Figure 6

Hypothetical mechanism of regulation by alternative splicing of histone dimethyltransferase G9a function. (A) Experimental evidence indicates that histone dimethyltransferase G9a plays an important role in the silencing of neuronal genes in non-neuronal tissues [9]. In the proposed mechanism [9], shown here with red arrows, in non-neuronal tissues the transcription factor NFSP (shown in magenta) recruits the fully functional isoform of G9a (shown here with two domains: a binding domain in blue, and a catalytic domain in yellow) to a series of target genes that are subsequently silenced by G9a dimethylation of lysine-9 from histone H3. This mechanism may be inhibited/modulated by expression of the G9a short isoform (which only retains the NFSP transcription factor binding domain, Table 2), as shown here with green arrows. This isoform may behave as a dominant-negative inhibitor, as shown by the green arrows, blocking the access of the catalytically active isoform to the chromatin of the target gene. Absence of methylation marks in histone H3's lysine-9 would then result in an active gene. (B) The expression state of the target genes in both the nervous system (active, green colour) and in other tissues (silenced, red colour), as a result of the silencing, combined, action of NFSP and G9a. Co-expression of both the long and the short isoforms may result in the modification of the expression state of the target genes in non-neuronal tissues. These target genes may now show varying degrees of activity, as a result of the dominant-negative inhibitor role played by the short isoform (described in (A)).