Barrier function at HMR.

The silenced HMR domain is restricted from spreading by barrier elements, and the right barrier is a unique t-RNA(THR) gene. We show that sequences immediately flanking the silenced domain were enriched in acetylated, but not methylated, histones, whereas the barrier element was associated with a nucleosome-free region. Surprisingly, the SAGA acetyltransferase resided across the entire region. We further demonstrate that a mutation in the barrier eliminated the nucleosome-free gap but only subtly altered the distribution of SAGA. Interestingly, neither reformation of the nucleosome nor mutations in chromatin-modifying enzymes alone led to an unrestricted spread of silenced chromatin. Double mutations in the t-RNA barrier and these complexes, on the other hand, led to a significant spread of Sir proteins. These results suggest two overlapping mechanisms function to restrict the spread of silencing: one of which involves a DNA binding element, whereas the other mechanism involves specific chromatin-modifying activities.

Growth rate-dependent control, feedback regulation and steady-state mRNA levels of the threonyl-tRNA synthetase gene of Escherichia coli.

The expression of the gene thrS encoding threonyl-tRNA synthetase is under the control of two apparently different regulatory loops: translational feedback regulation and growth rate-dependent control. The translational feedback regulation is due to the binding of threonyl-tRNA synthetase to a site located in the leader RNA of thrS, upstream of the initiation codon, which mimics the anticodon stem and loop of tRNA(Thr). This binding competes with that of the ribosome and thus inhibits translation initiation. Here, we investigate the mechanism of growth rate-dependent control, i.e. the mechanism by which the synthetase accumulates at high growth rates. We show that growth rate-dependent control acts at the level of translation and requires feedback regulation since mutations that abolish feedback regulation also abolish growth rate-dependent control. We also show that tRNA(Thr), which accumulates at high growth rates, is one of the effectors of growth rate-dependent control since its accumulation can cause derepression independently of growth rate. We show that this tRNA(Thr)-dependent derepression is also dependent on feedback regulation since mutations which abolish feedback also prevent derepression. Based on these results and previous data concerning the mechanism of translational feedback regulation, we propose that threonyl-tRNA synthetase growth rate-dependent control is the consequence of the accumulation at high growth rates of two effectors, the ribosome and tRNA(Thr). We also study the growth rate-dependence of the steady state level of thrS mRNA and show that the steady state level of thrS mRNA increases at high growth rates. This increase is dependent on the translational feedback regulation and can also be detected, independently of growth rate, when thrS mRNA translation is derepressed. Consistently with the model of growth rate-dependent control above, we propose that at high growth rates, the mRNA is well translated and thus stabilised and that, at low growth rates, because of its low translation, thrS mRNA is rapidly degraded.