SET overexpression decreases cell detoxification efficiency: ALDH2 and GSTP1 are downregulated, DDR is impaired and DNA damage accumulates

Authors

  • Luciana O. Almeida,

    1. Departamento de Análises Clínicas, Toxicológicas e Bromatológicas, Faculdade de Ciências Farmacêuticas de Ribeirão Preto, Universidade de São Paulo, Ribeirão Preto - SP, Brazil
    Search for more papers by this author
  • Renata N. Goto,

    1. Departamento de Análises Clínicas, Toxicológicas e Bromatológicas, Faculdade de Ciências Farmacêuticas de Ribeirão Preto, Universidade de São Paulo, Ribeirão Preto - SP, Brazil
    Search for more papers by this author
  • Cezar R. Pestana,

    1. Departamento de Física e Química, Faculdade de Ciências Farmacêuticas de Ribeirão Preto, Universidade de São Paulo, Ribeirão Preto - SP, Brazil
    Search for more papers by this author
  • Sérgio A. Uyemura,

    1. Departamento de Análises Clínicas, Toxicológicas e Bromatológicas, Faculdade de Ciências Farmacêuticas de Ribeirão Preto, Universidade de São Paulo, Ribeirão Preto - SP, Brazil
    Search for more papers by this author
  • Silvio Gutkind,

    1. Oral and Pharyngeal Cancer Branch, National Institute of Dental and Craniofacial Research, National Institutes of Health, Bethesda, MD, USA
    Search for more papers by this author
  • Carlos Curti,

    1. Departamento de Física e Química, Faculdade de Ciências Farmacêuticas de Ribeirão Preto, Universidade de São Paulo, Ribeirão Preto - SP, Brazil
    Search for more papers by this author
  • Andréia M. Leopoldino

    Corresponding author
    • Departamento de Análises Clínicas, Toxicológicas e Bromatológicas, Faculdade de Ciências Farmacêuticas de Ribeirão Preto, Universidade de São Paulo, Ribeirão Preto - SP, Brazil
    Search for more papers by this author

Correspondence

A. M. Leopoldino, Departamento de Análises Clínicas, Toxicológicas e Bromatológicas, Faculdade de Ciências Farmacêuticas de Ribeirão Preto, Universidade de São Paulo, Ave. Café s/n, 14040-903 Ribeirão Preto, SP, Brazil

Fax: +55 16 3602 4725

Tel: +55 16 3602 0659

E-mail: andreiaml@usp.br

Abstract

Alcohol and tobacco consumption are risk factors for head and neck squamous cell carcinoma (HNSCC). Aldehyde dehydrogenase 2 (ALDH2) and glutathione S–transferase pi 1 (GSTP1) are important enzymes for cellular detoxification and low efficiencies are implicated in cancer. We assessed the potential role of SET protein overexpression, a histone acetylation modulator accumulated in HNSCC, in gene regulation and protein activity of ALDH2 and GSTP1. SET was knocked down in HN13, HN12 and Cal27, and overexpressed in HEK293 cells; ethanol and cisplatin were the chemical agents. Cells with SET overexpression (HEK293/SET, HN13 and HN12) showed lower ALDH2 and GSTP1 mRNA levels and trichostatin A increased them (real-time PCR). Ethanol upregulated GSTP1 and ALDH2 mRNAs, whereas cisplatin upregulated GSTP1 in HEK293 cells. SET-chromatin binding revealed SET interaction with ALDH2 and GSTP1 promoters, specifically via SET NAP domain; ethanol and cisplatin abolished SET binding. ALDH2 and GSTP1 efficiency was assessed by enzymatic and comet assay. A lower ALDH2 activity was associated with greater DNA damage (tail intensity) in HEK293/SET compared with HEK293 cells, whereas HN13/siSET showed ALDH2 activity higher than HN13 cells. HN13/siSET cells showed increased tail intensity. Cisplatin-induced DNA damage response showed negative relationship between SET overexpression and BRCA2 recruitment. SET downregulated repair genes ATM, BRCA1 and CHEK2 and upregulated TP53. Cisplatin-induced cell-cycle arrest occurred in G0/G1 and S in HEK293 cells, whereas HEK293/SET showed G2/M stalling. Overall, cisplatin was more cytotoxic for HN13 than HN13/siSET cells. Our data suggest a role for SET in cellular detoxification, DNA damage response and genome integrity.

Abbreviations
ALDH2

aldehyde dehydrogenase 2

ATM

ataxia telangiectasia mutated

BRCA1

breast cancer 1

BRCA2

breast cancer 2

Cal27

oral adenosquamous carcinoma cell line

CHEK2

checkpoint kinase 2

ChIP

chromatin immunoprecipitation assay

DDR

DNA damage response

FANC

Fanconi anemia

FANCD2

Fanconi anemia, complementation group D2

GSTP1

glutathione S-transferase pi 1

γH2A.X

phosphorylated histone γH2A.X

H2B

histone H2B

H4

histone H4

HDAC

histone deacetylase

HEK293

human embryonic kidney cell line

HN12

head and neck cell line derived from metastatic squamous cell carcinoma

HN13

head and neck cell line derived from tongue squamous cell carcinoma

HNSCC

head and neck squamous cell carcimona

I2PP2A

inhibitor-2 of protein phosphatase-2A

qPCR

quantitative polymerase chain reaction

siSET

small interfering RNA to SET knockdown

TSA

trichostatin A

Introduction

More than 650 000 new head and neck squamous cell carcinoma (HNSCC) cases are diagnosed each year [1], and alcohol and tobacco consumption are the primary risk factors. Tobacco consumption is strongly associated with laryngeal cancer, whereas alcohol consumption is associated with pharynx and oral cavity cancer [2]. The accumulation of carcinogens in the aerodigestive tract depends on their activation/detoxification by phases I and II biotransformation enzymes. Carcinogen accumulation, in turn, can promote DNA adduct mutations in tumor suppressor genes/proto-oncogenes and cell malignization [3].

Ethanol does not have carcinogenic activity, but acetaldehyde, a primary metabolite of ethanol, accumulates in the saliva and is carcinogenic. Although some microorganisms of the oral flora metabolize ethanol into acetaldehyde, their capacity to detoxify the metabolite is limited [4]. Tobacco contains acetaldehyde, which becomes dissolved in the saliva during smoking and is distributed to the pharynx, esophagus and stomach on swallowing the saliva [5]. Aldehydes are highly reactive molecules whose accumulation promotes DNA damage [6]. The mechanisms of cell protection against DNA damage include the combined action of DNA repair and detoxification by enzymes. Initiation of the DNA damage response (DDR) is associated with p53 activation and the recruitment of a cascade of proteins including Fanconi anemia group (FANC), histone H2A.X, breast cancer 1 (BRCA1, early onset) and breast cancer 2 (BRCA2, early onset) [7]. The failure of DDR leads to the accumulation of DNA damage and promotes new mutations in tumor suppressor genes/proto-oncogenes and, consequently, cancer development and progression. Moreover, cancer cells are resistant to death, and this resistance has been associated with Akt signaling, apoptosis blockage, increased DDR and autophagy [8]

Aldehyde dehydrogenase 2 (ALDH2), which catalyzes the oxidation of acetaldehyde to acetic acid and other reactive aldehydes, is an important detoxifying enzyme [9]; in HNSCC [10] and other cancer types [11] a decrease in ALDH2 activity impairs cell detoxification mechanisms. Another important detoxifying enzyme family, glutathione S-transferase (GST), comprises phase II enzymes for xenobiotic elimination through conjugation reactions [12], which subsequently yield mercapturic acids and facilitate excretion. Detoxification of cigarette-derived xenobiotics is catalyzed by GST pi 1 (GSTP1), which is the most abundant type of GST in head and neck mucosal tissues [13] and is decreased in HNSCC [14].

SET protein [template-activating factor-I β or inhibitor-2 of protein phosphatase-2A (I2PP2A)] is an established inhibitor of protein phosphatase 2A (PP2A) [15] and of histone acetylation [16]. SET was first identified in renal development and Wilm's tumor [17], and its presence has been reported in HNSCC [18]. Recent studies from our laboratory have demonstrated that SET accumulation in HNSCC increases cell survival under oxidative stress in association with Akt phosphorylation/activation [19] and that SET overexpression in HEK293T cells may act as either a cell survival signal or an oxidative stress sensor for cell death [20]. In this study, we addressed the potential role of SET overexpression in the gene regulation and protein activity of ALDH2 and GSTP1. The HNSCC cell lineages HN13, HN12 and Cal27, with or without SET knockdown (HNSCC/siSET and HNSCC, respectively), and the human embryonic kidney cell line HEK293, with or without overexpression of the SET protein (HEK293/SET and HEK293, respectively), were used as experimental models; ethanol and cisplatin were the chemical agents.

Results

SET overexpression downregulates ALDH2 and GSTP1 gene expression by interacting with chromatin and promoting histone hypoacetylation

SET overexpression in HEK293 (HEK293/SET) cells decreased ALDH2 and GSTP1 mRNA expression levels; consistent with this effect, SET knockdown in HNSCC cell lineages, in which SET is constitutively overexpressed, slightly increased ALDH2 and GSTP1 mRNA expression levels (Table 1). Subsequently, we tested whether this effect might result from histone hypoacetylation by using trichostatin A (TSA, 100 ng·mL−1 for 4 h), which inhibits histone deacetylases by promoting histone hyperacetylation and derepression of silenced genes. TSA treatment was efficient at reversing the overexpressed SET action in the ALDH2 and GSTP1 gene expression (Fig. 1A,B). These results indicate that SET downregulates ALDH2 and GSTP1 gene expression and suggest that this effect is through a direct interaction of SET with the inhibitor of histone acetyltransferase complex, which remodels chromatin and promotes histone hypoacetylation.

Table 1. ALDH2 and GSTP1 gene expression is regulated by SET accumulation in HNSCC and in HEK293 cells. qPCR analyses of ALDH2 and GSTP1 mRNA levels were performed in HN13, HN13/siSET, HN12, HN12/siSET, Cal27, Cal27/siSET, HEK293 and HEK293/SET. The cells were transfected with either a siRNA against SET or SET cDNA vector. Relative mRNA expression levels were calculated using the 2−ΔΔCT method
Cell line ALDH2 GSTP1
ControlsiSETControlsiSET
HN130.6951.0450.6061.073
HN120.0221.8130.6161.152
Cal270.0261.5163.8760.756
 ControlSETControlSET
HEK29310.67510.643
Figure 1.

TSA reverses SET action in ALDH2 and GSTP1 gene expression regulation. qPCR analyses of (A) ALDH2 and (B) GSTP1 gene expression levels were performed in HEK293, HEK293/SET, Cal27, HN12 and HN13 cells. Cells were transfected with a SET full-length cDNA vector and after 24 h, were treated (+) with TSA (100 ng·mL−1) or untreated (−) for 24 h. Relative mRNA expression levels were calculated using the 2−ΔΔCT method. The data are the means ± SD of three independent experiments in triplicate.

Ethanol and cisplatin are suitable tools for assessing ALDH2 and GSTP1 enzyme activity, respectively. In HEK293 cells, which present endogenous SET, ALDH2 gene expression was upregulated in the presence of ethanol (Fig. 2A), which is consistent with the ALDH2-mediated conversion of acetaldehyde, an alcohol metabolism byproduct, into acetate. In addition, GSTP1, whose corresponding protein is responsible for cisplatin detoxification, was upregulated in the presence of ethanol as well as cisplatin. These results indicate that ALDH2 and GSTP1 upregulation are involved in the cell detoxification process in nontumor cells. However, in HN13 cells, which present high SET levels, no significant difference in ALDH2 and GSTP1 gene expression after cisplatin or ethanol treatment was observed; these data reinforce that ALDH2 and GSTP1 genes are regulated by the high SET levels found in HNSCC cells (Fig. 2A). SET mRNA expression levels decreased in HEK293 cells in the presence of ethanol and TSA, whereas cisplatin induced a slight increase; in HN13 cells, cisplatin, ethanol and TSA treatments increased SET mRNA levels (Fig. 2B).

Figure 2.

HEK293 and HN13 cells respond differently to ethanol and cisplatin regarding ALDH2, GSTP1 and SET gene expression. (A) HEK293 and HN13 cells were treated with 50 μm cisplatin or 5% ethanol for 24 h, and ALDH2 and GSTP1 mRNA expression levels were determined by qPCR. (B) SET gene expression levels were analyzed in HEK293 and HN13 cells after exposure to 50 μm cisplatin, 5% ethanol or 100 ng·mL−1 TSA. Relative mRNA expression levels were calculated using the 2−ΔΔCT method. The data are the means ± SD of three independent experiments in triplicate.

To verify whether the downregulation of ALDH2 and GSTP1 gene expression levels associated with SET overexpression occurred through histone acetylation control, we determined H2B and H4 acetylation levels in HEK293/SET cells by immunofluorescence (Fig. 3A,B) and immunoblotting (Fig. 3C). SET overexpression decreased H2B and H4 acetylation, and this effect was reversed by TSA. Chromatin immunoprecipitation (ChIP) was used to assess whether SET interacts directly with ALDH2 and GSTP1 promoter-derived chromatin. The SET overexpression in HEK293 cells caused increased SET–DNA interaction in the ALDH2 and GSTP1 promoters (Fig. 3D). These results are in agreement with those presented in Table 1 and indicate that SET negatively regulates ALDH2 and GSTP1 transcription through histone hypoacetylation.

Figure 3.

H2B and H4 histone hypoacetylation in SET-overexpressing HEK293 cells and SET binding in the ALDH2 and GSTP1 promoters. HEK293 and HEK293/SET cells were treated with TSA (100 ng·mL−1) or untreated for 24 h. Immunofluorescence was obtained using acetyl-H2B (K20) IgG (A), acetyl-H4 (K12) IgG (B) and SET (SET/I2PP2A) IgG (A,B). Merge is the overlap between acetyl-histone and anti-SET fluorescence; the nuclei were stained with a 4′,6–diamidino-2-phenylindole dye. The cells were visualized at a ×400 magnification. Five fields were analyzed per slide; the assays were performed in duplicate. (C) HEK293 (SET-) and HEK293/SET (SET+) cells that were treated with TSA (+) or untreated (−) for 24 h were analyzed by immunoblotting. The samples were stained with anti-(acetyl-H2B) (K20), acetyl-H4 (K12) IgG and β-actin IgG. (D) ChIP was performed using an SET IgG and DNA obtained from HEK293 and HEK293/SET cells. The immunoprecipitated ALDH2 and GSTP1 promoter regions were amplified by PCR. The amplicons were analyzed by electrophoresis in 2% agarose gel and stained with ethidium bromide. Input corresponds to total DNA and ChIP to specific DNA–SET protein binding. The immunoprecipitated DNA amount was quantified by qPCR using the primers from ALDH2 and GSTP1 promoters. The enrichment of specific ChIP DNA over input DNA was calculated as a ratio.

ALDH2 and GSTP1 detoxification efficiencies are lower with high SET levels

The detoxification efficiency of ALDH2 and GSTP1 was assessed through the comet assay, considering that xenobiotic accumulation causes DNA damage, whose level is associated with comet tail intensity. This technique detects DNA strand breaks, alkali-labile sites, oxidative base damage or DNA–DNA/DNA–protein/DNA–drug cross-links. HN13 and HN13/siSET cells were exposed to ethanol (Fig. 4A) or cisplatin (Fig. 4B). Greater tail intensity was observed in HN13/siSET cells than HN13 cells, which suggests that SET protein protects DNA integrity; this occurs through the SET ability to hypoacetylate histones and to preserve chromatin in the close status [21]. However, HN13/siSET cells exposed to ethanol exhibited a decrease in DNA damage levels when compared with HN13, suggesting an increase in the ALDH2 activity in association with SET knockdown. In agreement, SET knockdown in HN13 cells (HN13/siSET) increased ALDH2 activity, whereas SET overexpression in HEK293 cells (HEK293/SET) decreased ALDH2 activity (Fig. 4C). These results demonstrate that SET accumulation is associated with DNA damage via decreased ALDH2 activity levels.

Figure 4.

Decrease in the efficiency of detoxification by ALDH2 and GSTP1 proteins in HEK293/SET and HN13 cells. (A) A comet assay was performed in HN13 and HN13/siSET cells that were treated with ethanol for 24 h. Nuclei were stained with a 4′,6–diamidino-2-phenylindole dye and visualized at a ×200 magnification by fluorescence microscopy. One hundred cells were analyzed per slide; the assays were performed in duplicate, and the amount of DNA damage was associated with tail intensity. (B) The comet assay was performed in HN13 and HN13/siSET cells treated with cisplatin for 24 h. (C) ALDH2 enzymatic activity was determined in HEK293, HEK293/SET, HN13 and HN13/siSET cells, based on the oxidation of acetaldehyde to acetic acid in the presence of NAD+. (D) The viabilities of HN13 and HN13/siSET cells treated with cisplatin for 24 h were assessed in quintuplicate by the 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxy phenyl)-2-(4-sulfophenyl)-2H-tetrazolium salt assay. (E) ChIP of the GSTP1 promoter region was performed using an SET IgG in HEK293 and HN13 cells that were treated with cisplatin for 24 h as described in Fig. 3D and the ChIP/INPUT ratio was quantified by qPCR. The values are the mean ± SD. *P < 0.05; **P < 0.005; ***P < 0.001.

GSTP1 detoxification efficiency [22] was assessed through the viability of HN13 and HN13/siSET cells that were exposed to cisplatin for 24 h (Fig. 4D). In the presence of 250 μm cisplatin, HN13/siSET cell viability was greater than that of HN13 cells, which suggests that SET is involved in GSTP1 regulation. The comet assay data confirmed the SET-mediated effect by demonstrating that cisplatin-induced DNA damage was greater in HN13 cells than HN13/siSET cells (Fig. 4B). Moreover, ChIP demonstrated that in cisplatin-exposed cells, SET lost its affinity to the GSTP1 promoter region (Fig. 4E), which was probable because of increased GSTP1 gene expression in the presence of cisplatin. These results are in agreement with the upregulation of GSTP1 mRNA that is observed in HEK293 cells (Fig. 2A).

SET overexpression impairs DDR by reducing BRCA2 recruitment and promotes cell-cycle disturbances in HEK293 cells

Decreased ALDH2 and GSTP1 activity levels affect acetaldehyde/carcinogen detoxification and DDR, which causes DNA damage accumulation; this observation has been associated with susceptibility to head and neck cancers [23].

H2A.X phosphorylation (γH2A.X) is an initial event in DDR and is followed by FANCD2 and BRCA2 recruitment to damaged chromatin sites before repair initiation [7]. In this context, we assessed γH2A.X and BRCA2 using immunofluorescence. Both cisplatin-exposed HEK293 and HEK293/SET cells exhibited an increased amount of γH2A.X (Fig. 5A) consistent with DNA damage promoted by cisplatin treatment. By contrast, both cisplatin-exposed and unexposed HEK293/SET cells exhibited less nuclear BRCA2 recruitment compared with HEK293 cells (Fig. 5B). Therefore, SET overexpression in HEK293 cells may inhibit DDR by reducing BRCA2 recruitment, which is consistent with the higher DNA damage level that is observed in HN13 cells compared with HN13/siSET cells after cisplatin treatment (Fig. 4B). Notably, ethanol and cisplatin exhibited different profiles with respect to DDR signaling that involved SET.

Figure 5.

Failure of DDR and cell-cycle arrest in HEK293/SET cells. (A) HEK293 and HEK293/SET cells were exposed to 50 μm cisplatin or 5% ethanol for 24 h and immunostained with γH2A.X IgG (A) or BRCA2 (B) IgG. The cells were visualized at a ×400 magnification using fluorescence microscopy. Five fields were counted per slide for a positive nuclei; the assays were performed in duplicate. (C) cDNA was prepared from HEK293 and HEK293/SET cells and the gene expression of ATM, BRCA1, CHEK2 and TP53 was analyzed by PCR Array Systems (Sabiosciences–Qiagen). Relative mRNA expression levels were calculated using the 2−ΔΔCT method. (D) Immunoblotting analyses of HEK293 cells incubated with 50 μm cisplatin, 5% ethanol or 40 nm acetaldehyde. Total proteins were analyzed using SET, p53, HDAC1 and β-tubulin IgG. (E) Flow cytometry analysis of the cell cycle in HEK293 and HEK293/SET cells exposed to 50 μm cisplatin or 5% ethanol for 24 h. The assays were performed in triplicate for two independent experiments. The values are the mean ± SD. *P < 0.05; **P < 0.005; ***P < 0.001.

ATM, BRCA1, CHEK2 and p53 are important proteins involved in DNA damage repair mechanisms [7, 8]; hence, mRNA of their genes was quantified by qPCR in HEK293 and HEK293/SET cells. SET decreased mRNA levels of ATM, BRCA1 and CHEK2 and increased mRNA levels of TP53 (Fig. 5C).

The p53 protein is involved in the cellular response to DNA damage, apoptosis control, DNA repair and cell-cycle progression [8, 24]. Histone deacetylases (HDACs), in turn, are involved in chromatin modification and downregulation of target gene expression. In this context, we assessed p53 and HDAC1 proteins using immunoblotting (Fig. 5D). Cisplatin-treated HEK293 cells exhibited an increased amount of p53 protein and no increase in HDAC1 levels; no effects were observed with either ethanol or acetaldehyde (40 ng·μL−1) treatment. Cisplatin forms DNA adducts that activate p53 and CHK2 proteins, which either remove DNA adducts [25] or accumulate γH2A.X [24] or induce G0/G1 cell-cycle arrest associated with p21 activation [26].

SET-mediated action in cell-cycle control has been reported and attributed to its cooperation with p21Cip1 in the inhibition of cyclin B–CDK1 activity [27]. This activity was monitored by flow cytometry in HEK293 and HEK293/SET cells exposed to cisplatin or ethanol (Fig. 5E). SET overexpression was sufficient to disturb the cell cycle, which was characterized by a decrease in the G0/G1 phase and an increase in the G2/M phase. The cell-cycle response to cisplatin was different in the presence and absence of SET overexpression. Cisplatin increased the G0/G1 and S phases and decreased G2/M phase, but in the presence of SET overexpression, it decreased G0/G1 and S arrest and increased the cell count in the G2/M phase. HEK293 and HEK293/SET cells presented similar responses in the presence of ethanol, i.e. an increased cell count in the S and G2/M phases. These results indicate that SET overexpression interferes with the cellular response to cisplatin by both abrogating cell-cycle arrest in the G0/G1 phase and increasing the cell count in the G2/M phase.

SET accumulation impairs DDR in HNSCC cell lineages

The DDR mechanism in HNSCC cells exhibited a profile that is concordant with that of HEK293/SET cells, i.e. cisplatin exposure increased γH2A.X (Figs 6A and S2A) and BRCA2 (Figs 6B and S2B) levels in the HNSCC cell lineages with SET knockdown. This effect reinforces the role of SET in DDR and the negative relationship between SET and BRCA2 recruitment. By contrast, in HN13 cells, SET knockdown decreased γH2A.X levels in the presence of ethanol (Fig. 6A), which implies a lower amount of DNA damage in association with higher BRCA2 recruitment.

Figure 6.

The DDR in HN13 and HN13/siSET cells. HN13 control and SET knockdown cells were treated with 50 μm cisplatin or 5% ethanol for 24 h and incubated with γH2A.X IgG (A) or BRCA2 (B) IgG. The cells were visualized at a ×400 magnification by fluorescence microscopy. Five fields per slide were analyzed; the assays were performed in duplicate. The ATM, BRCA1, CHEK2 and TP53 gene expression was obtained through analyses of HN13 and HN13/siSET cDNAs by PCR Array Systems (Sabiosciences–Qiagen). Relative mRNA expression levels were calculated using the 2−ΔΔCT method (C). The values are given as mean ± SD. *P < 0.05, **P < 0.005, ***P < 0.001.

The mRNA levels of ATM, BRCA1, CHEK2 and TP53 were assessed in HN13 and HN13/siSET cells. The ATM, BRCA1 and CHEK2 mRNA levels were increased and TP53 mRNA levels were decreased by SET knockdown (Fig. 6C). This result agrees with what was observed in HEK293 cells where SET overexpression downregulated ATM, BRCA1 and CHEK2 and upregulated TP53 mRNA levels (Fig. 5C), and suggests a SET role in the regulation of DNA repair genes.

The NAP domain of the SET protein is associated with ALDH2 and GSTP1 gene regulation in HEK293 cells

The SET protein bears distinct regions for chromatin interaction, among them the NAP domain, which mediates histone fold-specific binding and the C–terminal region (acidic motif), which targets the basic N–terminal histone tails [28]. We performed qPCR to assess whether one of these SET regions (Fig. S1B,C) was specifically involved in the regulation of ALDH2 and GSTP1 gene expression. Figure 7A indicates that the NAP domain rather than the acidic motif is implicated in the decreased ALDH2 and GSTP1 mRNA expression levels. Indeed, ChIP results (Fig. 7B) demonstrated that the NAP domain could interact with the ALDH2 and GSTP1 promoter regions.

Figure 7.

The NAP domain of SET and regulation of ALDH2 and GSTP1 gene expression. (A) qPCR analysis was performed to evaluate ALDH2 gene expression in HEK293 cells overexpressing SET (HEK293/SET), NAP domain (HEK293/NAP) or acidic region (HEK293/Ac). (B) ChIP of ALDH2 and GSTP1 promoters was performed and quantified as described in Fig. 3D. DNA was extracted from cells and immunoprecipitated with an SET IgG. Input corresponds to total DNA and ChIP to specific DNA–SET binding.

Discussion

Several risk factors and genetic susceptibility to alcohol and tobacco consumption are implicated in the development of many cancer types, HNSCC in particular. ALDH2 detoxifies acetaldehyde, which is a primary metabolite of ethanol [29]. GSTP1, in turn, detoxifies carcinogens that are derived from tobacco and other environmental toxins [3]. A lack or decreased amount of ALDH2 and GSTP1, which implies a lower enzymatic activity, has been associated with many cancer types including lung, bladder and HNSCC [23]. In this study, we report decreases in both ALDH2 and GSTP1 gene expression and ALDH2 and GSTP1 protein activity in HNSCC cell lineages, which accumulate SET constitutively, and in SET-overexpressing HEK293 cells, which indicates SET involvement in the regulation of these enzymes (i.e. the amount, which reflects in the activity level).

SET protein (template-activating factor-I β/I2PP2A) is involved in many cellular functions that include the following: (a) inhibition of active demethylation of DNA [30], (b) interaction and cooperation with p21 blocking cyclin B–CDK1 activity [27], (c) stimulation of cell motility in association with activated Rac1 [31], (d) apoptosis signaling in Alzheimer's disease [32], and (e) cell death and survival under oxidative stress [19, 20]. In addition, SET is a member of the inhibitor of histone acetyltransferase complex that is responsible for inhibiting histone–histone acetyltransferase protein interaction. Consequently, SET influences the histone acetylation state and affects chromatin structure, which promotes transcriptional repression/epigenetic alterations that are responsible for gene silencing [30]. In this context, we demonstrated that SET is involved in the transcriptional repression of ALDH2 and GSTP1 genes and promotes H2B and H4 histone hypoacetylation. Remarkably, decreased ALDH2 and GSTP1 activity levels in the presence of accumulated SET were correlated with the downregulation of their respective mRNA levels and SET–DNA interactions in the promoter regions under the same experimental conditions. The observed reversal of the SET-mediated effects by TSA, a potent inhibitor of histone deacetylases, suggests that SET interacts preferentially with histones that are hypo- or deacetylated through the association of SET with HDAC I and II.

Acetaldehyde is one of the most important carcinogens in oral tumorigenesis. Acetaldehyde accumulation increases the frequency of sister chromatid exchanges and induces chromosomal aberrations, DNA interstrand and DNA–protein cross-linking [4]. The majority of the acetaldehyde that is generated by alcohol ingestion is processed by the ALDH2 enzyme. However, we observed lower ALDH2 activity in both HN13 and HEK293/SET cells, which have SET overexpressed; this finding suggests a role for SET in ALDH2 regulation and, consequently, in the accumulation of acetaldehyde in these cells. Because acetaldehyde promotes DNA damage, this parameter was also assessed. In the presence of ethanol, increased DNA damage was observed in HN13 cells compared with decreased DNA damage in HN13/siSET cells, which indicates that SET overexpression downregulates ALDH2, thereby increasing the carcinogenic potential of ethanol. In addition, the low activity of GSTP1 in HN13 compared with HN13/siSET cells was correlated with a lower protecting capability against cisplatin-induced DNA damage, which indicates that SET overexpression influences GSTP1-promoted cell detoxification. Moreover, it has been reported that the accumulation of DNA damage caused by an excess of carcinogens, such as those from ethanol or tobacco consumption, promotes cell transformation and tumor progression [33].

The accumulation of chemical carcinogens such as cisplatin (antitumor chemotherapy) and acetaldehyde (ethanol metabolism and tobacco) promotes DNA damage that causes apoptosis and cancer development; one way to avoid apoptosis is through cell-cycle arrest, which provides sufficient time for DDR. During cell-cycle arrest, cells activate and recruit a cascade of proteins, such as Fanconi anemia proteins, histone H2A.X, BRCA1 and BRCA2, which are required for complete DNA repair [7]. The BRCA2 profile was assessed after cells were treated with either cisplatin or ethanol. Lower recruitment of BRCA2 was observed in HN13 cells compared with HN13/siSET cells. Accordingly, BRCA2 recruitment was reduced in SET-overexpressing HEK293 cells that were exposed to cisplatin. A BRCA2 deficiency sensitizes cells to accumulate carcinogenic effects, such as cisplatin-induced DNA double-strand breaks [34]. Remarkably, SET overexpression was sufficient to reduce BRCA2 recruitment, which suggests that SET accumulation inhibits BRCA2-mediated DDR with potential accumulation of genome alterations and lead to tumor development.

Phosphorylation of H2A.X (γH2A.X) plays a key role in the recruitment and location of BRCA1 and BRCA2 [24]. We observed that γH2A.X levels decreased in ethanol-exposed HEK293 cells, HEK293 cells that overexpress SET and, in particular, in HN13/siSET cells. Cells with decreased γH2A.X levels can fail to recruit BRCA1 and BRCA2 and proceed with DDR [24]. Also, cells with deficient or truncated BRCA2 exhibit a progressive proliferative impediment with arrest in the G2/M phase [7]. In this regard, we found a negative relationship between SET accumulation and BRCA2 recruitment, as well as the loss of G0/G1 cell-cycle arrest with G2/M checkpoint stalling.

SET, a multifunctional protein has the following two main domains: the N–terminal (NAP), which is responsible for dimerization and histone binding; and the C–terminal acidic tail, which is implicated with chaperone activity [28]. In this study, we demonstrated that the SET NAP domain is specifically responsible for decreasing ALDH2 and GSTP1 gene expression through its interaction with ALDH2 and GSTP1 chromatin, which promotes histone hypoacetylation and decreases ALDH2 and GSTP1 enzymatic activity. The ensuing decrease in detoxification efficiency causes cytotoxic and carcinogenic agent accumulation that either leads to cell death or facilitates tumor development and progression (Fig. 8). This study is the first to report a role for SET in cellular detoxification, DDR and genome integrity, whose accumulation might contribute to either tumorigenesis in the presence of toxic/carcinogenic agents, or sensitivity of tumor cells to cisplatin therapy.

Figure 8.

A proposed model for the role of SET in decreasing cell detoxification efficiency and DDR. (A) In cells with SET overexpression (such as HN13 and HEK293/SET), the SET NAP domain, specifically, binds to chromatin of ALDH2 and GSTP1 promoters and blocks gene expression; this action decreases ALDH2 and GSTP1 expression levels, which affects their activities. (B) The ensuing decrease of detoxification efficiency causes acetaldehyde accumulation and DNA damage, which, in association with the SET-mediated reduction in BRCA2 recruitment, drives apoptosis or tumor development/progression. (C) GSTP1 expression is downregulated by SET overexpression in association with reduced BRCA2 recruitment and p53 protein levels and promotes cell death.

Materials and methods

Cell culture and treatments

HNSCC Cal27 (ATCC, Manassas, VA, USA), HN13 and HN12 human cell lineages [35], as well as HEK293 (ATCC), were maintained in Dulbecco's modified Eagle's medium (Sigma, St. Louis, MO, USA) supplemented with 10% fetal bovine serum (Life Technologies, Rockville, MD, USA). The cells were treated with TSA, LY294002, ethanol, acetaldehyde or cisplatin (Sigma).

Plasmids and reagents

For SET overexpression, a SET human full-length cDNA clone (pCMV SPORT.6; containing coding sequence NM_003011.3) was purchased from Invitrogen (Carlsbad, CA, USA) and transferred to a pcDNA 3.1 vector (Fig. S1A). NAP (1–660 nucleotide position) and acidic domain (660–834 nucleotide position) coding cDNAs of the human SET clone were amplified by PCR and subcloned into a pTZ/57 vector (Fermentas - Thermo Fisher Scientific, Waltham, MA, USA) (Fig. S1B,C). Then, the Gateway system (Invitrogen) was utilized to transfer them into the pDONR and pcDNA 3.1 vectors using BP and LR recombinases. After DNA sequencing, the HEK293 cells were transfected with the DNA constructions using the PolyFect Transfection Reagent (Qiagen, Valencia, CA, USA) following the manufacturer's instructions. The recombinant protein production was confirmed by immunoblotting. For SET siRNA, HNSCC cell lineages were transfected with double-stranded RNA oligonucleotides directed against SET (GS6418; Qiagen) using the HiPerFect Transfection Reagent (Qiagen) following the manufacturer's instructions. Four different double-stranded RNA oligonucleotides were evaluated, and the oligonucleotide that exhibited the greatest efficiency was used in the assays. Optimal siRNA oligonucleotide concentrations and incubation times were determined by constructing a dilution curve; protein knockdown was monitored by immunoblotting. The siCONTROL AllStars siRNA Negative Control (Qiagen) was used as a negative siRNA control (Fig. S1A).

RNA extraction and cDNA

RNA was isolated from cultured cells with TRIzol reagent (Invitrogen) according to the manufacturer's protocol. RQ1 RNase-Free DNase (Promega, Madison, WI, USA) was utilized to treat and precipitate the RNA. Subsequent cDNA synthesis adhered to the SuperScript III Reverse Transcriptase (Invitrogen) protocol.

Quantitative real-time PCR

The mRNA sequences were obtained from the NCBI database (www.ncbi.nlm.nih.gov), and the primers were constructed using the gene runner program (version 3.05; Hastings Software, Inc., Edison, NJ, USA). The sequences were as follows: ALDH2 (NM_000690.3) F–GAGAGTGACCTTGGAGCTGGGGG and R–CTCCGCTCCACAAACTCATCATA, GSTP1 (NM_000852.3) F–GCAAATACATCTCCCTCATCTACAC and R–AGCAGGTTGTAGTCAGCGAAGGAG, β-globin (NM_001101.3) F–GCCTCGCTGTCCACCTTCCA and R–AGAAAGGGTGTAACGCAACTAAG, and GAPDH (NM_002046.4) F–GACTTCAACAGCGACACCCACTC and R–GTCCACCACCCTGTTGCTGTAG. The reactions were performed in a final volume of 10 μL that contained 5 μL of Fast EvaGreen Master Mix (Uniscience, Cambridge, UK), 0.3 μm per primer and 100 ng of cDNA. The cycling conditions were as follows: 2 min at 96 °C, followed by 40 cycles of 15 s at 96 °C, 10 s at 59–65 °C and 25 s at 72 °C. Repair genes ATM, BRCA1, CHEK2 and TP53 mRNA levels were analyzed by PCR Array Systems (Sabiosciences–Qiagen, Valencia, CA, USA) according to the manufacturer's instructions.

ALDH2 activity assay

ALDH2 activity was determined through oxidation of acetaldehyde to acetic acid in the presence of the oxidized form of nicotinamide adenine dinucleotide (NAD+) as cofactor. The assay medium contained 33 mm sodium pyrophosphate, 15 μm acetaldehyde, 0.8 mm NAD+ and 0.1 mL cellular extract (250 μg protein) at 25 °C. NADH absorbance was monitored at 340 nm [36].

Comet assay

The comet alkaline assay, which detects low levels of DNA damage such as strand breaks, alkali-labile sites, oxidation and DNA repair deficiency, was performed as previously described [37] with modifications. The slides were stained with 4′,6–diamidino-2-phenylindole (0.5 ng·mL−1). The assay was performed in duplicate, and for each treatment, the tail intensity of 200 cells was analyzed using the TriTek CometScore (TriTek Corp., Sumerduck, VA, USA). Digital images were obtained using a Zeiss Axiovert 40 CFL Microscope and Zeiss axiovision 4.8.2 software.

Cell viability assay

Cells were seeded in 24-well plates (0.25 × 104 cells), and the rapid colorimetric 3–(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxy phenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt assay (CellTiter96 AQ non-radioactive cell proliferation kit; Promega) was used. Cell viability was determined by the reduction in 3–(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxy phenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt that only detects metabolically active viable and early apoptotic cells. The assay was performed according to the manufacturer's protocol.

ChIP

Cells were cross-linked with 1% formaldehyde for 20 min, washed, lysed and sonicated. The DNA–protein complexes were immunoprecipitated with SET IgG (Santa Cruz Biotechnology, Santa Cruz, CA, USA) and protein G agarose beads (GE HealthCare, Piscataway, NJ, USA). DNA was extracted with phenol/chloroform post-reversing of the DNA–protein cross-links. The DNA was amplified by PCR using the promoter primers of genes ALDH2 (forward CACCTCGTTCATCTCCTTCACC and reverse GCTGCCTCTACCATTCCTCGG) and GSTP1 (forward CCTCTCCCCTGCCCTGTGAAG and reverse GGCGAAACTCCAGCGAAGGC), and the PCR reaction product was analyzed using a 1.5% agarose gel. The ChIP/INPUT ratio was determined by quantitative real-time PCR using 10 ng of each sample, 0.5 μm of the primers sequences described above and 5 μL of Fast EvaGreen Master Mix (Uniscience) in a final volume of 10 μL.

Cell-cycle analysis

Cells were trypsinized and fixed in 70% cold ethanol overnight, stained with propidium iodide (50 μg·mL−1), treated with RNase (100 μg·mL−1) for 30 min at 37 °C, and analyzed by flow cytometry (BD FACSCalibur Analyzer, BD Biosciences - San Jose, CA, USA). Data acquisition and cell-cycle analysis were performed using BD facsdiva software Version 6.1.3.

Immunofluorescence assay

Cells were placed on glass coverslips in a 12- or 24-well plate, fixed with absolute methanol at −20 °C for 6 min, and blocked with 0.5% (v/v) Triton X–100 in NaCl/Pi and 3% (w/v) BSA. Cells were then incubated with anti-acetyl-H4 (K12) (#2591; Cell Signaling, Danvers, MA, USA), anti-acetyl-H2B (K20) (#2571; Cell Signaling), anti-I2PP2A (#sc-5655; Santa Cruz), anti-phospho-histone H2A.X (Ser139) (#05-636; Millipore, Billerica, MA, USA), BRCA2 (#sc-1817; Santa Cruz) IgG overnight and washed three times with NaCl/Pi. After incubation with an Alexa Fluor 546-, fluorescein isothiocyanate- or tetramethylrhodamine isothiocyanate (TRITC)-conjugated secondary antibody for 1 h, the cells were stained with 4′,6–diamidino-2-phenylindole (Sigma) and visualized using a Zeiss Axiovert 40 CFL Microscope and Zeiss axiovision 4.8.2 software.

Immunoblotting analysis

Histone proteins (30 μg) were extracted in a buffer that contained 250 mm NaCl, 50 mm Hepes, pH 7.0, 5 mm EDTA, 0.1% NP–40, 0.5 mm dithiothreitol, 100 ng·mL−1 TSA, 5 mm NaF and 1 mm Na3VO4. All reagents were purchased from Sigma. Total proteins (30 μg) were extracted using CelLytic M cell lysis reagent (Sigma) according to the manufacturer's protocol. The proteins were separated by 12 or 15% SDS/PAGE and transferred to a polyvinyl difluoride membrane (GE Healthcare). The membrane was blocked in 0.1 m Tris (pH 7.5), 0.9% NaCl and 0.05% Tween–20 with 5% nonfat dry milk for 1 h at room temperature and incubated in Tris-buffered saline that contained primary acetyl-H2B (K20), acetyl-H4 (K12), I2PP2A, p53 (#2527; Cell Signaling), histone deacetylase 1 (HDAC1) (#2062; Cell Signaling), β–actin (#47778; Santa Cruz Biotechnology) and β–tubulin (#05-661; Millipore) IgG in 0.1% Tween. Appropriate secondary antibodies conjugated with horseradish peroxidase (Sigma) were used, and the reaction was visualized by enhanced chemiluminescence (ECL, GE HealthCare or SuperSignal West Pico Substrate; Pierce Biotechnology, Rockford, IL, USA).

Statistical analyses

Statistical analyses were performed using graphpad prism software (version 5.0). Student's t-test or ANOVA was used to examine the association between media and treatments. Values of P < 0.05 were considered significant. Relative quantification analysis of real-time PCR assays were calculated using the 2−ΔΔCT method (2−[(CT sample − CT sample housekeeping gene) − (CT calibrator − CT calibrator housekeeping gene)]) [38]. This analysis is used to determine the gene expression profile of a treated sample in relation to a calibrator (control sample).

Acknowledgements

This work was supported by FAPESP (research grants: 2010/20384-0 and 2009/52228-0; fellowships: 2009/10783-7, 2010/20536-4 and 2010/08328-7) and CNPq. The authors thank Cristiana Bernadelli Garcia for her excellent technical assistance.

Ancillary