Epigenetic Heterogeneity of Human Chromosome 15 Duplication Syndrome Brain Samples

Background: Chromosome 15q11-13 contains an imprinted cluster of genes necessary for normal mammalian neurodevelopment. Paternal and maternal deficiencies of the 15q11-13 alleles results in Prader-Willi (PWS) and Angelman (AS) syndromes, respectively. A distinct maternal duplication of 15q11-13 (dup15), often associated with autism spectrum disorders (ASD), occurs as both interstitial duplications [int dup(15)] and supernumerary pseudodicentric chromosome 15 [idic(15)]. Our previous analysis of two post-mortem brain samples with idic(15) showed heterogeneity between the two cases, with one showing gene expression in 15q11-13 GABA_A receptor genes, UBE3A, and SNRPN in a manner not predicted by copy number or parental imprint.

Objectives: Determine the effect of 15q11-13 duplication on 15q11-13 transcript levels in additional brain samples and investigate potential epigenetic differences to explain heterogeneity in gene expression patterns between different chromosome 15 duplication cases.

Methods: Frozen human postmortem cerebral samples were obtained from the Autism Tissue Program, including 6 chromosome 15 duplication cases, 6 age- and sex-matched controls, and 2 Prader-Willi and Angelman samples with paternal or maternal 15q11-13 deletion. RNA was isolated by TriZol reagent and cDNA reverse transcribed with a mix of random hexamer and oligo-d(T) primers. Quantitative reverse transcriptase polymerase chain reaction (qRT-PCR) was performed for 15q11-13 transcripts SNRPN, UBE3A, and GABRB3 and two housekeeping genes GAPDH and HPRT1. In addition, fluorescence in situ hybridization (FISH) using probes to each of the 15q11-13 gene loci was used to determine homologous pairing, chromatin decondensation, and potential mosaicism in each brain sample.

Results: Prior experiments with two idic15 brain samples revealed two striking patterns with one sample showing increased and one showing decreased expression of 15q11-13 genes. Preliminary results with additional dup15 samples confirms a pattern of two distinct categories of gene expression: one subset of samples with lower levels of SNRPN and GABRB3 but no difference in UBE3A compared to controls and another subset with higher GABRB3 and UBE3A compared to controls but no difference in SNRPN. Previous work showed the idic15 chromosomes interacting non-selectively with maternal and paternal chromosome 15 alleles. Ongoing experiments are being performed to investigate homologous pairing and chromatin structure in the dup15 samples to investigate epigenetic heterogeneity associated with this syndrome.

Conclusions: These findings suggest that allelic expression within 15q11-13 is not based on entirely copy number or parental origin but can be influenced by epigenetic mechanisms that may create the clinical heterogeneity observed in 15q11-13 duplication syndromes.