Derivation of a minimal functional XIST by combining human and mouse interaction domains

Abstract

X-inactive specific transcript (XIST) is a 17–19 kb long non-coding ribonucleic acid (RNA) critical for X-chromosome inactivation. Tandem repeats within the RNA serve as functional domains involved in the cis-limited recruitment of heterochromatic changes and silencing. To explore the sufficiency of these domains while generating a functional mini-XIST for targeted silencing approaches, we tested inducible constructs integrated into 8p in a male cell line. Previous results suggested silencing could be accomplished with a transgene comprised of the repeat A, which is highly conserved and critical for silencing; the repeat F that overlaps regulatory elements and the repeat E that contributes to XIST localization by binding proteins such as CIZ1 (AFE). As polycomb-repressive complex 1 (PRC1) is recruited through HNRNPK binding of repeats B–C–D, we included a second ‘mini-XIST’ comprising AFE with the mouse Polycomb Interaction Domain (PID), a 660-nucleotide region known to recruit PRC1. Silencing of an adjacent gene was possible with and without PID; however, silencing more distally required the addition of PID. The recruitment of heterochromatic marks, evaluated by immunofluorescence combined with RNA fluorescence in situ hybridization, revealed that the AFE domains were sufficient only for CIZ1 recruitment. However, mini-XIST transgene recruited all marks, albeit not to full XIST levels. The ability of the PID domain to facilitate silencing and heterochromatic mark recruitment was unexpected, and inhibition of PRC1 suggested that many of these are PRC1 independent. These results suggest that the addition of this small region allowed the partial recruitment of all the features induced by a full XIST, demonstrating the feasibility of finding a minimal functional XIST.