Three-dimensional positioning of genes in mouse cell nuclei

Abstract

To understand the regulation of the genome, it is necessary to understand its three-dimensional organization in the nucleus. We investigated the positioning of eight gene loci on four different chromosomes, including the β-globin gene, in mouse embryonic stem cells and in in vitro differentiated macrophages by fluorescence in situ hybridization on structurally preserved nuclei, confocal microscopy, and 3D quantitative image analysis. We found that gene loci on the same chromosome can significantly differ from each other and from their chromosome territory in their average radial nuclear position. Radial distribution of a given gene locus can change significantly between cell types, excluding the possibility that positioning is determined solely by the DNA sequence. For the set of investigated gene loci, we found no relationship between radial distribution and local gene density, as it was described for human cell nuclei. We did find, however, correlation with other genomic properties such as GC content and certain repetitive elements such as long terminal repeats or long interspersed nuclear elements. Our results suggest that gene density itself is not a driving force in nuclear positioning. Instead, we propose that other genomic properties play a role in determining nuclear chromatin distribution.

Positions of pluripotency genes and hepatocyte-specific genes in the nucleus before and after mouse ES cell differentiation

“…The correlation coefficient (R) value in the figure was calculated to be -0.62. This value is comparable to those reported in the above cited study, in which the R values were almost the same between the ES cells and macrophages, -0.61 and -0.63, respectively (Hepperger et al, 2008). Our study showed that in hepatocytes, gene localization correlated very well with not only the local GC content but also gene density, and the R value of the former was -0.92 ( Figure 3D) and that of the latter -0.71 ( Figure 3B).…”

“…Gene localization is reportedly correlated with some genomic properties (Murmann et al, 2005;Hepperger et al, 2008;Jost et al, 2011). Therefore, we examined whether such a correlation exists for the genes under analysis.…”

“…Permeabilization was performed as previously reported (Hepperger et al, 2007(Hepperger et al, , 2008. Briefly, the cells were immersed in 0.5% Triton X-100/PBS for 15 min, and then incubated in 20% glycerol/PBS for at least 1 h. The cells were then subjected to five rounds of freeze/thaw treatment in liquid nitrogen.…”

“…Evaluations of gene density and GC content in Mus musculus were performed according to Hepperger et al (2008). The genes involved in the 2 Mbp region centered at each target gene were counted using the NCBI Map viewer, developed at the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/mapview, Build 38.1 from January 2012), and the sequence of the region was downloaded.…”

“…Using the genes encoding β-actin, albumin, β-globin, CD11b, lysozyme, MYB, MPO and OCT4 and the DNA region homologous to the human Beckwith-Wiedeman syndrome region, Hepperger et al (2008) showed that gene localization correlates with GC content but not gene density, in both mouse ES cells and the macrophages differentiated from the ES cells. Our results for the pluripotency genes and the hepatocyte-specific genes in the mouse ES cells agreed with this report, although the P value for the correlation was ~0.1 ( Figure 3C).…”