The DNA hypermethylation phenotype of colorectal cancer liver metastases resembles that of the primary colorectal cancers

Abstract

Background: Identifying molecular differences between primary and metastatic colorectal cancers-now possible with the aid of omics technologies-can improve our understanding of the biological mechanisms of cancer progression and facilitate the discovery of novel treatments for late-stage cancer. We compared the DNA methylomes of primary colorectal cancers (CRCs) and CRC metastases to the liver. Laser microdissection was used to obtain epithelial tissue (10 to 25 × 10⁶ μm²) from sections of fresh-frozen samples of primary CRCs (n = 6), CRC liver metastases (n = 12), and normal colon mucosa (n = 3). DNA extracted from tissues was enriched for methylated sequences with a methylCpG binding domain (MBD) polypeptide-based protocol and subjected to deep sequencing. The performance of this protocol was compared with that of targeted enrichment for bisulfite sequencing used in a previous study of ours.

Results: MBD enrichment captured a total of 322,551 genomic regions (249.5 Mb or ~ 7.8% of the human genome), which included over seven million CpG sites. A few of these regions were differentially methylated at an expected false discovery rate (FDR) of 5% in neoplastic tissues (primaries: 0.67%, i.e., 2155 regions containing 279,441 CpG sites; liver metastases: 1%, i.e., 3223 regions containing 312,723 CpG sites) as compared with normal mucosa samples. Most of the differentially methylated regions (DMRs; 94% in primaries; 70% in metastases) were hypermethylated, and almost 80% of these (1882 of 2396) were present in both lesion types. At 5% FDR, no DMRs were detected in liver metastases vs. primary CRC. However, short regions of low-magnitude hypomethylation were frequent in metastases but rare in primaries. Hypermethylated DMRs were far more abundant in sequences classified as intragenic, gene-regulatory, or CpG shelves-shores-island segments, whereas hypomethylated DMRs were equally represented in extragenic (mainly, open-sea) and intragenic (mainly, gene bodies) sequences of the genome. Compared with targeted enrichment, MBD capture provided a better picture of the extension of CRC-associated DNA hypermethylation but was less powerful for identifying hypomethylation.

Conclusions: Our findings demonstrate that the hypermethylation phenotype in CRC liver metastases remains similar to that of the primary tumor, whereas CRC-associated DNA hypomethylation probably undergoes further progression after the cancer cells have migrated to the liver.

Keywords: Colorectal cancer; CpG islands; CpG sites; DNA methylation; Differentially methylated regions; Liver metastasis; MBD capture; Methyl-binding domain; Normal colorectal mucosa.

Fig. 1

Genomic distribution and functional classification of MBD-captured regions within the genome. a. Schematic showing components of an "intragenic" genomic region. Areas outside such regions were classified as "extragenic". b. MBD-captured regions (total: 322,551) classified as extragenic vs. intragenic. c. Distribution of the 206,258 intragenic MBD-captured regions by subregions: gene regulatory (including 5'UTRs and regions 5 kb upstream from the TSS); gene body (including exonic and intronic sequences between the end of the 5'UTR and the beginning of the 3'UTR); and the 3'UTRs. d. Schematic showing components of an “sSISs” region (i.e., a CpG island and flanking CpG shores and shelves). Areas outside such regions were defined "extra-sSISs". Schematics in this panel and in panel A were adapted from diagrams contained in the R-package annotatr vignette.e. MBD-captured regions classified as sSISs vs. extra-sSISs. f. Distribution of the 53,521 sSISs MBD-captured regions by subregions (CpG islands, shores, and shelves). N.B. After exclusion of the X and Y chromosomes, the human genome contains 26,361 canonical CpG islands, 15,423 of which were covered by MBDE. (See Discussion).

Fig. 2

Genomic regions in primary CRCs or CRC liver metastases displaying differential methylation relative to that in normal colon mucosa (NM). a and b. Differentially methylated regions (DMRs) characterized by hypermethylation and hypomethylation (vs. NM) present in primary CRCs, metastatic CRCs, or both. c, d, e, and f. Distributions of hypermethylated and hypomethylated DMRs in the extragenic vs. intragenic genomes; among the intragenic genome components; between the sSISs and extra-sSISs genomic segments, and among the sSISs components, respectively. (See Figure 1 for topography of genomic segments.)

Fig. 3

Performance of MBD enrichment (MBDE) and targeted enrichment (TE) in identifying differentially methylated CpGs or regions (DMRs) in primary CRCs (vs. normal mucosa). a. Compared with TE, MBDE captured 3 times more genomic base pairs (including 33.4% of TE-captured base pairs) and almost 3 times more CpGs (including 41.3% of TE-captured sites). b. First three density plots, from left: Compared with CpGs missed by both methods (mean O:E CpG ratios: 0.35 for MBDE, 0.39 for TE), the CpGs captured by each method were preferentially located in genomic areas with relatively high CpG densities (mean O:E CpG ratios: 0.58 for MBDE, 0.65 with TE). Fourth density plot: CpGs captured by MBDE only, by TE only or by both methods tended to be located in genomic areas with similar CpG densities (mean O:E CpG ratios: 0.56 for MBDE, 0.65 for TE, 0.66 for both methods). c. Methylation changes (log fold change in MBDE, beta-values in TE) at CpGs captured by both methods in low-, medium-, and high-CpG-density areas. (See Methods for calculation of differential methylation with each method and for calculation and classification of CpG density.) Intermethod correlation values are shown for each density area. For most of the CpGs captured by both methods (gray bars), no methylation differences in primary CRCs (vs. normal mucosa) were identified with either MBDE or TE. The number of hypermethylated CpGs detected by both methods (red bars) steadily increased with increasing CpG rates. In addition, in medium- and high-CpG-rate areas of the genome, MBDE captured hypermethylated CpGs that were not found to be significantly hypermethylated with TE (blue vs. green bars). Hypomethylated CpGs were more frequently identified by TE than MBDE (yellow vs. orange bars). d. The CpG site trends were confirmed by analysis of DMRs. Over 400 hypermethylated DMRs were identified with both methods (in addition to the 1584 identified only by MBDE and the 827 detected only with TE). However, the overlap of commonly identified hypermethylated DMRs is likely underestimated because of the use of different analytical packages to identify them with MBDE or TE (see example in Supplementary Figure 3 C). Hypomethylated DMRs were in contrast identified almost exclusively by TE (examples in Supplementary Figure 3 D and E).