Effects of astragalus injection on different stages of early hepatocarcinogenesis in a two-stage hepatocarcinogenesis model using rats

Abstract

To clarify the suppressive effects of astragalus injection (AI) on different stages of early hepatocarcinogenesis induced by weak promotion, SD rats initiated with a single intraperitoneal (i.p.) injection of N-diethylnitrosamine (DEN) at 200 mg/kg body weight and promoted with 0.5% piperonyl butoxide (PBO) in diet were repeatedly administered AI at 5 ml/kg body weight/day in the early postinitiation (EPI) or late postinitiation (LPI) period for 2 or 8 weeks, respectively. The number and area of glutathione S-transferase placental form (GST-P)-immunoreactive ⁽⁺⁾ foci tended to increase in the DEN+PBO group compared with the DEN-alone group. Among the PBO-promoted groups, number and area of GST-P⁺ foci did not visibly change in the DEN+PBO+AI-EPI group compared with the DEN+PBO group. In contrast, number and area of GST-P⁺ foci tended to decrease in the DEN+PBO+AI-LPI group compared with the DEN+PBO group. Number of Ki67⁺ cells was increased in the DEN+PBO group compared with the DEN-alone group and was decreased in both AI-administered groups compared with the DEN+PBO group. Gene expression analysis revealed that the DEN+PBO+AI-LPI group showed increased transcript levels of Ccne1, Cdkn1b, Rb1, Bax, Bcl2, Casp3, and Casp9 compared with the DEN+PBO group; however, the DEN+PBO+AI-EPI group did not show changes in the transcript levels of any genes examined compared with the DEN+PBO. These results suggest that AI administration during the LPI period caused weak suppression of hepatocarcinogenesis under weak promotion with a low PBO dose by the mechanism involving facilitation of cell cycle suppression causing G1/S arrest and apoptosis via the mitochondrial pathway. In addition, the results suggest that AI administration during the EPI period has no effect on weakly promoted hepatocarcinogenesis.

Introduction

Astragalus is the dried root of Astragalus membranaceus (Fisch.) Bge. Var. Mongholicus (Bge.) Hsia or Astragalus membranaceus (Fisch.) Bge., also known as Astragali Radix or Huangqi, and it has been widely used as a traditional Chinese medicine for treatment of various diseases for more than 2000 years. According to the information from traditional Chinese medicine, astragalus has effects that tonify Qi (Qi refers to the driving force for maintaining normal functions of the human body) and lifting Yang (Yang mainly refers to body function), strengthening the exterior (this is not body’s exterior) and reducing sweat, as well as promoting the secretion of saliva or body fluids. Astragalus comprises flavonoids, saponins, polysaccharides, various amino acids, and trace elements^{1, 2}. Previous studies have shown that astragalus increases telomerase activity and has antioxidant, anti-inflammatory, immune-regulatory, anticancer, hypolipidemic, antihyperglycemic, hepatoprotective, expectorant, and diuretic effects^{3, 4, 5, 6, 7, 8}.

Astragalus injection (AI) is prepared from astragalus by the following steps: water extraction, alcohol precipitation, alkali addition, and adsorption of pigments by activated carbon. Purified AI contains 18 active ingredients, including pratensein 7-O-glucopyranoside, calycosin-7-O-β-D-glucoside, astragaloside I, astragalin, 4´-hyadroxysoflavone-7-O-β D-glucoside, pratensein, pterocarpane-3-glucoside, methylnissolin-3-O-glucoside, isomucronulatol, isorhamnetin, isorhamnetin-3-O-gentiobioside, medicarpin, 6˝-acetyl-calycosin-7D-glucoside, astragaloside V, astragaloside VI, astragaloside VII, astragaloside IV, and lrioresinol B⁹. Studies on the anticancer effects of astragalus and its main components have been performed, and abundant evidence has shown that astragalus and its main components inhibit cell proliferation, inhibit oxidative stress, and induce apoptosis in various cancer cells^{10, 11, 12}. For example, astragalosides inhibit human colon cancer cell proliferation by inducing cell cycle arrest in the S and G2/M phases¹³. Astragalus mongholicus injection inhibits MCF-7 breast cancer cell proliferation by inducing G0/G1 and S phase arrest¹⁴. In addition, flavonoids from astragalus inhibit the development of breast cancer by promoting cancer cell apoptosis¹². Astragaloside IV inhibits hepatic stellate cell activation by inhibiting oxidative stress and associated p38 MAPK activation¹⁵. Furthermore, flavonoids from Astragalus complanatus induce human hepatocarcinoma cell apoptosis by increasing caspase-3, caspase-8, BAX, P21^CIP1/WAF1, and p27^KIP1 protein levels¹⁶. Moreover, we have found that the decoction products of Astragalus membranaceus have a certain inhibitory effect on the promotion stage of early hepatocarcinogenesis in SD rats (unpublished data). Overall, these studies suggest that astragalus inhibits the development of tumors by inhibiting cell proliferation and inducing apoptosis in vitro and in vivo. However, it remains unclear if AI has the same effects on hepatocellular proliferative lesions of early hepatocarcinogenesis in animal models.

The two isomers of 4-hydroxy-5-hydroxymethyl-[1,3]dioxolan-2,6’-spirane-5’,6’,7’,8’-tetrahydro-indolizine-3’-carbaldehyde (HDTIC)-1 and HDTIC-2 from astragali radix slow the telomere shortening rate by attenuating oxidative stress and increasing DNA repair ability in human foetal lung diploid fibroblast cells¹⁷. Limagne et al. reported that the active ingredients of astragalus have a repairing effect on DNA damage and that there is a close relationship between the cancerization of normal cells and DNA damage¹⁸. In addition, astragalus polysaccharide relieves oxidative damage induced by H₂O₂ by promoting APE/Ref-1 and TRX expression in damaged MRC-5 cells¹⁹. These studies suggest that astragalus inhibits DNA damage in cancer cells in vitro. Thus, astragalus and its main components not only relieve DNA damage but also inhibit cell proliferation of various tumor cells and induce cancer cell apoptosis. However, it remains unclear if AI has these effects on the initiation or promotion phase of early hepatocarcinogenesis in SD rats.

The present study was performed to examine the modifying effect of a clinically applied level of AI on different stages of early hepatocarcinogenesis, consisting of the early postinitiation (EPI) period showing transition from extensive liver cell injury after strong genotoxic stimulation by a tumor-initiation treatment to liver cell regeneration and the late postinitiation (LPI) period showing selective proliferation of initiated cells, using a rat two-stage hepatocarcinogenesis model. For this purpose, we selected two different postinitiation periods, i.e., the EPI period comprising the 2 weeks just after the tumor initiation treatment, showing transition from extensive liver cell injury by initiation treatment to liver cell regeneration, and the LPI period for following the 8 weeks following the EPI period. Because the present study was aimed at investigating the inhibitory effect of a clinically applied level of AI, we decided to create a cellular environment for tumor promotion to initiate the carcinogenic process as slowly as possible to mimic clinically relevant situations.

Materials and Methods

Chemicals

AI was purchased from Chiatai Qingchunbao Pharmaceutical Co., Ltd. (Hangzhou, China). N-diethylnitrosamine (DEN) was purchased from TCI Chemicals Co., Ltd. (Shanghai, China). Piperonyl butoxide (PBO) was purchased from Aladdin^® (Shanghai, China). Anti-glutathione S-transferase placental form (GST-P) pAb was purchased from Medical and Biological Laboratories Co., Ltd. (1:1000; Nagoya, Japan). Anti-Ki67 antibody [SP6] was purchased from Abcam (1:500; Shanghai, China). The real-time PCR primers listed in Table 1 were designed and synthesized by Thermo Fisher Scientific (Chengdu, China).

Table 1. Sequence of Primers Used for Real-time Reverse Transcription-polymerase Chain Reaction (RT-PCR) Analysis.

Experimental design

The protocol for this study was approved by the Animal Care and Utilization Committee of Southwest University (Chongqing, China). A total of 40 male SD rats aged 5 weeks were purchased from Chongqing Academy of Chinese Materia Medica (Chongqing, China). SD rats were housed in polycarbonate cages (3 or 4 rats per cage) with hardwood chips for bedding in a conventional animal facility and were maintained under conditions of controlled temperature (22 ± 2°C), humidity (55 ± 5%), ventilation (12 air changes per hour), and lighting (12 h light/dark cycle). SD rats were allowed free access to a basal diet (Ensiveier, Co., Chongqing, China) and tap water.

After a 1 week acclimatization period, animals were subjected to a tumor-initiation treatment, promotion treatment, and partial hepatectomy following the protocol of a medium-term liver carcinogenesis bioassay²⁰. Forty five-week-old male SD rats were divided randomly into the following 4 groups with 10 animals per group: DEN-alone, DEN+PBO, DEN+PBO+AI-EPI, and DEN+PBO+AI-LPI (Fig. 1). All animals received an intraperitoneal (i.p.) injection of DEN at a dose of 200 mg/kg body weight, and all animals in the DEN+PBO, DEN+PBO+AI-EPI, and DEN+PBO+AI-LPI groups were fed a diet containing 0.5% PBO for 8 weeks starting 2 weeks after DEN initiation²¹. To enhance hepatocellular proliferation, all animals were subjected to a two-thirds partial hepatectomy 1 week after PBO treatment. To investigate the suppressive effects of AI on early hepatocarcinogenesis in rats, animals of the DEN+PBO+AI-EPI group received i.p. injections of AI at a dose of 5 ml/kg every day for 2 weeks after DEN treatment. In addition, animals of the DEN+PBO+AI-LPI group received i.p. injections of AI at a dose of 5 ml/kg every day for 8 weeks with PBO treatment. Three rats died due to an insufficiency caused by hepatectomy treatment. Body weight and food consumption were measured once a week. At the end of the experiment, rats were euthanized by ether anesthesia, and livers were excised and weighed. Sliced liver samples were fixed in 10% phosphate buffered formalin for histopathology and immunohistochemistry analyses. The remaining liver pieces were frozen in dry ice and stored at −80°C until further analysis.

Fig. 1.

Experimental design.

Histopathology and immunohistochemistry

The fixed liver slices were dehydrated in graded ethanol, embedded in paraffin, sectioned, and stained with hematoxylin and eosin (HE) for histopathological examinations. Immunohistochemical staining of GST-P and Ki-67 was performed. Deparaffinized liver sections were treated with 0.3% H₂O₂ in methanol for 30 min to block endogenous peroxidase, and tissue sections were boiled in 0.01 M citrate buffer for antigen retrieval (for Ki67 only) and then blocked with 1% normal goat serum (SouthernBiotech, Birmingham, AL, USA) or horse serum (Vector Laboratories, Burlingame, CA, USA) for 30 min at room temperature. Sections were then incubated overnight at 4°C with anti-GST-P pAb (1:1,000 dilution; Medical and Biological Laboratories Co., Ltd., Nagoya, Japan) or anti-Ki67 antibody (1:500 dilution, Abcam, Shanghai, China). After incubation with a biotin-conjugated secondary antibody, sections were incubated with VECTASTAIN^® ABC Reagent (PK-6100, Vector Laboratories, Burlingame, CA, USA) and 3,3-diaminobenzidine (DAB; Aladdin^®, Shanghai, China). All specimens were lightly counterstained with hematoxylin. The numbers and areas of GST-P-positive (GST-P⁺) liver cell foci with a diameter ≥200 μm^{22, 23}, as well as total areas of liver sections, were measured using Sunny Digital microscope lab system computer software (Ningbo Sunny Instruments Co., Ltd., China). Ki67⁺ cells were examined in a total of 30 fields at 40×10 magnification in different regions (approximately 2,000–5,000 hepatocytes in each field) per animal with the free Katikati2 software, and cells with strongly positive nuclei were considered for the Ki67⁺ cell ratio.

RNA isolation and quantitative real-time RT-PCR for mRNA expression

Briefly, total RNA was isolated from four or five rats in each group using TRIzol reagent (Thermo Fisher Scientific, Walthan, MA) according to the manufacturer’s instructions. Total RNA was reverse transcribed using TheromoScript Reverse Transcriptase (SuperScript III First-Strand Synthesis System; Invitrogen, Thermo Fisher Scientific, Waltham, USA). All PCR assays were performed using SYBR Green Master Mix (Applied Biosystems, Thermo Fisher Scientific, Waltham, USA) and were performed under the following conditions using a C1000-CFX96 Real-Time PCR System (Bio-Rad Laboratories, Hercules, CA, USA): 50°C for 2 min, 95°C for 10 min, and 45 cycles for 15 seconds at 95°C and 60°C for 1 min. The relative differences in gene expression were calculated using threshold cycle (Ct) values normalized to Actb (endogenous control in the same sample) and then compared to a control Ct value using the 2^−ΔΔt method²⁴. Data represent the average fold changes with standard deviation.

Statistical analysis

All data are expressed as the mean ± standard deviation. The statistical significance of differences between the DEN alone and PBO-treated group or between the DEN+PBO group and AI-treated groups was determined by Dunnett’s test or Williams test. A P-value less than 0.05 was considered statistically significant.

Results

General condition and body and liver weights

Body weight gains were suppressed in the PBO-treated rats from 2 weeks after the partial hepatectomy to the end of the experiment (data not shown). Final body weight did not change between the DEN-alone and DEN+PBO groups (Table 2). Among PBO-promoted groups, the body weight of DEN+PBO+AI-LPI animals was significantly decreased compared with that of the DEN+PBO group. Regarding the liver weights, the absolute and relative liver weights in the DEN+PBO group were significantly increased compared with those of the DEN-alone group (Table 2). Among PBO-promoted groups, the DEN+PBO+AI-LPI group showed a decreased absolute weight compared with the DEN+PBO group. No changes were observed in both absolute and relative liver weights in the DEN+PBO+AI-EPI group compared with the DEN+PBO group. No significant difference in food consumption was observed between the DEN-alone and DEN+PBO groups or between the DEN+PBO and each of the AI-treated groups (data not shown).

Table 2. Body and Liver Weights.

Effects of AI treatment on histopathology, GST-P⁺ foci, and cell proliferation

Histopathological analysis showed that PBO-treated animals induced hepatocellular vacuolation, which was significantly reduced in AI-treated groups (data not shown). The number and area of GST-P⁺ foci in the DEN+PBO group showed an increasing tendency compared with those in the DEN-alone group (Fig. 2). Among PBO-promoted groups, number and area of GST-P⁺ foci did not visibly change in the DEN+PBO+AI-EPI group compared with the DEN+PBO group. In contrast, number and area of GST-P⁺ foci tended to decrease in the DEN+PBO+AI-LPI group compared with the DEN+PBO group. The effect of AI on cell proliferation was evaluated by immunohistochemistry for Ki67 (Fig. 3). The number of Ki67⁺ hepatocytes was significantly increased in the DEN+PBO group compared with the DEN-alone group and was decreased in DEN+PBO+AI-EPI and DEN+PBO+AI-LPI groups compared with the DEN+PBO group.

Fig. 2.

The effect of astragalus injection (AI) on the area and number of GST-P-positive (GST-P⁺) foci in the liver of rats given piperonyl butoxide (PBO) after N-diethylnitrosamine (DEN) initiation. Values are expressed as means ± SD for rats in the DEN-alone group (n=9), DEN+PBO group (n=9), DEN+PBO+AI-EPI group (n=9) and DEN+PBO+AI-LPI (n=10). EPI, early postinitiation; LPI, late postinitiation.

Fig. 3.

The effect of astragalus injection (AI) on the quantitation of Ki67⁺ cells. Values are expressed as means ± SD for rats in the DEN-alone group (n=9), DEN+PBO group (n=9), DEN+PBO+AI-EPI group (n=9) and DEN+PBO+AI-LPI (n=10). **Significantly different from the DEN-alone group by Dunnett’s test or Williams test (P<0.01). ^## Significantly different from the DEN+PBO group by Dunnett’s test or Williams test (P<0.01). EPI, early postinitiation; LPI, late postinitiation.

Transcript level of cell cycle-related genes

To evaluate the effect of AI treatment on the EPI and LPI phases in the two-stage hepatocarcinogenesis assay, liver tissues were subjected to real-time RT-PCR analysis. The main components of astragalus inhibit cancer cell proliferation¹⁵. Therefore, the mRNA expression of cell cycle checkpoint regulatory genes (Pcna, Ccnd1, Ccne1, Cdk2, Cdk4, Cdkn1b, and Rb1) was investigated (Fig. 4). The Pcna mRNA expression level in the DEN+PBO group was significantly increased compared with the DEN-alone group, but the level in the AI-treated groups tended to decrease compared with the DEN+PBO group. There were no mRNA expression changes of other genes between the DEN-alone and DEN+PBO groups. Among PBO-promoted groups, mRNA expression levels of Ccne1, Cdkn1b, and Rb1 were significantly increased in the DEN+PBO+AI-LPI group compared with the DEN+PBO group. There were no significant changes in the mRNA expression levels of Ccnd1, Cdk2, or Cdk4 between the DEN+PBO and AI-treated groups (Fig. 4).

Fig. 4.

The effect of astragalus injection (AI) on mRNA expression levels of cell cycle-related genes in the livers of SD rats given piperonyl butoxide (PBO) after N-diethylnitrosamine (DEN) initiation. Relative values of mRNA expression levels (normalized by Actb) are expressed as means ± SD. * Significantly different from the DEN group by Dunnett’s test or Williams test (P<0.05). ^# Significantly different from the DEN+PBO group by Dunnett’s test or Williams test (P<0.05). ^## Significantly different from the DEN+PBO group by Dunnett’s test or Williams test (P<0.01). EPI, early postinitiation; LPI, late postinitiation.

Transcript level of apoptosis-related genes

Previous studies have shown that the main components of astragalus induce cancer cell apoptosis¹³. Therefore, mRNA expression levels of apoptosis-related genes (Tp53, Bax, Bcl2, Casp3, Casp8, and Casp9) were examined in the liver tissue (Fig. 5). The mRNA expression levels of these genes in the DEN+PBO group were not significantly changed compared with the DEN-alone group. Among PBO-promoted groups, mRNA expression levels of Bax, Bcl2, Casp3, and Casp9 in the DEN+PBO+AI-LPI were significantly higher than those of the DEN+PBO group. In addition, there were no significant changes between the DEN+PBO and AI-EPI groups.

Fig. 5.

The effect of astragalus injection (AI) on mRNA expression levels of apoptosis-related genes in the livers of SD rats given piperonyl butoxide (PBO) after N-diethylnitrosamine (DEN) initiation. Relative values of mRNA expression levels (normalized by Actb) are expressed as means ± SD. * Significantly different from the DEN group by Dunnett’s test or Williams test (P<0.05). ** Significantly different from DEN group by Dunnett’s test or Williams test (P<0.01). ^#Significantly different from the DEN+PBO group by Dunnett’s test or Williams test (P<0.05). ^##Significantly different from the DEN+PBO group by Dunnett’s test or Williams test (P<0.01). EPI, early postinitiation; LPI, late postinitiation.

Discussion

As tumors are characterized by dysregulation of cellular proliferation, reducing abnormal cell proliferation plays an important role in host defence and protection from cancer development²⁵. Many studies have suggested that induction of cell cycle arrest, apoptosis, and anti-oxidative effects in cancer cells is a promising approach for the treatment of cancer^{26, 27, 28}. Astragalus membranaceus has been used for centuries in China to treat liver diseases. In recent years, it has been reported that astragalus or its active ingredients can inhibit the proliferation of hepatoma cell lines and induce apoptosis of cancer cells^{10, 29, 30}. We previously found that astragalus decoction products suppress GST-P⁺ foci and cell proliferation activity in the early stages of hepatocarcinogenesis in SD rats (unpublished data). Therefore, it is reasonable to clarify the possibility of AI as a chemopreventive agent of hepatocellular tumors. It is also important to elucidate whether AI exerts an inhibitory effect at the early stage of hepatocarcinogenesis and also the action stage of AI during early hepatocarcinogenesis. Based on the aforementioned information of AI as a candidate therapeutic model, the present study conducted an anticancer experiment using AI, a purified product consisting of 18 active ingredients extracted from astragalus, to confirm the action stage during the postinitiation period and the suppression mechanisms of AI in the early stage of hepatocarcinogenesis in SD rats.

Flavonoids from Astragalus complanatus reduce DNA injury and mutation in vitro³¹. In addition, HDTIC, an isomer extracted from Astragalus membranaceus, improves DNA repair ability¹⁷. These results suggest that astragalus has a protective effect on DNA damage. In contrast, the rhamnocitrin 4-β-D-galactopyranoside flavonoid obtained from leaves of astragalus prevents induction of hepatocellular carcinoma accompanying suppression of the marked increase in the levels of serum marker enzymes, and it suppresses free radicals by scavenging hydroxyl radicals³². In addition, swainsonine, an extract from Astragalus membranaceus, significantly inhibits MHCC97-H cell growth by causing cell cycle arrest at the G0/G1 phase and inducing apoptosis³³. In the present study, AI treatment in the LPI stage not only decreased the body weight but also weakly inhibited the number and area of GST-P⁺ foci weakly increased by PBO promotion in early hepatocarcinogenesis. It is reported that oral administration of 1 ml/kg AI for 28 days can inhibit body weight gain of exhaustively exercised rats by enhancing the activity of metabolic enzymes³⁴. In the present study, AI treatment reduced hepatocellular vacuolation, which suggested that it had an ameliorating effect on PBO-induced hepatotoxicity. Therefore, the decrease of body weight in the DEN+PBO+AI-LPI group may be due to increased metabolic activity rather than hepatocellular toxicity caused by AI, suggesting that the possibility of AI-induced toxicity to affect a tumor-promoting effect is rather low. In addition, relative liver weight in the DEN+PBO+AI-LPI group was unchanged compared with the DEN +PBO group. Therefore, a decrease in absolute liver weight may be accompanied by the body weight decrease due to increased metabolic activity.

In the present study, Ki-67 immunostaining showed that AI treatment in both the EPI and LPI periods significantly inhibited the Ki67⁺ cell ratio. In addition, the expression levels of PCNA in both AI-treated groups tended to decrease compared with the DEN+PBO group. The Ki67 nuclear antigen is a proliferative marker and is a good indicator of the proliferative and differentiation ability of carcinoma cells³⁵. These results indicated that AI has a strong inhibitory effect on cell proliferation during the early stage of hepatocarcinogenesis. However, we observed no influence of AI treatment during the EPI period on the area and number of GST-P⁺ foci promoted by PBO, in contrast to the weak suppression of tumor promotion by AI during the LPI period. However, it remains unclear whether the antiproliferative effect is involved in the anticancer mechanism of AI during the LPI period. An antiproliferative effect involving the GST-P⁺ liver cell population probably occurs as a result of AI-LPI but dose not occur in DEN-initiated cells as a result of EI-LPI. To clarify the suppression mechanism of AI on hepatocellular proliferative lesions during early hepatocarcinogenesis, further studies may be necessary on the temporal change in target cells of AI during the postinitiation period.

The present study showed that AI not only significantly inhibited the Ki67⁺ cell ratio in the liver but also reduced the expression level of Pcna increased by PBO promotion. Both PCNA and Ki67 are nuclear antigens in proliferating cells, and they are essential nuclear proteins for DNA synthesis in eukaryotic cells^{36, 37}. Ki67 accumulates in the late G1/S phase and rapidly degrades in the late phase of mitosis³⁸. The abnormal expression of PCNA can be used to evaluate the degree of malignancy and proliferation potential of tumors, and PCNA shows different expression levels in tumors of various stages of development. The expression of PCNA begins at the S phase and reaches the highest peak at the G1 phase, but it decreases at the G2/M phase^{39, 40}. Therefore, we further examined the expression levels of genes related to the cell cycle G1/S phase. As a result, we found upregulation of Cdkn1b (p27^KIP1), Rb1, and Ccne1 in the DEN+PBO+AI-LPI group compared with the DEN+PBO group. p27^KIP1, a cell cycle inhibitor, has been demonstrated to inhibit CDK4 or CDK2 bound to cyclin D1 or cyclin E, and the upregulation of the Cdkn1b gene induces G1 phase arrest⁴¹. In addition, Rb is a tumor suppressor that negatively regulates the G1/S transition of the cell cycle, and it is present in 70% of all tumors^{42, 43}. Our results suggest that AI may inhibit cell proliferation by inducing G1/S phase arrest in the promotion period during early hepatocarcinogenesis. On the other hand, cyclin E and CDK2 form a complex and can phosphorylate and inactivate Rb, resulting in E2F activation and a shortened G1 phase, thereby expelling cells into the S phase, which promotes cell proliferation⁴⁴. However, the mRNA expression level of CDK2 was not significantly increased in the present study. This result indicated that the upregulation of cyclin E may be due to an increase in the number of cells in the G1 phase resulting from the AI-induced G1/S phase arrest due to imbalance of CDK2 and cyclin E.

In the present study, we observed that increased gene expression levels of apoptosis-related genes (Bax, Bcl2, Casp3, Casp9) were observed by real-time RT-PCR. The BCL2 protein family comprises large apoptosis regulatory proteins, and increases of the BCL2 protein family member, BAX, lead to the release of cytochrome C and other proapoptotic molecules from the intermembranous space to the cytosol, resulting in activation of downstream caspases⁴⁵. In the mitochondrial-dependent apoptotic pathway, the membrane potential of mitochondria decreases significantly in the early stage of apoptosis, releasing cytochrome C in mitochondria, which cooperates with caspase-9 to further activate caspase-3⁴⁶. In addition, it has been reported that various anticancer drugs induce apoptosis by activating the caspase-9/caspase-3 pathway and increasing BCL2 protein family members in various cancer cells^{47, 48}. Although there was no significant difference, the number and area of GST-P⁺ foci in the DEN+PBO-AI-LPI group showed a tendency to decrease compared with the DEN+PBO group, suggesting that AI treatment during the LPI period contributes to facilitation of the mitochondrial-dependent apoptotic pathway in GST-P⁺ foci. Furthermore, AI treatment inhibited the Ki-67⁺ cell ratio increased by PBO promotion, possibly as a result of AI-induced G1/S arrest. The obtained results in the present study indicated that AI induces hepatocyte apoptosis and G1/S arrest in the LPI (promotion) period of hepatocarcinogenesis and that these effects may be closely linked to events leading to inhibition in the development of preneoplastic lesions. Further study on the temporal relationships of apoptosis, cell proliferation, and cell cycle-related molecules with GST-P⁺ foci will be necessary to be addressed to clarify the AI-induced suppression effect on hepatocarcinogenesis.

In conclusion, the present study investigated the suppression effects of AI on different stages of early hepatocarcinogenesis initiated with DEN and promoted with PBO. The overall data suggested that AI treatment in the promotion stage of early hepatocarcinogenesis can inhibits cell proliferation activity and induces apoptosis through facilitation of G1/S phase arrest and caspase-9/3 pathway activation, leading to suppression the development of preneoplastic lesions on the promotion period. In addition, treatment with AI during the early postinitiation period can inhibit cell proliferation but cannot suppress the hepatocarcinogenesis. Further studies are necessary to clarify the potential suppression mechanism of AI in early hepatocarcinogenesis.

Disclosure of Potential Conflicts of Interest

We have no conflicts of interest to be declared.