Human beta-globin locus control region HS5 contains CTCF- and developmental stage-dependent enhancer-blocking activity in erythroid cells

Abstract

The human beta-globin locus contains five developmentally regulated beta-type globin genes. All five genes depend on the locus control region (LCR), located at the 5' end of the locus, for abundant globin gene transcription. The LCR is composed of five DNase I-hypersensitive sites (HSs), at least a subset of which appear to cooperate to form a holocomplex in activating genes within the locus. We previously tested the requirement for proper LCR polarity by inverting it in human beta-globin yeast artificial chromosome transgenic mice and observed reduced expression of all the beta-type globin genes regardless of developmental stage. This phenotype clearly demonstrated an orientation-dependent activity of the LCR, although the mechanistic basis for the observed activity was obscure. Here, we describe genetic evidence demonstrating that human HS5 includes enhancer-blocking (insulator) activity that is both CTCF and developmental stage dependent. Curiously, we also observed an attenuating activity in HS5 that was specific to the epsilon-globin gene at the primitive stage and was independent of the HS5 CTCF binding site. These observations demonstrate that the phenotype observed in the LCR-inverted locus was in part attributable to placing the HS5 insulator between the LCR HS enhancers (HS1 to HS4) and the promoter of the beta-globin gene.

FIG. 1.

Schematic representation of the experimental system. (A) An extra copy of the HS5 fragment (2.6-kbp, inverted orientation, middle) was floxed (solid triangles) and introduced between HS1 and the ɛ-globin gene in human β-globin YAC (A201F4.3; wild type, top) by homologous recombination in S. cerevisiae. Putative cis-DNA elements within LCR-HS5 are schematically represented (GATAx7, MAR, and CTCF). The YAC DNA was purified and used to generate transgenic mice (TgM). A single-copy, intact YAC transgenic mouse (mutant) was then mated with Cre-expressing transgenic mice to recreate the wild-type locus by deleting the ectopic HS5 fragment (wild-type with single loxP footprint) (B) Wild-type (HS5, upper) and mutant (HS5/dCTCF, lower) HS5 sequences surrounding the putative CTCF binding site are in italics. To facilitate screening for proper homologous recombination in S. cerevisiae, two nucleotides (underlined) outside the CTCF consensus motif were also mutated, which created an AccI restriction enzyme site.

FIG. 2.

Structural analysis of the β-globin YAC in transgenic mice. (A) Schematic representation of the mutant human β-globin YAC. The positions of the β-like globin genes are shown relative to the LCR. SfiI restriction enzyme sites are located 5′ to HS5, between HS4 and HS3, and in the right arm of the YAC. Probes (solid rectangles) used for long-range fragment analysis and expected restriction enzyme fragments with their sizes are shown. (B) Long-range structural analysis of transgenes. The whole β-globin locus is contained within two SfiI fragments (10 and 100 kbp, as in A). DNA from thymus cells was digested with SfiI in agarose plugs, separated by pulsed-field gel electrophoresis, and hybridized separately to probes (shown in A) from the β-globin locus or from the right YAC vector arm. (C) Schematic representation of the transgene locus around HS1. Cre-loxP-mediated HS5 deletion removes the 2.6-kbp insert from the mutated loci (middle), which creates a 7.7-kbp BglII fragment in the locus (bottom). Solid arrowheads, loxP sequences. G, BglII; H, HindIII. (D) Tail DNAs from wild-type (lines 134 and 31), mutant (lines 60-HS5 and 518-HS5/dCTCF), and loxP footprint (lines 60- and 518-loxP) transgenic mouse lines was digested with BglII, separated on an agarose gel, and transferred to a nylon membrane. Hybridization was performed with the HS1-3′ probe (solid rectangle in C). The bands representative of wild-type (7.7 kbp), mutated (10.3 kbp), and footprint (7.7 kbp) loci are indicated.

FIG. 3.

RT-PCR analysis of β-like globin expression in erythroid cells of HS5 YAC transgenic mice. (A) Total RNA was prepared from the spleens of 1- to 2-month-old anemic mice and subjected to cDNA synthesis by reverse transcriptase (RT). The relative expression levels of the β-like globin genes, after normalization to that of the endogenous mouse α-globin gene, were determined by RT-PCR analysis. Data were collected from two individuals for each line and the average and standard deviation, determined by three sets of PCRs, are graphically depicted. Representative results of RT-PCR for human β (hβ) and mouse α (mα) amplicons are shown below each panel. (B and C) Total RNA was prepared from the liver (E14.5) or yolk sacs (E9.5) of two fetuses in two litters (lit.1 and lit.2) derived from the intercross of male transgenic and female wild-type animals. Representative results of RT-PCR for human ɛ (hɛ), γ (hγ), β (hβ), and mouse α (mα) are shown below each panel.

FIG. 4.

RT-PCR analysis of β-like globin expression in erythroid cells of HS5/dCTCF YAC transgenic mice. Total RNA was prepared and analyzed as in Fig. 3.