Chromatin boundaries, insulators, and long-range interactions in the nucleus

Abstract

Within the genome, expressed genes marked by "open" chromatin are often adjacent to silent, heterochromatic regions. There are also regions containing neighboring active genes with different programs of expression. In both cases, DNA sequence elements may function as insulators, either providing barriers that prevent the incursion of heterochromatic signals into open domains or acting to block inappropriate contact between the enhancer of one gene and the promoter of another. The mechanisms associated with insulation are diverse: Enhancer-blocking insulation is largely associated with the ability to stabilize the formation of loop domains within the nucleus. Barrier insulation is often associated with the ability to block propagation of silencing histone modifications. Here, we provide examples of both kinds of insulator action, derived initially from studies of the compound insulator element at the 5' end of the chicken β-globin locus. Such elements appear to have more general regulatory roles in the genome that have been exploited to provide insulator function where necessary to demarcate separate domains within the nucleus.

Figure 1.

Properties of the chicken β-globin 5′HS4 compound insulator element (IE). (A) Diagram of the locus, showing the position of the insulator relative to the globin gene cluster, the condensed chromatin region upstream of the cluster, and further upstream the erythroid-specific folate receptor gene. (B) Enhancer-blocking assay using a locus control region (LCR) element as an enhancer to drive a neomycin-resistance gene (Neo) in the presence of an ~250-bp element from 5′HS4 (INS) or a small noninsulating DNA fragment derived from λ bacteriophage. Increased insulator activity results in fewer neomycin-resistant colonies (Chung et al. 1993). (C) Barrier function assay using fluorescence-activated cell sorter (FACS) analysis to detect the expression levels of a fragment of the IL2 receptor. The fragment was expressed from a reporter construct carrying a globin-specific promoter and enhancer, stably transformed in each case into the avian erythroid cell line 6C2. (Left panel) Uninsulated reporter, (right panel) reporter surrounded by two copies on each side of the ~250-bp IE from 5′HS4 (Pikaart et al. 1998). (D) The five distinct protein-binding domains within the IE, the proteins that bind to them, and their functions. At FIV, the USF1/USF2 heterodimer binds to DNA and recruits a protein complex containing SET1 and another containing PRMT1 (Huang et al. 2007). Histone acetyltransferases are also present in each complex, and the complexes are probably not bound simultaneously. The overlaps between factors shown are not intended to signify physical contacts between them.

Figure 2.

Schematic representation of the role of CTCF binding to the imprinted control region (ICR) at the Igf2/H19 locus in mice. On the maternally transmitted allele, four CTCF sites are occupied and constitute a strong insulator that blocks the ability of downstream enhancers to activate Igf2 expression. On the paternal allele, CpG sites within the ICR are methylated and CTCF does not bind. There is no insulator activity and Igf2 expression is activated on this allele. Similar mechanisms are found at the human locus (see Bell and Felsenfeld 2000; Hark et al. 2000; Kanduri et al. 2000).

Figure 3.

Analysis of the splice variants of Dnmt3b; Dnmt3b transcript is decreased in Vezf1^−/− cells. (A) Map of Dnmt3b gene showing three characterized splicing events. To analyze alternative splicing events, total RNA from wild-type and Vezf1^−/− ES cells was reverse-transcribed and amplified by PCR for 27 cycles. (B) Catalytic domain splicing. (C) 5′-end splicing. (D) Relative quantitative reverse-transcriptase (RT)–PCR analysis for the measurement of DNA methylases (MTases) using TaqMan expression assay. The Ct values for all MTases are normalized to that of β-actin. (Reprinted, with permission, from Gowher et al. 2008.)

Figure 4.

Effects of TSA and Dicer on structure of the 16-kb heterochromatin domain upstream of the β-globin cluster (see Fig. 1A). (A) Sucrose gradient sedimentation pattern of the 16-kb chromatin fragment (see Fig. 1A), liberated from 6C2 cell nuclei digested with Msp1. (Vertical arrow) Peak position of the intact fragment that has not been cut internally by the enzyme. Approximately 31% of the total chromatin is in this peak (see F). (B) Effects of treatment with trichostatin A (TSA) on levels of histone H4 acetylation over the 16-kb heterochromatin domain, showing large increases in acetylation. TSA was then removed and 17 h later, acetylation levels were again measured. (Light gray bars) Untreated, (dark gray bars) TSA added, (white bars) TSA removed. (C) Transcript RNA levels over the 16-kb domain in wild-type 6C2 cells. (D) Increase in transcript RNA after TSA treatment. (E) Effect of Dicer knockdown on transcript levels across the 16-kb domain. (F) Results of sucrose gradient experiments such as that in A, showing the fraction of Msp1-resistant intact 16-kb fragment present after knockdown of Ago2 or Dicer, compared to results for a mock knockdown experiment (shown in A). (Reprinted or replotted, with permission, from Giles et al. 2010 [Nature Publishing Group].)