Non-coding RNAs in the epigenetics of human centromere formation. We will examine the epigenetics of the formation and function of human centromeres, the fundamental chromosomal component responsible for proper chromosome segregation during cell division. Defects in chromosome segregation leads to genome instability, which represents a major cause of birth defects and cancer. Human centromeres contain highly repetitive alpha satellite DNA. Transfection of alpha satellite DNA into cells can result in de novo centromere and human artificial chromosome (HAC) formation, albeit at low frequency. Dicentric chromosomes contain inactivated centromeres but retain the array of alpha satellite. Neocentromeres are fully functional human centromeres that have formed on low copy non-repetitive DNA with no alpha satellite DNA. Thus, inactive centromeres and neocentromeres demonstrate that alpha satellite DNA is neither sufficient nor even necessary, respectively, for centromere formation and function. Instead, centromere formation is a largely sequence independent epigenetic process dependent on the formation of multiple distinct chromatin domains characterized by known epigenetic marks. The kinetochore chromatin contains CENtromere Protein A (CENP-A), a centromere-specific histone H3 variant, interspersed with Histone H3 dimethylated at lysine 4 (H3K4diMe). Surrounding the CENPA domain is the centromeric heterochromatin, characterized by histone H3 lysine 9 methylation (H3K9me) and heterochromatin protein 1 (HP1). Remarkably, it has been shown in S. pombe, Drosophila, Arabidopsis and mouse that assembly of centromeric heterochromatin actually requires bidirectional cis transcription of centromeric DNA into dsRNA, which directs the correct epigenetic modifications for heterochromatin formation and establishment of CENP-A chromatin. However, the exact nature of these centromeric transcripts are unknown, especially in mammalian and human cells. Thus, we propose to discover and perform functional analysis of the novel non-coding centromeric RNA-based epigenetic marks that are responsible for human centromere formation and function, as described in the following three specific aims. 1) We will discover the non-coding RNA transcribed from human endogenous centromeres in HT1080 cells using RT-PCR, examining chromosome 17 alpha satellite as a model. Detection and detailed characterization of specific centromeric alpha satellite DNA transcripts will provide insights into their role in epigenetic silencing of centromeric heterochromatin in human cells. 2) We will determine the role of RNAi in establishment and maintenance of de novo centromeres in using novel HAC vectors engineered to provide the proper alpha satellite-specific RNAi signals to initiate centromeric heterochromatin assembly. The effect of Dicer depletion on mitotic stability of HACs and endogenous chromosomes will be investigated. 3) We will investigate the role of RNAi and heterochromatin and mode of sister chromosome cohesion at human neocentromeres, using our unique collection of patient-derived cell lines with neocentric chromosomes.
Centromeres are the critical chromosomal component responsible for proper sister chromatid segregation during cell division. Metazoan centromeres are epigenetically determined, involving transcription of centromeric repeats and RNAi-based establishment of heterochromatin. These epigenetic marks are currently unknown in human cells, and their discovery will provide great insight into the requirements for centromere formation and function.
