A comprehensive overview of genomic imprinting in breast and its deregulation in cancer

Abstract

Genomic imprinting plays an important role in growth and development. Loss of imprinting (LOI) has been found in cancer, yet systematic studies are impeded by data-analytical challenges. We developed a methodology to detect monoallelically expressed loci without requiring genotyping data, and applied it on The Cancer Genome Atlas (TCGA, discovery) and Genotype-Tissue expression project (GTEx, validation) breast tissue RNA-seq data. Here, we report the identification of 30 putatively imprinted genes in breast. In breast cancer (TCGA), HM13 is featured by LOI and expression upregulation, which is linked to DNA demethylation. Other imprinted genes typically demonstrate lower expression in cancer, often associated with copy number variation and aberrant DNA methylation. Downregulation in cancer frequently leads to higher relative expression of the (imperfectly) silenced allele, yet this is not considered canonical LOI given the lack of (absolute) re-expression. In summary, our novel methodology highlights the massive deregulation of imprinting in breast cancer.

Fig. 1

Mixture distributions of (non-)imprinted SNPs. Observed (red) and modelled (blue) fraction of alternative alleles for two significantly imprinted SNP positions, i.e. a IGF2 (rs2585, adj. p-value < detection limit (LRT), î = 0.97) b SNRPN (rs705, adj. p-value = 1.81E-71 (LRT), î = 0.99), and a non-imprinted SNP, i.e. c CHMP6 (rs1128687, î = 0)

Fig. 2

SNP positions differentially imprinted between normal and cancer samples. a MEST (rs10863, adj. p-value = 0.022). b H19 (rs2839704, adj. p-value = 0.069). c H19 (rs2839703, adj. p-value = 0.082). d HM13 (rs6059873, adj. p-value = 0.062)

Fig. 3

SNP positions differentially imprinted between normal and cancer subtypes. a MEST (rs10863). b H19 (rs2839704). c HM13 (rs6059873)

Fig. 4

Kaplan–Meier plot of the Cox proportional hazards model for survival as a function of differential imprinting. Differential imprinting is implemented continuously as the allelic ratio (AR) of the least expressed allele over the most expressed allele (with AR as categorical variable for the Kaplan–Meier plot: AR ≤ 0.2 (blue curve) and AR > 0.2 (red curve)), and age. a rs3732084 (ZDBF2) b rs1975597 (ZDBF2)

Fig. 5

Graphical representation of rationale of the PMF. a The PMF is defined as a mixture model of genotype-dependent binomial distributions and describes the probability of observing specific RNA-seq coverages for each allele for a specific SNP locus. In these binomial probabilities, sequencing error rates, degree of imprinting (i) as well as the specific genotype are taken into account. For non-imprinted loci, the PMF results in two homozygous peaks and one heterozygous peak. For imprinting, on the other hand, no heterozygous can be detected on RNA-level and this peak is hence eliminated. Heterogeneous data leads to the detection of partial imprinting. b PMF for different degrees of imprinting. In this mixture model, the genotype-dependent binomial distributions have weights corresponding to their Hardy–Weinberg theorem derived expected chances