Controlling long-range genomic interactions at a native locus by targeted tethering of a looping factor

Abstract

Chromatin loops juxtapose distal enhancers with active promoters, but their molecular architecture and relationship with transcription remain unclear. In erythroid cells, the locus control region (LCR) and β-globin promoter form a chromatin loop that requires transcription factor GATA1 and the associated molecule Ldb1. We employed artificial zinc fingers (ZF) to tether Ldb1 to the β-globin promoter in GATA1 null erythroblasts, in which the β-globin locus is relaxed and inactive. Remarkably, targeting Ldb1 or only its self-association domain to the β-globin promoter substantially activated β-globin transcription in the absence of GATA1. Promoter-tethered Ldb1 interacted with endogenous Ldb1 complexes at the LCR to form a chromatin loop, causing recruitment and phosphorylation of RNA polymerase II. ZF-Ldb1 proteins were inactive at alleles lacking the LCR, demonstrating that their activities depend on long-range interactions. Our findings establish Ldb1 as a critical effector of GATA1-mediated loop formation and indicate that chromatin looping causally underlies gene regulation.

Figure 1. ZF-mediated targeting of Ldb1 to the β-globin locus

(A) Experimental model. Top: wild-type scenario in which GATA1 and the TAL1 complex recruit Ldb1 to promote chromatin looping. Middle: lack of GATA1 leads to loss of Ldb1 at the promoter, impaired looping, and reduced transcriptional activation. Bottom: ZF-mediated Ldb1 tethering to the β-globin promoter is examined for its ability to restore looping and transcription activation. (B) Top: P-ZF and L-ZF target the β-major promoter (red triangle) and HS2 of the LCR (red oval), respectively. (B-D) Anti-HA ChIP in cells expressing P-Ldb1 (B), L-Ldb1 (C), and L-Ldb1+P-Ldb1 (D). L-Ldb1 binds selectively to HS2 of the LCR. Of note, P-Ldb1 binds to the promoter but additionally associates with HS 1, 2, 3 of the LCR but not to other regions, including the εy, βH1, and βmin genes, an intervening region (IVR16) or an inactive gene (CD4). N≥3; error bars denote standard deviation. See also Figure S1.

Figure 2. Activation of β-globin transcription in GATA1 null cells by tethered Ldb1 or its SA domain

(A) β-major mRNA levels as measured by RT-qPCR with primer pairs for exon 2 in G1E cells and derivatives expressing indicated ZF and ZF-Ldb1 constructs. (B) Data in (A) were re-plotted next to those obtained from induced G1E-ER4 cells (G1E+GATA1). Note that β–major expression achieved by P-Ldb1 or L-Ldb1+P-Ldb1 amounts to approximately 20% of that induced by GATA1. (C) Relative expression of indicated erythroid genes as determined by RT-qPCR. (D) Top: schematic of Ldb1. SA, self association domain, LID, LIM interaction domain. Bottom: β-major mRNA levels in G1E cells expressing indicated ZF fused to the SA domain of Ldb1. Transcript levels were normalized to β-actin. N≥3; error bars denote standard deviation. See also Figure S2.

Figure 3. Chromatin looping by the tethered Ldb1 self-association domain

(A-D) 3C assay measuring locus wide cross-linking frequencies in G1E cells (blue) or induced G1E-ER4 cells expressing GATA1 (A, red), or G1E cells containing P-SA (B, red), L-SA (C, red) or L-SA+P-SA (D, red). The murine β-globin locus is depicted on top of each graph. The X-axis indicates distances (kb) from the εy gene, which represents zero. Black bar denotes the HS2-containing BglII fragment serving as anchor. Grey bars denote analyzed BglII fragments. N=3 (A,B,D), and N=2 (C). Error bars indicate standard-error-of-mean. See also Figure S3.

Figure 4. Restoration of Pol II recruitment and serine 5 phosphorylation by ZF-SA

(A) Location of amplicons (black bars). Prom, promoter; numbers indicate exons. (B, C) ChIP with antibodies against total Pol II (B), ser5ph (C) using G1E cells or G1E cells expressing GATA1 or P-SA. Note that while total Pol II binding at the promoter matched that induced by GATA1, Pol II levels in the body of the gene were only partially restored in P-SA cells, consistent with incomplete rescue of transcriptional elongation (compare with figure 2B). N=3; error bars denote standard deviation. See also Figure S4.

Figure 5. ZF-SA enhances β-globin expression in primary early progenitor cells

(A) Staging of E13.5 fetal liver erythroid cells by Ter119, CD71 profiling. (B) mRNAs from FACS purified R1 cells transduced with ZF constructs were examined by RT-qPCR with primers for the indicated genes. Negative controls (Neg Ctrl), cells expressing empty vector. Results were normalized to GAPDH. N=3; error bars denote standard deviation. See also Figure S5.

Figure 6. LCR-dependence of β-globin induction by ZF-SA proteins

(A) Experimental concept. The LCR deleted allele is on the background of the β-major D haplotype while the wild type allele is of the β-major S haplotype. (B) β–major transcript levels as measured by allele-specific RT-qPCR in R1 cells from WT/ΔLCR or ΔLCR/ΔLCR fetal livers expressing indicated ZF-SA proteins. Transcript levels were normalized to GAPDH. N=3; error bars denote standard deviation. See also Figures S6 and S7.

Figure 7. Hypothetical model functionally integrating chromatin looping and transcription activation

Recruitment of Ldb1 to the β-globin promoter either by ZF proteins or GATA1 promotes LCR-promoter looping. Forced chromatin looping by ZF-Ldb1 efficiently restores PIC assembly, Pol II recruitment, Pol II serine 5 phosphorylation, and transcription initiation. In the absence of GATA1, diminished recruitment of P-TEFb and likely additional GATA1 co-factors accounts for inefficient transcription elongation. Therefore, chromatin looping can trigger transcription initiation and can occur independently of full transcription elongation.