Highly restricted localization of RNA polymerase II within a locus control region of a tissue-specific chromatin domain

Abstract

RNA polymerase II (Pol II) can associate with regulatory elements far from promoters. For the murine beta-globin locus, Pol II binds the beta-globin locus control region (LCR) far upstream of the beta-globin promoters, independent of recruitment to and activation of the betamajor promoter. We describe here an analysis of where Pol II resides within the LCR, how it is recruited to the LCR, and the functional consequences of recruitment. High-resolution analysis of the distribution of Pol II revealed that Pol II binding within the LCR is restricted to the hypersensitive sites. Blocking elongation eliminated the synthesis of genic and extragenic transcripts and eliminated Pol II from the betamajor open reading frame. However, the elongation blockade did not redistribute Pol II at the hypersensitive sites, suggesting that Pol II is recruited to these sites. The distribution of Pol II did not strictly correlate with the distributions of histone acetylation and methylation. As Pol II associates with histone-modifying enzymes, Pol II tracking might be critical for establishing and maintaining broad histone modification patterns. However, blocking elongation did not disrupt the histone modification pattern of the beta-globin locus, indicating that Pol II tracking is not required to maintain the pattern.

FIG. 1.

Highly restricted localization of Pol II within the β-globin LCR. (A) Organization of the murine Hbb^d β-globin domain. The embryonic (Ey and βH1) and adult (βmajor and βminor) β-globin genes are shown as boxes, and HSs are shown as spheres. The locations of selected real-time PCR amplicons used throughout the study are indicated below the locus. Intervening regions (IVR) represent sequences containing no known functional elements and were analyzed in subsequent ChIP experiments. (B) High-resolution analysis of Pol II binding at the LCR. The relative levels of Pol II binding at 20 sites within the LCR were determined by ChIP with DMSO-induced MEL cells. Black bars represent amplicons within the murine HS cores, as defined by Hardison et al. (18). End points for these amplicons are at positions 258 to 308 (4core), 5700 to 5750 (3core), 11974 to 12046 (2core), and 15551 to 15626 (1core). Stacked gray (or black) and white bars represent the signals obtained with an anti-Pol II antibody and control preimmune serum, respectively. Arrows indicate positions that were analyzed when enrichments for Pol II and preimmune sera did not exceed 0.0002.

FIG. 2.

Pol II binds the LCR in primitive EryP colonies derived from murine ES cells. (A) Primitive EryP colonies were generated by in vitro differentiation of murine ES cells. (B) Real-time PCR analysis of the expression of embryonic β-globin gene βH1 in EryP and MEL cells. The graph shows the relative levels of βH1 mRNA transcripts normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA transcripts. Note that βH1 expression in EryP cells exceeded expression in MEL cells by more than 6 orders of magnitude. (C) ChIP analysis of Pol II binding to HS2, HS3, and HS4, the active Ey promoter, and intergenic site IVR5. The graph shows the relative intensities of PCR products from two (IVR5) or three independent experiments (mean and standard error of the mean).

FIG. 3.

Blocking of transcriptional elongation eliminates Pol II from βmajor but does not redistribute LCR-bound Pol II. MEL cells were differentiated with 1.5% DMSO for 4 days and then treated with 100 μM DRB for 1 h prior to ChIP analysis. (A) Elimination of Pol II-generated genic and extragenic transcripts by DRB treatment. cDNAs were prepared from total cellular RNAs from MEL cells treated with DRB or with vehicle (DMSO). Primers specific to a region within the second intron of βmajor (βmaj I2) or to extragenic sites within the β-globin locus were used to detect unprocessed primary transcripts. To control for potential contaminating DNA, RT was omitted from control reactions (−RT). RNA was analyzed in two or three independent experiments. (B) ChIP analysis of Pol II binding at the β-globin locus following DRB treatment. The positions of PCR primers to amplify LCR sequences are shown in Fig. 1. Pol II binding at the βmajor promoter (βmaj pro), within the second intron (βmaj I2), and within the third exon (βmaj 3′) was examined. The graphs show the relative intensities of PCR products from three or four independent experiments. PI, rabbit preimmune serum. Data are the means and standard errors of the means.

FIG. 4.

High-resolution analysis of histone modifications at the LCR. (A) Positions of HS core primers within the murine β-globin domain are shown. (B to D) The relative levels of diacetylated H3 (B), H3-meK4 (C), and multiacetylated H4 (D) within the LCR were determined by ChIP with DMSO-induced MEL cells. The positions examined by PCR were the same as those examined for Pol II binding in Fig. 1B. As in Fig. 1B, black bars represent amplicons within the murine HS cores. Signals obtained upon ChIP with control preimmune serum are the same as those in Fig. 1B, as these experiments were conducted simultaneously. The graphs show the relative intensities of PCR products from three or more independent experiments.

FIG. 5.

Lack of correlation between the distributions of Pol II binding and specific histone modifications. (A to C) Linear regression analyses of values obtained from Fig. 1B and 4B to D for Pol II binding and either H3 diacetylation (A), H3-K4 dimethylation (B), or multisite H4 acetylation (C) were carried out for the 20 sites within the LCR shown in Fig. 1B. (D) For comparison, a correlation plot is shown for H3 diacetylation and H4 acetylation.

FIG. 6.

Pol II binding at the LCR is independent of Pol II recruitment to the βmajor promoter. Comparison of Pol II binding by ChIP at β-globin gene promoters and HS cores in p45/NF-E2-null CB3 cells. CB3 cells were cultured with 1.5% DMSO for 4 days prior to ChIP analysis. The ∼2-fold enrichment of Pol II compared to preimmune serum that was observed at the promoters was also evident at extragenic sites IVR3 and IVR16 (for the positions of these sites, see Fig. 1A). The graph shows the relative intensities of PCR products from three or more independent experiments (mean and standard error of the mean).

FIG. 7.

Maintenance of the erythroid-specific composite histone modification pattern of the β-globin locus does not require ongoing transcriptional elongation by Pol II. Samples analyzed by ChIP for Pol II binding in Fig. 3 were also used to detect DRB-induced changes in histone acetylation and H3-meK4 at the βmajor gene and at the LCR. The graphs show the relative intensities of PCR products from three or four independent experiments (mean and standard error of the mean). PI, rabbit preimmune serum.