Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Inhibition of histone deacetylase 3 stimulates apoptosis induced by heat shock under acidic conditions in human maxillary cancer

Abstract

To elucidate the molecular mechanisms for the enhancement of heat-induced apoptosis on exposure to acidic conditions, human maxillary carcinoma IMC-3 cells were heat-shocked at 44°C for 30 min at either pH 7.4 or 6.7. Analyses with cDNA arrays, the reverse transcription–polymerase chain reaction (RT–PCR), and Western blotting were performed. We found that histone deacetylase 3 (HDAC3) was specifically induced after hyperthermia at 44°C for 30 min at pH 6.7. Although the cytotoxicity of heating at 44°C for 30 min was enhanced by decreasing the pH from 7.4 to 6.7, it was enhanced even more by antisense RNA oligonucleotides for HDAC3. The induction of G2/M arrest after heating occurred earlier at pH 6.7 than at pH 7.4. The inhibition of HDAC3 by the antisense RNA oligonucleotides suppressed partially the induction of G2/M arrest, resulting in an enhancement of the apoptosis caused by the heating under acidic conditions. Antisense RNA oligonucleotides for HDAC3 enhanced apoptosis 48 h after hyperthermia at 43°C for 30 min in vivo. Analyses of p65 activity suggested that NF-κB is involved in this enhancement of hyperthermia. HDAC3 may be a novel target enhancing hyperthermia and combined treatment with hyperthermia and HDAC inhibitors is a possible modality for cancer therapy.

Introduction

Cancer thermotherapy, that is hyperthermia, has been adopted as a modality for treating various kinds of cancer. Several randomized clinical trials showed hyperthermia to be effective for the treatment of human cancer, particularly when combined with radiotherapy (Overgaard et al., 1995; Sneed et al., 1998). The combined use of hyperthermia and chemotherapeutic drugs (cisplatin (CDDP), irinotecan (CPT-11), and mitomycin C (MMC)) has had a synergistic effect in both cultured cancer cells in vitro (Matsumoto et al., 1998a, 1998b; Jin et al., 2004) and human advanced cancer in vivo (Dahl and Mella, 2002; Takahashi et al., 2002). Human head and neck cancer is the most suitable for hyperthermia because it is easy to elevate the temperature of the tumor without injuring normal tissue (Engin et al., 1995).

Temperature is very important to the mode of action for cell killing by heating. Relatively mild temperatures induce apoptosis, but higher temperatures induce necrosis (Harmon et al., 1990; Matsumoto et al., 1997). The process of apoptosis induced by heating involves numerous biochemical reactions such as the degradation of proteins by certain caspases (Ohnishi and Ohnishi, 2001). It is clear that such activity, as well as other intracellular enzymatic reactions, can be influenced by environmental acidity (Gatenby and Gillies, 2004). It is well established that acidic conditions enhance heat-induced apoptosis in a variety of human cancer cells (Takasu et al., 1998; Ohtsubo et al., 2000, 2001). Acidic conditions were reported to enhance heat-induced apoptosis but not radiation-induced apoptosis (Ohtsubo et al., 2001). Moreover, this enhancement was independent of p53 status (Ohtsubo et al., 2000) and is expected to be a new strategy for the treatment of human cancer because numerous cancers bear a mutated p53 gene. However, the mechanisms behind the interactive effect of heating and acidic conditions on cell killing or induction of apoptosis remain unclear because of the large number of substances involved. It is important to research substances relative to heat and acidity to use hyperthermia to treat human malignancies more effectively.

In the present study, we analysed the genes whose expression was newly induced or upregulated after heating at 44°C for 30 min under acidic conditions using a cDNA array. We found that the expression of histone deacetylase 3 (HDAC3) was newly induced by heating under acidic conditions, resulting in G2/M arrest. Here, we show the potentiation of heat-induced apoptotic cell killing by inhibition of HDAC3 in human head and neck cancer, suggesting that combined therapy with hyperthermia and HDAC (especially HDAC3) inhibitors may be applicable to the treatment of human head and neck cancer.

Results

Detection of upregulated or induced genes after heating under acidic conditions

We examined the gene expression 6 h after heating at 44°C for 30 min at either pH 7.4 or 6.7 using a cDNA array. Table 1 shows the genes whose expression was upregulated or induced after heating at pH 6.7 as compared with pH 7.4. Five and three genes were upregulated and newly induced after heating at pH 6.7, respectively. We focused on HDAC3, one of the three newly induced genes, because HDAC3 may be a factor in signal transduction for survival and contribute to regulation of the cell cycle (Pena et al., 1997; Glaser et al., 2003). A total of 76 genes were lost and three genes were downregulated, after heating at pH 6.7, among the 1176 genes analysed (data not shown). They included genes for signal transducers, and cell cycle regulators as well as genes related to the intracellular kinase network.

Table 1 Genes upregulated 6 h after hyperthermia at 44°C for 30 min at pH 6.7

Reverse transcription (RT)–PCR analysis and Western blotting of HDAC3 after heating under acidic conditions

We confirmed the gene expression of HDAC3 6 h after heating at 44°C for 30 min using a RT–PCR analysis. The expression of HDAC3 mRNA was found only after heating at pH 6.7 (Figure 1a). The finding was consistent with that obtained with the cDNA array. Moreover, the level of HDAC3 protein increased 6 h after heating at pH 6.7 in the analysis using Western blotting (Figure 1b, lane 4).

Figure 1
figure1

Expression of HDAC3 gene and protein. (a) RT–PCR analysis of HDAC3. The left and right lanes show the expression of HDAC3 6 h after hyperthermia at 44°C for 30 min at pH 7.4 and 6.7, respectively. β-Actin is a control. (b) Western blot analysis of HDAC3. Lane 1 is a control (not treated). Lane 3 shows weak expression of HDAC3 6 h after hyperthermia at 44°C for 30 min at pH 7.4. Lane 4 shows the expression of HDAC3 6 h after hyperthermia at 44°C for 30 min at pH 6.7. GAPDH is an internal reference

Suppression of HDAC3 protein expression using antisense RNA

We designed eight types of antisense RNA oligonucleotides for HDAC3. Oligonucleotide No. 4 suppressed HDAC3 protein expression almost completely 6 h after heating at 44°C for 30 min at pH 6.7 (Figure 2). The other seven antisense RNA oligonucleotides and a scrambled oligonucleotide did not suppress HDAC3 protein expression.

Figure 2
figure2

Suppression of HDAC3 protein with antisense RNA oligonucleotide. IMC-3 cells were heat-treated at 44°C for 30 min in an acidic environment (pH 6.8) with antisense RNA oligonucleotide against HDAC3 (Nos.1–8) and a scrambled oligonucleotide as a control. Western blot analysis was conducted 6 h after heat treatment. The No.4 antisense RNA oligonucleotide completely repressed the expression of HDAC3. GAPDH is an internal reference

Flow cytometric analysis of cell cycle distribution after heating under various conditions

Figure 3 shows the profiles of flow cytometric analysis of cells after heating at 44°C for 30 min with or without antisense RNA oligonucleotide for HDAC3 at either pH 7.4. or 6.7. The treatment with antisense RNA oligonucleotide for HDAC3 at 37°C little affected the sub-G1 population (data not shown). A G2/M arrest was induced 24 h after heating at 44°C for 30 min at pH 6.7, whereas the arrest occurred 60 h after heating at 44°C for 30 min at pH 7.4. The G2/M arrest induced after heating at pH 6.7 was partially suppressed in association with an increase in the size of the sub-G1 population on treatment with antisense RNA oligonucleotides for HDAC3 (7.5–14.0%). The sub-G1 population, which was 33.3% after 60 h heating at pH 6.7, further increased to 49.5% on treatment with the antisense RNA oligonucleotides. The scrambled oligonucleotide did not enhance the sub-G1 population after heating at pH 6.7. The other antisense RNA oligonucleotides for HDAC3 (Nos.1–3 and 5–8, Figure 2) caused no increase in the size of the sub-G1 population 48 h after heating at 44°C for 30 min at pH 6.7 (data not shown).

Figure 3
figure3

Flow cytometric analysis of cell cycle population after hyperthermia. (a) Cell cycle population after hyperthermia at 44°C for 30 min at pH 7.4; (b) cell cycle population after hyperthermia at 44°C for 30 min at pH 6.7; (c) cell cycle population after hyperthermia at 44°C for 30 min at pH 6.7 with HDAC3 antisense oligonucleotide; (d) cell cycle population after hyperthermia at 44°C for 30 min at pH 6.7 with scrambled oligonucleotide. The percentages on the left of the panels denote sub-G1 and G2/M populations after each treatment

Clonogenic cell survival from heat shock under various conditions

Clonogenic cell survival was examined using a standard colony-forming assay. The antisense RNA oligonucleotides for HDAC3 had no cytotoxic effect compared to the control (Figure 4, column 1). A decrease in pH from 7.4 to 6.7 caused a reduction in cell survival after heating at 44°C for 30 min (Figure 4, columns 2 and 3, surviving fraction: 0.50–0.21). Moreover, administration of the antisense RNA oligonucleotides to culture medium at 5 μg/ml significantly reduced cell survival following heating at 44°C for 30 min at pH 6.7 (Figure 4, columns 3 and 4, 0.21–0.018). The scrambled oligonucleotide had no effect on cell survival after heating at 44°C for 30 min at pH 6.7 (Figure 5, column 5).

Figure 4
figure4

Surviving fraction of IMC-3 cells following hyperthermia with or without an acidic environment and HDAC3 antisense RNA. Column 1, cultured with HDAC3 antisense RNA oligonucleotide; column 2, after hyperthermia at 44°C for 30 min at pH 7.4; column 3, after hyperthermia for 30 min at pH 6.7; column 4, after hyperthermia for 30 min at pH 6.7 with HDAC3 antisense RNA oligonucleotide. The surviving fraction was calculated by normalizing to the plating efficiency of unheated cells. Values shown are the means of three independent experiments ±s.e.

Figure 5
figure5

Analysis of apoptosis with Hoechst 33342 staining. (a) Images of apoptotic IMC-3 cells stained with Hoechst 33342. Arrows show apoptotic cells. Addition of HDAC3 antisense RNA oligonucleotide induced apoptosis after hyperthermia at 44°C for 30 min at pH 6.7. (b) The percentage of apoptotic cells after hyperthermia with or without an acidic environment and HDAC3 antisense RNA oligonucleotide in IMC-3 cells. The types of treatment under the lanes are compatible with those in Figure 5a. Apoptotic IMC-3 cells were counted after treatment with Hoechst 33342. Column 1, control; column 2, 16 h after hyperthermia at 44°C for 30 min at pH 7.4; column 3, 16 h after culture at pH 6.7 for 30 min; column 4, 16 h after hyperthermia at 44°C for 30 min at pH 6.7; column 5, 16 h after hyperthermia at 44°C for 30 min at pH 6.7 with HDAC3 antisense RNA oligonucleotide

Induction of apoptosis after heating under various conditions

To confirm the findings of the flow cytometric analysis, we examined morphological changes in cells using fluorescence staining with Hoechst 33342. Figure 5a shows the images of apoptotic cells taken with a fluorescence microscope at × 400 after heating under various conditions. In all cases, three assistants, who were blind to the source of the specimen, counted over 10 000 cells in random fields. Under acidic conditions at pH 6.7 for 30 min, apoptosis was hardly observed without heating (Figure 5b, column 2, 0.1%). The percentage of apoptotic cells induced by heating (44°C, 30 min at pH 7.4) alone was 0.61% (Figure 5b, column 3). A decrease in pH from 7.4 to 6.7 increased the percentage of apoptotic cells after the heating (1.35%, Figure 5b, column 4). The addition of antisense RNA oligonucleotides for HDAC3 to the medium at 5 μg/ml enhanced the induction of apoptosis by heating at pH 6.7 (3.32%, Figure 5b, column 5, P<0.05 compared to that without the antisense RNA oligonucleotides). The scrambled oligonucleotide showed no enhancement of percentage of apoptotic cells after heating at 44°C for 30 min at pH 6.7 (Figure 5b, column 6).

Enhancement of heat-induced apoptosis by inhibition of HDAC3 expression in vivo

We examined the effect of antisense RNA oligonucleotides for HDAC3 on apoptotic cell killing by hyperthermia in IMC-3 tumors transplanted into nude mice using an immunohistochemical analysis (TUNEL method) (Figure 6a). A few TUNEL-positive apoptotic cells were found in untreated tumors (Figure 6b, column 1, 1.0%). The incidence of apoptosis increased significantly to about sixfold 48 h after hyperthermia alone (43°C for 30 min) as compared with the control (Figure 6b, column 2, 6.4%,). The combined treatment with the hyperthermia and antisense RNA oligonucleotides for HDAC3 further increased the incidence of apoptosis as compared with hyperthermia alone (Figure 6b, column 3, 6.4 vs 18.0%). The scrambled oligonucleotide did not upregulate apoptosis after hyperthermia (Figure 6b, column 4). However, necrotic areas were also found after the combined treatment. Antisense RNA oligonucleotides for HDAC3 had no cell-killing effect in themselves (data not shown). We confirmed the expression of HDAC3 in vivo using immunohistochemical staining. Hyperthermia (43°C for 30 min)-induced HDAC3 expression 48 h after treatment in vivo was confirmed and administration of our antisense oligonucleotide suppressed it (data not shown).

Figure 6
figure6

Apoptosis 48 h after hyperthermia at 43°C for 30 min in vivo. (a) Microscopic images of treated cells stained by the TUNEL method. The arrow indicates an apoptotic cell. (b) Percentage of apoptotic cells with or without HDAC3 antisense oligonucleotide. The addition of HDAC3 antisense RNA oligonucleotide increased the apoptosis caused by heating at 43°C for 30 min

Analysis of p65 activity after heating under acidic conditions

Finally, we analysed the activity of p65, which is a subunit of NF-κB, after heating under acidic conditions in order to investigate the relation between NF-κB and enhancement of heat-induced apoptosis under acidic conditions by HDAC3 suppression. Heating (44°C, 30 min at pH 6.7) upregulated the activity of p65 24 h after treatment (Figure 7, column 2) and antisense oligonucleotide for HDAC3 reduced it to below normal levels (Figure 7, column 3). This result provides for the possibility that suppression of HDAC3 enhanced apoptosis via suppression of NF-κB, which acts for cell survival. The scrambled oligonucleotide did not reduce p65 activity (Figure 7, column 4).

Figure 7
figure7

Analysis of p65 activity in vitro. Hyperthermia at 44°C for 30 min at pH 6.7 enhanced the activity of p65 compared to the control (columns 1 and 2). Antisense oligonucleotide for HDAC3 reduced the activity of p65 significantly (column 3). Activity of p65 is indicated as intensity of chemiluminescence

Discussion

In this study, we found that the expression of HDAC3 was newly induced 6 h after heating at 44°C for 30 min under acidic conditions at pH 6.7. The effect of the heating at pH 6.7 on cell killing was enhanced by the administration of antisense RNA oligonucleotides for HDAC3 in vitro. The treatment of transplanted tumors with these oligonucleotides also enhanced apoptosis 48 h after hyperthermia at 43°C for 30 min in vivo.

HDACs have been implicated in various cellular phenomena and divided into three classes based on sequence homology with yeast orthologs. Class I HDACs include HDAC1, 2, 3, 8 and 11 Class II HDACs include HDAC4, 5, 6, 7, 9, and 10. Class III mammalian HDACs are proteins homologous to yeast Sir 2 (Khochbin et al., 2001). The cDNA arrays used in this study contained class I HDACs (HDAC1, 2, and 3). Only the HDAC3 gene was expressed after heating under acidic conditions. Gene expression of HDAC1 and 2 was not induced by heating under acidic conditions (data not shown). Using Western blotting, we researched the expression of other class II HDACs (HDAC4,5, and 7), which have been reported to be involved in apoptosis (Huang et al., 2002; Bakin and Jung, 2004; Liu et al., 2004). The results showed that expression of HDAC4, 5, and 7 did not change 6 h after hyperthermia, under acidic conditions (44°C for 30 min at pH 6.7; data not shown). Although roles for other HDACs (HDAC6, 8, 9, and 10) are not excluded, these results suggest that HDAC3 is mainly responsible for the enhancement of heat-induced apoptosis under acidic conditions.

Previously, we reported that an acidic environment increases the thermosensitivity of mammalian cells by promoting heat-induced apoptosis (Ohtsubo et al., 2001). The mechanism by which acidic conditions enhance heat-induced apoptosis may be related to an apoptotic cascade originating from the cell membrane. Pena et al. (Takasu et al., 1998) suggested that various kinds of stress such as ionizing radiation, UV radiation, H2O2, and heat-shock act directly upon the membrane and activate acidic pH-dependent sphingomyelinase (ASMase), generating ceramide and initiating signaling through the stress-activated protein kinase (SAPK)/c-jun kinase (JNK) pathway leading to the activation of caspase 9. However, we expect that there is a lot of unknown signal transduction for hyperthermia under acidic conditions. We chose HDAC3 based on the results obtained with the cDNA arrays because it was newly induced and is known to promote G2/M arrest of the cell cycle. The enhancement of thermosensitivity on exposure to acidic conditions was limited (Figure 4), suggesting that cells in G2/M may be rescued during G2/M arrest induced after the heating, under acidic conditions, and that HDAC3 possibly plays an important role in this (Figure 3). We expected the suppression of HDAC3 to enhance the cell-killing effect of hyperthermia at a low pH. The inhibition of HDAC3 by the antisense RNA oligonucleotides reduced the size of the population arrested in G2/M 24 h after heating under acidic conditions, while expanding the sub-G1 population. A previous report showed that the inhibition of HDAC3 by siRNA caused hyperacetylation of histone and increased the percentage of apoptotic cells among HeLa cells (Glaser et al., 2003). However, our flow cytometric analysis proved that repression of HDAC3 in itself was not lethal to IMC3 cells (data not shown). Several investigators also reported that FK228, an inhibitor of multiple HDACs, itself induced apoptosis in malignant cells (Murata et al., 2000; Sasakawa et al., 2002). Although the toxicity differed between the HDAC inhibitors and antisense RNA oligonucleotides, the inhibition of HDAC3 enhanced the apoptosis induced by heating and a low pH. In addition, we presented the effect of HDAC3 on the induction of apoptosis in vivo. The intratumor pH in our model is around 6.8 (Rhee et al., 1984). Therefore, we examined the induction of apoptosis after hyperthermia at 43°C for 30 min in the presence or absence of the antisense oligonucleotides for HDAC3 using transplanted IMC-3 tumors in nude mice. The conditions in vivo were equivalent to the low pH and hyperthermia in vitro. Consequently, the same results were obtained from the two experiments in vitro and in vivo. The antisense RNA oligonucleotides for HDAC3 further increased the incidence of apoptosis compared with the hyperthermia alone, in vivo. The incidence of apoptosis in vivo (Figure 6) was higher than that in vitro (Figure 5), suggesting that the environment in tumors may be more suitable for enhancement of heat-induced apoptosis by inhibiting HDAC3 than the conditions in culture.

A recent report showed that HDAC3 deacetylated p65 (Rel A), a subunit of NF-κB, which has an important role in cell survival. Acetylation decreases the ability of p65 to bind κβ-DNA. Finally, acetylation of p65 facilitates its release from DNA and Iκβα consequently exports it from the nucleus (Kiernan et al., 2003). The acetylated and exported p65 is deacetylated by HDAC3 and prepared for activation with p300 or PCAF. Suppression of HDAC3 leads to acetylation of p65 (Kiernan et al., 2003) and reduces the transcriptional function of NF-κB to rescue damaged cells and possibly reduces cell survival after hyperthermia and other lethal treatments. In contrast, opposite phenomena were reported in which HDAC3-mediated deacetylation of p65 induced binding to Iκβ and Iκβ-dependent nuclear export of the complex through chromosomal region maintenance-1 (CRM-1) (Chen et al., 2001). Thus, HDAC3-mediated deacetylation of p65 is controversial since it has the possibility of either eliminating or promoting the activity of NF-κB. Our experiments showed that the activity of p65 was upregulated by hyperthermia under acidic conditions. The activity of p65 enhanced by hyperthermia under acidic conditions was reduced below normal levels with antisense RNA oligonucleotides for HDAC3. Our results are compatible with the former report (Kiernan et al., 2003) and HDAC3 is expected to act as a promoter of NF-κB and provide resistance to cancer cells against hyperthermia under acidic conditions. Although different experimental approaches may cause the two opposite results concerning the activation of p65 by HDAC3, other unknown mechanisms are expected to be involved in the enhancement of apoptosis via suppression of HDAC3.

A previous report showed that FK228, a class I HDAC inhibitor, acetylated heat–shock protein 90 (Hsp90), which is known to bind and stabilize oncoproteins (Yu et al., 2002). Acetylation of Hsp90 with FK228 reduced the binding of oncoproteins such as mutant p53 and Raf-1 to Hsp90 and led to the degradation of p53 and Raf-1. Since the activation of Raf-1 and following MEK/ERK pathway inhibits apoptosis in cancer cells (Cleveland et al., 1994; Xia et al., 1995), FK228 possibly has an anticancer effect by suppressing oncoproteins. Similarly, suppression of HDAC3 with antisense RNA oligonucleotides could lead to degradation of oncoproteins such as Raf-1 via acetylation of Hsp90 and induce apoptosis after hyperthermia under acidic conditions.

Hyperthermia is known to induce the generation of nitric oxide and oxidative stress in various kinds of cells (Matsumoto et al., 1999). Recently, trichostatin A (TSA, class I HDAC inhibitor) was shown to induce the expression of endothelial nitric oxide synthase (eNOS) in non-endothelial cells such as HeLa cells through inhibition of HDACs (Gan et al., 2005). Therefore, it is suggested that the suppression of HDAC3 by antisense RNA oligonucleotides evokes intracellular oxidative stress through the generation of nitric oxide by eNOS produced after the treatment, resulting in the enhancement of heat-induced apoptosis under acidic conditions. Further investigation of these possible pathways between enhancement of heat-induced apoptosis and suppression of HDAC3 is needed.

Previous research and clinical practice regarding the enhancement of hyperthermia have been performed in combination with anticancer drugs, which prevent cancer cells from obtaining thermo-resistance after hyperthermia. Radiotherapy has been used to enhance hyperthermia because radio-sensitivity is greatest in cells in the G2/M phase and thermosensitivity is greatest in cells in the S phase. Thus, radiotherapy and hyperthermia complement each other. However, for the enhancement of hyperthermia, mechanisms related to G1/S arrest have been targeted because of high thermosensitivity, but little notice has been taken of G2/M arrest in vitro or in vivo. We enhanced the hyperthermic effect by preventing the rescue of cells during G2/M arrest using antisense RNA oligonucleotides for HDAC3. Thus, our research into the enhancement of hyperthermia targeting G2/M could be epochmaking.

Inhibitors of multiple HDACs are considered therapeutic agents for malignancy, but they potentially injure normal tissue because they repress many different HDACs. FK 228, a class I HDAC inhibitor, has been reported to have an antitumor effect by itself or with other factors, in malignant lymphoma cells (Sasakawa et al., 2002). TSA is known to suppress class I HDACs and have anticancer effects like FK228. If we use FK228 or TSA instead of antisense RNA oligonucleotides for HDAC3, similar results to those obtained in the present study are expected because they suppress HDAC3. Although further investigation using HDAC inhibitors is needed, this study provides a new concept of combining HDAC inhibitors such as FK228 and TSA with thermotherapy to improve the therapeutic potential of these inhibitors. However, it remains unclear which of the HDACs is effective for the treatment of cancer cells (Marshall et al., 2002). According to our findings, specific inhibitors of HDAC3 possibly enhance the antitumor effects of hyperthermia without any severe side effects. Taken together, our findings suggest that HDAC3 may be a candidate for a target of hyperthermia and that combined treatment with hyperthermia and HDAC inhibitors (drugs or antisense oligonucleotides) may be a novel strategy for human cancer therapy.

Materials and methods

Cells and cell culture

Human maxillary squamous cell carcinoma IMC-3 cells (Mizoguchi et al., 1991) were used in the present study. Cells were cultured in RPMI-1640 medium (Nissui Pharmaceutical Co. Ltd, Tokyo, Japan) supplemented with 10% heat-inactivated fetal calf serum (FCS; Gibco Laboratories, Life Technologies Inc., Grand Island, NY, USA), 0.29 mg/ml of glutamine, 100 U/ml of penicillin, and 100 μg/ml of streptomycin, at 37°C in a conventional humidified CO2 incubator.

Heat treatment

The cells were seeded in 25-cm2 screw-capped polystyrene flasks (Falcon; Becton Dickinson, Bedford, MA, USA) with 6 ml of a medium of normal pH (pH 7.4), and cultured for at least 48 h to reach a logarithmic stage of growth. After the incubation, heat treatment was carried out by immersing the flasks, with fully tightened screw caps in a water bath (BK-43, Yamato Kagaku Co., Tokyo, Japan) preset at 44±0.05°C. In the heating under acidic conditions, cells were exposed to a medium of pH 6.7 just before heating. Immediately after heating, the treated cells were rinsed twice with the medium of pH 7.4 and refed fresh medium (pH 7.4) for further incubation.

Isolation of RNA and cDNA array hybridization

We used human Atlas™ cDNA expression arrays (Clontech Laboratories, Palo Alto, CA, USA) to determine the differences in gene expression, using the procedure described in the clontech manual, with a minor modification (Narita et al., 2002). The Atlas™ Human Cancer 1.2 Array contains 1176 known human genes. Cells heated as described above and cultured at 37°C for 6 h were harvested, pelleted and immediately frozen at −70°C. In brief, total RNA was isolated from the frozen cell pellets and treated with DNase, using a NucleoSpin™ RNA II Kit (Clontech). The mRNA was isolated from total RNA using oligo(dT)-conjugated magnetic beads. Radiolabeled cDNA probes were generated by RT of 0.5–1 μg of mRNA (the same amount for stimulated and control cells in each case) using Moloney murine leukemia virus polymerase in the presence of [α-33P]dATP. Hybridization of the cDNA probes to Atlas arrays took place overnight at 68°C in ExpressHyb™ (Clontech). After a high-stringency wash, the hybridization patterns were quantified by phosphor imaging and then analysed using AtlasImage™ software (Clontech). This process was performed twice using mRNA isolated from independent experiments on two different filter sets. Nine housekeeping genes and three negative controls were spotted on The Atlas™ Human Cancer 1.2 array. The identities of the cDNAs on the array are available on the Internet (http://atlasinfo.clontech.com/). The AtlasImage™ software package subtracts the array background, normalizes each result based on the signal strength obtained with one or more of the housekeeping genes selected by the operator, and finally computes the ratio of the expression between two arrays. As for the criteria to determine whether gene expression changes were significant, we adopted a ratio greater than 2.0 in this study according to previous reports (Levenson et al., 2000; Narita et al., 2002).

RT–PCR analysis

Dnase-treated total RNA was extracted as described above. Using 2 μg of total RNA, first-strand cDNA was synthesized using a First-Strand cDNA synthesis kit (Amersham Pharmacia Biotech, Piscataway, NJ, USA). Then, 1 μl of the resulting first-strand cDNA was used for each PCR. The primers for human HDAC3 and β-actin (Fujieda et al., 1995) were used in the reactions. The primers were as follows: HDAC3 upstream, 5′-IndexTermGTCGATGTTATTTCCCCAGC-3′ (nucleotides 79–98 of human HDAC3 cDNA), HDAC3 downstream, 5′-IndexTermCCGATTTGGTGATGGGTGTT-3′ (nucleotides 862–881). The reaction mixture in a final volume of 25 μl consisted of 1 × Taq DNA polymerase buffer, 0.2 mM dNTP, 1.5 mM MgCl2, 0.5 μ M of each primer, and 2.5 U of Taq DNA Polymerase (Amersham Pharmacia Biotech) according to the instructions for the Taq DNA Polymerase. A preamplification denaturation for 10 min at 94°C was performed and amplification was carried out for 35 cycles. For β-actin, PCR amplification was performed for 25 cycles. Each cycle was 94°C for 45 s, 60°C for 30 s, and 72°C for 90 s. The final extension step was performed for 10 min at 72°C. Samples (10 μl) of the PCR products were analysed on 2% agarose gel in 1 × Tris-acetate-EDTA (TAE) buffer and bands were visualized by ethidium bromide staining.

Western blot analysis

Cells were washed twice with ice-cold PBS and dissolved in solubilizing buffer (pH 7.4, 1% Triton X-100, 1% deoxycholic acid sodium salt, 0.1% SDS, 20 mM Tris-HCl, 150 mM NaCl, 5 mM EDTA, 1 mM PMSF, 10 μg/ml pepstatin, and 10 μg/ml leupeptin). Each aliquot of protein (50 μg) was subjected to Western blot analysis. After electrophoresis on 12.5% polyacrylamide gels, the protein was transblotted to Hybond-P (Amersham Life Sciences, Inc., Arlington Heights, IL, USA) in transfer buffer (192 mM glycine, 25 mM Tris, 2.5 mM SDS, and 10% methanol). The blots were blocked with 3% nonfat dry milk in pH 7.4 TBST, and then incubated with anti-HDAC3 antibody (1 : 200; Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA), and then secondarily incubated with an HRP-conjugated (horse radish peroxidase-conjugated) anti-rabbit or -mouse IgG antibody (DACO Corp., Carpinteria, CA, USA). Subsequently, the blots were developed with chemiluminescence Western blotting detection reagents (DACO Corp., Carpinteria, CA, USA) according to the manufacturer's instructions.

Design and use of antisense RNA for HDAC3

We designed eight antisense RNA oligonucleotides for HDAC3 from the open reading frame of human HDAC3 (nucleotides 2–82) referring to a previous report (Yang et al., 1997). These antisense RNA oligonucleotides for HDAC3 were purchased from Nisshinnbo Co., Japan. We adopted RNA oligonucleotides expected to bind and suppress mRNA of HDAC3 efficiently referring to a classical method (Moroni et al., 1992; Mukhopadhyay and Roth, 1996; Ru et al., 1999). We did not use chemical modifications because they reduce sequence-specific effects of antisense oligonucleotides on the target mRNA. Although our No.4 antisense RNA oligonucleotide includes a uracil, it is very similar to sequences of DNA oligonucleotides. Through pilot experiments, it was proved to be stable and appropriate for suppressing the expression of HDAC3 protein. We added antisense RNA oligonucleotides to the culture medium at 5 μg/ml, 48 or 96 h before the heat treatment. We added antisense RNA oligonucleotides at 5 μg/ml when we exchanged the medium after heat-treatment. The sequences of the antisense RNA oligonucleotides for HDAC3 are as follows: No.1 5′-IndexTermCGGCGGCCGCGGGCGGCGGG-3′; No.2 5′-IndexTermCGGCGGAGGUGCGGGGCCUG-3′; No.3 5′-IndexTermCUCCCGCCGGCACCAUGGCC-3′; No.4 5′-IndexTermGGGCGGCGGGCGGCGGAGGU-3′; No.5 5′-IndexTermCGGCGGAGGUGCGGGGCCUG-3′; No.6 5′-IndexTermGCGGGGCCUGCUCCCGCCGG-3′; No.7 5′-IndexTermCUCCCGCCGGCACCAUGGCC-3′; No.8 5′-IndexTermCACCAUGGCCAAGACCGUG-3′.

The sequence of the scrambled oligonucleotide for the control is 5′-IndexTermGGCCGGCCGGCGACGGCGGU-3′.

Flow cytometric analysis of cell cycle population

The cell cycle distribution of cells after the heating at either pH 6.7 or 7.4 was analysed by flow cytometry, as previously reported (Park et al., 1999). Cells were washed twice with ice-cold PBS, fixed in 70% (v/v) ethanol, and stored at 4°C. The cells were stained with propidium iodide (50 μg/ml), treated with DNase-free RNase (10 μg/ml), and subjected to an analysis of DNA content using an EPICS flow cytometer (Coulter, Hialeah, FL, USA). The sizes of the sub-G1 and G2/M fractions were calculated using the Coulter flow cytometry software.

Clonogenic survival assay

The surviving cell fractions were determined in colony-forming units and values were corrected using the plating efficiency of the controls. Three replicate flasks were used, and three replicate experiments were conducted. Obtained colonies were fixed with methanol and stained with a 2% Giemsa solution (Merck & Co., Inc., Rahway, NJ, USA). Visible colonies composed of more than 50 cells after 10–14 days were counted as having grown from surviving cells.

Apoptotic cell staining

The detection of apoptotic cells after the heating was carried out using Hoechst 33342 (SIGMA CHEMICAL Co., St Louis, MO, USA) staining. Cells were harvested 16 h after the heating and fixed with freshly prepared 1% glutaraldehyde (Nakalai Tesque, Kyoto, Japan) overnight at 4°C. After two washes with PBS, cells were stained with 1 mM Hoechst 33342. Apoptotic cells were counted using fluorescence microscopy at × 400.

Analysis of apoptosis in vivo

IMC-3 cells (2 × 107) were transplanted into the right legs of nude mice (Balb/cA-Jcl nu/nu). Three mice were prepared in each treatment group. When the tumor reached 5 mm in diameter, 300 μg of antisense RNA oligonucleotide for HDAC3 was injected into it. Hyperthermia was carried out by immersing the right legs bearing tumors in a water bath (Thermominder EX, TAITEC Co., Ltd., Saitama, Japan) preset at 43±0.1°C 24 h after the injection, according to a previous report (Takahashi et al., 2003). We adopted 43°C because it has been revealed that hyperthermia at 44°C reduces the survival of mice. Tumors were taken out 48 h after hyperthermia and fixed as frozen sections. Immunohistochemical staining was carried out, with an ApopTag in situ Detection Kit® (Intergen Co., Perchase, NY, USA) and anti-HDAC3 antibody, to analyse the incidence of apoptosis and HDAC3 after the treatment with hyperthermia and/or antisense RNA oligonucleotides for HDAC3 in vivo, respectively (Sugimoto et al., 1999). In all cases, three individuals who were blind to the source of the specimen counted over 1000 cells in three random fields. Apoptotic cells were counted and compared with other conditions.

Analysis of p65 activity

Proteins contained in nuclear compartments of IMC-3 cells were extracted 24 h after heating (44°C, 30 min at pH 6.7), with or without antisense RNA oligonucleotides for HDAC3 using a Nuclear Extract Kit (Active Motif Co., Carlsbad, CA, USA). In total, 2 μg of protein extracted from nuclei was used to analyse the activity of p65 with Trans AM™ (Active Motif Co.). The activity of p65 was measured with chemiluminescence using a plate reader (FUSION™, PerkinElmer, Wellesley, MA, USA).

References

  1. Bakin RE and Jung MO . (2004). J. Biol. Chem., 279, 51218–51225.

  2. Chen Lf, Fischle W, Verdin E and Greene WC . (2001). Science, 293, 1653–1657.

  3. Cleveland JL, Troppmair J, Packham G, Askew DS, Lloyd P, Gonzalez-Garcia M, Nunez G, Ihle JN and Rapp UR . (1994). Oncogene, 9, 2217–2226.

  4. Dahl O and Mella O . (2002). Int. J. Hyperther., 18, 25–30.

  5. Engin K, Leeper DB, Tupchong L and Waterman FM . (1995). Clin. Cancer Res., 1, 139–145.

  6. Fujieda S, Zhang K and Saxon A . (1995). J. Immunol., 155, 2318–2328.

  7. Gan Y, Shen YH, Wang J, Wang X, Utama B, Wang J and Wang XL . (2005). J. Biol. Chem., 280, 16467–16475.

  8. Gatenby RA and Gillies RJ . (2004). Nat. Rev. Cancer, 4, 891–899.

  9. Glaser KB, Li J, Staver MJ, Wei RQ, Albert DH and Davidsen SK . (2003). Biochem. Biophys. Res. Commun., 310, 529–536.

  10. Harmon BV, Corder AM, Collins RJ, Gobe GC, Allen J, Allan DJ and Kerr JF . (1990). Int. J. Radiat. Biol., 58, 845–858.

  11. Huang Y, Tan M, Gosink M, Wang KK and Sun Y . (2002). Cancer Res., 62, 2913–2922.

  12. Jin ZH, Matsumoto H, Hayashi S, Hatashita M, Ohtsubo T, Shioura H, Kitai R and Kano E . (2004). Int. J. Radiat. Oncol. Biol. Phys., 59, 852–860.

  13. Khochbin S, Verdel A, Lemercier C and Seigneurin-Berny D . (2001). Curr. Opin. Genet. Dev., 11, 162–166.

  14. Kiernan R, Bres V, Ng RW, Coudart MP, El Messaoudi S, Sardet C, Jin DY, Emiliani S and Benkirane M . (2003). J. Biol. Chem., 278, 2758–2766.

  15. Levenson VV, Davidovich IA and Roninson IB . (2000). Cancer Res., 60, 5027–5030.

  16. Liu F, Dowling M, Yang XJ and Kao GD . (2004). J. Biol. Chem., 279, 34537–34546.

  17. Marshall JL, Rizvi N, Kauh J, Dahut W, Figuera M, Kang MH, Figg WD, Wainer I, Chaissang C, Li MZ and Hawkins MJ . (2002). J. Exp. Ther. Oncol., 2, 325–332.

  18. Matsumoto H, Hayashi S, Hatashita M, Ohnishi K, Ohtsubo T, Kitai R, Shioura H, Ohnishi T and Kano E . (1999). Cancer Res., 59, 3239–3244.

  19. Matsumoto H, Hayashi S, Shioura H, Ohtsubo T, Kitai R, Ohnishi K, Hayashi N, Ohnishi T and Kano E . (1998a). Int. J. Radiat. Oncol. Biol. Phys., 41, 915–920.

  20. Matsumoto H, Hayashi S, Shioura H, Ohtsubo T, Nishida T, Kitai R, Ohnishi T and Kano E . (1998b). Int. J. Oncol., 13, 741–747.

  21. Matsumoto H, Takahashi A, Wang X, Ohnishi K and Ohnishi T . (1997). Int. J. Radiat. Oncol. Biol. Phys., 38, 1089–1095.

  22. Mizoguchi H, Komiyama S, Matsui K, Hamanaka R, Ono M and Kiue A . (1991). Int. J. Cancer., 49, 738–743.

  23. Moroni MC, Willingham MC and Beguinot L . (1992). J. Biol. Chem., 267, 2714–2722.

  24. Mukhopadhyay T and Roth JA . (1996). Crit. Rev. Oncog., 7, 151–190.

  25. Murata M, Towatari M, Kosugi H, Tanimoto M, Ueda R, Saito H and Naoe T . (2000). Jpn. J. Cancer Res., 91, 1154–1160.

  26. Narita N, Noda I, Ohtsubo T, Fujieda S, Tokuriki M, Saito T and Saito H . (2002). Int. J. Radiat. Oncol. Biol. Phys., 53, 190–196.

  27. Ohnishi K and Ohnishi T . (2001). Int. J. Hyperther., 17, 415–427.

  28. Ohtsubo T, Igawa H, Saito T, Matsumoto H, Park HJ, Song CW, Kano E and Saito H . (2001). Int. J. Radiat. Oncol. Biol. Phys., 49, 1391–1398.

  29. Ohtsubo T, Park HJ, Lyons JC, Ohnishi T and Song CW . (2000). Int. J. Hyperther., 16, 481–491.

  30. Overgaard J, Gonzalez Gonzalez D, Hulshof MC, Arcangeli G, Dahl O, Mella O and Bentzen SM . (1995). Lancet, 345, 540–543.

  31. Park HJ, Lyons JC, Ohtsubo T and Song CW . (1999). Br. J. Cancer, 80, 1892–1897.

  32. Pena LA, Fuks Z and Kolesnick R . (1997). Biochem. Pharmacol., 53, 615–621.

  33. Rhee JG, Kim TH, Levitt SH and Song CW . (1984). Int. J. Radiat. Oncol. Biol. Phys., 10, 393–399.

  34. Ru K, Schmitt S, James WI and Wang JH . (1999). Oncol. Res., 11, 505–512.

  35. Sasakawa Y, Naoe Y, Inoue T, Sasakawa T, Matsuo M, Manda T and Mutoh S . (2002). Biochem. Pharmacol., 64, 1079–1090.

  36. Sneed PK, Stauffer PR, McDermott MW, Diederich CJ, Lamborn KR, Prados MD, Chang S, Weaver KA, Spry L, Malec MK, Lamb SA, Voss B, Davis RL, Wara WM, Larson DA, Phillips TL and Gutin PH . (1998). Int. J. Radiat. Oncol. Biol. Phys., 40, 287–295.

  37. Sugimoto C, Fujieda S, Seki M, Sunaga H, Fan GK, Tsuzuki H, Borner C, Saito H and Matsukawa S . (1999). Int. J. Cancer, 82, 860–867.

  38. Takahashi A, Ota I, Tamamoto T, Asakawa I, Nagata Y, Nakagawa H, Kondo N, Ohnishi K, Furusawa Y, Matsumoto H and Ohnishi T . (2003). Int. J. Hyperther., 19, 145–153.

  39. Takahashi I, Emi Y, Hasuda S, Kakeji Y, Maehara Y and Sugimachi K . (2002). Surgery, 131, 78–84.

  40. Takasu T, Lyons JC, Park HJ and Song CW . (1998). Cancer Res., 58, 2504–2508.

  41. Xia Z, Dickens M, Raingeaud J, Davis RJ and Greenberg ME . (1995). Science, 270, 1326–1331.

  42. Yang WM, Yao YL, Sun JM, Davie JR and Seto E . (1997). J. Biol. Chem., 272, 28001–28007.

  43. Yu X, Guo ZS, Marcu MG, Neckers L, Nguyen DM, Chen GA and Schrump DSJ . (2002). Natl. Cancer Inst., 94, 504–513.

Download references

Acknowledgements

This study was supported by a Grant-in-Aid for Scientific Research and Scientific Research on Priority Area (C) from the Ministry of Education, Culture, Sports, Science and Technology, Japan. We thank Drs C Sugimoto, H Sunaga, H Igawa, Y Kimura and H Yamamoto (Department of Otorhinolaryngology, Fukui Medical University) for a critical review of this study. We are also grateful to Ms Kazumi Uno for excellent technical assistance.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Norihiko Narita.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Narita, N., Fujieda, S., Tokuriki, M. et al. Inhibition of histone deacetylase 3 stimulates apoptosis induced by heat shock under acidic conditions in human maxillary cancer. Oncogene 24, 7346–7354 (2005). https://doi.org/10.1038/sj.onc.1208879

Download citation

Keywords

  • cDNA array
  • hyperthermia
  • low pH
  • HDAC3
  • head and neck cancer

Further reading

Search

Quick links