The selective Aurora B kinase inhibitor AZD1152 as a novel
treatment for hepatocellular carcinoma
Arihiro Aihara, Shinji Tanaka*, Mahmut Yasen, Satoshi Matsumura, Yusuke Mitsunori, Ayano Murakata, Norio Noguchi, Atsushi Kudo, Noriaki Nakamura, Koji Ito, Shigeki Arii
Department of Hepato-Biliary-Pancreatic Surgery, Graduate School of Medicine, Tokyo Medical and Dental University, 1-5-45 Yushima,
Bunkyo-ku, Tokyo 113-8519, Japan
Background & Aims: We previously identified that high Aurora B expression was associated with hepatocellular carcinoma (HCC) recurrence due to tumor dissemination. In this preclinical study, a novel inhibitor of Aurora B kinase was evaluated as a treatment for human HCC.
Methods: AZD1152 is a selective inhibitor of Aurora B kinase. Twelve human HCC cell lines were analyzed for Aurora B kinase expression and the in vitro effects of AZD1152. The in vivo effects of AZD1152 were analyzed in a subcutaneous xenograft model and a novel orthotopic liver xenograft model.
Results: Aurora B kinase expression varied among the human HCC cell lines and was found to correlate with inhibition of cell prolifer- ation, accumulation of 4N DNA, and the proportion of polyploid cells following administration of AZD1152-hydroxyquinazoline- pyrazol-anilide (AZD1152-HQPA). AZD1152-HQPA suppressed histone H3 phosphorylation and induced cell death in a dose- dependent manner. Growth of subcutaneous human HCC xeno- grafts was inhibited by AZD1152 administration. In an orthotopic hepatoma model, treatment with AZD1152 significantly deceler- ated tumor growth and increased survival. Pharmacobiological analysis revealed that AZD1152 induced the rapid suppression of phosphohistone H3, followed by cellular apoptosis in the liver tumors but not in the normal tissues of the orthotopic models. Conclusions: Our preclinical studies indicate that AZD1152 is a promising novel therapeutic approach for the treatment of HCC. ti 2009 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.
Introduction
Hepatocellular carcinoma (HCC) is one of the most common malig- nancies worldwide, accounting nearly for 1 million deaths per year
[1], and the incidence is still increasing [2]. The primary curative treatment for HCC is surgical resection, and there has been limited improvement in the availability of alternative treatments in the last decade [3]. A major obstacle for the treatment of HCC is the high frequency of tumor recurrence after curative resection. In fact, it is the recurrence pattern, rather than the recurrence itself, that critically affects patient prognosis [4]. The systemic treatment of HCC using conventional anticancer agents has provided little clin- ical benefit or prolonged survival for patients with advanced HCC [5]. A recent clinical trial by Llovet et al. [6] revealed a molecular- targeted inhibitor, sorafenib, as the first agent that demonstrated an improved overall survival in patients with advanced HCC. The increased understanding of the molecular mechanisms regulating cancer progression has led to the development of novel targeted therapies [7,8]. In order to fulfill this promise, there is an urgent need to identify the optimal targets for treatment.
In our previous studies in HCC patients after curative resec- tion, the aggressive recurrence exceeding Milan criteria showed extremely poor prognosis [9]; moreover, a genome wide micro- array profiling analysis identified the over-expression of Aurora B kinase as the only independent factor predictive of the aggres- sive recurrence [10]. The Aurora kinase family of serine–threo- nine kinases control chromosome assembly and segregation during mitosis. Aberrant expression of the Aurora kinases has been reported in a variety of solid tumors including prostate [11], colon [12], pancreas [13], lung [14], breast [15], and thyroid [16]. These findings have led to an interest in these kinases as molecular targets for cancer treatment [17,18]. Several small- molecule inhibitors of Aurora kinases have been developed as potential anticancer treatments. According to the recent review on Aurora inhibitors [19], ZM447439, Hesperadin, and MK0457/
VX680 were the first to be described and to have similar potency versus Aurora A, Aurora B, and Aurora C. Currently, MLN8054 and MLN8237 are being developed as selective Aurora A kinase inhib- itors. AZD1152 is a selective inhibitor of Aurora kinase activity
Keywords: Hepatocellular carcinoma; Aurora B kinase; AZD1152; Orthotopic model; Molecular-targeted agent.
Received 9 April 2009; received in revised form 2 July 2009; accepted 13 July 2009; available online 29 October 2009
* Corresponding author. Tel.: +81 3 58035928; fax: +81 3 58035264. E-mail address: [email protected] (S. Tanaka).
Abbreviations: cCasp-3, cleaved caspase-3; HE, hematoxylin and eosin; HCC, hepatocellular carcinoma; HQPA, hydroxyquinazoline-pyrazol-anilide; IC50, half- maximal inhibitory concentration; PBS, phosphate-buffered saline; PhH3, phos- phohistone H3.
with specificity for Aurora B kinase [20,21]. AZD1152 is a prodrug that is rapidly converted to the active moiety AZD1152-hydroxy- quinazoline-pyrazol-anilide (AZD1152-HQPA) in plasma. Thus, AZD1152 is used for in vivo studies, while AZD1152-HQPA is used for in vitro work.
The importance of the role of the organ microenvironment in cancer is being increasingly understood [22]. This is particularly true for HCC, an organotropic cancer in which the liver-specific
Journal of Hepatology 2010 vol. 52 j 63–71
microenvironment may play a critical role in HCC tumor develop- ment, cellular apoptosis, and drug sensitivity [23]. Additionally, hepatic tumors reside within the liver parenchyma, where drug metabolism and transformation occur. Thus, the pharmacody- namics of drug therapy for intrahepatic tumors may vary signif- icantly from those drugs targeted at tumors in peripheral tissues. Several attempts have been made to generate a model of intrahe- patic HCC via intraportal or intrahepatic injection of tumor cells in mice; however, frequent cancer dissemination makes it partic- ularly difficult to generate a single quantitative tumor. A recent report describes development of a novel orthotopic liver tumor xenograft model that could be used in quantitative investigations of a single tumor within its native microenvironment [24]. This might provide a system in which the tumor’s biological response to therapeutic agents more closely mimics that observed in liver tumors in patients [25]. The in vivo efficacy of Aurora kinase inhibitors in orthotopic xenograft models of solid cancers has not been reported to date [20,26,27].
Outcome of HCC patients is determined by combination of two distinct types of HCC recurrence, and the aggressive recurrence is driven by malignant characteristics of the tumor [4,9]. Because Aurora B kinase was found to be associated with the aggressive recurrence exceeding Milan criteria [10], it makes sense to target Aurora B kinase to treat the tumor. In this regard, the Aurora B kinase-specific inhibitor AZD1152 might be an attractive candi- date for HCC therapy. This investigation evaluates the in vitro and in vivo effects and pharmacodynamics of AZD1152 in a num- ber of preclinical liver tumor models, including an orthotopic model that more closely mimics the human disease.
Materials and methods
Reagents
AZD1152-HQPA and its prodrug AZD1152 were provided by AstraZeneca Pharma- ceuticals (Macclesfield, UK).
Cell culture
The human HCC cell lines SK-Hep1, Hep3B, and PLC/PRF/5 were obtained from the American Type Culture Collection (Manassas, VA, USA). Other human HCC cell lines—JHH-1, JHH-2, JHH-4, HuH-1, HuH-6, HuH-7, HLE, HLF, and HepG2—were obtained from the Human Science Research Resources Bank (Osaka, Japan). Cul- ture media were RPMI-1640 (SK-Hep1, Hep3B, HuH-7, and HepG2), Dulbecco’s modified Eagle’s medium (PLC/PRF/5, HuH-1, HuH-6, HLE, and HLF), and Wil- liam’s E medium (JHH-1, JHH-2, and JHH-4), supplemented with 5% fetal bovine serum (FBS) for HLF cells or 10% FBS for the remaining cell lines. All media sup- plemented 100 U/mL of penicillin and 100 lg/mL of streptomycin; all cell lines were cultivated in a humidified incubator at 37 ti C in 5% carbon dioxide and har- vested with 0.25% trypsin–0.03% EDTA.
Analysis of cell proliferation and cell viability
All cell lines were cultured in logarithmic growth phase in the presence of various concentrations of AZD1152-HQPA (0.3–1000 nM) for 72 h. Cells were seeded at
4 ti 104 cells in six-well plates with the appropriate control medium. After 24 h, plates were treated with compound and incubated for 72 h at 37 tiC. At the end of the incubation time, cells were detached from each plate, and viable cells were counted using a hemocytometer. Half-maximal inhibitory concentra- tion (IC50) values were calculated with BioDataFit v.1.02 software using the four-parameter logistic model. The mean values and standard deviations of IC50 were calculated in triplicate for each cell line. To investigate cell viability, tripli- cate samples of SK-Hep1, Hep3B, and HLF cells were cultured in the presence of various concentrations of AZD1152-HQPA (1–100 nM) for 72 h. The number of nonviable cells was assessed using a hemocytometer and trypan blue dye exclusion.
Western blotting
Total protein was extracted from each cell line, as described previously [28]. Pro- tein levels of Aurora B kinase, phosphohistone H3 (PhH3), and alpha-tubulin (control) were detected using standard western blot analysis on 8–15% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS–PAGE). Blots were incu- bated overnight at 4 ti C with the primary antibody antihuman Aurora B (1:1000; Abcam, Cambridge, UK; Catalog No. ab2254) or antihuman PhH3 (1:200; Santa Cruz, CA, USA; Catalog No. sc-8656-R), then at room temperature for 1 h with anti-alpha-tubulin (1:5000; Sigma–Aldrich, St. Louis, MO, USA; Catalog No. T9026). Appropriate secondary antibodies were added for 2 h, and protein expression was visualized with enhanced chemiluminescence by the ECL western blotting detection system (GE Healthcare, Buckinghamshire, UK). The expression ratio of Aurora B kinase to the control was analyzed using Multi-Gage software (FUJIFILM, Tokyo, Japan).
Flow cytometry
Samples of all cell lines in logarithmic growth phase were exposed to AZD1152- HQPA 100 nM for 24 h, and then fixed in 70% ethanol at ti 20 ti C overnight. Cells were rehydrated in phosphate-buffered saline (PBS), and then resuspended in PBS containing RNase 100 lg/mL (Sigma) and propidium iodide 10 lg/mL. Cellu- lar DNA content was analyzed on a FACS Caliber flow cytometer (Becton & Dick- inson Biosciences, San Jose, CA, USA). For detection of apoptosis, cells were labeled with the Annexin V-FITC Kit (Miltenyi Biotec, Bergisch Gladbach, Ger- many; Catalog No. 130-092-052) at room temperature for 15 min, followed by analysis on a FACS Caliber flow cytometer.
Immunocytochemistry and immunohistochemistry
SK-Hep1, Hep3B, and HLF cells were cultured on glass slides coated with silane in the presence of various concentrations of AZD1152-HQPA (1–100 nM) for 4 h. They were then fixed using 3.7% formalin for 10 min and permeabilized using 100% methanol for 20 min for immunocytochemical detection of PhH3.
Xenograft tumor tissue was harvested, formalin fixed, and paraffin embed- ded. The primary antibodies, PhH3 (Upstate Cell Signaling Solution, Danvers, MA, USA; Catalog No. 9701) and anti-cleaved caspase-3 (anti-cCasp-3; Upstate Cell Signaling Solution; Catalog No. 9661), were used at 1:100 and 1:400 dilution, respectively, in PBS containing 1% bovine serum albumin. The tissue sections and slides were stained with an automated immunostainer (BenchMark XT; Ventana Medical Systems, Tucson, AZ, USA) using heat-induced epitope retrieval and a standard diaminobenzidine detection kit (Ventana).
In vivo studies in a subcutaneous tumor xenograft model
A subcutaneous tumor model was used to analyze the in vivo activity of AZD1152, as described previously [29]. Five-week-old female nude mice (nu/nu) were pur- chased from Japan SLC (Shizuoka, Japan) and kept under pathogen-free condi- tions, fed standard food, and given free access to sterilized water. In all experiments, mice were anesthetized by 100 mg/kg Nembutal intraperitoneal injection. Subcutaneous xenografts were established by inoculating 1 ti 107 SK- Hep1 cells into the right dorsal flank. Palpable tumors were confirmed on day 5 following inoculation, and mice were randomized into treatment groups to receive AZD1152 or the control Tris-buffered saline. AZD1152 was prepared in Tris-buffered saline (pH 9) and administered by intraperitoneal injection. Tumor size was measured using calipers as frequently as every other day for 2 weeks,
and tumor volumes were calculated as AB2 ti 0.5 (A, length; B, width). The Animal Care Committee of Tokyo Medical and Dental University School of Medicine approved the experimental protocols in accordance with its institutional guidelines.
In vivo studies in a novel orthotopic xenograft model
An orthotopic xenograft model was created by direct intrahepatic inoculation of SK-Hep1 and Hep3B cells, as described by Lu et al. [25]. With the mice fully anes- thetized, a small transverse incision was made below the sternum to expose the liver. Then, 2.5 ti 106 cells suspended in 25 lL of RPMI-1640 and 25 lL of Matri- gel (Becton & Dickinson Biosciences) were slowly injected at a 30ti angle into the upper left lobe of the liver using a 28-gauge needle. After injection, a small piece of sterile gauze was placed on the injection site, and light pressure was applied for 1 min to prevent bleeding. The abdomen was then closed with a 6–0 silk suture. Pilot studies confirmed development of liver tumors in 6 of 6 mice at
14 days after inoculation. AZD1152 (100 mg/kg) or the control Tris buffer was administered to mice by intraperitoneal injection on 2 consecutive days per week for 2 weeks starting on day 14 after inoculation. In both cell lines at 4 weeks after initiation of treatment, mice were sacrificed to assess the antitumor effects of AZD1152. The survival end points were defined as ascites formation in the hepa- toma-bearing mice [30]. Animal survival data were entered in the Kaplan–Meier Life Table format and presented as the cumulative survival plot. Statistical differ- ences were analyzed by Mantel–Cox log-rank test. All in vivo procedures were approved by the Animal Care Committee of Tokyo Medical and Dental University (Permission No. 090235).
The pharmacobiological effects of AZD1152 treatment in the orthotopic liver tumors were assessed by immunohistochemical analysis of PhH3 and cCasp-3 expression in control tumors and in those harvested 3 and 5 days after initiation of AZD1152 treatment.
Results
Aurora B kinase expression and in vitro effects of AZD1152-HQPA in human hepatocellular carcinoma cells
Evaluation of Aurora B kinase protein in 12 human HCC cell lines revealed a variety of expression levels, as shown in Fig. 1A
(Aurora B/tubulin expression ratio: JHH-1, 0.120; JHH-2, 0.039; JHH-4, 0.059; HuH-1, 0.078; HuH-6, 0.220; HuH-7, 0.243; HLE, 0.040; HLF, 0.032; PLC/PRF/5, 0.083; SK-Hep1, 0.107; Hep3B, 0.079; HepG2, 0.044). Expression of Aurora B kinase was approx- imately 7-fold higher in HuH-7 and HuH-6 cells than in JHH-2 and HLF cells. To evaluate the growth inhibitory effects of AZD1152-HQPA, cell proliferation assays were conducted in these HCC cell lines. AZD1152-HQPA showed potent antiproliferative activity in all HCC cell types with IC50 values (JHH-1, 17.4 ± 1.0 nM; JHH-2, 218.0 ± 10.8 nM; JHH-4, 155.6 ± 16.8 nM; HuH-1, 27.3 ± 5.0 nM; HuH-6, 3.7 ± 0.6 nM; HuH-7, 6.8 ± 0.3 nM; HLE, 45.9 ± 6.4 nM; HLF, 126.1 ± 12.2 nM; PLC/PRF/5, 76.9 ± 9.9 nM; SK-Hep1, 21.9 ± 1.2 nM; Hep3B, 7.6 ± 1.2 nM; HepG2, 14.7 ± 1.7 nM) (Fig. 1A). Fig. 1B demonstrates the rela- tionship between Aurora B kinase expression and indexes of AZD1152-HQPA IC50 in the panel of cell lines tested (correlation coefficient: ti0.72738; R2 = 0.529; p = 0.0073).
Alterations in DNA ploidy in the human HCC cell lines were analyzed by flow cytometry (Fig. 1C). Accumulation of cells with
>4N DNA content was observed in all of the cell lines following
A
JHH
1
-
JHH
2
-
JHH
4
-
HuH
1
-
HuH
6
-
HuH
7
-
B1000
JHH-2
R 2 = 0.529 p = 0.0073
D 1000
R 2 = 0.443 p = 0.0129
Aurora B Tubulin
IC50 (nM) 17.4 218.0
± 1.0 ± 10.8
HLE HLF Aurora B
155.6 27.3 3.7 6.8
± 16.8 ± 5.0 ± 0.6 ± 0.3
/ 5 Hep1
PLCPRF /SK – Hep3B HepG2
100
10
JHH-4
HLF PLC/PRF/5 HLE
SK-Hep1
HuH-1
JHH-1
HepG2
Hep3B HuH-6
HuH-7
100
10
JHH-4
HLF
HLE
SK-Hep1
JHH-2
PLC/PRF/5 HuH-1
JHH-1
HepG2
Hep3B
HuH-7
HuH-6
Tubulin IC50 (nM)
45.9 ± 6-4
126. 1 ± 12.2
76.9 ± 9.9
21.9 ± 1.2
7.6 ± 1.2
14.7 ± 1.7
1 0 0.1 0.2 0.3
Aurora B expression (ratio)
1
0
20 40 60 80
Increasing rate of > 4N DNA (%)
CJHH – 1 JHH – 2 JHH – 4 HuH – 1 HuH – 6 HuH – 7
40.5 ± 3.6 24.2 ± 0.2 2.3 ± 0.1 32.0 ± 0.1 31.5 ± 0.8 53.7 ± 2.9
HLE HLF PLC / PRF / 5 SK – Hep1 Hep3B HepG2
24.9 ± 1.8 8.4 ± 1.7 37.3 ± 2.7 22.0 ± 2.5 45.4 ± 1.9 59.0 ± 0.9
Fig. 1. Expression of Aurora B kinase and AZD1152-hydroxyquinazoline-pyrazol-anilide (AZD1152-HQPA) activity in human hepatocellular carcinoma (HCC) cell lines. (A) Western blot analysis of Aurora B kinase and the control alpha-tubulin. The concentration that induced half-maximal inhibitory concentration (IC50) in 12 human HCC cell lines is indicated. (B) Relationship between the ratio of Aurora B kinase expression to control tubulin and the indexes of AZD1152 IC50 values in each human HCC cell line. Correlations were analyzed by Pearson two-tailed correlation. The level of statistical significance was p < 0.05. (C) Cellular DNA content was analyzed by flow cytometry in 12 human HCC cell lines after 24-h incubation with AZD1152-HQPA 100 nM (thick lines) or the control DMSO buffer (thin lines), and the increasing rate of >4N DNA (%) was indicated. (D) Relationship between the increasing rate of >4N cells and the indexes of AZD1152 IC50 values in each human HCC cell line. Correlations were analyzed by Pearson two-tailed correlation. The level of statistical significance was p < 0.05.
24-h incubation with AZD1152-HQPA 100 nM, with the excep- tion of JHH-2 and HLF, which showed AZD1152 insensitivity with low expression levels of Aurora B kinase. As shown in Fig. 1D, the increasing rate of >4N DNA by AZD1152-HQPA (JHH-1, 40.5 ± 3.6%; JHH-2, 24.2 ± 0.2%; JHH-4, 2.3 ± 0.1%; HuH-1, 32.0 ± 0.1%; HuH-6, 31.5 ± 0.8%; HuH-7, 53.7 ± 2.9%; HLE, 24.9 ± 1.8%; HLF, 8.4 ± 1.7%; PLC/PRF/5, 37.3 ± 2.7%; SK-Hep1, 22.0 ± 2.5%; Hep3B, 45.4 ± 1.9%; HepG2, 59.0 ± 0.9%) was corre- lated with the indexes of IC50 values (correlation coefficient: ti0.66534; R2 = 0.443; p = 0.0129). The accumulation of polyploid cells is consistent with failed cytokinesis following inhibition of Aurora B kinase activity.
Previously, cellular apoptosis in response to the pan-Aurora kinase inhibitor VX680 was limited in cells expressing wild-type p53 but was enhanced in cells lacking p53 [31]. The p53 point mutations have been reported in four HCC cell lines (HuH-7 at codon 220 Tyr-to-Cys; HLF at codon 244 Gly-to-Ala; HLE and PLC/PRF/5 at codon 249 Arg-to-Ser), and null expression of p53 was reported due to the deletion in the Hep3B cell line, while SK-Hep1 and HepG2 have wild-type p53 [32–34]. There was no significant correlation between the efficacy of AZD1152-HQPA and the p53 status of each cell line in our experiments.
In vitro effects of AZD1152-HQPA on phosphorylation of histone H3 and cell death in human hepatocellular carcinoma cell lines
In the previous studies by Mortlock et al. [35], AZD1152-HQPA is a selective Aurora B kinase inhibitor with more than 1000- to 10,000-fold selectivity for Aurora A kinase and various tyrosine kinases including kinase insert domain receptor (KDR), the Abel- son virus kinase (vABL), and epidermal growth factor receptor (EGFR). The inhibition of Aurora B kinase is determined by its specific cellular substrate histone H3 [36]. We investigated whether AZD1152-HQPA was able to inhibit PhH3 in the sensi- tive SK-Hep1 and Hep3B cells. As shown in Figs. 2 and 3a, AZD1152-HQPA 100 nM yielded a substantial reduction in the level of PhH3. This inhibition of histone H3 phosphorylation was shown to be dose dependent in SK-Hep1 and Hep3B cells treated with AZD1152-HQPA 1–100 nM (Fig. 3B). The cellular apoptosis was confirmed by analysis of Annexin-V binding (Fig. 3C). Cell death rates were measured and were also found be proportional to AZD1152-HQPA dose (Fig. 3D) [21]. These results indicate that inhibition of Aurora B kinase by AZD1152-HQPA can induce cell death in the SK-Hep1 and Hep3B cells in vitro. In contrast, the AZD1152-insensitive HLF
cells with a low expression of Aurora B kinase (Fig. 1) showed no significant effects on PhH3 and apoptosis compared with SK-Hep1 and Hep3B cells (Figs. 2 and 3).
In vivo effects of AZD1152 on subcutaneous xenografts of human hepatocellular carcinoma cells
The human HCC cell line SK-Hep1 (AZD1152 IC50: 21.9 nM) is known to be aggressively tumorigenic in vivo [37]. To investigate in vivo antitumor activity, AZD1152 100 mg/kg per day was administered to nude mice bearing established SK-Hep1 subcuta- neous xenografts on 2 consecutive days per week for 2 weeks (n = 10). Tumor volumes were measured every other day. As shown in Fig. 4A, significant regression of SK-Hep1 tumors was observed in the group of mice that received AZD1152 compared with control. The mean tumor volumes were substantially decreased by treatment with AZD1152 on day 14 following treat- ment, and tumor volumes in treated mice were 15.5% of those in control mice (Fig. 4B). None of the AZD1152-treated mice showed signs of wasting or other toxicity relative to control mice. AZD1152 was tolerated at the dose at which antitumor efficacy was observed.
In vivo effects of AZD1152 on orthotopic liver xenografts of human hepatocellular carcinoma cells
Anovel orthotopic xenograft model of liver tumors with Matri- gel was utilized to explore tumor growth inhibition in situ [25]
(Fig. 5A). AZD1152 100 mg/kg was administered to mice bearing SK-Hep1 orthotopic xenografts on 2 consecutive days per week for 2 weeks (n = 5). Histological analysis of the liver tumors was conducted within 4 weeks after treatment. Growth of liver tumors was found to be suppressed in all of the mice that had been treated with AZD1152 (Fig. 5B). After drug administra- tion, the mean liver tumor weight in those animals that had received AZD1152 was 10% of that in the control mice (Fig. 5C). Similar growth inhibition was observed in Hep3B orthotopic xenografts by administration of AZD1152 (Fig. 5B and C). In the orthotopic model, mouse survival was signifi- cantly enhanced by AZD1152 treatment in comparison with the control (p < 0.005; Fig. 5D). These results demonstrate that AZD1152 was able to significantly inhibit in vivo growth of a human HCC tumor in the liver microenvironment in mice. All of the host tissues examined, including liver, bone marrow, kid- ney, intestine, and lung, were histologically normal in all experiments.
SK - Hep1
Hep3B
HLF
Pharmacobiological effects of AZD1152 on orthotopic liver xenografts of human hepatocellular carcinoma cells
PhH3
Tubuline
The liver xenograft model described above was subjected to histological analysis by immunostaining to investigate the pharmacobiological effects of AZD1152 in the hepatic microen-
Control
AZD1152
- HQPA
Control
AZD1152
- HQPA
Control
AZD1152
- HQPA
vironment (Fig. 6). Three days after treatment with AZD1152, there was a substantial decrease in PhH3 (Fig. 6E and H) com- pared with the control (Fig. 6D and G), although after 5 days,
Fig. 2. Effects of AZD1152-hydroxyquinazoline-pyrazol-anilide (AZD1152- HQPA) on phosphorylation of histone H3 in human hepatocellular carcinoma (HCC) cell lines. Western blot analysis of phosphohistone H3 and the control alpha-tubulin in human HCC cell lines after 4-h incubation with AZD1152-HQPA 100 nM or the control DMSO buffer. Cell lines: SK-Hep1 (left), Hep3B (middle), and HLF (right).
PhH3 had recovered (Fig. 6F and I). Staining of tumor samples for apoptotic marker cCasp-3 showed gradually elevating levels following AZD1152 treatment (Fig. 6J–O). The hepatocytes from the host liver were histologically normal at all points following AZD1152 administration (Fig. 6A–C).
A
SK - Hep1
B
50
40
30
20
10
SK - Hep1
Hep3B
0 0 1 3 10 30 100 12
Hep3B
8
4
00 1 3 10 30 100
HLF
25
20
15
10
5
HLF
Control
AZD1152 HQPA (100nM)
0
13 10 30 100
AZD1152 - HQPA (nM)
C
SK - Hep1
1.38%
4.61%
9.14%
17.3%
D
50
40
30
20
10
SK - Hep1
0
0 1 3 10 30 100
Hep3B
HLF
4.70%
4.06%
2.83%
1.89%
11.2%
18.8%
2.70%
2.51%
50
40
30
20
10
0
50
40
30
20
0
1
Hep3B
3 10 30 100
HLF
Control
Annexin V
AZD1152
10
0
0
1
3 10 30 100
HQPA (100nM) AZD1152 - HQPA (nM)
Fig. 3. Dose-dependent effects of AZD1152-hydroxyquinazoline-pyrazol-anilide (AZD1152-HQPA) on phosphorylation of histone H3 and cell death of human hepatocellular carcinoma (HCC) cell lines. (A) Immunocytochemistry of phosphohistone H3 (PhH3) in human HCC cells after 4-h incubation with AZD1152-HQPA 100 nM or the control DMSO buffer. Magnification ti 200. Cell lines: SK-Hep1 (upper), Hep3B (middle), and HLF (lower). (B) Dose–response analysis showing percentage of PhH3- positive HCC cells analyzed by immunocytochemistry. Cell lines: SK-Hep1 (upper), Hep3B (middle), and HLF (lower). Columns, PhH3-positive cells (%); vertical bars, standard deviation. (C) AZD1152-HQPA induces apoptosis. SK-Hep1 (upper), Hep3B (middle), and HLF (lower) cells were treated with AZD1152-HQPA for 72 h, and apoptosis was assessed by flowcytometric analysis of cells labeled with Annexin V and propidium iodide. (D) Dose–response analysis showing percentage of nonviable cells in SK-Hep1, Hep3B, and HLF cell samples analyzed using a hemocytometer and trypan blue dye exclusion. Columns, dead cells (%); vertical bars, standard deviation.
A Control AZD1152 B
1.4
p < 0.05
Day 0
1.2
1.0
Control *
0.8
*
*
Day 7
0.6
*
*
0.4 *
Day 14
0.2
*
AZD1152
0 2 4 6 8 10 12 14
Day
Fig. 4. In vivo effects of AZD1152 on human hepatocellular carcinoma (HCC) growth in subcutaneous xenograft models. Established subcutaneous xenografts of SK- Hep1 were treated with intraperitoneal AZD1152 100 mg/kg or the control Tris buffer on 2 consecutive days per week for 2 weeks. (A) Subcutaneous SK-Hep1 tumors in mice on days 0, 7, or 14 following treatment with AZD1152 (right) or the control (left). (B) Tumor volumes were measured and plotted every other day in AZD1152-treated or control mice (n = 10). Arrows, timing of administration; vertical bars; standard error. Statistical analysis was done by two-tailed Student t test (tip < 0.05).
Discussion
The Aurora family of serine–threonine kinases has recently emerged as a key mitotic regulator required for genome stability [38]. In mammals, the Aurora family consists of three members: Aurora A and B kinases and the less well-characterized Aurora C kinase. Aurora B kinase has been clearly shown to regulate kine- tochore function, as it is required for correct chromosome align- ment and segregation, spindle-checkpoint function, and cytokinesis. As Aurora kinases are frequently overexpressed in various tumors [39], they have received much attention as poten- tial targets for novel anticancer therapeutics. Treatment with Aurora kinase inhibitors induces the accumulation of cells arrested in a pseudo-G1 state with >4N DNA content or the accu- mulation of cells with >4N DNA content, the latter population representing cells that exit mitosis and subsequently proceed through S phase in the absence of cell division [31]. Continued proliferation in the presence of aberrant mitosis and failed cyto- kinesis presumably gives rise to cells with higher DNA content due to an increase of the cell diameter, resulting in apoptosis [17,18,40,41]. The defective cytokinesis, as well as the inhibition of PhH3 by Aurora kinase inhibitors, suggests that the cellular effects of Aurora kinase inhibitors might be largely mediated by the disruption of Aurora B kinase function [18]. AZD1152 is a selective inhibitor of Aurora kinase with specificity for Aurora B kinase. AZD1152 has the potential to be efficacious in multiple tumor types and is currently undergoing phase 1 clinical evalua- tion as a treatment for a range of malignancies [20,21].
We have previously identified Aurora B kinase as the only independent predictor for the aggressive recurrence of human HCC [10]. In our present study, AZD1152-HQPA substantially reduced in vitro proliferation in a variety of human HCC cell lines. The extent of proliferation inhibition was correlated with Aurora
Bkinase expression levels (Fig. 1). As shown in Fig. 1C, significant DNA fragmentation in the form of a sub-G1 peak could not be detected after 24 h of treatment with AZD1152-HQPA, which is in line with data reported by Wilkinson et al. [20]. This inability to detect a sub-G1 population after AZD1152-HQPA treatment
may result because inhibition of Aurora B kinase induces poly- ploidy before apoptosis, in which case DNA fragmentation will occur in the >4N population, making it difficult to detect a sub- G1 population.
Treatment with AZD1152-HQPA also led to inhibition of PhH3 as well as failure of tumor cell division, and ultimately induced death of human HCC cells (Figs. 2 and 3). In vivo administration of AZD1152 suppressed the growth of human HCC tumors in established subcutaneous xenografts (Fig. 4). Although subcuta- neous xenograft models have the benefits of easy visualization and monitoring of tumor growth, the biological response to ther- apeutic agents in the natural microenvironment of the tumor should be analyzed using orthotopic xenograft models [30,42]. In this study, a novel model of intrahepatic inoculation with Matrigel was utilized to closely mimic HCC tumors in humans [24]. As shown in Fig. 5, AZD1152 inhibited in vivo growth of established liver tumors and increased survival in this model. Furthermore, pharmacobiological studies of AZD1152 confirmed in vivo suppression of PhH3 and induction of cellular apoptosis of human HCC (Fig. 6). AZD1152 was well tolerated at the dose required to elicit a potent and durable antitumor effect in mice. According to the previous report by Wilkinson et al. [20], mice were almost resistant to myelosuppression after AZD1152 treat- ment; the authors could not find any reductions in bone marrow nucleated cells at the end of the dosing period. In rats, there was a myelosuppressive effect of AZD1152 that was associated with a reduction in bone marrow nucleated cells to 34% of that seen in the controls at the end of the 48-h dosing period; however, the bone marrow nucleated cell content rapidly recovered such that it was 104.8% of control at the end of the study period. Although the phase 1 studies on the side effects of AZD1152 have not yet been reported in detail, humans might be more sensitive to the myelosuppressive effects compared to the experimental rodents. Further study should be required for clinical application to HCC patients, especially those with cirrhosis.
Clinical evidence exists indicating a significant relationship between Aurora B kinase expression and the aggressive progres- sion of HCC [10], and our preclinical studies indicated that
A B Control AZD1152
SK – Hep1
Hep3B
C
1.0
0.8
0.6
0.4
0.2
SK – Hep1
p < 0.05
D
100
80
60
40
20
SK - Hep1
AZD1152
Control
0.0
*
Control AZD1152
0
p < 0.005
0
10
20
30
40
Day
0.8
0.6
Hep3B
p < 0.05
100
80
60
Hep3B
AZD1152
0.4
40 Control
0.2
0.0
*
Control AZD1152
20
0
p < 0.005
0 10 20 30 40
Day
Fig. 5. In vivo effects of AZD1152 on human hepatocellular carcinoma (HCC) growth in orthotopic xenograft models. (A) Schematic representation of generation of orthotopic xenografts. A small transverse incision was made below the sternum to expose the liver, and 2.5 ti 106 cells with Matrigel were then slowly injected at a 30ti angle into the upper left lobe of the liver using a 28-gauge needle. After 14 days, the mice were treated intraperitoneally with AZD1152 100 mg/kg or the control Tris buffer on 2 consecutive days per week for 2 weeks. (B) The liver tumor in mice within 4 weeks after administration of AZD1152 (light) and the control (right). Cell lines: SK-Hep1 (upper) and Hep3B (lower). (C) Liver tumor weight was analyzed within 4 weeks after administration of AZD1152 or the control (n = 5). SK-Hep1 (upper), and Hep3B (lower). Vertical bars, standard error. Statistical analysis was done by two-tailed Student t test (tip < 0.05). (D) Results are expressed in terms of percent survival in experiment time. Arrows, timing of administration. Statistical differences were analyzed by Mantel–Cox log-rank test (p < 0.005).
AZD1152, a specific inhibitor of Aurora B kinase, is a promising novel therapeutic approach for the treatment of human HCC.
Urgent studies and clinical trials of AZD1152 will confirm its role in the treatment of HCC.
Day 0 Day 3 Day 5
A B C
T T T
H. E
x 40
N N N
D E F
PhH3
x 40
G H I
PhH3
x 200
cCasp3
x 40
cCasp3
x 200
J
M
K
N
L
O
Fig. 6. Pharmacobiological analysis of orthotopic xenograft models. Established orthotopic xenografts of human SK-Hep1 hepatocellular carcinoma (HCC) cells were treated intraperitoneally with AZD1152 100 mg/kg on 2 consecutive days. Mice were sacrificed humanely on day 0 prior to treatment and on days 3 and 5 after the first administration of AZD1152 (left, middle and right, respectively). (A–C) Transverse sections of liver tumor (T) or host normal liver (N) were stained with hematoxylin and eosin (HE; magnification ti 40). The same sections were analyzed for expression of phosphohistone H3 (PhH3): (D–F) magnification ti40 and (G–I) magnification ti200. The same sections were also analyzed for the apoptotic marker of cleaved caspase-3 (cCasp-3): (J–L) magnification ti 40 and (M–O) magnification ti200.
Acknowledgments
The authors who have taken part in this study declared that they do not have anything to disclose regarding funding from indus- tries or conflict of interest with respect to this manuscript.
This research was supported by the Ministry of Education, Sci- ence and Culture, Grant-in-Aid for Scientific Research.
We thank AstraZeneca for kindly providing us with AZD1152 and AZD1152-HQPA for experimental studies. We also thank Drs. Robert Wilkinson, Elizabeth Anderson (AstraZeneca) for helpful discussion, Sarah Mason (Mudskipper Bioscience) for editorial assistance, Kaoru Mogushi for statistical analysis, Akimoto Nim- ura for technical advice regarding flow cytometry, and Ayumi Shioya for technical assistance.
References
[1]Farazi PA, DePinho RA. Hepatocellular carcinoma pathogenesis: from genes to environment. Nat Rev Cancer 2006;6:674–687.
[2]Ince N, Wands JR. The increasing incidence of hepatocellular carcinoma. N Engl J Med 1999;340:798–799.
[3]Arii S, Yamaoka Y, Futagawa S, Inoue K, Kobayashi K, Kojiro M, et al. Results of surgical and nonsurgical treatment for small-sized hepatocellular carci- nomas: a retrospective and nationwide survey in Japan. The Liver Cancer Study Group of Japan. Hepatology 2000;32:1224–1229.
[4]Sherman M. Recurrence of hepatocellular carcinoma. N Engl J Med 2008;359:2045–2047.
[5]Llovet JM, Burroughs A, Bruix J. Hepatocellular carcinoma. Lancet 2003;362:1907–1917.
[6]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.
[7]Zhu AX. Development of sorafenib and other molecularly targeted agents in hepatocellular carcinoma. Cancer 2008;112:250–259.
[8]Tanaka S, Sugimachi K, Maehara S, Harimoto N, Shirabe K, Wands JR. Oncogenic signal transduction and therapeutic strategy for hepatocellular carcinoma. Surgery 2002;131:S142–S147.
[9]Tanaka S, Noguchi N, Ochiai T, Kudo A, Nakamura N, Ito K, et al. Outcomes and recurrence of initially resectable hepatocellular carcinoma meeting Milan criteria: rationale for partial hepatectomy as first strategy. J Am Coll Surg 2007;204:1–6.
[10]Tanaka S, Arii S, Yasen M, Mogushi K, Su NT, Zhao C, et al. Aurora kinase B is a predictive factor for the aggressive recurrence of hepatocellular carcinoma after curative hepatectomy. Br J Surg 2008;95:611–619.
[11]Lee EC, Frolov A, Li R, Ayala G, Greenberg NM. Targeting Aurora kinases for the treatment of prostate cancer. Cancer Res 2006;66:4996–5002.
[12]Bischoff JR, Anderson L, Zhu Y, Mossie K, Ng L, Souza B, et al. A homologue of Drosophila aurora kinase is oncogenic and amplified in human colorectal cancers. EMBO J 1998;17:3052–3065.
[13]Li D, Zhu J, Firozi PF, Abbruzzese JL, Evans DB, Cleary K, et al. Overexpression of oncogenic STK15/BTAK/Aurora A kinase in human pancreatic cancer. Clin Cancer Res 2003;9:991–997.
[14]Smith SL, Bowers NL, Betticher DC, Gautschi O, Ratschiller D, Hoban PR, et al. Overexpression of aurora B kinase (AURKB) in primary non-small cell lung carcinoma is frequent, generally driven from one allele, and correlates with the level of genetic instability. Br J Cancer 2005;93: 719–729.
[15]Tanaka T, Kimura M, Matsunaga K, Fukada D, Mori H, Okano Y. Centrosomal kinase AIK1 is overexpressed in invasive ductal carcinoma of the breast. Cancer Res 1999;59:2041–2044.
[16]Sorrentino R, Libertini S, Pallante PL, Troncone G, Palombini L, Bavetsias V, et al. Aurora B overexpression associates with the thyroid carcinoma undifferentiated phenotype and is required for thyroid carcinoma cell proliferation. J Clin Endocrinol Metab 2005;90:928–935.
[17]Gautschi O, Heighway J, Mack PC, Purnell PR, Lara Jr PN, Gandara DR. Aurora kinases as anticancer drug targets. Clin Cancer Res 2008;14:1639–1648.
[18]Keen N, Taylor S. Aurora-kinase inhibitors as anticancer agents. Nat Rev Cancer 2004;4:927–936.
[19]Macarulla T, Ramos FJ, Tabernero J. Aurora kinase family: a new target for anticancer drug. Recent Patents Anticancer Drug Discov 2008;3:114–122.
[20]Wilkinson RW, Odedra R, Heaton SP, Wedge SR, Keen NJ, Crafter C, et al. AZD1152, a selective inhibitor of Aurora B kinase, inhibits human tumor xenograft growth by inducing apoptosis. Clin Cancer Res 2007;13:3682–3688.
[21]Yang J, Ikezoe T, Nishioka C, Tasaka T, Taniguchi A, Kuwayama Y, et al. AZD1152, a novel and selective aurora B kinase inhibitor, induces growth arrest, apoptosis, and sensitization for tubulin depolymerizing agent or topoisomerase II inhibitor in human acute leukemia cells in vitro and in vivo. Blood 2007;110:2034–2040.
[22]Hoshida Y, Villanueva A, Kobayashi M, Peix J, Chiang DY, Camargo A, et al. Gene expression in fixed tissues and outcome in hepatocellular carcinoma. N Engl J Med 2008;359:1995–2004.
[23]Tredan O, Galmarini CM, Patel K, Tannock IF. Drug resistance and the solid tumor microenvironment. J Natl Cancer Inst 2007;99:1441–1454.
[24]Han C, Wu T. Cyclooxygenase-2-derived prostaglandin E2 promotes human cholangiocarcinoma cell growth and invasion through EP1 receptor-medi- ated activation of the epidermal growth factor receptor and Akt. J Biol Chem 2005;280:24053–24063.
[25]Lu YS, Kashida Y, Kulp SK, Wang YC, Wang D, Hung JH, et al. Efficacy of a novel histone deacetylase inhibitor in murine models of hepatocellular carcinoma. Hepatology 2007;46:1119–1130.
[26]Harrington EA, Bebbington D, Moore J, Rasmussen RK, Ajose-Adeogun AO, Nakayama T, et al. VX-680, a potent and selective small-molecule inhibitor of the Aurora kinases, suppresses tumor growth in vivo. Nat Med 2004;10:262–267.
[27]Soncini C, Carpinelli P, Gianellini L, Fancelli D, Vianello P, Rusconi L, et al. PHA-680632, a novel Aurora kinase inhibitor with potent antitumoral activity. Clin Cancer Res 2006;12:4080–4089.
[28]Tanaka S, Pero SC, Taguchi K, Shimada M, Mori M, Krag DN, et al. Specific peptide ligand for Grb7 signal transduction protein and pancreatic cancer metastasis. J Natl Cancer Inst 2006;98:491–498.
[29]Tanaka S, Sugimachi K, Yamashita Yi Y, Ohga T, Shirabe K, Shimada M, et al. Tie2 vascular endothelial receptor expression and function in hepatocellular carcinoma. Hepatology 2002;35:861–867.
[30]Raskopf E, Dzienisowicz C, Hilbert T, Rabe C, Leifeld L, Wernert N, et al. Effective angiostatic treatment in a murine metastatic and orthotopic hepatoma model. Hepatology 2005;41:1233–1240.
[31]Gizatullin F, Yao Y, Kung V, Harding MW, Loda M, Shapiro GI. The Aurora kinase inhibitor VX680 induces endoreduplication and apoptosis preferen- tially in cells compromised p53-dependent postmitotic checkpoint function. Cancer Res 2006;66:7668–7677.
[32]Bressac B, Galvin KM, Liang TJ, Isselbacher KJ, Wands JR, Ozturk M. Abnormal structure and expression of p53 gene in human hepatocellular carcinoma. Proc Natl Acad Sci USA 1990;87:1973–1977.
[33]Hsu IC, Tokiwa T, Bennett W, Metcalf RA, Welsh JA, Sun T, et al. P53 gene mutation and integrated hepatitis B viral DNA sequences in human liver cancer cell lines. Carcinogenesis (Lond.) 1993;14:987–992.
[34]Tanaka S, Toh Y, Adachi E, Matsumata T, Mori R, Sugimachi K. Tumor progression in hepatocellular carcinoma may be mediated by p53 mutation. Cancer Res 1993;53:2884–2887.
[35]Mortlock AA, Foote KM, Heron NM, Jung FH, Pasquet G, Lohmann JJ, et al. Discovery, synthesis, and in vivo activity of a new class of pyrazoloquinaz- olines as selective inhibitors of aurora B kinase. J Med Chem 2007;50:2213–2224.
[36]Hirota T, Lipp JJ, Toh BH, Peters JM. Histone H3 serine 10 phosphorylation by Aurora B causes HP1 dissociation from heterochromatin. Nature 2005;438:1176–1180.
[37]Maret A, Galy B, Arnaud E, Bayard F, Prats H. Inhibition of fibroblast growth factor 2 expression by antisense RNA induced a loss of the transformed phenotype in a human hepatoma cell line. Cancer Res 1995;55:5075–5079.
[38]Giet R, Petretti C, Prigent C. Aurora kinases, aneuploidy and cancer, a coincidence or a real link? Trends Cell Biol 2005;15:241–250.AZD1152-HQPA
[39]Hontz AE, Li SA, Lingle WL, Negron V, Bruzek A, Salisbury JL, et al. Aurora A and B overexpression and centrosome amplification in early estrogen- induced tumor foci in the Syrian hamster kidney: implications for chromo- somal instability, aneuploidy, and neoplasia. Cancer Res 2007;67:2957–2963.
[40]Warner SL, Bearss DJ, Han H, Von Hoff DD. Targeting Aurora-2 kinase in cancer. Mol Cancer Ther 2003;2:589–595.
[41]Carmena M, Earnshaw WC. The cellular geography of aurora kinases. Nat Rev Mol Cell Biol 2003;4:842–854.
[42]Yao X, Hu JF, Daniels M, Yien H, Lu H, Sharan H, et al. A novel orthotopic tumor model to study growth factors and oncogenes in hepatocarcinogen- esis. Clin Cancer Res 2003;9:2719–2726.