Chromatin architecture near a potential 3' end of the igh locus involves modular regulation of histone modifications during B-Cell development and in vivo occupancy at CTCF sites

Abstract

The murine Igh locus has a 3' regulatory region (3' RR) containing four enhancers (hs3A, hs1,2, hs3B, and hs4) at DNase I-hypersensitive sites. The 3' RR exerts long-range effects on class switch recombination (CSR) to several isotypes through its control of germ line transcription. By measuring levels of acetylated histones H3 and H4 and of dimethylated H3 (K4) with chromatin immunoprecipitation assays, we found that early in B-cell development, chromatin encompassing the enhancers of the 3' RR began to attain stepwise modifications typical of an open conformation. The hs4 enhancer was associated with active chromatin initially in pro- and pre-B cells and then together with hs3A, hs1,2, and hs3B in B and plasma cells. Histone modifications were similar in resting splenic B cells and in splenic B cells induced by lipopolysaccharide to undergo CSR. From the pro-B-cell stage onward, the approximately 11-kb region immediately downstream of hs4 displayed H3 and H4 modifications indicative of open chromatin. This region contained newly identified DNase I-hypersensitive sites and several CTCF target sites, some of which were occupied in vivo in a developmentally regulated manner. The open chromatin environment of the extended 3' RR in mature B cells was flanked by regions associated with dimethylated K9 of histone H3. Together, these data suggest that 3' RR elements are located within a specific chromatin subdomain that contains CTCF binding sites and developmentally regulated modules.

FIG. 1.

DNase I hypersensitive sites in the region downstream of hs4. (A) Schematic map of the murine Igh locus shows V, D, J, and constant region genes (solid bars) and enhancers and DNase I-hypersensitive sites (solid circles). The murine Igh locus is located on chromosome 12 and comprises hundreds of variable (V) genes, 16 diversity (D) genes, and 4 joining (J) genes spanning more than 3 Mb. Between Cμ and Cα, there are additional constant region genes: δ, γ3, γ1, γ2b, γ2a, and ɛ. The numbers represent the position, in kilobases, along the insert in BAC199M11, which begins downstream of the last exon of the Cα gene. EcoRI (RI) restriction enzyme sites are depicted, as are HindIII (H3) sites used for analysis of DNase I hypersensitivity. The horizontal bars above the line in the vicinity of hs4 represent the PstI-2.3 and hs4D probes used in this study. The PstI-2.3 probe detects a ∼22-kb RI fragment (BALB/c), while hs4D detects a ∼6.6-kb HindIII fragment (C57BL/6). Triangles are targets used for ChIP analysis. (B) EcoRI digestion of DNase I-treated nuclei from the 63-12 pro-B cell line, the 18-81 pre-B cell line, the A20 mature B cell line, and the 2017 pro-T cell line. The ∼22-kb fragment detected by PstI-2.3 includes hs4 as well as two novel DNase I-hypersensitive sites, hs5 and hs6 (*), located ∼4.5 and 3.5 kb, respectively, upstream of the distal EcoRI site. (C) HindIII digestion of DNase I-treated nuclei from resting and LPS-stimulated B cells revealed an additional hs4 downstream DNase I-hypersensitive site, hs7 (*).

FIG. 2.

Developmental analysis of AcH3 and AcH4 association at the 3′ RR. ChIP assays were performed on chromatin from the MEL cell line (A), 18-81 pre-B cell line (B), A20 mature cell line (C), and MPC11 plasma cell line (D). No antibody and IgG serve as experimental controls. While β-globin and CAD (the gene encoding carbamoyl-phosphate synthetase II [EC 6.3.5.5], aspartate transcarbamylase [EC 2.1.3.2], and dihydroorotase [EC 3.5.2.3]) serve as positive controls, Eμ is used as a negative control in the MEL cell line. CAD and the β-globin gene are used as a positive and a negative control for histone acetylation, respectively, in B cells. The primers used are indicated next to relevant gels. Shown are serial (1:3) dilutions of input and IP samples for the semiquantitative PCR analysis. The signal for hs1,2 in the input of MPC11 is absent due to a technical error.

FIG. 3.

Association of AcH3 and AcH4 at the 3′ end of the Igh locus in mouse primary pro-B-cell and splenic B-cell populations and a pre-B-cell line. Shown are real-time PCR analyses of a ChIP done with anti-AcH3 and with anti-AcH4 on pro-B cells isolated from the bone marrow of a RAG1-deficient mouse (A), the 3-1 pre-B-cell line (B), and splenic B cells (C). The primers used are indicated below the histogram bars. Data are plotted as a numerical comparison of signal in immunoprecipitated samples to that detected in the input, as described in Materials and Methods. A gray swath represents the 3′ RR extending from hs3A to hs4.

FIG. 4.

Reciprocal association of di-me K4 H3 and di-me K9 H3 at sequences flanking the extended 3′ RR and real-time PCR analysis of di-me K4 H3 and di-me K9 H3 modifications. Shown is the enrichment of di-me K4 H3 in pro-B cells (A) and in splenic B cells (B) and the enrichment of di-me K9 H3 in A20 cells (C) and in splenocytes (D).

FIG. 5.

Histone H3 and H4 acetylation as well as di-me K4 H3 at the Igh 3′ RR in cells undergoing CSR. ChIP assays with antibodies to AcH3, AcH4, and di-me K4 H3 were performed on freshly isolated splenic B cells (A) or after LPS stimulation for 48 h (B).

FIG. 6.

Identification and analysis of CTCF sites in the region downstream of hs4. (A) In-scale schematic of the 3′ RR showing 30 consecutive overlapping DNA fragments (fragment numbers intermittently indicated) used for EMSA. EMSA identified CTCF binding sites associated with hs4, hs5, and the region downstream of hs6, i.e., hs7. F, free DNA probe; C, control TnT reaction with luciferase template; ZF, 11 zinc finger region of CTCF; FL, full-length CTCF. (B) Assessment of insulator activity of hs5 and the region containing strong CTCF binding sites (hs7) (see Materials and Methods for exact fragments tested) by using a stable transfection assay that measures Neo^R colonies. The number of colonies detected with the construct containing both the T-cell receptor enhancer and the promoter ranged from ∼130 to 180 in different experiments. We set this number at 100 and normalized all the other data accordingly. BB identifies the BEAD-1 insulator and is used as a positive control.

FIG. 7.

ChIP analysis of CTCF binding in 3′ RR in 63-12 pro-B cells (A), 18-81 pre-B cells (B), and resting and LPS-stimulated normal B cells (C). The horizontal line above the x axis represents the highest level of signal detected with normal IgG.