Although most genes in a cell are expressed from both the maternal and paternal chromosome, there are exceptions. For example, in women, most X-linked genes are expressed from only one of the two X chromosomes, a phenomenon called X inactivation. In addition, there is a class of autosomal genes, termed imprinted genes, for which parental origin determines which allele is expressed. Finally, there are autosomal genes that appear at first glance to be bi-allelically expressed but actually show random monoallelic expression (sometimes termed allelic exclusion) at the single-cell level. These exceptions, examples of epigenetics, have proven to be of great interest for researchers because they shed light on gene regulation, chromatin structure, development, and the pattern of inheritance of certain genetic disorders.
My research program is focused on the potential role of allele-specific expression in development and function of the central nervous system (CNS). What is the evidence that genes likely to play a role in CNS function show such expression? Olfactory receptors, which are expressed in specialized cells of the CNS, show allelic exclusion, as does p120 catenin, which is involved in synapse formation. Intriguing recent work has shown that a number of factors involved in the immune response, including the genes for interleukin-2 and interleukin-4, also show allelic exclusion. Some of these genes are expressed in the CNS, and the possibility arises that other inflammation-sensitive genes in the CNS may show a similar pattern of expression. Using gene expression profiling, we discovered that, Cdkn1a, coding for the cell cycle regulator p21Waf1/Cip1, is inflammation-sensitive in the CNS as well as other tissues. While this gene is bi-allelically expressed, we expect to find additional immune response genes that do undergo monoallelic expression.
We have also developed an imprinting screen using expression microarrays. As a model system, we analyzed mice with imprinting defects in proximal chromosome 7; part of this region is analogous to human chromosome 15q11-q13, a locus associated with a number of behavioral and cognitive disorders including the well-studied Prader-Willi/Angelman Syndrome (PW/AS). Our analysis revealed the presence of two novel paternally expressed intergenic transcripts at the mouse PW/AS locus, in a region highly enriched in LINE-1 elements; the function of these transcripts is still unknown. In separate work, we discovered, in collaboration with Dr. Chauncey Bowers (Department of Neurosciences) that the dense LINE-1 elements in this region are organized in a uniquely asymmetric way, perhaps related to imprinting at the locus.
Our current work involves the identification and characterization of genes that are subject to random monoallelic expression in the CNS. We have developed a microarray-based assay for genes that are both silenced and active at the same locus as evidenced by a dual DNA methylation pattern. We further analyze candidate genes using SNP differences in cDNA of clonal neural stem cell lines derived from F1 hybrids of two different strains of mice. We have found a number of “hits” and are currently characterizing those that appear potentially most relevant to disorders of the CNS.