NAT10 is upregulated in hepatocellular carcinoma and enhances mutant p53 activity

The hepatoma cell lines Huh7 (mutant p53 Y220C) was obtained from COBIOER BIOSCIENCES CO., LTD (COBIOER, Nanjing, China) and HepG2 (wild-type p53), MHCC-97H (R249S), MHCC-97 L (R249S) and the normal hepatic cell line LO2 (wild-type p53) were gifts from Prof. Curtis C. Harris. Liver cancer cells were cultured and maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum at 37 °C in a humidified atmosphere containing 5% CO₂. Cells were transfected with plasmid DNA or siRNA duplexes by using Lipofectamine® 2000 (Invitrogen, CA, USA) according to the manufacturer’s protocol. For silencing NAT10 expression, a small interfering RNA (siRNA) targeting NAT10 (sequence: 5′-CAGCACCACUGCUGAG AAUAAGA-3′), Mdm2 (sequence: 5′-AAGCCAUUGCUUUUGAAGUUA-3′) was synthesized, together with the control siRNA (5′-ACUACCGUUGUUAUAGGUG-3′; Shanghai GenePharma Co., Ltd).

FLAG-tagged NAT10 and NAT10 mutants were cloned into pCI-neo. Anti-p53 (DO-1), anti-actin (C-11), anti-Mdm2 (SMP14) and anti-p21 (817) antibodies were purchased from Santa Cruz Biotechnology, Inc. Anti-Flag (F3165) antibody was purchased from Sigma. Anti-NAT10 antibody was a gift from Dr. B. Zhang.

Preparation of cellular extracts was performed as described previously [26]. In Brief, cells were harvested and washed with PBS. Then, cells were lysed in ice-cold H lysis buffer A (10 mM Tris-HCl pH 7.4, 10 mM KCl, 2 mM MgCl₂, 0.05% Triton™ X-100, 1 mM DTT, 1 mM EDTA and fresh proteinase inhibitors). Next, the nuclear pellet was collected after centrifugation for 10 min at 2000 rpm, and the supernatant was collected as the cytoplasmic extract. The crude nuclear pellet was suspended in T Lysis Buffer (20 mM HEPES pH 7.9, 420 mM NaCl, 0.2 mM EDTA, 1.5 mM MgCl₂, 0.5 mM DTT, and 10% glycerol with protease inhibitor mixture) and swollen at 4 °C for 30 min. The homogenate was centrifuged for 15 min at 12000 rpm. Nuclear and cytoplasmic fractions were subjected to Western blot using the indicated antibodies.

Total proteins were extracted and immunoblotting was performed as the standard procedures. Then, the immunoreactivity was detected with ECL Western blot Detection Reagent (GE Healthcare).

Immunoprecipitation assay was performed as described previously [27]. In brief, Huh7 cellular lysates were prepared in lysis buffer A (25 mM Tris-HCl pH 7.5, 100 mM KCl, 1 mM DTT, 2 mM EDTA, 0.5 mM phenylmethylsulfonyl fluoride, and 0.1% Nonidet P-40). Cellular extracts were obtained by centrifugation for 30 min at 12000 rpm. Specific antibodies were incubated with 15 ul protein A and G beads (Amersham Biosciences) in IPP500 (500 mM NaCl, 10 mM Tris-HCl pH 8.0, 0.1% Nonidet P-40). Coupled beads were incubated with cellular extracts for 2 h at 4 °C. After extensive washes, the precipitated proteins were subjected to Western blotting.

Cell growth curve was analyzed using the Cell Counting Kit-8 (CCK-8, Dojindo) according to the manufacturer’s directions. Briefly, Huh7 or MHCC-97 L cells were transfected with the indicated siRNAs (500 cells per well) and grew in 10% serum containing media. Cell numbers were estimated at day 0, 1, 2, 3, 4 and 5. The growth curve shows the mean ± standard deviation from three technical replicates.

Human HCC tissues and adjacent noncancerous tissues for western blotting were obtained from 19 patients with HCC who underwent tumor resection at the Beijing Cancer Hospital. After resection, specimens were rinsed thoroughly in ice-cold normal saline and stored in liquid nitrogen.

Sections were obtained from 119 formalin-fixed, paraffin-embedded human HCC tissues and corresponding non-cancerous tissues of the same patient undergoing surgical resection without prior neoadjuvant therapy between January 2003 and October 2006 in the Beijing Cancer Hospital. The clinico-pathological patient characteristics are summarized in Table 1.

Sections (4 μm thick) were dewaxed in xylene and gradually rehydrated gradually. After antigen retrieval, endogenous peroxidases were blocked with 3% hydrogen peroxide. Then, the sections were incubated with 10% goat serum for 30 min at room temperature. Sections were incubated with rabbit anti-NAT10 polyclonal antibody at 4 °C overnight and then with HRP-conjugated goat anti-rabbit IgG (Zhongshan Golden bridge Biotechnology, Beijing, China) at 37 °C for 30 min. Immunocomplexes were visualized by using 3,3-diaminobezidine (DAKO, CA, USA). Slides were counterstained with light hematoxylin, dehydrated, and cover-slipped.

Histological slides were assessed by two independent observers, including an experienced pathologist, blinded to all clinical, pathological, and outcome information. The score discrepancies were discussed to achieve a consensus. Immunostaining was categorized into four groups: negative (0 score), 0%–10% positive cells; faintly positive (1 score), 10%–25% positive cells; moderately positive (2 scores), 25–50% positive cells; and highly positive (3 scores), ≥50% positive cells.

To determine the significance of NAT10 in hepatocellular carcinomas, we performed immunoblotting on human HCC tissues and their matched noncancerous tissues. Fourteen of 19 (73.7%) tumor samples showed increased NAT10 protein levels compared with their respective paired noncancerous tissue (Fig. 1a). These data indicated a positive correlation of NAT10 expression with HCC.

Next, we investigated the correlation of NAT10 expression with HCC progression. Immunohistochemical staining was performed to evaluate NAT10 expression on primary human tumors from a large cohort of HCC patients (n = 119). Among these 119 patients, all biopsy specimens contained both tumors and matched non-tumorous tissues. Consistent with our previous study, NAT10 was expressed in the nuclei of human HCC tumor cells (Fig. 2a). For further evaluation of the expression level of NAT10, the staining level was graded and scored from 0 to 3. According to the staining score, all patients was subgrouped as weak expression (staining score 0–1) versus strong expression (staining score 2–3). Strong expression of NAT10 was detected in 101 of 119 cases (84.8%) of HCC tumor tissues, whereas NAT10 expression was not detected in their benign counterparts (Fig. 2b and c). Thus, NAT10 expression was significantly upregulated in HCC tumor tissues compared with their non-tumorous counterparts.

The correlation between NAT10 expression and clinico-pathological variables was analyzed using SPSS version 17. As shown in Table 1, NAT10 expression was significantly correlated with vascular invasion (p < 0.05). However, NAT10 expression did not correlate significantly with age, α-fetoprotein (AFP) levels, capsular formation, tumor number, margin status and Edmondson-Steiner grade. In addition, as indicated by Kaplan-Meier analysis, high level expression of NAT10 was associated with shorter overall survival (OS; Fig. 2d) in our cohort (p < 0.01). According to this result, we further investigated whether NAT10 expression can affect the prognosis of HCC patients independently. Univariate analysis by Cox-regression revealed that 5 prognostic factors affecting OS: NAT10 expression level, tumor size, tumor number, microvascular invasion and lymph node metastasis. Multivariate analysis by cox-regression revealed that NAT10 expression, tumor number, microvascular invasion, and lymph node metastasis were independent prognostic factors of OS (Table 2). These data demonstrated that NAT10 is an independent prognostic factor for HCC patients.

A previous study had demonstrated that NAT10 regulates p53 activation [26] and that p53 is frequently mutated in HCC [14]. Therefore, we next investigated whether NAT10 regulates mutant p53 activity in HCC. We compared the NAT10 and p53 protein levels from surgically removed human HCC samples by using immunoblotting. As shown in Fig. 3a and b, p53 was upregulated in 16 of 19 (84.2%) tumor samples, indicating that these tumor samples carry p53 mutations. Notably, we found that NAT10 and p53 levels was positively correlated (r² = 0.4, p = 0.03) in the tumor samples with co-upregulation of NAT10 and p53 (Fig. 3c). NAT10 protein levels were also positively correlated with p53 in the HCC cell lines (Fig. 3d). Together, the results above indicate that increased NAT10 expression is correlated with p53 level in HCC.

To understand the molecular mechanism by which NAT10 regulates mutant p53 in HCC, we investigated whether NAT10 interacts with mutant p53. Extract from cytoplasm and nucleus were fractionated and subjected to Western blotting to evaluate NAT10 expression. As shown in Fig. 4a, NAT10 was detected in the nuclear extracts. Immunofluorescence staining showed that NAT10 was partially colocalized with p53 in the nucleoli (Fig. 4b). Co-immunoprecipitation confirmed that NAT10 bound to mutant p53 in the HCC cell line Huh7 carrying the p53 mutation (Fig. 4c). These findings indicate that NAT10 interacts with mutant p53.

Given the fact that NAT10 regulates p53 activity and in light of our findings that NAT10 level is correlated with mutant p53 level in HCC, we hypothesized that NAT10 also regulates mutant p53 stability. In agreement with our hypothesis, depletion of NAT10 decreased mutant p53 levels; however, no alteration of p21 levels was observed (Fig. 4d). This result was consistent with previous reports that mutant p53 loses the ability to activate wild-type p53 target genes [28, 29]. Furthermore, knockdown of NAT10 increased ubiquitination of mutant p53 (Fig. 4e), indicating that NAT10 regulates mutant p53 ubiquitination and stability. Recent reports suggest that mutant p53 is still under the regulation of Mdm2 [30, 31] and NAT10 modulates Mdm2 activity. Thus, we next analyzed whether NAT10 regulates p53 stability by counteracting Mdm2 action. As shown in Fig. 4f, ectopic expressed Mdm2 could enhance mutant p53 ubiquitination, indicating that Mdm2 could still target mutant p53 to decompose. Importantly, coexpression of NAT10 counteracted the Mdm2-induced ubiquitination of mutant p53 (Fig. 4f). Moreover, the deletion mutant NAT10-D5, which lost the ability to inhibits Mdm2 activity, failed to do so (Fig. 4f, lane 3 vs. lane 4). In addition, NAT10 had no effect on mutant p53 stability in Mdm2-depleted cells (Fig. 4g). The interaction between Mdm2 and NAT10 was further verified by co-immunoprecipitation (Fig. 4h). Taken together, these data indicated that NAT10 regulates mutant p53 stability through counteracting Mdm2 action. Given that mutant p53 often displays acquisition of the ability to potentiates cell proliferation [32], we investigated whether increased NAT10 in cells with mutant p53 could be advantageous to cell proliferation, in contrast to cells with wild-type p53. As expected, downregulation of NAT10 in Huh7 cells resulted in decreased cell proliferation (Fig. 4i). The results were confirmed in MHCC-97 L cells which also carry mutant p53. Overexpression of NAT10 in MHCC-97 L cells enhanced cell proliferation, while ectopic expression of NAT10 had little effect on cell proliferation in p53-depleted MHCC-97 L cells (Fig. 4j). Thus, NAT10 promotes cell proliferation in cells expressing mutant p53. Together, our findings demonstrated that NAT10 regulates mutant p53 and promotes cell proliferation in cells carrying mutant p53.