Abstract / Summary
Clear cell renal cell carcinoma (ccRCC) harbours recurrent epigenetic alterations, but locus-specific DNA methylation haplotype architecture remains poorly defined. Conventional methylation analyses report average methylation levels and cannot resolve CpG arrangements on individual DNA molecules. We investigated whether MUC1 promoter methylation haplotypes differ between neoplastic and adjacent non-neoplastic renal tissues. We performed bisulfite amplicon sequencing of the MUC1 promoter in 53 paired neoplastic and non-neoplastic ccRCC tissue regions and quantified methylation haplotypes at single-molecule resolution. Haplotype complexity was assessed using Shannon, Simpson, Chao1, and ACE indices, and tissue-associated architecture was evaluated by consensus t-SNE and PERMANOVA. MUC1 methylation haplotype composition was markedly remodelled in ccRCC. Neoplastic regions showed reduced haplotype complexity across all α-diversity indices (all p ≤ 0.001), with enrichment of unmethylated haplotypes, including 5′-UUUUUUUU-3′, and depletion of highly methylated haplotypes, including 5′-MMMMMMMM-3′. β-diversity analysis demonstrated separation between neoplastic and adjacent non-neoplastic tissue regions (PERMANOVA R2 = 0.483, p < 0.001), indicating distinct tissue-associated epigenetic architecture. Exploratory in silico pseudo-mixture analysis showed that the tissue-derived signature retained discriminatory information at low simulated fractions. These findings identify loss of MUC1 methylation haplotype complexity as a distinct epigenetic feature of ccRCC tissues and support haplotype-level profiling for dissecting cancer-associated epigenetic heterogeneity.