Decoupling of DNA methylation and activity of intergenic LINE-1 promoters in colorectal cancer

Abstract

Hypomethylation of LINE-1 repeats in cancer has been proposed as the main mechanism behind their activation; this assumption, however, was based on findings from early studies that were biased toward young and transpositionally active elements. Here, we investigate the relationship between methylation of 2 intergenic, transpositionally inactive LINE-1 elements and expression of the LINE-1 chimeric transcript (LCT) 13 and LCT14 driven by their antisense promoters (L1-ASP). Our data from DNA modification, expression, and 5′RACE analyses suggest that colorectal cancer methylation in the regions analyzed is not always associated with LCT repression. Consistent with this, in HCT116 colorectal cancer cells lacking DNA methyltransferases DNMT1 or DNMT3B, LCT13 expression decreases, while cells lacking both DNMTs or treated with the DNMT inhibitor 5-azacytidine (5-aza) show no change in LCT13 expression. Interestingly, levels of the H4K20me3 histone modification are inversely associated with LCT13 and LCT14 expression. Moreover, at these LINE-1s, H4K20me3 levels rather than DNA methylation seem to be good predictor of their sensitivity to 5-aza treatment. Therefore, by studying individual LINE-1 promoters we have shown that in some cases these promoters can be active without losing methylation; in addition, we provide evidence that other factors (e.g., H4K20me3 levels) play prominent roles in their regulation.

L1 chimeric transcripts are expressed in healthy brain and their deregulation in glioma follows that of their host locus

“…Our study also bring some insights into the role of DNA methylation in the control of L1-ASP activity. In line with a comprehensive study on two L1-ASPs in various cancer cell lines ( 33 ), but unlike what described for the L1- MET LCT ( 17 ), we did not observe any inverse correlation between L1 promoter methylation and L1-ASP activity (evaluated by measuring the relative LCT expression level). It must be noted that in all studies on this question ( 17 , 33 ), including the present work, DNA methylation was assessed at CpG sites located in a CpG island that overlaps with the internal L1 sense promoter region, but downstream of the L1-ASP transcription starting site.…”

DNA methylation of intragenic CpG islands depends on their transcriptional activity during differentiation and disease

“…For the ZFαAS time course, amplified products were cloned, sequenced, and analyzed as described by Vafadar-Isfahani et al (49).…”

The Tumor-Specific Expression of L1 Retrotransposons Independently Correlates with Time to Relapse in Hormone-Negative Breast Cancer Patients

“…A 60% methylation is often used as a cut-off between promoter methylation and demethylation [ 30 , 31 ]. However, (i) some expressive tumors may have high methylation levels [ 38 ], as confirmed in our study; (ii) there is no clear-cut correlation between demethylation and transcription in tumor tissues [ 39 ]; and (iii) the deamination of cytosine into uracil, the central part of the sodium–bisulfite conversion protocol [ 40 , 41 ], represents one of the main artifacts of FFPE tissues [ 40 ], leading to potential errors.…”