplan-Meier survival curve of patients with CypB-negative (CypB [−]) and CypB-positive (CypB[+]) in HCC (E) and colon cancer (F) patients. P < 0.01.

Download figure to PowerPoint

Table 1. Relative Expression Levels of CypB in the HCC and Colon Cancer Tissues Obtained From Patients
Tumor GradeGendernCypB expression
Negative (%) (−)Weak (%) (+)Moderate (%) (++)Strong (%) (+++)Total Positive (%)
  1. The multivariate model was adjusted for tumor grade, gender, and pathologic stage.

  2. Only specimens exhibiting ++ immunoreactivity were considered positive.

HCC tissues7817 (22)  61 (78)
1M70 (0)0 (0)5 (71)2 (29)7 (100)
F30 (0)0 (0)3 (100)0 (0)3 (100)
2M195 (26)0 (0)6 (32)8 (42)14 (74)
F60 (0)1 (17)2 (33)3 (50)5 (83)
3M295 (17)2 (7)13 (45)9 (31)22 (76)
F111 (9)2 (18)5 (46)3 (27)8 (73)
4M30 (0)1 (33)1 (33)1 (33)2 (66)
FN/AN/AN/AN/AN/AN/A
Colon cancer tissues12311 (9)112 (91)
1M90 (0)0 (0)7 (78)2 (22)9 (100)
F60 (0)1 (17)4 (66)1 (17)5 (73)
2M301 (3)0 (0)13 (43)16 (54)29 (97)
F101 (10)1 (10)3 (30)5 (50)8 (80)
3M472 (4)1 (2)23 (49)21 (45)44 (94)
F181 (6)2 (10)10 (56)5 (28)15 (84)
4M30 (0)1 (33)2 (67)0 (0)2 (67)
FN/AN/AN/AN/AN/AN/A

To investigate the association between CypB expression level and 5-year survival, we evaluated HCC and colon cancer patients using the Kaplan-Meier method. We examined survival information of 40 cases of HCC among 78 cases and 123 cases of colon cancer. Unfortunately, we lost survival information for 38 HCC cases, because we got the specimen of HCC patients from multiple hospitals. The Kaplan-Meier survival curve, with a follow-up period of 60 months, demonstrated that patients with lower expression of CypB (CypB [−]) survive significantly longer than those with higher expression of CypB (CypB [+]) in both cancer patients (Fig. 7E,F).

Discussion

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

Currently, the only available treatment for HCC is either surgical resection or liver transplantation. However, as many HCCs involve scattered tumors, they cannot be removed surgically. Therefore, most patients with HCC receive only palliative treatments, including transarterial chemoembolization (TACE), anticancer drugs, and antiangiogenic agents. However, TACE eventually results in hypoxia, leading to HIF-1α activation and thus chemoresistance and radioresistance in HCC. Furthermore, anticancer and antiangiogenic agents are ineffective in patients with HCC because of multidrug resistance, resulting from the induction of diverse factors such as multidrug resistance-associated protein, glutathione, and glutathione S-transferase as well as apoptosis-related genes, including bcl-2, c-myc, p53, and protein kinase C.27-29 Therefore, the development of a more effective treatment would clearly have a tremendous benefit.

We previously demonstrated that CypB is transcriptionally induced by ER stress-related transcription factor ATF6.3 In this study, we noted that the direct transcriptional induction of CypB by HIF-1α. HIF-1α binds to the HRE3 site located at −266 bp upstream of the transcriptional initiation site within the CypB promoter, containing both an HBS and an HAS. Therefore, HIF-1α regulates the induction of CypB during hypoxia. Considering that CypB promoters of the mouse, rat, and monkey also harbor typical consensus HRE sequences, we believe that the transcriptional regulation of CypB by HIF-1α occurs in a variety of mammalian systems. Interestingly, our results reveal that the decay of CypB reduces the mRNA levels of HIF-1α and the expression of a variety of hypoxia-inducible genes, including VEGF, EPO, and GLUT1. Furthermore, we showed that CypB interacts with STAT3, then transactivates the HIF-1α promoter. On the basis of these results, we suggest that CypB regulates the expression level and transactivation of HIF-1α in a positive feedback loop with HIF-1α via interaction with STAT3.

Hypoxia results in the accumulation of ROS from mitochondrial electron transport chain complex III.30, 31 Hypoxia also causes inflammation through the activation of nuclear factor kappa light-chain enhancer of activated B cells, which, in turn, results in profound ROS.32-34 These ROS induce hypoxic cell death. However, hypoxia also stabilizes HIF-1α via ROS-mediated inhibition of prolyl hydroxylase domain-containing proteins and thus prevents apoptosis via the up-regulation of anti-apoptotic molecules.35-38 Furthermore, HIF-1α enhances the proliferation of tumor cells in HCC and increases blood supply by regulating the transcription of several angiogenesis-associated genes, particularly in the hypoxic regions.39 Therefore, HCC is extremely resistant to a variety of therapeutic modalities, including TACE, anticancer drugs, and antiangiogenic agents. Our results demonstrate that CypB protects cells against apoptosis induced by hypoxia, cisplatin, and oxidative stresses. In addition, CypB regulates angiogenesis via the regulation of VEGF production, and overexpressed CypB increases tumor growth and renders resistance to cisplatin in vivo. Because the protective role of CypB was also observed in p53-defective HCC cells, we believe that CypB is a candidate target for effective treatment of chemoresistant tumors, including p53-defective HCC.

In the IHC analysis, CypB was overexpressed in 61 (78%) of the 78 HCC samples and 112 (91%) of the 123 colon cancer samples, suggesting that CypB plays important roles in tumorigenesis in other cancers as well as HCC. Furthermore, CypB overexpression in these cancers reduced patient survival. Interestingly, we could not find any correlation of CypB overexpression with the tumor grade or development. Therefore, we think that CypB expression is mainly regulated by hypoxia and not by tumor invasiveness or metastasis.

In summary, our findings evidence that CypB enhances tumorigenesis. We demonstrated that CypB is induced directly by hypoxia and contributes to the effects of HIF-1α as the result of cross-talk between CypB and HIF-1α. Our results also demonstrated that CypB up-regulation suppresses hypoxia-induced apoptosis, which, consequently, stimulates tumor growth, and that CypB overexpression is a major cause of resistance to anticancer drugs. Therefore, we propose CypB as an important potential target for developing chemotherapeutic agents against HCC and colon cancer.

References

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

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials 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_24539_sm_suppfig1A-B.tif1080KSupporting Figure 1A-B.
HEP_24539_sm_suppfig1C-D.tif1838KSupporting Figure 1C-D.
HEP_24539_sm_suppfig2A-C.tif1487KSupporting Figure 2A-C.
HEP_24539_sm_suppfig2D.tif433KSupporting Figure 2D.
HEP_24539_sm_suppfig3A.tif201KSupporting Figure 3A.
HEP_24539_sm_suppfig3B.tif1944KSupporting Figure 3B.
HEP_24539_sm_suppinfo.doc71KSupporting Information.
HEP_24539_sm_supptable1.doc94KSupporting Table 1.

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.