Methylation target array for rapid analysis of CpG island hypermethylation in multiple tissue genomes

Abstract

Hypermethylation of multiple CpG islands is a common event in cancer. To assess the prognostic values of this epigenetic alteration, we developed Methylation Target Array (MTA), derived from the concept of tissue microarray, for simultaneous analysis of DNA hypermethylation in hundreds of tissue genomes. In MTA, linker-ligated CpG island fragments were digested with methylation-sensitive endonucleases and amplified with flanking primers. A panel of 468 MTA amplicons, which represented the whole repertoire of methylated CpG islands in 93 breast tumors, 20 normal breast tissues, and 4 breast cancer cell lines, were arrayed on nylon membrane for probe hybridization. Positive hybridization signals detected in tumor amplicons, but not in normal amplicons, were indicative of aberrant hypermethylation in tumor samples. This is attributed to aberrant sites that were protected from methylation-sensitive restriction and were amplified by PCR in tumor samples, while the same sites were restricted and could not be amplified in normal samples. Hypermethylation frequencies of the 10 genes tested in breast tumors and cancer cell lines were 60% for GPC3, 58% for RASSF1A, 32% for 3OST3B, 30% for HOXA5, 28% for uPA, 25% for WT1, 23% for BRCA1, 9% for DAPK1, and 0% for KL. Furthermore, hypermethylation of 5 to 7 loci of these genes was significantly correlated with hormone receptor status, clinical stages, and ages at diagnosis of the patients analyzed. This novel approach thus provides an additional avenue for assessing clinicopathological consequences of DNA hypermethylation in breast cancer.

Figure 1.

A: Methylation-sensitive restriction test. The efficiency of digestion in CpG fragments in BRCA1 and p16^INK4 was examined by subjecting duplicate tubes of genomic DNA, one doubly-digested with BstUI and HpaII (D) and the other mock-digested without enzymes (U). Both types of DNA were amplified under identical conditions for 17 to 35 cycles and the resultant products separated on 1.5% agarose gel. The product bands are visualized with ethidium bromide. A molecular weight ladder (M) is shown on left. A differential presence of PCR products was clearly demonstrated between the digested templates and the mock-digested templates in these two genes. B: Amplification of gene-specific methylation fragments in breast cancer. Pooled genomic DNA from 20 to 30 normal (N) controls or 20 to 30 breast tumors (T) was digested with a 4-base frequent cutter, ligated to end linkers, and further treated with methylation-sensitive BstUI and HpaII (see detailed description in the text). Subsequent PCR (20 cycles) was conducted with primers specific for promoter CpG islands (listed at top). PCR products were separated on 1% to 1.5% agarose gels. Genomic fragments containing methylated sites were protected from the digestion and could be amplified, while fragments containing unmethylated sites were restricted away and not present in the amplified samples.

Figure 2.

The content of repetitive sequences in different methylation targets. A: Serial dilution of different cutter-generated (MseI, Tsp509I, NlaIII, or BfaI) methylation targets (15 to 500 ng) and an identical amount of human Cot-1 DNA were dotted on the nylon filter and hybridized with radiolabeled Cot-1 probe. B: The hybridization signals shown in A were quantified by a densitometer and subsequently standardized by its respective Cot-1 spots, as discussed in the text.

Figure 3.

Map locations of restriction sites in the 10 genes analyzed by MTA. The shaded horizontal bar indicates that the first exon region and its position designation is relative to the transcription start site (+1). The short vertical bars indicate the flanking 4-base endonuclease recognition sites of NlaIII (N), Tsp509I (T), BfaI (Bf), and MseI (M), and the methylation-sensitive HpaII (H) or BstUI (B) sites in between. The horizontal line with flanking black circles represents the area analyzed by methylation-specific PCR. The thick horizontal line denotes the probe location for methylation target array analysis. Note that less GC-rich regions within the selected fragments were chosen as probes to prevent non-specific hybridization. The thin horizontal line under the schematics on right represents the distance of 100-bp.

Figure 4.

A: Methylation Target Array (MTA). The MTA chart is used to indicate the location of each arrayed target. T, tumor; N, normal control; Pos, positive control, ie, target prepared without methylation-sensitive restriction; and Neg, negative control, ie, water. T77 to T80 represent targets prepared from breast cancer cell lines ZR-75, T47D, MDA-MB-468, and Hs578t (see detailed description in the text). Shown here are the results of MTA filters hybridized with radiolabeled p16^INK4 and Cot-1. Samples marked by triangle indicate DNA hypermethylation independently confirmed by methylation-specific PCR (MSP). Except for T23, the asterisks indicate no methylation detected by MSP. B: Methylation-specific PCR (MSP). PCR of bisulfite-treated genomic DNA was amplified with a methylated pair of primers (M) or an unmethylated pair of primers (U). Positive (+) and negative (−) controls were the Universal Methylated DNA (Intergen), respectively. PCR products were separated on 1% to 1.5% agarose gels. Breast tumors showing hypermethylation are marked in italics, while the remaining tumors show no methylation. C: Representative results of methylation target array (MTA) analysis. The restriction map of each gene is shown in Figure 3 ▶ . Methylation targets were prepared as described in the text and hybridized with a ³²P-labeled probe. Positive hybridization signals, which are indicative of DNA methylation for a specific gene, are presented in each MTA panel. The grid location of each arrayed target is the same as presented in A.

Figure 5.

Profiles of Methylation Target Array in breast cancer. A: The methylation status of 10 promoter CpG islands in 93 breast tumors, 4 breast cancer cell lines, and 20 normal breast tissues. A filled box indicates the presence of methylation and an open box indicates the absence of methylation. CpG island loci are indicated above each column. The order of cell lines is arranged from top to bottom - T47-D, MDA-MB-231, ZR75–1, and Hs578t. B: Concurrent DNA hypermethylation was found in tumors hypermethylated in GPC3 and RASSF1A: subgroup a had additional methylation in HOXA5, WT1, and uPA genes and subgroup b had additional methylation in 3OST3B, BRCA1, and DAPK1 genes.

Figure 6.

Methylation and expression analysis of the 3OST3B gene. Breast cancer cell lines were cultured in the absence or presence of 5-aza-2′-deoxycytidine (DeoxyC; 750 nmol/L) for 6 days and harvested for DNA and RNA isolation. A: Methylation-specific PCR. PCR of bisulfite-treated genomic DNA was amplified with a methylated primer (M) or an unmethylated primer (U); see Table 1 ▶ for primer information. Positive (+) and negative (–) controls were the Universal Methylated DNA (Intergen) and blood, respectively. PCR products were separated on 1% to 1.5% agarose gels. B: Reverse-transcription PCR. PCR was conducted using primers (see Table 1 ▶ ) derived from the 3′-end of 3OST3B. The levels of 3OST3B mRNA were normalized with the level of β-actin mRNA. C: Northern blot analysis. Ten μg of RNA from DexoyC-treated and untreated samples were electrophoresed on a 1.5% agarose gel and subjected to Northern analysis using a 3OST3B cDNA probe.

Figure 7.

Correlation of CpG island hypermethylation with clinicopathological features of breast tumors. The x axis indicates the total number of hypermethylated loci (low for 0 to 2 loci, median for 3 to 4 loci and high for 5 to 7 loci). A: Correlation with ER/PR status. B and C: Correlation with clinical stages and age at diagnosis, respectively. The statistical P values are also shown in each panel.