Spreading of silent chromatin: inaction at a distance

Key Points A common feature of heterochromatin is that it can 'spread' over long distances and inactivate multiple genes along a chromosome. A survey of spreading in diverse model eukaryotes reveals evidence for three of the classical modes of action-at-a-distance: 'looping' (contact between distant sites), 'sliding' (tracking along a chromosome) and 'oozing' (binding of one silencing protein facilitates adjacent binding of the next, and so on). Long-range oozing was first proposed in the 1930s, and is favoured by textbooks, despite observations of 'skipping' and a general lack of experimental support. Oozing seems to be the mechanism of short-range SIR (silent information regulator)-dependent silencing in budding yeast. Looping can explain cooperative effects in silencing, and is thought to bring distant regions together to help create or maintain regions of high concentration that would favour heterochromatin assembly. Sliding by DNA or RNA polymerases is an attractive mechanism for spreading, because these enzymes move processively along DNA and must profoundly disrupt chromatin to gain access to DNA for copying. We propose that a common mechanism for spreading is 'hopping', whereby a histone-modifying enzyme locally diffuses from a source site to nearby sites of low affinity, residing long enough to modify nearby histone tails. Several models have been proposed to explain the spreading of heterochromatin, including looping, sliding and oozing. A review of studies from diverse model eukaryotes allows the authors to evaluate the existing models and leads them to propose a common, ancestral mechanism for spreading. One of the oldest unsolved problems in genetics is the observation that gene silencing can 'spread' along a chromosome. Although spreading has been widely perceived as a process of long-range assembly of heterochromatin proteins, such 'oozing' might not apply in most cases. Rather, long-range silencing seems to be a dynamic process, involving local diffusion of histone-modifying enzymes from source binding sites to low-affinity sites nearby. Discontinuous silencing might reflect looping interactions, whereas the spreading of continuous silencing might be driven by the processive movement of RNA or DNA polymerases. We review the evidence for the spreading of silencing in many contexts and organisms and conclude that multiple mechanisms have evolved that silence genes at a distance.