Abstract
Genome-wide association studies (GWAS) have ascertained numerous trait-associated common genetic variants localized to regulatory DNA. The hypothesis that regulatory variation accounts for substantial heritability has undergone scarce experimental evaluation. Common variation at BCL11A is estimated to explain ∼15% of the trait variance in fetal hemoglobin (HbF) level but the functional variants remain unknown.
We use chromatin immunoprecipitation (ChIP), DNase I sensitivity and chromosome conformation capture to evaluate the BCL11A locus in mouse and human primary erythroblasts. We extensively genotype 1,263 samples from the Collaborative Study of Sickle Cell Disease within three HbF-associated erythroid DNase I hypersensitive sites (DHSs) at BCL11A. We pyrosequence heterozygous erythroblasts to assess allele-specific transcription factor binding and gene expression. We conduct transgenic analysis by mouse zygotic microinjection and genome editing with transcription activator-like effector nucleases (TALENs) and clustered, regularly interspaced, short palindromic repeats (CRISPR)/CRISPR-associated nuclease 9 (Cas9) RNA-guided nucleases.
Common genetic variation at BCL11A associated with HbF level lies in noncoding sequences decorated by an erythroid enhancer chromatin signature. Fine-mapping this putative regulatory DNA uncovers a motif-disrupting common variant associated with reduced GATA1 and TAL1 transcription factor binding, modestly diminished BCL11A expression and elevated HbF. This variant, rs1427407, accounts for the HbF association of the previously reported sentinel SNPs. The composite element functions in vivo as a developmental stage-specific lineage-restricted enhancer. Genome editing reveals that the enhancer is required in erythroid but dispensable in B-lymphoid cells for expression of BCL11A. We demonstrate species-specific functional components of the composite enhancer in mouse as compared to human erythroid precursor cells. The mouse sequences homologous to the human DHS sufficient to drive reporter activity are dispensable from the mouse composite element, whereas the adjacent DHS, whose human homolog does not direct reporter activity, is absolutely required for BCL11A expression.
We describe a comprehensive and widely applicable approach, including chromatin mapping followed by fine-mapping, allele-specific ChIP and gene expression studies, and functional analyses, to reveal causal variants and critical elements. We assert that functional validation of regulatory DNA ought to include perturbation of the endogenous genomic context by genome editing and not solely rely on in vitro or ectopic surrogate assays. These results validate the hypothesis that common variation modulates cell type-specific regulatory elements, and reveal that although functional variants themselves may be of modest impact, their harboring elements may be critical for appropriate gene expression. We speculate that species-level functional differences in components of the composite enhancer might partially account for differences in timing of globin gene expression among animals. We suggest that the GWAS-marked BCL11A enhancer represents a highly attractive target for therapeutic genome editing for the major b-hemoglobin disorders.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
