AUTHORS
Nasren Eiza, PhD, Annie Luong, BS2, Justin Le, MD3, Miguel Villa, BS4, Jacqueline Bergseid, PhD5, Chintan Parekh, MD6
AFFILIATIONS
1 City of Hope National Medical Center, Duarte, California, United States
2 City of Hope National Medical Center, Duarte, California, United States
3 Philadelphia College of Osteopathic Medicine, Philadelphia, United States
4 City of Hope National Medical Center, Duarte, California, United States
5 City of Hope National Medical Center, Duarte, California, United States
6 City of Hope National Medical Center, Duarte, California, United States
ABSTRACT
Incomplete understanding of the transcription factor (TF) and co-factor mechanisms governing human T-lineage fate remains a barrier to the generation of T-cells from stem cell sources for immunotherapy. The TF BCL11B is essential for repressing alternative (non-T) fates in humans and mice; however, BCL11B is uniquely required in humans for the induction of T-cell genes in hematopoietic stem and progenitor cells (HSPC). BCL11B is also needed for group 2 innate lymphoid cell (ILC2) development. BCL11B recruits the Nucleosome Remodeling and Deacetylase (NuRD) complex, a context specific transcriptional activator or repressor. However, which BCL11B driven fate decisions require NuRD is unknown. We aimed to investigate whether BCL11B activation enhances T-cell generation from human HSPC and define mechanisms underlying distinct fate decisions.
Human cord blood HSPC transduced with control GFP (control), wildtype (WT) BCL11B, or a NuRD-binding deficient mutant BCL11B (R3S) expression lentivirus were cultured in an artificial thymic organoid (ATO) model (with DLL1 NOTCH ligand).
Relative to control HSPC, BCL11B WT HSPC showed strikingly faster differentiation into early T-cell precursors, T-lineage committed cells, and polyclonal naïve mature T-cells (p<0.05, 12 donors); BCL11B overexpression (OE) increased the output of CD8+ T-cells (2.5-fold, p<0.05). BCL11B-HSPC derived T-cells showed enhanced central memory differentiation (mean, BCL11B vs control, 30% vs 4%, p<0.05) and more sustained proliferation upon repeated activation.
Mathematical modeling indicated BCL11B OE enhanced the earliest stages of T-cell differentiation. Remarkably, RNA-Seq prior to NOTCH exposure in ATO showed rapid upregulation of T-cell genes and downregulation of HSPC and alternative lineage genes in BCL11B OE HSPC; these effects were enhanced after exposure to NOTCH signaling in ATO (5 donors). To define which programs can be orchestrated by BCL11B independent of the key T-cell factor NOTCH, we analyzed WT BCL11B or control HSPC cultured in ATO or organoids without NOTCH signaling (3 donors). Remarkably, without NOTCH, BCL11B was sufficient to induce 80% of T-cell genes and repress stem, erythroid, mast cell, and dendritic programs. However, expression of recombination genes (RAG1, RAG2) and the E2A TF TCF12 as well as repression of NK and monocyte programs was NOTCH dependent. BCL11B without NOTCH established open chromatin at 85% of the regions that gain chromatin accessibility during T-cell differentiation; these sites were enriched for key T-cell TF motifs (BCL11B, TCF7). In contrast, NOTCH dependent sites, including recombination genes, were enriched for the E2A motif (ATAC-Seq).
To elucidate lineage-specific NuRD requirements, we expressed a BCL11B variant with a mutation (R3S) that abrogates NuRD recruitment (4 donors). R3S HSPC ATO showed reduced early T-cell precursors (R3S vs WT BCL11B, 6% vs. 24%, p < 0.05). CITE-seq showed both WT and R3S BCL11B suppressed erythroid, granulocyte/monocyte, and dendritic fates; however, repression of mast cell fate was impaired in R3S ATO. Strikingly, disrupting the BCL11B-NuRD interaction induced robust ILC2 emergence (absent in WT). T-lineage cells in R3S ATO showed aberrant co-expression of stem and T-cell genes, reduced TCR complex gene expression, prematurely expressed RAG genes, and upregulated innate lymphoid genes. R3S and WT BCL11B cells shared 70% of the sites that gained accessibility relative to control HSPC; the GATA 3 motif (an ILC2 TF) was enriched in open sites unique to R3S. Cas9 knockout of the core NuRD subunit CHD4 recapitulated many aspects of R3S cells.
Overall, BCL11B OE expedites the generation and increases the yield of functional T-cells from human HSPC; an effect uniquely driven by a single TF. BCL11B is sufficient for the induction of many parts of the T-cell program without NOTCH; however, NOTCH dependent inputs including potentially TCF12 are needed for full T-cell differentiation. We found lineage specific differences in NuRD dependence of BCL11B effects; while repression of stem cell genes and upregulation of TCR complex genes occurs via NuRD, induction of many T-cell genes and repression of multiple myeloid fates is NuRD independent. Critically, loss of NuRD recruitment converts BCL11B from a T-cell factor into a driver of ILC2 fate. These novel findings could inform the targeting of specific mechanisms in the BCL11B network for optimal T-cell generation from stem cells.
DISCLOSURES
Nasren Eiza, PhD: No relevant conflicts of interest to declare.