Tetramer Formation of BCL11A Is Required for Both Protein Stability and Fetal Hemoglobin Silencing

The zinc-finger (ZnF) repressor protein BCL11A regulates the switch from fetal (HbF, α ₂γ ₂) to adult hemoglobin (HbA, α ₂β ₂) through direct binding at the γ-globin promoter. Clinical trials of gene-based reduction of BCL11A with ex vivo modified hematopoietic stem cells (HSCs) have yielded transformative outcomes in patients with sickle cell disease and β-thalassemia, establishing BCL11A as a therapeutic target. Unless in vivo methods for delivery of gene-based therapies become efficient enough to modify a substantial fraction of host HSCs, small molecule drugs will be required to reduce overall disease burden.

With the goal of targeting BCL11A with small molecules, we have characterized its native state. The protein is largely unstructured, except for 7 ZnFs, numbered 0-6. Previously we showed that BCL11A protein appears stable in cells with a T _1/2 of ~ 24 hrs. Here we report the unexpected finding that the protein assembles and functions as a tetramer. Recombinant BCL11A behaves as a tetramer in biophysical assays. Gene editing of the endogenous BCL11A locus in CD34-derived, immortal HUDEP2 cells has been employed to correlate protein structure and expression with consequences of a given modification. We found that BCL11A lacking ZnF0 is unstable, leading to protein loss and derepression of γ-globin. Recombinant ZnF0 self-assembles into a stable tetramer, whose X-ray structure reveals a hydrophobic interface between subunits. Mutation of residues in the hydrophobic core disrupts tetramerization and markedly reduces steady-state protein in cells. Reciprocal immunoprecipitation in cells reveals that ZnF0 is both necessary and sufficient to mediate BCL11A mutlimer assembly. Collectively, these results indicate that BCL11A tetramerization is required for protein stability. In cells expressing BCL11A lacking ZnF0, the steady-state protein level is partially restored by addition of proteasome inhibitors, suggesting that mutant protein is actively degraded by the ubiquitin-proteasome system. In contrast to findings with ZnF0, discrete removal of ZnF1 within the context of full-length BCL11A results in a marked increase in steady-state protein in cells. In vitro purified ZnF1 forms a dimer and only dimeric ZnF1 can be stably expressed in cells. These findings re consistent with a model in which assembly of the BCL11A tetramer is required to mask a degron in ZnF1 that becomes obscured on its dimerization. Masking of a degron in ZnF1 predicts that discrete removal of both ZnF0 and ZnF1 within full-length BCL11A will lead to production of a monomer. Indeed, HUDEP2 cells expressing BCL11A lacking both ZnFs exhibit an elevated level of steady-state, monomeric protein, consistent with the degron masking model. However, despite an elevated level, cells expressing monomeric BCL11A are defective in HbF silencing, indicating that the tetramer structure is critical for γ-globin repression.

Our findings provide novel insights into BCL11A action and suggest new strategies for therapeutic targeting. Combining structural, biochemical, and cellular approaches, we have shown that ZnF-mediated tetramer formation shields BCL11A from proteolytic degradation, and that the tetramer state itself is required for effective γ-globin repression. Of relevance to BCL11A as a therapeutic target, the tetramer of ZnF0 constitutes a structured, stable domain in an otherwise largely unstructured protein, and thereby offers possibilities for small molecule targeting. The multimer state of BCL11A is a vulnerability for HbF repression.