RESUMEN
Stem cells have lower facultative heterochromatin as defined by trimethylation of histone H3 lysine 27 (H3K27me3) compared to differentiated cells, however, the underlying mechanism for this observation has been unknown. Because H3K27me3 levels are diluted two-fold in every round of replication and then restored through the rest of the cycle, we reasoned that the cell cycle length could determine the time available for setting total H3K27me3 levels. Here, we demonstrate that a fast cell cycle sets low levels of H3K27me3 in serum-grown murine embryonic stem cells (mESCs). Extending the G1 phase leads to an increase in global H3K27me3 in mESCs due to the formation of de novo Polycomb domains, and the length of the G1/S block correlates with the extent of gain in H3K27me3, arguing that levels of the modification depend on the time available between successive rounds of replication. Similarly, accelerating the cell cycle in HEK293 cells decreases H3K27me3 levels. Finally, we applied this principle in tumor cells that, due to the dominant negative effect of the sub-stoichiometric H3K27M mutant, have reduced H3K27me3. Here, extending G1 using Palbociclib increased H3K27me3 levels, pointing to an unexpected means to rescue the effect of oncohistones. Our results suggest cell cycle length as a universal mechanism to modulate heterochromatin formation and, thus, cellular identity.
RESUMEN
Human matrix attachment regions (MARs) can insulate transgene expression from chromosomal position effects in Drosophila melanogaster. To gain insight into the mechanism(s) by which chromosomal insulation occurs, we studied the expression phenotypes of Drosophila transformants expressing mini-white transgenes in which MAR sequences from the human apoB gene were arranged in a variety of ways. In agreement with previous reports, we found that a single copy of the insulating element was not sufficient for position-independent transgene expression; rather, two copies were required. However, the arrangement of the two elements within the transgene was unimportant, since chromosomal insulation was equally apparent when both copies of the insulator were upstream of the mini-white reporter as when the transcription unit was flanked by insulator elements. Moreover, experiments in which apoB 3' MAR sequences were removed from integrated transgenes in vivo by site-specific recombination demonstrated that MAR sequences were required for the establishment but not for the maintenance of chromosomal insulation. These observations are not compatible with the chromosomal loop model in its simplest form. Alternate mechanisms for MAR function in this system are proposed.
Asunto(s)
Apolipoproteínas B/genética , Cromosomas/ultraestructura , Color del Ojo , Regulación del Desarrollo de la Expresión Génica , Regulación de la Expresión Génica , Técnicas Genéticas , Transgenes , Animales , Núcleo Celular/metabolismo , Drosophila melanogaster , Genes Reporteros , Vectores Genéticos , Humanos , Regiones de Fijación a la Matriz , Modelos Biológicos , Modelos Genéticos , Células Fotorreceptoras de Invertebrados/embriología , Unión Proteica , Recombinación GenéticaRESUMEN
The human serine protease inhibitor (serpin) gene cluster at 14q32.1 is a useful model system for studying the regulation of gene activity and chromatin structure. We demonstrated previously that the six known serpin genes in this region were organized into two subclusters of three genes each that occupied approximately 370 kb of DNA. To more fully understand the genomic organization of this region, we annotated a 1-Mb sequence contig from data from the Genoscope sequencing consortium (http://www.genoscope.cns.fr/ ). We report that 11 different serpin genes reside within the 14q32.1 cluster, including two novel alpha1-antiproteinase-like gene sequences, a kallistatin-like sequence, and two recently identified serpins that had not been mapped previously to 14q32.1. The genomic regions proximal and distal to the serpin cluster contain a variety of unrelated gene sequences of diverse function. To gain insight into the chromatin organization of the region, sequences with putative nuclear matrix-binding potential were identified by using the MAR-Wiz algorithm, and these MAR-Wiz candidate sequences were tested for nuclear matrix-binding activity in vitro. Several differences between the MAR-Wiz predictions and the results of biochemical tests were observed. The genomic organization of the serpin gene cluster is discussed.