Regulation of HK2 expression through alterations in CpG methylation of the HK2 promoter during progression of hepatocellular carcinoma - PubMed (original) (raw)

Regulation of HK2 expression through alterations in CpG methylation of the HK2 promoter during progression of hepatocellular carcinoma

Hyun Gyu Lee et al. Oncotarget. 2016.

Abstract

Hexokinase 2 (HK2) is a rate-determining enzyme in aerobic glycolysis, a process upregulated in tumor cells. HK2 expression is controlled by various transcription factors and epigenetic alterations and is heterogeneous in hepatocellular carcinomas (HCCs), though the cause of this heterogeneity is not known. DNA methylation in the HK2 promoter CpG island (HK2-CGI) and its surrounding regions (shore and shelf) has not previously been evaluated, but may provide clues about the regulation of HK2 expression. Here, we compared HK2 promoter methylation in HCCs and adjacent non-cancerous liver tissues using a HumanMethylation450 BeadChip array. We found that, while the HK2-CGI N-shore was hypomethylated, thereby enhancing HK2 expression, the HK2-CGI was itself hypermethylated in some HCCs. This hypermethylation suppressed HK2 expression by inhibiting interactions between HIF-1α and a hypoxia response element (HRE) located at -234/-230. HCCs that were HK2negative and had distinct promoter CGI methylation were denoted as having a HK2-CGI methylation phenotype (HK2-CIMP), which was associated with poor clinical outcome. These findings indicate that HK2-CGI N-shore hypomethylation and HK2-CGI hypermethylation affect HK2 expression by influencing the interaction between HIF 1α and HRE. HK2-CGI hypermethylation induces HK2-CIMP and could represent a prognostic biomarker for HCC.

Keywords: HIF-1α; HK2-CIMP; hexokinase 2; hypoxia response element; HumanMethylation450 BeadChip.

PubMed Disclaimer

Conflict of interest statement

The authors have declared no conflicts of interest in relation to this manuscript.

Figures

Figure 1

Figure 1. Two different alterations of CpG DNA methylation in the HK2 promoter: hypomethylation at the _HK2_-CGI N-shore and hypermethylation at the _HK2_-CGI observed only in HK2negative_HK2_-CIMP HCCs

(A) A comparison of HK2 promoter methylation status between HCC and Adj-NCL tissues. The _HK2_-CGI located at the range of −388 to +500 bp from the transcriptional start site (TSS: +1) was predicted by Methprimer (

http://www.urogene.org/cgi-bin/methprimer/methprimer.cgi

, lower panel). (B) The methylation status of the HK2 promoter according to HK2 expression in HCC tissues. HK2positive or HK2negative was defined by immunoblot. The β-values of CpGs in the HK2 promoter were plotted. Two vertically dashed line indicted the borders of the _HK2_-CGI. Values represent the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ns, not significant.

Figure 2

Figure 2. Hypermethylation of the _HK2_-CGI caused by the crosstalk of DNMTs and HMTs suppresses HK2 expression in HCC cell lines

(A) HK2 protein expression in HCC cell lines (Hep3B, SNU475, and SNU449 cells) was evaluated by immunoblot and RT-PCR. (B) Methylation status of HK2 promoter among HCC cell lines. The _HK2_-CGI is indicated by dashed lines. (C) Bisulfite sequencing of the _HK2_-CGI in Hep3B, SNU475, and SNU449 cells. Each circle indicates an individual CpG site. An individual row represents a single clone. The open and closed circles denote unmethylated and methylated CpG sites, respectively. (D) DNMTs (DNMT1, DNMT3A, and DNMT3B) and HMTs (EZH2 and G9A) were evaluated in HCC cell lines by immunoblot. (E) The chromatin status of indicated cell lines was evaluated by H3K4me3-, H3K9me3-, and H3K27me3-ChIP assay.

Figure 3

Figure 3. Identification of −234/−230 HRE, a key region in the HK2 promoter regulated by methylation

(A) The promoter activity of HK2 promoter-deleted luciferase constructs was evaluated in HCC cell lines under normoxic and hypoxic conditions. Luciferase activity of all test constructs was normalized to that of the null construct. The relative luciferase activity was plotted as the percentage of the -1921 construct under normoxic or hypoxic conditions. (B) The promoter activity of HK2 promoter-deleted luciferase constructs under normoxic or hypoxic conditions using Hep3B cells (rectangle; putative HRE, circle; putative Sp1-binding site). (C) The HK2 promoter site-specific mutant luciferase constructs for putative HIF-1α and Sp1 binding sites were constructed as described in the upper panel. The promoter activity of each mutant construct under normoxic or hypoxic conditions is shown as the luciferase activity relative to the −175 construct (red closed circle; putative Sp1-binding site, green open rectangle; putative HRE). (D) The luciferase activity of the −965W and −965M under normoxic (N) or hypoxic (H) conditions. (E) The interaction between the −234/−230 HRE and HIF-1α was evaluated by ChIP assay. Sp1 was used as a positive control. The −302/−114 region and a non-relevant region (480/792) designed as shown in the upper panel were amplified by PCR. The specific interaction was plotted as the percentage of the input in the lower panel. (F) EMSA experiment involving the −234/−230 HRE and the mutant version (HREm) on the human HK2 promoter. The oligonucleotides shown in the upper panel were labeled and incubated with nuclear extracts from Hep3B cells. NS, non-specific bands.

Figure 4

Figure 4. The induction of HK2 expression in HK2negative SNU449 cells by treatment with 5-Aza-CdR and hypoxia

(A) Hypoxia-mediated HK2 expression. (B) The suppression of HK2 expression by HIF-1α silencing. (C) The methylation status of the HK2 promoter CpGs plotted for the 5-Aza-CdR-treated SNU475 cells and SNU449 cells. The difference in methylation frequency between 5-Aza-CdR-treated cells and untreated cells is shown in each lower panel. (D) The induction of HK2 expression in SNU475 and SNU449 cells by treatment with 5-Aza-CdR for 2 d, followed by hypoxic stimuli for 1 d. In all experiments, the expression of HIF-1α and HK2 were evaluated by immunoblot. (E) The interaction between the −234/−230 HRE and HIF-1α following 5-Aza-CdR treatment was evaluated using a ChIP assay.

Figure 5

Figure 5. _HK2_-CIMP, a distinct subgroup of HCCs

(A) The methylation status of _HK2_-CGI and its N-shore, according to HK2 expression. Unsupervised hierarchical clustering of (B) the top 1,417 significantly differentially methylated CpG sites between 13 HK2positive and 10 HK2negative HCCs and (C) further selected 289 CpG sites located in the promoter CGI which corresponds to TSS1500, TSS200, 5′UTR and 1st Exon. (D) Overall survival after surgery was compared using Kaplan-Meier analysis (Left panel) between the HK2 upper 25th percentile (n = 8) and HK2 lower 75th percentile (n = 24), and (Right panel) between HK2 upper 25th percentile (n = 8) and HK2 lower 75th percentile except _HK2_-CIMP HCCs (n = 19), which were plotted separately as an _HK2_-CIMP group (n = 5).

Figure 6

Figure 6. A model for opposing regulatory mechanisms of HK2 expression, according to the methylation status of the _HK2_-CGI and its N-shore

In normal liver cells, the HK2 promoter is initially hypermethylated (closed circles) at the _HK2_-CGI N-shore and hypomethylated (open circles) at the _HK2_-CGI. In HCCs with HK2 expression, the _HK2_-CGI N-shore is progressively hypomethylated, while the _HK2_-CGI remains unmethylated, as evidenced by active chromatin marks (H3K4me3 high, H3K9me3 low, and H3K27me3 low). These modifications enable HIF-1α to access the HRE in the _HK2_-CGI, resulting in HK2 expression. In _HK2_-CIMP HCCs, the _HK2_-CGI is methylated and is accompanied by the poised status of chromatin (H3K4me3 high and H3K27me3 high), which inhibits HIF-1α from binding to the HRE in the _HK2_-CGI, resulting in HK2 suppression. The dashed circles, triangles, and stars denote that the methylation status of H3 lysine was not determined.

References

    1. Wilson JE. Isozymes of mammalian hexokinase: structure, subcellular localization and metabolic function. J Exp Biol. 2003;206:2049–2057. - PubMed
    1. Mathupala SP, Ko YH, Pedersen PL. Hexokinase II: cancer's double-edged sword acting as both facilitator and gatekeeper of malignancy when bound to mitochondria. Oncogene. 2006;25:4777–4786. - PMC - PubMed
    1. Peng SY, Lai PL, Pan HW, Hsiao LP, Hsu HC. Aberrant expression of the glycolytic enzymes aldolase B and type II hexokinase in hepatocellular carcinoma are predictive markers for advanced stage, early recurrence and poor prognosis. Oncol Rep. 2008;19:1045–1053. - PubMed
    1. Gong L, Cui Z, Chen P, Han H, Peng J, Leng X. Reduced survival of patients with hepatocellular carcinoma expressing hexokinase II. Med Oncol. 2012;29:909–914. - PubMed
    1. Kwee SA, Hernandez B, Chan O, Wong L. Choline kinase alpha and hexokinase-2 protein expression in hepatocellular carcinoma: association with survival. PloS one. 2012;7:e46591. - PMC - PubMed

MeSH terms

Substances

LinkOut - more resources