Histone deacetylase 3 participates in self-renewal of liver cancer stem cells through histone modification

Cancer Letters 339 (2013) 60–69

Contents lists available at SciVerse ScienceDirect

Cancer Letters journal homepage: www.elsevier.com/locate/canlet

Histone deacetylase 3 participates in self-renewal of liver cancer stem cells through histone modification Chungang Liu a,c,1, Limei Liu a,1, Juanjuan Shan a, Junjie Shen a, Yanmin Xu a, Qianzhen Zhang a, Zhi Yang a, Lin Wu b, Feng Xia b, Ping Bie b, Youhong Cui a, Xia Zhang a, Xiuwu Bian a, Cheng Qian a,⇑ a b c

Institute of Pathology and Southwest Cancer Center, Southwest Hospital, Third Military Medical University, Chongqing, China Institute of Hepatobiliary Surgery, Southwest Hospital, Third Military Medical University, Chongqing, China Department of Pathology, School of Medicine, Zhejiang University, Hangzhou, China

a r t i c l e

i n f o

Article history: Received 22 April 2013 Received in revised form 11 July 2013 Accepted 15 July 2013

Keywords: Cancer stem cells Self-renewal TSA HDAC3 Histone modification HCC

a b s t r a c t Understanding molecular mechanisms in self-renewal of cancer stem cells (CSCs) is important for finding novel target in therapy of cancer. In this study, we explored potential effects of histone deacetylase (HDAC) on liver CSCs. Our data showed that HDAC inhibitors suppressed self-renewal and induced differentiation of liver CSCs. Furthermore, we demonstrated that HDAC3 was selectively expressed in liver CSCs and participated in self-renewal of liver CSCs via regulating expression of pluripotency factors. Overexpression of HDAC3 was associated with poor outcome of liver cancer. HDAC inhibitors could render liver CSCs sensitive to sorafenib. Taken together, our data suggest that HDAC3 plays a critical role in regulating self-renewal of liver CSCs. Ó 2013 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Hepatocellular carcinoma (HCC) is one of the most frequent malignancies with high mortality rate [1]. A recent concept about tumorigenesis is cancer stem cells (CSCs). In the CSCs model, there is a small subset of CSCs which constitute a reservoir of self-sustaining cells with the exclusive ability to self-renew and maintain the tumor. These CSCs have the capacity of expanding the CSCs pool and differentiating into the heterogeneous nontumorigenic cancer cell types that in most cases appear to constitute the bulk of the cancer cells within the tumor [2]. It has been shown that high-grade tumors are enriched with a high content of CSCs [3]. In liver cancer, CSCs have been isolated by several cell surface antigens such as CD133 [4], CD90 [5], EpCAM [6], CD13 [7] and CD24 [8]. It has been demonstrated that these CSCs are responsible for tumor initiation, resistance to therapeutic regimens and tumor recurrence after surgical removal of primary tumors [2,9,10]. CD24 molecule can drive self-renewal and tumor initiation of liver CSCs though STAT3-mediated Nanog regulation [8]. Our previous study has identified transcription factor Nanog as a novel marker for liver CSCs and demonstrated that Nanog ⇑ Corresponding author. Tel.: +86 23 68765957; fax: +86 23 68752247. 1

E-mail address: [email protected] (C. Qian). These authors contributed equally to this work.

0304-3835/$ - see front matter Ó 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.canlet.2013.07.022

plays an important role in regulating self-renewal of liver CSCs via IGF signaling pathway [11]. However, the key components and molecular mechanisms contributing to self-renewal of CSCs are largely unknown. Previous studies have indicated that chromatin structure including histone modifications and DNA methylation are features of cancers, early embryonic development, and differentiation of stem cells [12–16]. Histone acetylation is a dynamic process resulting from the balance between histone acetyltransferases and histone deacetylases (HDACs) [16]. Carcinogenesis is associated with a relative decrease in histone acetylation by numerous molecular mechanisms [16]. It has been reported that the altered expression of HDACs is associated with a number of human cancers including HCC and HDACs play a critical role in the development of cancers [17–20]. Epigenetic regulation by histone lysine acetylation plays a key role in regulating self-renewal capability of embryonic and adult stem cells [12,21]. HDAC inhibitors can induce inhibition of proliferation and induction of apoptosis in many types of cancer cell [22,23]. Some HDAC inhibitors are being tested clinically as anticancer agents for the treatment of leukemia and a variety of solid tumors [24]. However, little is known about the function of HDAC inhibitors on liver CSCs and which member of HDAC family to regulate self-renewal of liver CSCs. In this study, we investigated the effect of HDAC inhibitors on liver CSCs. We found that HDAC3 participated in the self-renewal

C. Liu et al. / Cancer Letters 339 (2013) 60–69

61

of liver CSCs through histone modification and HDACs inhibitors rendered liver CSCs sensitive to the therapy of sorafenib.

Immunohistochemical staining was performed as previously described [11]. The clinical and pathological characteristics of the patients were summarized (Supplementary Table 1).

2. Materials and methods

2.8. Cell proliferation assay

2.1. Tissue samples and cell lines

Cell proliferation was assessed by using CCK-8 assay according to the manufacturers’ recommendations. Briefly, cells were plated at 1000 cells/well on a 96-well plate. The cells were treated with HDAC inhibitors at different concentrations in DMEM medium with 10% of FBS and the cell survival was measured at different time points after treatment.

Fresh and paraffin fixed tumor specimens were obtained with informed consent from all patients according to the protocols approved by the Institutional Review Board of the Southwest Hospital, Third Military Medical University. All patients underwent surgical resection of primary HCC at the Institute of Hepatobiliary Surgery, Southwest Hospital, Third Military Medical University. The human HCC cell lines (Huh7 and PLC/PRF/5) were purchased from Shanghai Cell Collection (Shanghai, China). Patient-derived primary HCC cultures of tumor cells (T1115 and T1224) were prepared in our laboratory as described previously [11]. All cells were cultured in DMEM (Gibco) medium with 10% FBS (Gibco) at 37 °C in a humidified atmosphere containing 5% CO2. 2.2. Cell sorting by flow cytometry HCC cell lines and patient-derived primary HCC cultures of tumor cells were infected with Lv-PNanog-GFP at MOI of 10 as described previously [11]. GFP expression in Lv-PNanog-GFP infected cells was analyzed by flow cytometry. Dead cells were excluded by 7-AAD staining. Top high (<5%) and top low (<5%) were chosen for sorting NanogPos and NanogNeg cells. To obtain CD133+ cells, HCC cells were marked with PE-labeled anti-CD133 antibody (Miltenyi Biotec) at 4 °C for 15 min. Top high (<5%) and top low (<5%) were chosen for sorting CD133+ and CD133 cells. The stained cells were analyzed and sorted with FACS Aria II (BD Biosciences). Purity of the sorted cells was over 99%. 2.3. Sphere formation assay A total of 200 the sorted NanogPos, NanogNeg or CD133Pos, CD133Neg cells were plated into CostarÒ Ultra Low Cluster 24-well plates (Corning). The cells were cultured in the DMEM/F12 medium (Sigma) supplemented with B27 (Gibco), antibiotics, 20 ng/mL EGF, 20 ng/mL bFGF (Peprotech) and 10 ng/mL HGF (Peprotech) in the absence or presence of different concentrations of HDAC inhibitors, and 1% methyl cellulose was added to prevent cell aggregation. Cells were incubated at 37 °C for 14 days and numbers of spheres were counted. 2.4. Colony formation assays Briefly, 1  104 cells were seeded in 10-cm tissue culture plates or 200 cells were seeded in 24 well plates. The cells were cultured in the DMEM medium supplemented with 10% of FBS in the absence or presence of different concentrations of HDAC inhibitors for 14 days. The colonies were fixed with 4% formaldehyde and stained with 0.1% crystal violet (Sigma–Aldrich). Numbers of clones were counted.

2.9. Reverse transcription PCR analysis Total RNA was extracted from the cells with RNAiso (Takara) according to the manufacturer’s protocol. For mRNAs detection, reverse transcription was performed according to the protocol of RevertAid™ First Strand cDNA Synthesis Kits (Fermentas); qPCR was performed with SYBR premix Ex Taq (TaKaRa) on an Applied Biosystems 7300 Real Time PCR System supplied with analytical software (Applied Biosystems, USA). GAPDH mRNA was used to normalize RNA inputs. Primers for markers of stem cells and mature hepatocytes were shown at Supplementary Table 2. Data were processed by using the 2DDCt method. The results were represented as the means ± SD of three independent experiments. 2.10. Immunofluorescence analysis Cells were grown on glass cover-slips in a six-well plate and washed three times with PBS before fixed in 4% paraformaldehyde and permeabilized with 0.1% Triton X-100 in PBS for 5 min at room temperature (RT). Cells were blocked with 10% FBS (Gibco) in PBS for 30 min at RT. Cover-slips were then incubated with respective primary antibodies (Supplementary Table 3). Secondary antibodies were donkey anti-rabbit IgG-Alexa Fluor 647, or donkey anti-mouse IgG-Alexa Fluor 647 (Invitrogen). Cells were further washed in PBS and mounted with vectashield mounting medium containing 40 ,6-diamidino-2-phenylindole (DAPI) for counterstaining nuclei. Cells were analyzed by using fluorescence microscopy. 2.11. Western blot analysis Cells were harvested and lysed in the lysis buffer for 30 min at 4 °C. Total cell extracts were separated in 12% SDS–polyacrylamide gel electrophoresis (PAGE) and then transferred on PVDF membrane (Millipore). The membrane were then blocked in 5% milk for 2 h at RT and blotted with antibody overnight at 4 °C. Antibodies used in the study were listed in Supplementary Table 3. After washing with phosphate buffered saline with Tween-20 (PBST) and incubating with either antirabbit IgG or anti-mouse IgG horseradish peroxidase-conjugated secondary antibody at a dilution of 1:2000 in PBST, immunocomplexes were visualized by using SuperSignal West Femto Chemiluminescent Substrate (Pierce). For quantification, signals were densitometrically normalized to GAPDH by GeneTools image analysis program (SynGene).

2.5. Tumor formation assay SCID mice at age of 3–5 weeks, male, were maintained in pathogen-free conditions at animal facility of Third Military Medical University and received humane care according to the criteria outlined in the ‘‘Guide for the Care and Use of Laboratory Animals’’ prepared by the National Academy of Sciences. The different numbers of NanogPos after treatment with HDAC3 siRNA or scramble siRNA were resuspended in serum-free medium and mixed with Matrigel at the ratio of 1:1. The cells were subcutaneously injected into SCID mice. Tumor formation was evaluated regularly after injection by palpation of injection sites for 3 months.

2.12. Statistical analysis A Student’s t test was used to calculate the statistical significance of the experimental data. The Kaplan–Meier survival curves and log-rank test were used for estimation of survival and difference between groups. The level of significance was set as * P < 0.05 and **P < 0.01. All data were presented as the means ± standard deviation (SD). The software tools SPSS 10.0 and Microsoft Excel were used.

3. Results 2.6. Induction of HCC with diethylnitrosamine (DEN) Male Wistar rats (6 weeks old, 170 g) were maintained in pathogen-free conditions at the animal facility of Third Military Medical University and received humane care according to the criteria outlined in the ‘‘Guide for the Care and Use of Laboratory Animals’’ prepared by the National Academy of Sciences. An acclimatization period of 4 days was carried out. The weight of the rats was recorded every week. Animals received 10 mg/kg/day of DEN (Sigma) for 24 weeks. Rats were given the weekly dose of DEN in drinking water (0.01% v/v) corresponding to the estimated water consumption of 6 days. Once the animals consumed the administered DEN solution, they were given DEN-free water for the rest of the week. DENA solution was prepared each week. 2.7. Immunohistochemical staining Tissue specimens were obtained with informed consent from 76 patients who underwent hepatectomy for HCC at the Institute of Hepatobiliary Surgery, Southwest Hospital, Third Military Medical University, China. A tissue array block containing both HCC and non-HCC samples from these patients was constructed.

3.1. Inhibition of cell growth and self-renewal of liver CSCs by HDAC inhibitors To understand whether the altered HDAC activity regulated cell growth and self-renewal of liver CSCs, we tested the effect of HDAC inhibitors (TSA and SAHA) on the liver CSCs isolated from HCC cells either by CD133 biomarker or by our newly established method to obtain pluripotency transcription factor Nanog-positive liver CSCs [11]. Our results showed that TSA could inhibit cell growth of NanogPos or CD133Pos cells isolated from HCC cell line Huh7 and patient-derived primary HCC cultures of tumor cells T115 in doseand time-dependent manners (Fig. 1A and B, and Supplementary Fig. 1A). Similarly, TSA could also inhibit cell growth of NanogNeg cells from Huh7 cell line (Supplementary Fig. 2A). There was higher expression of proliferation marker Ki-67 in NanogPos cells than

62

C. Liu et al. / Cancer Letters 339 (2013) 60–69

NanogNeg cells. Ki-67 was dramatically reduced in NanogPos cells after treatment with TSA (Fig. 1C). However, apoptotic markers cleaved PARP and cleaved caspase-3 were not changed in NanogPos cells after treatment with TSA (Fig. 1D). Cell cycle analysis showed that treatment of NanogPos cells with TSA led to reductions of S and G2/M phases and increase of G0/G1 phase (Fig. 1E). These data indicate that the reduced cell growth of liver CSCs induced by TSA is due to the reduced cell proliferation mediated by cell cycle arrest, but not due to cell apoptosis. Furthermore, we examined the effect of TSA and SAHA on the self-renewal of NanogPos or CD133Pos cells. Our data showed that both of TSA and SAHA suppressed sphere formation efficiency (Fig. 2A, and Supplementary Fig. 1B and 3A) and clone formation efficiency (Fig. 2B, and Supplementary Fig. 1C and 3B) at the doses of 50 and 200 nM of TSA and 5 and 10 lM of SAHA. It has been demonstrated that low dose of TSA (10 nM) is capable of maintaining the pluripotent state of embryonic stem cell [13]. However, our results showed that treatment of NanogPos and NanogNeg cells with low dose of TSA (10 nM) had no significant effect on GFP expression (Supplementary Fig. 4A and B) and volume of sphere (Supple-

mentary Fig. 4C), as well as sphere formation efficiency (Fig. 2A). Similarly, low dose of SAHA (1 lM) had no effect on sphere formation efficiency of NanogPos cells (Supplementary Fig. 3A). Collectively, our data indicate that HDAC activity is essential for maintaining growth and self-renewal of liver CSCs. 3.2. Induction of differentiation of liver CSCs by TSA through histone modification We further investigated molecular mechanisms underlying the self-renewal of liver CSCs mediated by TSA. Our data showed that stem cell markers (SOX2, OCT4, Bmi-1 and c-Myc) were dramatically reduced in NanogPos cells after treatment with TSA, whereas mature hepatocyte marker albumin was significantly increased (Fig. 2C). Quantitative PCR analysis further confirmed that mature hepatocyte markers increased and stem cell markers decreased after treatment with TSA (Fig. 2D), indicating that TSA can induce liver CSCs differentiation. Next, we detected the level of acetylated histones H3 (H3Ac) and H4 (H4Ac) in NanogPos and NanogNeg cells or NanogPos cells after treatment with TSA. Our results showed that

Fig. 1. TSA inhibits cell growth of liver CSCs. (A and B) NanogPos CSCs from Huh7 and T1115 cells were cultured with TSA at different concentrations or ethanol as control and cell survival was measured by CCK8 assay. NanogNeg cells were used as a negative control. The data was presented as the mean ± SD from three independent experiments. (*p < 0.05, **p < 0.01). (C) NanogPos CSCs from Huh7 were treated with TSA (200 nM) and ethanol (0.05%) as control for two days and were then stained with antibody against Ki-67. Nuclei were counterstained by DAPI. All experiments were performed in triplicate and representative images were shown (scale bar: 100 lm). (D) Levels of apoptotic markers (cleaved PARP and cleaved caspase-3) were determined by western blot in lysates from NanogPos CSCs from Huh7 after treatment with TSA (200 nM) and ethanol (0.05%) for 2 days. (E) Cell cycle analysis was performed in NanogPos CSCs from Huh7 after with TSA (200 nM) and ethanol (0.05%) for 2 days. The cells were stained with PI and followed by FACS analysis.

C. Liu et al. / Cancer Letters 339 (2013) 60–69

levels of H3Ac and H4Ac were higher in NanogNeg cells than in NanogPos cells (Fig. 2E). Treatment of NanogPos cells with TSA increased the levels of H3Ac and H4Ac (Fig. 2E). Overall, our data suggest that HDAC inhibitor may induce differentiation of liver CSCs through histone modification.

3.3. Participation of HDAC3 in the self-renewal of liver CSCs To determine which member of HDAC family plays an important role in the self-renewal of liver CSCs, we examined expression profiles of HDACs in NanogPos or NanogNeg cells. As shown in Fig. 3A, HDACs were differentially expressed in NanogPos and NanogNeg cell from HCC cell line and patient-derived primary HCC cultures of tumor cells. Since HDAC activity is essential for maintaining growth and self-renewal of liver CSCs, we only concentrate to those HDACs whose expressions are up-regulated in NanogPos cells. By analysis of the changed expression of HDACs in these three HCC cells, we found that five HDACs were up-expressed in NanogPos cells. Of these five HDACs, expression of HDAC3 and HDAC7 were higher in NanogPos cells than NanogNeg cell in these three HCC cells (Fig. 3B). Next, we examined the correlation between expression of HDAC3 or HDAC7 and Nanog or CD133 as the biomarkers of liver CSCs in ten HCC tissues by quantitative PCR. Our results showed that HDAC3 expression was significantly correlated with both of Nanog and CD133 expression. However, there was no correlation between expression of HDAC7 and Nanog and CD133 (Fig. 3C). Furthermore, we focused on analyzing HDAC3 expression in NanogPos and NanogNeg cells by western blot and

63

immunofluorescence staining. Our results showed that HDAC3 expression was higher in NanogPos cells than NanogNeg cells (Fig. 3D and E). The higher expression of Nanog, OCT4 and SOX2 were also observed in NanogPos cells than NanogNeg cells (Fig. 3D). In order to determine whether HDAC3 contributes to the self-renewal of liver CSCs, we knocked down HDAC3 expression by a specific siRNA against HDAC3. Our data showed that HDAC3 was dramatically reduced in the NanogPos CSCs after treatment with HDAC3-siRNA (Fig. 4A, and Supplementary Fig. 5). However, treatment with HDAC3-siRNA did not affect expression of other members of HDAC family (Supplementary Fig. 5). Knock-down of HDAC3 in the NanogPos CSCs inhibited the expression of proliferation marker Ki-67 (Fig. 4A) and did not alter levels of cleaved PARP and cleaved caspase 3 (Fig. 4B). Moreover, our data showed that knock-down of HDAC3 in the NanogPos CSCs suppressed both of sphere formation (Fig. 4C) and clone formation (Fig. 4D). Tumor initiation assay showed that tumorigenicity was dramatically reduced in the NanogPos CSCs after knock-down HDAC3 expression (Fig. 4E). These data indicate that HDAC3 regulates self-renewal of liver CSCs.

3.4. Changes in histone modifications and gene expression in liver CSCs upon the knock-down of HDAC3 expression In order to know whether knock-down of HDAC3 expression could alter histone modification and gene expression, we examined status of histone modification and gene expression of stem cell markers in liver CSCs. Our results showed that knock-down of HDAC3 expression increased the levels of H3Ac and H4Ac in liver

Fig. 2. TSA inhibits self-renewal and induces of differentiation of liver CSCs through histone modification. (A and B) Sphere and clone formation efficiencies were determined in NanogPos CSCs from Huh7 and T1115 in the presence of TSA (10–200 nM) or ethanol (0.05%) in suspension condition or conventional condition. Numbers of spheres (A) and clones (B) were counted at the 14th day after plating. Sphere and clone formation efficiencies were presented as the percentages of numbers of spheres or clones per total numbers of plated-cells and shown as the mean ± SD from three independent experiments. (**p < 0.01). (C and D). NanogPos CSCs from Huh7 were cultured with TSA (200nM) or ethanol (0.05%) as control for 2 days. Stem cell- and mature hepatocyte-markers were determined by immunofluorescence staining (C) and quantitative PCR (D). The data was presented as the mean ± SD of three independent experiments. (*p < 0.05, **p < 0.01). (E) Immunostaining analysis of acetylated histone H3 (H3Ac) and H4 (H4Ac) in NanogPos CSCs after treatment with TSA (10–200 nM) for 2 days or ethanol (0.05%) as control. NanogNeg cells were used as a negative control. The treated cells were stained with anti-H3Ac or H4Ac antibody. Nuclei were counterstained by DAPI. (Scale bar: 100 lm). Representative photomicrographs were presented from two independent experiments.

64

C. Liu et al. / Cancer Letters 339 (2013) 60–69

Fig. 3. HDAC3 is highly expressed in liver CSCs-enriched populations. (A) Relative expression levels of HDAC1-11 in NanogPos and NanogNeg cells from Huh7, T1115 and T1224 were determined by quantitative PCR. The data was presented as the mean ± SD of three independent experiments (left panel). (*p < 0.05, **p < 0.01). The higher expression level of HDACs in NanogPos than NanogNeg cells was summarized (right panel). (B) Venn diagram shows overlap of the overexpressed HDACs in NanogPos cells from Huh7, T1115 and T1224. (C) Expression of HDAC3 and markers of CSCs (Nanog, OCT4 and SOX2) in NanogPos and NanogNeg cells from Huh7 and PLC/PRF/5 cells by western blot analysis. (D) Expression of HDAC3 in NanogPos and NanogNeg cells from Huh7 and T1115 was determined by immunofluorescence. Representative photomicrographs were presented from three independent experiments. (Scale bar: 100 lm).

CSCs (Fig. 4A). Furthermore, knock-down of HDAC3 expression in liver CSCs resulted in increase of H3K56Ac and H3K9Me3 and de-

crease of H3K27Me3, whereas the levels of H3K9Ac, H3K14Ac, H4K5Ac, H4K16Ac and H3K4Me3 were unaffected (Fig. 4F). In

C. Liu et al. / Cancer Letters 339 (2013) 60–69

65

Fig. 4. HDAC3 participates in the self-renewal of liver CSCs via changes of histone modifications and gene expression. (A) Expression of HDAC3 and levels of Ki-67, H3Ac and H4Ac were determined in NanogPos CSCs from Huh7 after treatment with HDAC3 siRNA and scramble siRNA by immunofluorescence. Representative photomicrographs were presented from three independent experiments. (Scale bar: 100 lm). (B) Levels of cleaved caspase-3 and PARP protein in NanogPos CSCs from Huh7 cells after treatment with HDAC3 siRNA and scramble siRNA were determined by western blot. (C and D) Sphere and clone formation efficiencies were determined in NanogPos CSCs from Huh7 after treatment with HDAC3 siRNA and scramble siRNA. Sphere (C) and clone (D) formation efficiencies were presented as percentages to scramble siRNA and shown as the mean ± SD from three independent experiments. (**p < 0.01). (E) NanogPos CSCs from Huh7 after treatment with HDAC3 siRNA and scramble siRNA were injected subcutaneously into SCID mice (n = 6 for each group). Tumor formation was observed until 3 months after implantation. (F) Levels of histone lysine acetylation and methylation were determined by western blot analysis of nuclear extracts prepared from NanogPos CSCs from Huh7 after treatment with HDAC3 siRNA and scramble siRNA. (G) Expression of stem cell genes was determined in NanogPos CSCs from Huh7 after treatment with HDAC3 siRNA and scramble siRNA by western blot analysis.

addition, knock-down expression of HDAC3 in liver CSCs resulted in the decreased expression of stem cell markers including Nanog, OCT4 and SOX2 (Fig. 4G). Collectively, our results indicate that HDAC3 plays an important role in the self-renewal of liver CSCs through both histone modifications and alternation of stemness gene expression. 3.5. Association of high level of HDAC3 expression with poor prognosis of patients with HCC To investigate whether HDAC3 expression has clinical implications in human cancers, we examined HDAC3 expression in 76 paraffin-embedded primary human HCC tissues by immunohistochemistry. Expression of HDAC3 was observed in 56 out of 76 cases in HCC tissues, whereas no HDAC3 expression was observed in non-HCC counterparts. HDAC3 protein was localized in the nuclei of cancer cells (Fig. 5A). Among 56 HDAC3-positive cases, there were 23, 13 and 20 cases with weak, moderate and strong HDAC3 expression, respectively (Fig. 5B). No significant association was found between HDAC3 expression and other parameters such as age, gender, tumor size, AFP level and liver injury (Supplementary

Table 1). Interestingly, we observed that HDAC3 expression was positively correlated with tumor stages and tumor recurrence (Supplementary Table 1). Kaplan–Meier analysis showed that HDAC3 expression in HCC was significantly correlated with overall and disease-free survival (Fig. 5C and D). These results indicate that high expression of HDAC3 correlates with poor prognosis of HCC patients. Furthermore, we investigated dynamic expression of HDAC3 in DEN-induced hepatocarcinogenesis. There was increased expression of HDAC3 in livers after 8 weeks of induction. At 24 weeks of induction, liver tumor was formed and expression of HDAC3 was dramatic increased in liver tumors (Supplementary Fig. 6). These data suggest that the increased expression of HDAC3 might be associated with hepatocarcinogenesis. 3.6. HDACs inhibitors render liver CSCs sensitive to therapy of sorafenib Our previous study has demonstrated that liver CSCs are resistant to sorafenib [11], an oral multikinase inhibitor which is used currently for treatment of HCC [25]. In order to know whether HDAC inhibitors could increase sensitivity of liver CSCs to sorafe-

66

C. Liu et al. / Cancer Letters 339 (2013) 60–69

Fig. 5. Expression of the HDAC3 in human HCC. (A and B) Expression of HDAC3 was analyzed by IHC in the paraffin-embedded formalin-fixed HCC tissues and paired nonHCC counterparts. Representative photomicrographs show negative () staining in non-HCC tissues and negative (), weak (+), moderate (++) and strong (+++) staining in HCC tissues. Original magnification was 200. (B) Various levels of HDAC3 expression in 76 HCC patients were summarized. (C and D) Kaplan–Meier curve for overall survival and disease-free survival were compared according to HDAC3 expression in HCC tissues.

nib, we treated liver CSCs by combination of HDAC inhibitors and sorafenib. Our data showed that survival of liver CSCs was significantly reduced when both of HDAC inhibitors TSA and SAHA at low dose was combined with low dose of sorafenib, as compared with single treatment alone (Fig. 6A and B). Furthermore, our results showed that treatment of liver CSCs with TSA, SAHA or sorafenib could induce drug resistant clones. In contrast, treatment of liver CSCs by combination of either of TAS or SAHA with sorafenib could completely abolish drug resistant clones (Fig. 6C). The complete elimination of drug resistant clones was also observed when liver CSCs were treated firstly with TSA or SAHA for 9 days and followed by 6 days of treatment with sorafenib. However, treatment of liver CSCs firstly with sorafenib for 9 days and followed by either of TAS or SAHA for another 6 days could only slight reduce drug resistant clones, as compared with treatment with single agent (Fig. 6D). To confirm whether the effect of HDAC inhibitors on drug resistance was through HDAC3, we applied sorafenib on the liver CSCs of which HDAC3 were knockdown. As shown in Fig. 6E, HDAC3 knockdown could abolish drug resistant clones in liver CSCs. Thus, our data indicate that HDAC inhibitors or HDAC3 inhibition could render liver CSCs sensitive to therapeutic agent sorafenib.

4. Discussion According to the CSCs hypothesis, CSCs are responsible for tumor initiation, invasion and metastasis, treatment resistance and relapse after therapy [2,9,10]. Thus, understanding molecular mechanisms and finding key components on regulation of CSCs have great implications for controlling cancers. It has been demonstrated that altering chromatin conformation by histone acetylation is involved in cancer development and biology of stem cells [12,14,16]. Thus, we postulate that histone acetylation is involved in the biology of CSCs.

In this study, we firstly studied the effect of HDAC inhibitors on the biology of NanogPos liver CSCs. Our data showed that HDAC inhibitors could inhibit the cell growth of liver CSCs in dose- and time- dependent manners. Inhibition of the cell growth of liver CSCs was due to the inhibition of cell proliferation mediated by cell cycle arrest, but not by cell apoptosis. However, previous studies have demonstrated that HDAC inhibitors could inhibit cell proliferation and induce cell apoptosis in HCC cells [26]. Since CSCs are only a small fraction of cell population, apoptotic effect of HDAC inhibitors may act on non-CSCs which exist as the majority of cell population. Our result confirmed that HDAC inhibitor could induce apoptosis of NanogNeg non-CSCs (Supplementary Fig. 2). Furthermore, we observed that treatment of CSCs with HDAC inhibitors could inhibit the self-renewal of CSCs. Previous studies have used different molecular markers to isolate liver CSCs [27]. Our data further confirm that HDAC inhibitor has similar effect on the liver CSCs isolated by CD133 marker. Furthermore, our data showed that TSA could induce liver CSC differentiation by upregulation of mature hepatocyte markers and downregulation of stem cell markers, which is consistent with the evidence that high dosage of TSA induced ESC differentiation [14]. As compared to non-CSCs with high levels of acetylated histones H3 (H3Ac) and H4 (H4Ac), liver CSCs exhibited low levels of H3Ac and H4Ac. Treatment with TSA could increase levels of H3Ac and H4Ac, suggesting that differentiation of liver CSCs was attributed to the increase of histone acetylation. A recent study has shown that HDAC inhibitors could stimulate expansion of breast CSCs through dedifferentiation of non-CSCs [28]. However, we did not observe this phenomenon in both NanogPos liver CSCs and NanogNeg non-CSCs that were treated with high and low doses of HDAC inhibitors. Low doses of TSA (10 nM) and SAHA (1 lM) did not the affect self-renewal of liver CSCs. Treatment with NanogNeg non-CSCs with TSA resulted in inhibition of cell growth, induction of cell apoptosis and inhibition of sphere and colony formation efficiency (Supplementary Fig. 2). Previous study showed that low

C. Liu et al. / Cancer Letters 339 (2013) 60–69

67

Fig. 6. HDACs inhibitors render liver CSCs sensitive to therapy of sorafenib. (A and B) Cell survival of NanogPos CSCs from Huh7 was treated with sorafenib (2.5 lM), TSA (50 nM), SAHA (2 lM) alone, and the combination of sorafenib and TSA (A) or SAHA (B). Cell viability was determined by CCK-8 assay at 48 h after treatment. Data was presented as the means ± SD of three independent experiments (*p < 0.05, **p < 0.01). (C) For drug resistant clone formation assay, NanogPos CSCs from Huh7 were either untreated or treated with sorafenib, TSA or SAHA alone or sorafenib plus TSA or SAHA for 15 days. (D) NanogPos CSCs from Huh7 were treated sorafenib for first 9 days and then TSA or SAHA for additional 6 days; or TSA or SAHA for 9 days and then sorafenib for additional 6 days. (E) NanogPos CSCs from Huh7 were treated either with scramble siRNA, HDAC3 siRNA, scramble siRNA plus sorafenib or HDAC3 siRNA plus sorafenib for 15 days. All experiments were performed in triplicate and representative images were shown.

dose of HDAC inhibitors promoted the self-renewal of ESCs and high dose induced the differentiation of ESCs [14]. Such discrepancy might be due to the difference of the resources of cell types and the defined doses of HDAC inhibitors. To identify which member of HDAC family is responsible for maintaining liver CSCs, we examined the expression patterns of HDACs. Our data showed that HDAC3 and HDAC7 were highly expressed in the liver CSCs. Furthermore, we found a significant correlation between HDAC3 expression and liver CSCs markers. To understand the function of HDAC3 in liver CSCs, we inhibited HDAC3 expression specifically by siRNA. Our data showed that knock-down of HDAC3 expression inhibited both proliferation and self-renewal of liver CSCs. Furthermore, inhibition of HDAC3 expression resulted in increase of H3K56Ac and H3K9Me3 and decrease of H3K27Me3, which was accompanied with the decreased expression of pluripotency factors. It has been proposed that the existence of an intriguing interplay between histone modifiers and pluripotency factors decides the cell fate of normal ESC [29]. Silencing of OCT4 and Nanog by H3K9Me2 and H3K9Me3 resulted in differentiation of ESC [30] by de novo DNA methylation of OCT4 and Nanog promoters mediated by the DNA methyltransferases DNMT3A and DNMT3B [29]. In addition, Nanog can also be silenced by H3K27Me3 via the Polycomb repressive complex (PRC) component enhancer of zeste homologue 2 (EZH2) [31]. Our previ-

ous study and other study have suggested that Nanog plays a critical role in the self-renewal of liver CSCs [8,11]. A recent study demonstrated that HDAC3 played an important role in the hypoxia-induced epithelial-mesenchymal transition (EMT) by H3K4Ac [32]. Induction of EMT could generate CSCs [33]. These data further support our observation that HDAC3 plays a critical role in the selfrenewal of liver CSCs. To the best of our knowledge, this is the first study to identify HDAC3 as a key molecule for promoting the selfrenewal of liver CSCs. However, the detail mechanisms of HDAC3 regulating histone modifications and pluripotency factors in liver CSC need further exploration. In addition, HDAC3 overexpression in the clinical HCC samples was significantly correlated with advanced tumor stage and early recurrence of HCC after surgery, suggesting the overexpression of HDAC3 in the liver CSCs could contribute poor prognosis of HCC. These data support a previous finding of which HDAC3 is a useful predictive marker for early tumor recurrence after liver transplantation in HBV-associated HCC [34]. We further demonstrated that expression of HDAC3 was gradually increased in the DEN-induced hepatocarcinogenesis, indicating that HDAC3 may be associated with initiation of HCC. However, relationship between Nanog positive liver CSCs and HDAC3 expression in both of human HCC patients and DEN-induced hepatocarcinogenesis in rat model should be investigated in more detail of future study.

68

C. Liu et al. / Cancer Letters 339 (2013) 60–69

Our previous study and many others have demonstrated that CSCs are resistant to therapeutic agents [2,4,9,11]. Since HDAC inhibitor could induce liver CSCs differentiation, we wonder whether HDAC inhibitors could render liver CSCs sensitive to sorafenib. Our data showed that combination of HDAC inhibitors and sorafenib could significantly inhibit the cell growth and completely abolish the drug resistant clones of liver CSCs. It is coincident with a previous observation that inhibition of HDAC activity prevents the development of drug resistance of cancer cells [35]. This data may also explain a recent study that additive antitumor activity could be achieved by combined therapy of HDAC inhibitor panobinostat with sorafenib in HCC [36]. Taken together, we demonstrate that HDAC inhibitors can inhibit proliferation and self-renewal of liver CSCs and HDAC3 participates in the self-renewal of liver CSCs through histone modification. HDAC inhibitors can render liver CSCs sensitive to sorafenib. Our data may provide an attractive therapeutic target for treatment of HCC by targeting liver CSCs.

Conflicts of interest No potential conflicts of interest were disclosed.

Acknowledgements We thanks for Ms. Qinghua Ma for her excellent technical assistance for sorting cells by FACS. This work was supported by funds from National Natural Sciences Foundation of China (Nos. 81090423, 81020108026, 81000966 and 81101630) and National Basic Research Program of China (973 Program, No. 2010C B529406).

Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.canlet.2013. 07.022.

References [1] J.M. Llovet, A. Burroughs, J. Bruix, Hepatocellular carcinoma, Lancet 362 (2003) 1907–1917. [2] J.E. Visvader, G.J. Lindeman, Cancer stem cells in solid tumours: accumulating evidence and unresolved questions, Nat. Rev. Can. 8 (2008) 755–768. [3] S. Pece, D. Tosoni, S. Confalonieri, G. Mazzarol, M. Vecchi, S. Ronzoni, L. Bernard, G. Viale, P.G. Pelicci, P.P. Di Fiore, Biological and molecular heterogeneity of breast cancers correlates with their cancer stem cell content, Cell 140 (2010) 62–73. [4] S. Ma, T.K. Lee, B.J. Zheng, K.W. Chan, X.Y. Guan, CD133+ HCC cancer stem cells confer chemoresistance by preferential expression of the Akt/PKB survival pathway, Oncogene 27 (2008) 1749–1758. [5] Z.F. Yang, D.W. Ho, M.N. Ng, C.K. Lau, W.C. Yu, P. Ngai, P.W. Chu, C.T. Lam, R.T. Poon, S.T. Fan, Significance of CD90+ cancer stem cells in human liver cancer, Cancer Cell 13 (2008) 153–166. [6] T. Yamashita, J. Ji, A. Budhu, M. Forgues, W. Yang, H.Y. Wang, H. Jia, Q. Ye, L.X. Qin, E. Wauthier, L.M. Reid, H. Minato, M. Honda, S. Kaneko, Z.Y. Tang, X.W. Wang, EpCAM-positive hepatocellular carcinoma cells are tumor-initiating cells with stem/progenitor cell features, Gastroenterology 136 (2009) 1012– 1024. [7] N. Haraguchi, H. Ishii, K. Mimori, F. Tanaka, M. Ohkuma, H.M. Kim, H. Akita, D. Takiuchi, H. Hatano, H. Nagano, G.F. Barnard, Y. Doki, M. Mori, CD13 is a therapeutic target in human liver cancer stem cells, J. Clin. Invest. 120 (2010) 3326–3339. [8] T.K. Lee, A. Castilho, V.C. Cheung, K.H. Tang, S. Ma, I.O. Ng, CD24+ liver tumorinitiating cells drive self-renewal and tumor initiation through STAT3mediated NANOG regulation, Cell Stem Cell 9 (2011) 50–63.

[9] A. Singh, J. Settleman, EMT, cancer stem cells and drug resistance: an emerging axis of evil in the war on cancer, Oncogene 29 (2010) 4741–4751. [10] S.T. Fan, Z.F. Yang, D.W. Ho, M.N. Ng, W.C. Yu, J. Wong, Prediction of posthepatectomy recurrence of hepatocellular carcinoma by circulating cancer stem cells: a prospective study, Ann. Surg. 254 (2011) 569–576. [11] J.J. Shan, J.J. Shen, L.M. Liu, F. Xia, C. Xu, G.J. Duan, Y.M. Xu, Q.H. Ma, Z. Yang, Q.Z. Zhang, L.N. Ma, J. Liu, S. Xu, X. Yan, P. Bie, Y.H. Cui, X.W. Bian, C. Qian, Nanog regulates self-renewal of cancer stem cell through IGF pathway in human hepatocellular carcinoma, Hepatology 56 (2012) 1004–1014. [12] J.H. Lee, S.R. Hart, D.G. Skalnik, Histone deacetylase activity is required for embryonic stem cell differentiation, Genesis 38 (2004) 32–38. [13] H. Lei, S.P. Oh, M. Okano, R. Jüttermann, K.A. Goss, R. Jaenisch, E. Li, De novo DNA cytosine methyltransferase activities in mouse embryonic stem cells, Development 122 (1996) 3195–3205. [14] K.W. McCool, X. Xu, D.B. Singer, F.E. Murdoch, M.K. Fritsch, The role of histone acetylation in regulating early gene expression patterns during early embryonic stem cell differentiation, J. Biol. Chem. 282 (2007) 6696–6706. [15] L. Sinkkonen, T. Hugenschmmidt, P. Berninger, D. Gaidatzis, F. Mohn, C.G. Artus-Revel, M. Zavolan, P. Svoboda, W. Filipowicz, MicroRNAs control de novo DNA methylation through regulation of transcriptional repressors in mouse embryonic stem cells, Nat. Struct. Mol. Biol. 15 (2008) 259–267. [16] P.M. Rodríguez, M. Esteller, Cancer epigenetics reaches mainstream oncology, Nat. Med. 17 (2011) 330–339. [17] M. Dokmanovic, C. Clarke, P.A. Marks, Histone deacetylase inhibitors: overview and perspectives, Mol. Cancer Res. 5 (2007) 981–989. [18] M.A. Glozak, E. Seto, Histone deacetylases and cancer, Oncogene 26 (2007) 5420–5432. [19] W.S. Xu, R.B. Parmigiani, P.A. Marks, Histone deacetylase inhibitors: molecular mechanisms of action, Oncogene 26 (2007) 5541–5552. [20] L. Blackwell, J. Norris, C.M. Suto, W.P. Janzen, The use of diversity profiling to characterize chemical modulators of the histone deacetylases, Life Sci. 82 (2008) 1050–1058. [21] B. Marije, R.G. Christian, B. Marc, M. Buitenhuis, P.J. Coffer, Histone deacetylase inhibition modulates cell fate decisions during myeloid differentiation, Haematological 95 (2010) 1052–1060. [22] J.E. Bolden, M.J. Peart, R.W. Johnstone, Anticancer activities of histone deacetylase inhibitors, Nat. Rev. Drug Discov. 5 (2006) 769–784. [23] V. Santini, A. Gozzini, G. Ferrari, Histone dezcetylase inhibitors: molecular and biology activity as a premise to clinical application, Curr. Drug Metab. 8 (2007) 383–393. [24] P.A. Marks, R.D. Breslow, Dimethyl sulfoxide to vorinostat: development of this histone deacetylase inhibitor as an anticancer drug, Nat. Biotechnol. 25 (2007) 84–90. [25] J.M. Llovet, S. Ricci, V. Mazzaferro, P. Hilgard, E. Gane, J.F. Blanc, A.C. de Oliveira, A. Santoro, J.L. Raoul, A. Forner, M. Schwartz, C. Porta, S. Zeuzem, L. Bolondi, T.F. Greten, P.R. Galle, J.F. Seitz, I. Borbath, D. Häussinger, T. Giannaris, M. Shan, M. Moscovici, D. Voliotis, J. Bruix, Sorafenib in advanced hepatocellular carcinoma, New Engl. J. Med. 359 (2008) 378–390. [26] J.P. Lai, C. Yu, C.D. Moser, I. Aderca, T. Han, T.D. Garvey, L.M. Murphy, M.M. Garrity-Park, V. Shridhar, A.A. Adjei, L.R. Roberts, SULF1 inhibits tumor growth and potentiates the effects of histone deacetylase inhibitors in hepatocellular carcinoma, Gastroenterology 130 (2006) 2130–2144. [27] S. Ma, K.H. Tang, Y.P. Chan, T.K. Lee, P.S. Kwan, A. Castilho, I. Ng, K. Man, N. Wong, K.F. To, B.J. Zheng, P.B. Lai, C.M. Lo, K.W. Chan, X.Y. Guan, MiR-130b promotes CD133(+) liver tumor-initiating cell growth and self-renewal via tumor protein 53-induced nuclear protein 1, Cell Stem Cell 7 (2010) 694– 707. [28] B.G. Debeb, L. Lacerda, W. Xu, R. Larson, T. Solley, R. Atkinson, E.P. Sulman, N.T. Ueno, S. Krishnamurthy, J.M. Reuben, T.A. Buchholz, W.A. Woodward, Histone deacetylase inhibitors stimulate dedifferentiation of human breast cancer cells through WNT/b-Catenin signaling, Stem Cells 30 (2012) 2366– 2377. [29] M. Hemberger, W. Dean, W. Reik, Epigenetic dynamics of stem cells and cell lineage commitment: digging Waddington’s canal, Nat. Rev. Mol. Cell Biol. 10 (2009) 526–537. [30] N. Hattori, Y. Imao, K. Nishino, N. Hattori, J. Ohgane, S. Yagi, S. Tanaka, K. Shiota, Epigenetic regulation of Nanog gene in embryonic stem and trophoblast stem cells, Genes Cells 12 (2007) 387–396. [31] L.A. Boyer, K. Plath, J. Zeitlinger, T. Brambrink, L.A. Medeiros, T.I. Lee, S.S. Levine, M. Wernig, A. Tajonar, M.K. Ray, G.W. Bell, A.P. Otte, M. Vidal, D.K. Gifford, R.A. Young, R. Jaenisch, Polycomb complexes repress developmental regulators in murine embryonic stem cells, Nature 441 (2006) 349–353. [32] S.A. Mani, W. Guo, M.J. Liao, E.N. Eaton, A. Ayyanan, A.Y. Zhou, M. Brooks, F. Reinhard, C.C. Zhang, M. Shipitsin, L.L. Campbell, K. Polyak, C. Brisken, J. Yang, R.A. Weinberg, The epithelial mesenchymal transition generates cells with properties of stem cells, Cell 133 (2008) 704–715. [33] M.Z. Wu, Y.P. Tsai, M.H. Yang, C.H. Huang, S.Y. Chang, C.C. Chang, S.C. Teng, K.J. Wu, Interplay between HDAC3 and WDR5 is essential for hypoxia-induced epithelial-mesenchymal transition, Mol. Cell 43 (2011) 811–822. [34] L.M. Wu, Z. Yang, L. Zhou, F. Zhang, H.Y. Xie, X.W. Feng, J. Wu, S.S. Zheng, Identification of histone deacetylase 3 as a biomarker for tumor recurrence following liver transplantation in HBV-associated hepatocellular carcinoma, Plos One 5 (2010) e14460.

C. Liu et al. / Cancer Letters 339 (2013) 60–69 [35] S.V. Sharma, D.Y. Lee, B. Li, M.P. Quinlan, F. Takahashi, S. Maheswaran, U. McDermott, N. Azizian, L. Zou, M.A. Fischbach, K.K. Wong, K. Brandstetter, B. Wittner, S. Ramaswamy, M. Classon, J. Settleman, A chromatin-mediated reversible drug-tolerant state in cancer cell subpopulations, Cell 141 (2010) 69–80.

69

[36] A. Lachenmayer, S. Toffanin, L. Cabellos, C. Alsinet, Y. Hoshida, A. Villanueva, B. Minguez, H.W. Tsai, S.C. Ward, S. Thung, S.L. Friedman, J.M. Llovet, Combination therapy for hepatocellular carcinoma: additive preclinical efficacy of the HDAC inhibitor panobinostat with sorafenib, J. Hepatol. 56 (2012) 1343–1350.