BACKGROUND: Large-scale screening methods are widely used to reveal cancer-specific DNA methylation markers. We previously identified non-satellite 3.3-kb repeats associated with facioscapulohumeral muscular dystrophy (FSHD) as hypermethylated in cervical cancer in genome-wide screening. To determine whether hypermethylation of 3.3-kb repeats is a tumor-specific event and to evaluate frequency of this event in tumors, we investigated the 3.3-kb repeat methylation status in human papilloma virus (HPV)-positive cervical tumors, cancer cell lines, and normal cervical tissues. Open reading frames encoding DUX family proteins are contained within some 3.3-kb repeat units. The DUX mRNA expression profile was also studied in these tissues. METHODS: The methylation status of 3.3-kb repeats was evaluated by Southern blot hybridization and bisulfite genomic sequencing. The expression of DUX mRNA was analyzed by RT-PCR and specificity of PCR products was confirmed by sequencing analysis. RESULTS: Hypermethylation of 3.3-kb repeats relative to normal tissues was revealed for the first time in more than 50% (18/34) of cervical tumors and in 4 HPV-positive cervical cancer cell lines. Hypermethylation of 3.3-kb repeats was observed in tumors concurrently with or independently of hypomethylation of classical satellite 2 sequences (Sat2) that were hypomethylated in 75% (15/20) of cervical tumors. We have revealed the presence of transcripts highly homologous to DUX4 and DUX10 genes in normal tissues and down-regulation of transcripts in 68% of tumors with and without 3.3-kb repeats hypermethylation. CONCLUSION: Our results demonstrate that hypermethylation rather than hypomethylation of 3.3-kb repeats is the predominant event in HPV-associated cervical cancer and provide new insight into the epigenetic changes of repetitive DNA elements in carcinogenesis.
No takes yet. Share an insight, caveat, or question.
Katargin et al. (2009) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: