Gene Silencing Repeats that Count

RIP is nucleus limited. We know this because the specialized cells that carry out RIP are heterokaryotic, with nuclei from both parents. When a sequence is duplicated in one nucleus but not the other, the lone copy is never silenced. The mechanism of RIP is unknown but probably involves DNA:DNA pairing (Figure 1) followed by methyla-Repeated DNA and cytosine methylation are common tion and deamination of some cytosines in the paired features of most eukaryotic genomes and are frequently region. Products of RIP are typically found methylated, found associated with one another. Much repeated even when they segregate away from all homologous DNA, such as the highly repeated satellite sequences, sequences. In some cases, the observed methylation is of dubious value, and some, such as transposable simply results from " maintenance methylation " (defined elements, may be purely parasitic. Methylation of such as methylation triggered by preexisting methylation), but sequences may reflect the action of a genome defense in many cases the sequences mutated by RIP serve as system that can silence repetitive DNA. Indeed, silenc-triggers for de novo methylation. ing mechanisms directed specifically at repeated DNA, A close, but interestingly different, relative of RIP, and involving methylation, have been identified in fungi. named MIP (methylation-induced premeiotically), oc-Furthermore, DNA methylation is frequently associated curs in the fungus Ascobolus immersus. MIP and RIP with silencing of multicopy sequences in plants and are virtually identical except that silencing by MIP does animals. The correlation between repetitive DNA and not involve mutations. Methylation resulting from both methylation is far from perfect, however, indicating that RIP and MIP causes transcriptional arrest. An Ascobolus repeats, per se, do not always trigger methylation and gene with features expected of a DNA methyltransferase silencing. Recent findings in plants, including those re-gene is required for efficient MIP but not for maintenance ported by Bender and colleagues in the April issue of of methylation in vegetative cells (Malagnac et al., 1997). Molecular Cell (Luff et al., 1999), suggest that certain Methylation resulting from MIP, like that associated with repeats count more than others: inverted repeats seem genes inactivated by RIP and that associated with gene to be especially potent transacting inducers of silencing inactivation in plants, is not limited to symmetrical sites and methylation. Interestingly, inverted repeats have such as CpGs and CpNpGs, which are thought to be also been implicated in methylation-independent silenc-required for maintenance methylation. As discussed …