SEARCH

SEARCH BY CITATION

Abstract

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

Hepatocellular carcinoma (HCC) is a leading cause of cancer death worldwide, yet effective therapeutic options for advanced HCC are limited. This study was aimed at assessing the antitumor effect of a novel phenylbutyrate-derived histone deacetylase (HDAC) inhibitor, OSU-HDAC42, vis-à-vis suberoylanilide hydroxamic acid (SAHA), in in vitro and in vivo models of human HCC. OSU-HDAC42 was several times more potent than SAHA in suppressing the viability of PLC5, Huh7, and Hep3B cells with submicromolar median inhibitory concentration (IC50) values. With respect to SAHA, OSU-HDAC42 exhibited greater apoptogenic potency, which was associated with reduced levels of the apoptotic regulators phosphorylated Akt B-cell lymphoma-xL, survivin, cellular inhibitor of apoptosis protein 1, and cellular inhibitor of apoptosis protein 2. The in vivo efficacy of OSU-HDAC42 versus SAHA was assessed in orthotopic and subcutaneous xenograft tumor models in athymic nude mice. Daily oral treatments with OSU-HDAC42 and SAHA, both at 25 mg/kg, suppressed the growth of orthotopic PLC5 tumor xenografts by 91% and 66%, respectively, and of established subcutaneous PLC5 tumor xenografts by 85% and 56%, respectively. This differential tumor suppression correlated with the modulation of intratumoral biomarkers associated with HDAC inhibition and apoptosis regulation. Moreover, the oral administration of OSU-HDAC42 at 50 mg/kg every other day markedly suppressed ectopic tumor growth in mice bearing large tumor burdens (500 mm3) at the start of treatment. Conclusion: OSU-HDAC42 is a potent, orally bioavailable inhibitor of HDAC with a broad spectrum of antitumor activity that includes targets regulating multiple aspects of cancer cell survival. These results suggest that OSU-HDAC42 has clinical value in therapeutic strategies for HCC. (HEPATOLOGY 2007.)

Hepatocellular carcinoma (HCC) is the fifth most common human cancer and third most frequent cause of cancer death worldwide.1 Although over 80% of new HCC cases will occur in Eastern Asia and sub-Saharan Africa,2 the startling increase in HCC incidence in the United States over the past 20–25 years and the typically high mortality from this disease have established HCC as an important health concern in this country.2–6 The clinical management of HCC is complicated by typically late-stage disease at presentation and prevalent underlying liver dysfunction that can render patients ineligible for potentially curative surgical therapies, which are generally suitable for only 20%–30% of HCC patients.3, 5, 7 Although regional therapies, such as transarterial embolization and percutaneous treatments, are used in patients with nonresectable disease, their success is curtailed by recurrence as locally advanced or metastatic disease. For these patients, systemic therapy is indicated but has been largely unsuccessful.3, 4, 8–10 Thus, a clear need exists to develop effective, life-prolonging therapeutic strategies for the large number of HCC patients with advanced disease.

A major challenge in the systemic treatment of HCC is cellular resistance to conventional cytotoxic agents, which, at least in part, may be attributed to the heterogeneity of the genetic abnormalities acquired during the course of hepatocarcinogenesis, many of which dysregulate signaling pathways governing cell proliferation and survival.11, 12 Consequently, targeting molecular defects that allow HCC cells to evade apoptosis signaling represents a viable strategy to improve patient outcome. Accordingly, a number of small-molecule inhibitors targeting aberrant cellular growth and survival signaling pathways have been investigated for the treatment of HCC.4

Histone deacetylase (HDAC) inhibitors have been the focus of many recent preclinical and clinical investigations because of their ability to induce growth arrest, differentiation, and apoptosis in multiple types of human cancer cells,13–15 including those of the liver.16–24 The accepted model for the anticancer mechanism of HDAC inhibitors has been drug-induced hyperacetylation of core histones leading to chromatin remodeling and reactivated expression of genes regulating cell proliferation, cell cycle progression, and cell survival.13, 14, 25–28 In addition to this histone acetylation–dependent modulation of transcription, histone acetylation–independent mechanisms involving a number of nonhistone HDAC substrates have been implicated in the anticancer activities of HDAC inhibitors,15, 29, 30 many of which participate in important signaling pathways such as nuclear factor κB, signal transducer and activator of transcription 3, p53, and heat shock protein-90.31–37 Moreover, we recently reported HDAC inhibitor–induced disruption of HDAC–protein phosphatase 1 (PP1) complexes leading to the dephosphorylation of Akt (protein kinase B) and subsequent apoptosis in cancer cells.38 These findings reveal a complex mode of action for HDAC inhibitors that likely underlies the high potency of these agents in suppressing tumor growth in vitro and in vivo.30

We have recently developed a novel phenylbutyrate-based HDAC inhibitor, OSU-HDAC42, that possesses potent inhibitory activity against HDAC, cancer cell viability, and prostate tumor xenograft growth in vivo39, 40 (Fig. 1B). In light of the efficacy of HDAC inhibitors in both in vitro and in vivo preclinical models of HCC, the antitumor activity of OSU-HDAC42, in comparison with that of suberoylanilide hydroxamic acid (SAHA), an HDAC inhibitor recently approved for the treatment of cutaneous T cell lymphoma, was evaluated in HCC models in vitro and in vivo. Our findings show that OSU-HDAC42 is a potent inhibitor of HCC cell viability and induces a greater apoptotic response than SAHA. Moreover, in addition to inducing hallmark indicators of HDAC inhibition, OSU-HDAC42 also modulates multiple regulators of cell survival, including Akt, B-cell lymphoma xL (Bcl-xL), and members of the inhibitor of apoptosis protein (IAP) family. Finally, these in vitro results were extended to in vivo orthotopic and subcutaneous xenograft models of HCC tumor growth in which orally administered OSU-HDAC42 suppressed tumor growth and induced intratumoral hyperacetylation of histone and α-tubulin and reductions in Bcl-xL and cellular inhibitor of apoptosis protein 2 (cIAP-2) levels. OSU-HDAC42 is a novel orally bioavailable, phenylbutyrate-based HDAC inhibitor that has significant therapeutic potential for HCC and warrants further investigation in this regard.

thumbnail image

Figure 1. Antiproliferative effects of OSU-HDAC42 and SAHA in 3 HCC cell lines and normal hepatocytes. (A) Differential expression of HDAC isozymes in PLC5, Huh7, and Hep3B cells. Lysates were prepared from HCC cells cultured in 10% FBS–containing DMEM and then immunoblotted as described in the Materials and Methods section. (B) Dose-dependent effects of OSU-HDAC42 (left panel) and SAHA (right panel) on the viability of PLC5, Huh7, and Hep3B cells in comparison with normal human hepatocytes. The chemical structures of the agents are shown above the respective panels. The cells were treated at the indicated concentrations in a 10% FBS–supplemented medium for 72 hours. The cell viability was assessed with MTT assays. The points indicate means, and the bars indicate standard deviations (n = 6).

Download figure to PowerPoint

Materials and Methods

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

Reagents.

The HDAC inhibitors SAHA and OSU-HDAC42 {also known as (S)-(+)-N-hydroxy-4-(3-methyl-2-phenyl-butyrylamino)benzamide [(S)-HDAC-42]} were synthesized in our laboratory with purities exceeding 99% as shown by nuclear magnetic resonance spectroscopy (300 MHz). OSU-HDAC42 (NSC 736012) is a novel hydroxamate-tethered phenylbutyrate derivative39 that is currently undergoing preclinical evaluation. For in vitro studies, stock solutions of inhibitors were prepared in dimethyl sulfoxide (DMSO) and diluted in a 10% serum–containing culture medium for the treatment of cells (final concentration of DMSO < 0.1%). For in vivo studies, OSU-HDAC42 and SAHA were prepared as suspensions in a vehicle [0.5% methylcellulose (wt/vol) and 0.1% Tween 80 (vol/vol) in sterile water] for oral administration to xenograft-bearing athymic nude mice. The target proteins for the rabbit polyclonal antibodies used in the study and their commercial sources were as follows: Akt, p-473Ser-Akt, Bcl-xL, caspase 9, poly(ADP-ribose) polymerase (PARP), and various HDAC isozymes from Cell Signaling Technology Inc. (Beverly, MA); cellular inhibitor of apoptosis protein 1 (cIAP-1) and survivin from R&D Systems Inc. (Minneapolis, MN); cIAP-2 from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA); and acetylated histone H3 and phosphatase and tensin homolog (PTEN) from Upstate Biotechnology Inc. (Lake Placid, NY). Mouse monoclonal antibodies against p21 (Santa Cruz), β-actin (ICN Biomedicals, Irvine, CA), and acetylated α-tubulin (Sigma-Aldrich, St. Louis, MO) were also used.

Cell Culture.

The Huh7 HCC cell line was obtained from the Health Science Research Resources Bank (Osaka, Japan; JCRB0403). The Hep3B and PLC5 cell lines were obtained from the American Type Culture Collection (Manassas, VA). These 3 cell lines vary with respect to their p53 functional status. Huh7 cells express a mutant p53 protein with a longer half-life as a result of a point mutation at codon 220, whereas Hep3B and PLC5 cells lack functional p53 because of a deletion and a mutation at codon 249, respectively. In addition, these cell lines are characterized by high expression levels of Bcl-xL but low levels of B-cell lymphoma 2.41, 42 The HCC cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Life Technologies, Grand Island, NY) supplemented with 10% fetal bovine serum (FBS; Life Technologies). Normal human hepatocytes were obtained from Cambrex Bioscience-Walkersville, Inc. (Walkersville, MD) and maintained in the defined hepatocyte culture medium according to the vendor's recommendation. All cell types were cultured at 37°C in a humidified incubator containing 5% CO2. Cells in log phase growth were harvested by trypsinization for use in various assays and in vivo studies.

Cell Viability Assay.

The cell viability was assessed with a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Tokyo Chemical Industry Inc., Tokyo, Japan) assay in 6 replicates. Cancer cells and normal human hepatocytes were seeded at 4000 cells per well in 96-well, flat-bottom plates. After 24-72 hours, the cells were treated with HDAC inhibitors at the indicated concentrations in 10% FBS–supplemented DMEM or a hepatocyte culture medium. The control cells received DMSO at a concentration equal to that of the drug-treated cells. At the indicated intervals, one-fifth of the volume of 5× MTT (2.5 mg/mL) was added to each well, and the cells were incubated at 37°C for 2 hours. The medium was removed, and the reduced MTT dye was solubilized in DMSO (200 μL/well). The absorbances were determined at 570 nm.

Flow Cytometry/Apoptosis Assays.

Approximately 5 × 105 cells were plated onto 10-cm dishes and incubated at 37°C overnight. The cells were treated with DMSO or OSU-HDAC42 at different concentrations for 48 hours, collected by trypsinization, washed twice with phosphate-buffered saline, and then fixed in ice-cold ethanol at −20°C overnight. After centrifugation at 400g for 5 minutes at room temperature, the cells were stained with propidium iodide (50 μg/mL) in the presence of ribonuclease A (100 U/mL). The cell cycle phase distributions were determined on a FACSort flow cytometer and analyzed with ModFitLT V3.0 software.

Immunoblotting.

Lysates of Hep3B, PLC5, and Huh7 cells treated with HDAC inhibitors at the indicated concentrations for 72 hours were prepared for the immunoblotting of acetylated histone H3, acetylated α-tubulin, β-actin, caspase 9, PARP, Bcl-xL, Bax, survivin, cIAP-1, cIAP-2, and p21. A western blot analysis was performed as previously reported.40 For the immunoblotting of biomarkers in PLC5 tumor xenografts, tumor tissue homogenates were prepared, and immunoblotting was performed as described.40

Semiquantitative Reverse-Transcription Polymerase Chain Reaction (RT-PCR).

The total RNA was isolated from HCC cells with TRIZOL reagent (Invitrogen, Carlsbad, CA) and chloroform extraction. The first-strand complementary DNA (cDNA) was synthesized with the iScript cDNA synthesis kit (Bio-Rad, Hercules, CA) according to the manufacturer's protocol. Reverse transcription and polymerase chain reaction (PCR) were performed in a Peltier thermal cycler (PTC-200, MJ Research, Waltham, MA) with the following primers: p21 gene, 5′-tcaccgagacaccactggag-3′/5′-tggagtggtagaaatctgtc-3′, and β-actin gene, 5′-acactgtgcccatctacgagg-3′/5′-aggggccggactcgtcatact-3′.

Chromatin Immunoprecipitation (ChIP) Assay.

PLC-5 cells were plated into 100-mm–diameter culture dishes for overnight incubation, and this was followed by a treatment with 1 μM OSU-HDAC42 for 12 hours. The ChIP assay was performed with the EZ CHIP kit (Millipore-Upstate, Billerica, MA) according to the supplier's protocol. Briefly, treated cells were fixed in a serum-free medium containing 1% formaldehyde for 10 minutes at room temperature to immobilize histone on DNA. The cells were then harvested, lysed, and sonicated to shred genomic DNA into fragments of 200–1000 base pairs. An aliquot of the chromatin preparation was saved as input DNA. The remaining chromatin preparation was immunoprecipitated with anti–acetyl histone H3 antibody (Millipore-Upstate, Billerica, MA), and this was followed by DNA purification. An immunoprecipitation step using nonspecific immunoglobulin G was performed in parallel as a negative control. Both immunoprecipitated and input DNA samples were analyzed by PCR to determine the relative amounts of DNA from the p21 gene promoter region present in the samples. Two primer pairs were used for amplification by PCR: p1, 5′-ctgtctgcaccttcgctcct-3′/5′-cgtggtggtggtgagctaga-3′, and p2, 5′-ggtgtctaggtgctccaggt-3′/5′-gcactctccaggaggacaca-3′.43

Transfection.

In order to facilitate serial assessments of tumor engraftment and growth in an orthotopic xenograft tumor model by bioluminescent imaging, luciferase-expressing PLC5 (PLC5-luc) cells were generated by stable transfection with the luciferase-expressing phCMV plasmid containing cDNA encoding the firefly luciferase gene. PLC5 cells were transfected with Lipofectamine 2000 (Life Technologies, Inc., Gaithersburg, MD) according to the manufacturer's protocol. Stable transfectants were selected with a medium containing 700 μg/mL geneticin (Invitrogen-Gibco, Carlsbad, CA).

In Vivo Studies.

Female NCr athymic nude mice (5–7 weeks of age) were obtained from the National Cancer Institute (Frederick, MD) for the assessment of OSU-HDAC42 in both orthotopic and ectopic xenograft tumor models of HCC. All HDAC inhibitor treatments were administered to mice orally by gavage (10 μL/g body weight) for the duration of the study. Body weights were measured weekly. In both models, tumors were harvested at sacrifice. One half of each tumor was snap-frozen in dry ice and stored at −80°C until it was used for western blot analysis of the relevant biomarkers, and the other half was fixed in 10% neutral buffered formalin and processed for histopathological and immunohistochemical evaluations. All mice were group-housed under conditions of a constant photoperiod (12 hours of light and 12 hours of darkness) with ad libitum access to sterilized food and water. All experimental procedures using these mice were performed in accordance with protocols approved by The Ohio State University Institutional Animal Care and Use Committee.

Orthotopic Xenograft Tumor Model.

Orthotopic tumors were established by the direct intrahepatic injection of PLC5-luc cells with a previously published procedure with slight modifications.44 Briefly, the left hepatic lobe of each mouse under isoflurane anesthesia was exposed through a small (8–10–mm) transverse incision made in the left cranial abdomen beginning 2–3 mm below the xyphoid process and extending to the left. One million PLC5-luc cells, in a total volume of 0.02 mL of a serum-free medium containing 50% Matrigel (BD Biosciences, Bedford, MA), were slowly injected into the liver through the diaphragmatic surface with a 28-gauge needle. The needle was inserted at a shallow angle so that a gentle injection produced a visible, translucent, subcapsular bleb on the liver surface. After injection, a sterile cotton swab was placed on the needle insertion site as the needle was withdrawn, and gentle pressure was applied for 1 minute to ensure hemostasis. The abdominal musculature and skin were then closed with absorbable suture material and sterile surgical clips, respectively. The establishment and growth of tumors were monitored by bioluminescent imaging with the IVIS imaging system (Xenogen Corp., Alameda, CA) according to a protocol similar to those described.45–48 Briefly, mice were administered firefly luciferin through intraperitoneal injection (100 μL/mouse). Approximately 5 minutes later, while under isoflurane anesthesia, mice were imaged with a highly sensitive, cooled charged-coupled device camera in the light-tight specimen chamber of the IVIS system. Data acquisition and analysis were achieved with Living Image software (Xenogen).

One week after the tumor cell injections, mice with established intrahepatic tumors, as confirmed by bioluminescent imaging, were randomized into 3 groups (n = 6) that received the following treatments daily for 21 days: (1) the methylcellulose/Tween 80 vehicle, (2) OSU-HDAC42 at 25 mg/kg of body weight, and (3) SAHA at 25 mg/kg of body weight. Mice were imaged weekly during the course of the treatments. The tumor size was also assessed after the removal of the livers at the conclusion of the study by measurement with calipers, and the volumes was calculated with a standard formula: width2 × length × 0.52.

Subcutaneous Ectopic Xenograft Tumor Model.

The efficacy of OSU-HDAC42 was tested in mice bearing low and high ectopic tumor burdens. For the low tumor burden group, each mouse received a subcutaneous injection containing 1 × 106 PLC5 cells in a total volume of 0.1 mL of a serum-free medium containing 50% Matrigel (BD Biosciences) under isoflurane anesthesia. As the tumors became established (mean starting tumor volume = 47 ± 7 mm3), the mice were randomized into 3 groups (n = 5–8) that received the following treatments daily: (1) the methylcellulose/Tween 80 vehicle, (2) OSU-HDAC42 at 25 mg/kg of body weight, and (3) SAHA at 25 mg/kg of body weight. For the high tumor burden group, each mouse was inoculated with 1 × 106 PLC5 cells in a total volume of 0.1 mL of a serum-free medium containing 50% Matrigel (BD Biosciences). As the tumors reached a volume of approximately 500 mm3, the mice were randomized into 2 groups (n = 3) that received the methylcellulose/Tween 80 vehicle or OSU-HDAC42 at 50 mg/kg of body weight every other day. The tumors were measured weekly with calipers, and their volumes were calculated with a standard formula: width2 × length × 0.52.

Histological Evaluations.

After fixation for up to 48 hours, tumor tissues were embedded in paraffin blocks according to routine procedures. Five-micrometer sections were cut and stained with hematoxylin-eosin for histopathological evaluation. Apoptotic cells were detected in representative sections of tumor tissue through staining with the terminal deoxynucleotidyl transferase biotin-dUTP nick-end labeling (TUNEL) technique according to the manufacturer's instructions (Chemicon, Temecula, CA). The expression of acetylated histone H3 was also examined by the immunohistochemical staining of representative sections of tumor tissues using a specific antibody and the EnVision Plus staining kit (Dako, Carpinteria, CA).

Statistical Analysis.

The tumor volume data met the assumptions of normality and homogeneity of variance for parametric analysis; thus, group means at 21 days of treatment were compared with a one-way analysis of variance followed by Fisher's least significant difference method for multiple comparisons. Tumor growth data are expressed as the mean tumor volumes ± the standard error, and in vitro data are expressed as the mean ± the standard deviation. Differences were considered significant at P < 0.05. Statistical analysis was performed with SPSS for Windows (SPSS Inc., Chicago, IL).

Results

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

OSU-HDAC42 Induces Apoptosis in Different HCC Cell Lines Regardless of Differential HDAC Isozyme Expression.

The antitumor effects of OSU-HDAC42 in relation to SAHA were assessed in 3 human HCC cell lines: PLC5, Huh7, and Hep3B. These 3 cell lines are known to be resistant to cytotoxic drugs because of a lack of functional p53 and/or expression of high levels of Bcl-xL.41, 42 In this study, a western blot analysis indicated distinct expression profiles of HDAC isozymes among these cell lines (Fig. 1A). Of the 6 isozymes examined, HDAC1 was highly expressed in all cell lines evaluated, whereas other isozymes, especially HDAC3 and HDAC5, were differentially expressed among these cells. Despite these differences, PLC5, Huh7, and Hep3B exhibited comparable susceptibility to the antiproliferative effects of each of the HDAC inhibitors tested. OSU-HDAC42 was several times more potent than SAHA in suppressing cell viability in these 3 cell lines, with submicromolar IC50 values after 72 hours of treatment (PLC5, 0.72 ± 0.04 μM; Huh7, 0.58 ± 0.03 μM; Hep3B, 0.71 ± 0.04 μM) in comparison with 2–4 μM for SAHA (Fig. 1B). An evaluation of the effects on nonmalignant cells showed that normal hepatocytes were 9–12 times less sensitive to OSU-HDAC42–mediated suppression of cell viability (IC50 = 6.82 + 0.08 μM; Fig. 1B).

Our results suggest that this antiproliferative activity of OSU-HDAC42 was, at least in part, attributable to apoptosis induction. A western blot analysis of drug-treated PLC5 and Hep3B cells revealed dose-dependent increases in caspase 9 activation and PARP cleavage (Fig. 2A). As shown, these changes in apoptotic biomarkers were markedly higher in OSU-HDAC42–treated cells than those observed after SAHA treatment, and this is consistent with the greater antiproliferative potency of OSU-HDAC42 described previously. To assess apoptosis quantitatively, sub-2N DNA contents were measured by a flow cytometric analysis of propidium iodide–stained nuclei. The exposure of PLC5 to OSU-HDAC42 for 48 hours led to a dose-dependent increase in apoptotic cells (sub-G1), whereas the same treatment caused both G2/M arrest and apoptosis in Hep3B cells (Fig. 2B). This finding suggests differences between these 2 cell lines in the signaling mechanisms that regulate cell cycle arrest and apoptosis in response to OSU-HDAC42.

thumbnail image

Figure 2. Apoptosis-inducing effects of OSU-HDAC42 and SAHA in PLC5 and Hep3B cells. (A) Western blot analysis of the dose-dependent effects of OSU-HDAC42 and SAHA on biomarkers associated with HDAC inhibition (histone H3 acetylation) and apoptosis (caspase 9 activation and PARP cleavage) in PLC5 and Hep3B cells after 48 hours of exposure to inhibitors at the indicated concentrations in 10% FBS supplemented DMEM. (B) Flow cytometric analysis of apoptosis in the PCL5 and Hep3B cell lines after the treatment with the DMSO vehicle or the indicated concentrations of OSU-HDAC42 for 48 hours. The drug-treated cells were fixed and stained with propidium iodide. The extent of apoptosis was assessed by the quantification of sub-2N (sub-G1) DNA by flow cytometry. The histograms are representative of 2 independent experiments. Ac-H3 indicates acetylated histone H3; H3B, Hep3B; and HDAC42, OSU-HDAC42.

Download figure to PowerPoint

OSU-HDAC42 Facilitates Apoptosis by Targeting Signaling Pathways Mediated by Akt, Bcl-xL, and IAP Family Members.

To examine the relationship between OSU-HDAC42–induced apoptosis and HDAC inhibition and survival signaling, changes in the relevant biomarkers over time in drug-treated cells were assessed by western blot analysis. As shown in Fig. 3A, OSU-HDAC42 at 1 μM induced significant apoptosis after 48 hours of treatment in PLC5, Hep3B, and Huh7 cells as determined by PARP cleavage. This apoptotic event, however, was preceded by changes in a multitude of biomarkers associated with HDAC inhibition and survival pathways 24 hours after the treatment. Hallmark features associated with intracellular HDAC inhibition, including up-regulated expression of the cyclin-dependent kinase inhibitor p21 and hyperacetylation of histone H3 and α-tubulin, were induced by OSU-HDAC42 in all 3 cell lines, albeit to various degrees. This differential effect of OSU-HDAC42 on histone H3 acetylation status across the cell lines was paralleled by the changes induced in p21 protein expression levels. Semiquantitative RT-PCR revealed that this OSU-HDAC42–induced increase in p21 protein expression in PLC-5 cells was paralleled by effects at the transcriptional level as steady-state levels of p21 messenger RNA (mRNA) were also elevated after the drug treatment (Fig. 3B, upper panel). To examine the effect of OSU-HDAC42 on the accumulation of acetylated histone H3 in chromatin associated with the p21 gene promoter, ChIP was performed on PLC-5 cells treated with a 1 μM concentration of the drug for 12 hours. As shown in Fig. 3B (lower panel), using 2 different primer sets for the amplification of different regions of the p21 promoter, we found that the treatment with OSU-HDAC42 increased the amount of p21 promoter DNA that was associated with hyperacetylated histone H3 in comparison with the same regions isolated from untreated cells. These findings provide direct evidence of the involvement of HDAC inhibition in the observed increase in p21 expression.

thumbnail image

Figure 3. Effects of OSU-HDAC42 on biomarkers of apoptosis (PARP cleavage), HDAC inhibition (histone H3 and α-tubulin acetylation, p21 expression), and survival signaling (phospho-Akt, Bcl-xL, and IAP family members survivin, cIAP-1, and cIAP2) in PLC5, Hep3B, and Huh7 cells. (A) Time-dependent effects of OSU-HDAC42 on the indicated biomarkers in the HCC cell lines. The cells were treated with 1 μM OSU-HDAC42 for the indicated times in 10% FBS–containing DMEM, and cell lysates were immunoblotted as described in the Materials and Methods section. (B) Involvement of HDAC inhibition in the induction of p21 mRNA expression by OSU-HDAC42. The upper panel shows the RT-PCR analysis of the time-dependent effects of OSU-HDAC42 on p21 mRNA levels. The PLC5 cells were treated with 1 μM OSU-HDAC42 for the indicated times in 10% FBS–containing DMEM. The isolated total RNA was subjected to RT-PCR analysis as described in the Materials and Methods section. The lower panel shows ChIP assay results for the accumulation of acetylated histone H3 in chromatin associated with the p21 gene promoter in OSU-HDAC42–treated HCC cells. Chromatin was isolated from PLC5 cells treated with or without 1 μM OSU-HDAC42 for 12 hours and then immunoprecipitated with the anti–acetyl histone H3 antibody. Two PCR primer sets for different regions of the p21 gene promoter were used to amplify DNA isolated from the immunoprecipitated chromatin as described in the Materials and Methods section. (C) Endogenous expression levels of PTEN and p-Akt in Huh7, PLC5, and Hep3B cells. The HCC cells were cultured in 10% FBS–containing DMEM, and the cell lysates were immunoblotted for PTEN, p-Akt, and Akt as described in the Materials and Methods section. PC-3 prostate cancer cells, which are PTEN-null, were used as negative controls. Ac-H3 indicates acetylated histone H3; anti–Ac-H3, antibody against acetylated histone H3; p-Akt, phospho-473Ser-Akt; p1, primer set 1; and p2, primer set 2.

Download figure to PowerPoint

Previously, we reported that HDAC inhibitors mediated antiproliferative effects through both histone acetylation–dependent and histone acetylation–independent mechanisms. Specifically, OSU-HDAC42 caused Akt dephosphorylation in PTEN-defective U87MG glioblastoma and PC-3 prostate cancer cells by disrupting complex formation between HDAC and PP1, leading to increased PP1-Akt association.38 As the HCC cell lines used here contained functional PTEN (Fig. 3C), they exhibited undetectable (PLC5) or low (Hep3B and Huh7) levels of Akt phosphorylation (Fig. 3A). Nevertheless, the treatment of Hep3B and Huh7 with 1 μM OSU-HDAC42 led to observable reductions in phospho-Akt within 24 hours of treatment and its complete disappearance after 48 hours. It is noteworthy that this Akt deactivation was accompanied by the concomitant down-regulation of Bcl-xL and IAP family members survivin, cIAP-1, and cIAP-2. Bcl-xL and the IAP family of proteins have been implicated in oncogenesis, cancer progression, and therapeutic resistance, at least in part through their well-known antiapoptotic functions of maintaining mitochondrial integrity and inhibiting caspase activity. Together, these findings suggest that the ability of OSU-HDAC42 to block survival signaling pathways at multiple levels underlies its high potency in triggering apoptotic death in HCC cells.

OSU-HDAC42 Suppresses Orthotopic HCC Xenograft Growth In Vivo.

In order to assess the in vivo efficacy of OSU-HDAC42 in HCC, the agent was tested in both orthotopic and ectopic tumor xenograft models. In order to facilitate assessments of the engraftment and growth of intrahepatic tumors in the orthotopic HCC xenograft model, the PLC5-luc cells were used to establish intrahepatic tumors. Athymic nude mice bearing established orthotopic PLC5-luc tumors, as confirmed by bioluminescent imaging, were treated orally for 21 days with OSU-HDAC42 or SAHA at 25 mg/kg daily or with the vehicle (n = 6). As shown in Fig. 4A, the growth of orthotopic PLC5-luc tumors in vehicle-treated mice exhibited a positive correlation with time, giving rise to a 4000-fold increase in bioluminescence after 21 days. However, the treatment of PLC5-luc tumor-bearing mice with either HDAC inhibitor resulted in significantly higher bioluminescent intensities, presumably because of the drugs' epigenetic effects on the transcriptional activation of the luciferase gene, rendering this type of imaging inapplicable for monitoring tumor growth in HDAC inhibitor–treated animals (data not shown). Thus, the effects of the agents on tumor growth were evaluated by direct measurements of the tumors at the end of the study.

thumbnail image

Figure 4. Effects of oral OSU-HDAC42 and SAHA, each at 25 mg/kg per day, on orthotopic PLC5 tumors in athymic nude mice. Mice bearing established intrahepatic, PLC5-luc tumor xenografts were randomized into 3 groups (n = 6) that received treatments by gavage for the duration of the study as described in the Materials and Methods section. (A) Bioluminescent imaging of the time course of orthotopic PLC5-luc tumor growth in a single mouse from the vehicle-treated group. The anesthetized mouse was positioned in dorsal recumbency and showed increasing bioluminescence, which was indicative of PLC5-luc tumor growth within the liver, over time. (B) Orthotopic PLC5-luc tumor size after 21 days of treatment with oral OSU-HDAC42 or SAHA, each at 25 mg/kg/day, with respect to that after the vehicle treatment. The upper panel shows photographs of livers dissected from representative mice treated with the vehicle (left), OSU-HDAC42 (middle), or SAHA (right). In the vehicle-treated liver, a very large lobulated tumor mass occupied the entire left lobe and was accompanied by atrophy of the right lobe. In the OSU-HDAC42–treated mouse, a small tumor nodule on the left hepatic lobe was noted, whereas a slightly larger mass was observed in the left lobe from the SAHA-treated mouse. The lower panel shows the mean tumor volumes for each treatment group after 21 days of treatment. The columns indicate the means, and the bars indicate the standard error. (C) Histological features of representative orthotopic PLC5-luc tumors from mice in each of the treatment groups. Five-micrometer sections of paraffin-embedded tumors harvested at the end of the study were hematoxylin-eosin–stained and examined with light microscopy. The arrows indicate sinusoid-like capillary formation. The boxes enclose nonmalignant liver tissue adjacent to the xenograft tumor. The original magnification was ×100. (D) Western blot analysis of intratumoral biomarkers of drug activity in the homogenates of 3–4 representative PLC5-luc tumors from each treatment group and their respective final volumes. The examined biomarkers include indicators of HDAC inhibition (acetylation status of histone H3 and α-tubulin) and expression levels of cell survival regulators (Bcl-xL and cIAP-1). Ac-H3 indicates acetylated histone H3, and HDAC42, OSU-HDAC42.

Download figure to PowerPoint

As shown in Fig. 4B, PLC5-luc tumor volumes after 21 days of treatment with OSU-HDAC42 and SAHA were reduced by 91% and 66%, respectively, with respect to those of vehicle-treated controls (P < 0.01). The treatment with OSU-HDAC42 was well tolerated by the mice, which did not exhibit overt signs of toxicity, significant loss of body weight, or abnormalities in serum chemistry variables in comparison with the vehicle-treated group. Moreover, although 2 mice in the control group and 1 mouse in the SAHA-treated group experienced tumor-related intra-abdominal bleeding reminiscent of the notorious complications encountered in human HCC cases, none of the OSU-HDAC42–treated mice exhibited this problem. Histological differences between drug-treated and control tumors were also clearly evident (Fig. 4C). Tumors from vehicle-treated mice (left panel) were characterized by prominent basophilic cytoplasm, enlarged nuclei, and frequent mitotic figures indicating rapid cell proliferation, which was accompanied by abundant sinusoid-like capillary formation and compression of the adjacent nonmalignant liver tissue. In contrast, the tumors from OSU-HDAC42–treated animals (center panel) contained cells that were smaller than those in the vehicle-treated tumors and exhibited distinctly fewer capillary structures. Mild compression of adjacent nonmalignant liver tissue was observed in SAHA-treated animals (right panel), whereas that in OSU-HDAC42–treated animals was apparently unchanged.

To correlate the tumor-suppressive response observed in vivo with mechanisms identified in vitro, the effects of HDAC inhibitors on representative intratumoral biomarkers of drug activity were evaluated through the immunoblotting of PLC5-luc tumor homogenates collected after 21 days of treatment. The treatments with OSU-HDAC42 and SAHA increased the acetylation levels of histone H3 and α-tubulin in PLC5-luc tumors with respect to those of the vehicle-treated controls, thereby confirming HDAC inhibition in vivo (Fig. 4D). Down-regulation of intratumoral Bcl-xL and cIAP-2 was also observed after treatment with either HDAC inhibitor; however, OSU-HDAC42 induced greater reductions than SAHA. This differential effect on Bcl-xL and cIAP-2 reflected the relative in vivo potencies of these drugs in tumor suppression.

OSU-HDAC42 Suppresses Subcutaneous HCC Xenograft Growth In Vivo.

In the orthotopic xenograft tumor model, serial assessments of drug-mediated suppression of PLC5-luc tumor growth through bioluminescent imaging were prevented by the transactivation of the luciferase gene by HDAC inhibitors. Thus, to further assess the antitumor potential of OSU-HDAC42, athymic nude mice bearing established subcutaneous PLC5 tumor xenografts (mean tumor volume ± standard error = 47 ± 7 mm3) were treated orally for 21 days with 25 mg/kg OSU-HDAC42 or SAHA or with the vehicle daily. As shown in Fig. 5A, OSU-HDAC42 exhibited a quick onset in suppressing PCL5 tumor growth, with 79% and 85% inhibition after 14 and 21 days of treatment, respectively, in comparison with the vehicle-treated controls (P < 0.01). The nearly complete remission in 1 of 6 OSU-HDAC42–treated mice was reflected microscopically as multifocal liquefactive necrosis with replacement by fibrous tissue (Fig. 5B). In contrast, the suppressive effect of SAHA did not achieve statistical significance until day 21, when tumor growth was reduced by 56% (P < 0.01). Despite the presumed differences in the pharmacokinetics of drug therapy between intrahepatic and peripheral tissues, the in vivo efficacy data for OSU-HDAC42 and SAHA in the 2 xenograft models were consistent.

thumbnail image

Figure 5. Effects of oral OSU-HDAC42 and SAHA on subcutaneous PLC5 tumors in athymic nude mice bearing (A-D) low or (E) high tumor burdens. Athymic nude mice bearing established subcutaneous PLC5 tumors (mean tumor volumes of approximately 50 and 500 mm3 for low and high tumor burdens, respectively) were randomized into groups that received treatments by gavage for 21 days as described in the Materials and Methods section. (A) Mean tumor volumes as a function of the day of treatment for groups with low tumor burdens receiving oral OSU-HDAC42 or SAHA, each at 25 mg/kg per day, and the vehicle (n = 5–8). The points indicate means, and the bars indicate the standard error. *P < 0.05. (B) Histological features of a PLC5 tumor that underwent nearly complete regression after the OSU-HDAC42 treatment. The hematoxylin-eosin–stained section was characterized by multifocal liquefactive necrosis with replacement by fibrous tissue. (C) Differential effects of OSU-HDAC42 versus SAHA on histone H3 acetylation and apoptosis induction in PLC5 tumors. The upper panels show immunohistochemical staining of paraffin-embedded tumor sections with anti–Ac-H3 antibodies. Nuclear immunostaining for Ac-H3 was present in tumors from mice treated with either OSU-HDAC42 or SAHA. OSU-HDAC42 induced the strongest Ac-H3 immunopositivity. For the lower panels, a TUNEL assay was used to assess apoptosis in paraffin-embedded tumor sections. Tumors from OSU-HDAC42–treated mice showed a larger proportion of cells with TUNEL-positive nuclei than tumors from vehicle-treated or SAHA-treated mice. The original magnification was ×200; counterstaining was performed with hematoxylin. (D) Western blot analysis of intratumoral biomarkers of drug activity in the homogenates of 3 representative PLC5 tumors from each treatment group and their respective final volumes. The assessed biomarkers were the same as those examined in orthotopic PLC5 tumors (Fig. 4). (E) Mean tumor volumes as a function of the day of treatment for groups with high tumor burdens receiving oral OSU-HDAC42 at 50 mg/kg every other day or the vehicle (n = 3) as described in the Materials and Methods section. The points indicate the means, and the bars indicate the standard error. Ac-H3 indicates acetylated histone H3, and HDAC42, OSU-HDAC42.

Download figure to PowerPoint

To examine whether the differential tumor suppressive potencies of OSU-HDAC42 and SAHA were reflective of their respective activities on HDAC and survival signaling, relevant biomarkers were assessed in xenograft tumors by immunohistochemistry, TUNEL assay, and western immunoblotting. As shown in Fig. 5C, immunohistochemical staining for acetylated histone H3 revealed stronger nuclear immunopositivity in PLC5 xenograft tumors from mice treated with OSU-HDAC42 than in those from SAHA-treated mice. Similarly, the treatment with OSU-HDAC42 resulted in a greater apoptotic fraction in the PLC5 tumors than SAHA (Fig. 5C). Although western immunoblotting of tumor homogenates did not detect a distinct difference in intratumoral levels of acetylated histone H3 between OSU-HDAC42–treated and SAHA-treated groups, the level of α-tubulin hyperacetylation was clearly greater in tumors from the OSU-HDAC42 treatment group (Fig. 5D). Moreover, OSU-HDAC42 reduced intratumoral expression levels of both Bcl-xL and cIAP-2 in contrast to SAHA, which induced a sizeable reduction in Bcl-xL only.

The therapeutic potential of OSU-HDAC42 was further assessed by an examination of its ability to contain the growth of large subcutaneous xenograft tumors. The oral administration of OSU-HDAC42 at 50 mg/kg every other day was commenced when the tumor volumes reached approximately 500 mm3. Despite the large tumor burden, OSU-HDAC42 markedly suppressed tumor growth, whereas the tumor volume approximately doubled every 10 days in the vehicle-treated controls (Fig. 5E).

Discussion

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

The recent US Food and Drug Administration approval of the HDAC inhibitor SAHA for the treatment of cutaneous T-cell lymphoma firmly establishes HDAC as a valid target for cancer therapy. Here, we report the in vitro and in vivo efficacy of a novel phenylbutyrate-based HDAC inhibitor, OSU-HDAC42, in HCC. With respect to SAHA, OSU-HDAC42 exhibited substantially higher low-micromolar potency in inducing apoptosis in 3 different HCC cell lines, PLC5, Huh7, and Hep3B, as indicated by caspase 9 activation, PARP cleavage, and flow cytometric analysis. Mechanistic evidence indicates that this high apoptogenic activity is attributable not only to chromatin remodeling by HDAC inhibition but also to the modulation of multiple regulators of apoptosis. Specifically, OSU-HDAC42 reduced Akt phosphorylation and the expression levels of Bcl-xL and IAP family members survivin, cIAP1, and cIAP2 in all 3 HCC cell lines. The concerted actions of OSU-HDAC42 on p21 up-regulation in conjunction with the inhibition of multiple antiapoptotic signaling targets involved in mitochondrial integrity and caspase activity underlie its apoptogenic activity in these HCC cells.

We previously reported that OSU-HDAC42 and trichostatin A, another HDAC inhibitor, facilitated Akt dephosphorylation in cancer cells by altering the dynamics of HDAC-PP1 complexes independently of histone acetylation.38 These findings coincide with evidence that some HDAC inhibitors may exert their antitumor effects through histone acetylation–independent mechanisms.15, 29, 30 Whether the suppressive effects of OSU-HDAC42 on Bcl-xL and IAP family members observed in this study were also independent of histone acetylation is not clear. Nonetheless, the down-regulation of Bcl-xL, survivin, cIAP-1, and cIAP-2 expression levels by OSU-HDAC42 warrant attention for 2 reasons. First, these antiapoptotic proteins provide cellular mechanisms by which cancer cells resist apoptotic stimuli and acquire a drug-resistant phenotype. Second, high expression levels of Bcl-xL and IAP family members, in particular survivin, have been found in the majority of surgically resected HCC specimens,42, 49, 50 which may be clinically relevant with respect to HCC progression and prognosis. Hence, the activity of OSU-HDAC42 against these apoptotic regulators suggests its potential clinical efficacy against HCC.

The differential in vitro antiproliferative activities of OSU-HDAC42 and SAHA were reflected in our in vivo studies, in which OSU-HDAC42 exhibited substantially higher potency than SAHA in suppressing both orthotopic and subcutaneous PLC5 xenograft tumor growth. The mechanistic basis for this differential potency is suggested by our assessment of intratumoral biomarkers by western blotting, which revealed that the greater tumor suppressive effects of OSU-HDAC42 paralleled its greater ability to mediate α-tubulin acetylation and to inhibit the expression of representative apoptosis biomarkers, Bcl-xL and cIAP-2. This differential hyperacetylation of α-tubulin, a nonhistone substrate of HDAC 6, suggests that, among HDAC isoforms, HDAC 6 may be a predominant target for OSU-HDAC42 in PLC5 HCC cells. This is noteworthy because HDAC 6 complexes with PP1, thereby regulating its availability to dephosphorylate its substrates, which include phospho-Akt. As mentioned previously, we have previously shown that the disruption of such HDAC-PP1 complexes by HDAC inhibitors facilitates the dephosphorylation of Akt in prostate cancer and glioblastoma cells.38

In both orthotopic and ectopic tumor models, OSU-HDAC42 was equipotent in suppressing tumor growth, despite the predicted differences in the pharmacokinetics and pharmacodynamics of drug therapy in intrahepatic tissues versus subcutaneous tissues. Thus, the oral administration of OSU-HDAC42 resulted in the delivery of sufficient quantities of the drug to these disparate tissue compartments to inhibit HCC tumor growth. To the best of our knowledge, this represents the first report that directly compared in vivo therapeutic efficacy between these 2 models. It is also worth commenting that, although we developed PLC5-luc cells in the hope of serially monitoring tumor growth through bioluminescence, this approach failed as the administration of either drug elevated the luminescent signal from the tumor cells, despite clear reductions in the tumor size as determined by direct measurements at the end of the study. Presumably, this occurred as a result of HDAC inhibitor–induced hyperacetylation of core histones leading to chromatin remodeling and activated expression of the firefly luciferase gene. Similar phenomena have been reported, not in the context of therapeutic models but in studies aimed at reversing reporter gene silencing after long-term culture or in vivo transplantation of stem cells.51 It is apparent that imaging systems based on the ectopic expression of reporter genes are likely to be inappropriate for the assessment of therapeutic agents, such as HDAC inhibitors and other epigenetic modulators, that alter gene expression through transcriptional activation.

In conclusion, our results show that the novel orally bioavailable, phenylbutyrate-derived HDAC inhibitor, OSU-HDAC42, potently inhibits HDAC and targets, such as Akt, Bcl-xL, and IAP family members, that regulate multiple aspects of cancer cell survival. These findings are consistent with, and extend, those previously reported for this agent in in vitro and in vivo models of human prostate cancer.40 OSU-HDAC42′s broad spectrum of activity, which underlies its potent apoptogenic and antitumor activities, and its efficacy in models of multiple cancer types support its clinical promise as a component of therapeutic strategies for human cancers, including advanced HCC, for which systemic therapies have been largely unsuccessful.

References

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