Targeting Inflammation in Tet2 Mutant Clonal Hematopoiesis Via Inhibition of Brd4 Demonstrates Activity in Both Murine and Human Models

Loss-of-function mutations in TET2 comprise 1/3^rd of clonal hematopoiesis (CH) cases. To date, there are no approved therapies to slow progression of CH to MDS/AML. Malignant progression in TET2 CH is driven by dysregulated HSC function, myeloid expansion, and increased inflammation. Here, we investigate the role of the epigenetic reader protein bromo- and extra-terminal (BET) domain 4 (Brd4) in a novel murine model of Tet2 CH and demonstrate that BET inhibition effectively reduces inflammatory signaling in human CCUS.

We first identified pathways in Tet2 CH amendable to pharmacologic targeting by performing RNA-seq on sorted Lin^-cKIT⁺ (LK) and Lin^-cKIT⁺Sca-1⁺ (LSK) progenitors from Tet2 KO versus WT mice, observing significant upregulation of hallmark JAK/STAT3 and TNF-α signaling in Tet2 KO cells. Because Brd4 interacts with NF-κB and STAT3 to promote transcription of inflammatory programs, we hypothesized Brd4 inhibition would reverse abnormal hematopoiesis in Tet2 CH via reduction in inflammatory signaling. To test this, we generated hematopoietic-specific knockout of both Brd4 and Tet2 using Mx1-Cre conditional Brd4 Tet2 double knockout (DKO) mice. While deletion of Brd4 during fetal hematopoiesis is lethal, we have successfully induced deletion of Brd4 both alone and in combination with Tet2 without any bone marrow failure.

Brd4 Tet2 DKO marrow showed reduced colony formation and serial replating compared to Tet2 KO (p<0.0001, n=5) or WT (p<0.05, n=5) mice. Similarly, the BET inhibitor PLX51107 abrogated the increased colony formation (p<0.0001, n=5) and replating (p<0.01, n=5) of Tet2 KO marrow. Despite targeting related BET proteins Brd2 and Brd3, PLX51107 treatment of Brd4 single KO or Brd4 Tet2 DKO marrow did not alter colony formation, suggesting Brd4 has a non-redundant role in HSPCs.

Tet2 KO mice have an exaggerated increase of myeloid cells in the bone marrow and peripheral blood upon TLR agonist stimulation. Two days after a sublethal dose of LPS, Brd4 Tet2 DKO mice showed reduced emergency granulopoiesis, with significant reductions in peripheral blood CD11b+ Ly-6G+ neutrophils (p<0.0001), Ly-6Chi+ monocytes (p<0.001), and CD11b+ Gr-1+ myeloid cells (p<0.001) compared to Tet2 KO (n=7). One week after LPS challenge, frequencies of LSK (p<0.01), MPP1 (p<0.05), and MPP3 (p<0.05) populations were reduced in DKO vs Tet2 KO bone marrow (n=7). Frequencies of other progenitors including MPP4, CMP, MEP, and GMP were unchanged, suggesting Brd4 has developmentally specific roles in Tet2 CH. To identify Brd4 dependent transcriptional programs underlying inflammation and self-renewal in Tet2 KO, RNA-seq analysis of LSK cells from LPS-treated single KO, DKO, or control mice is ongoing.

Because monocytic cells are a predominant source of cytokines in Tet2 CH, we examined functional changes in inflammation in LPS-stimulated bone marrow derived macrophages (BMDMs) from Brd4 Tet2 DKO and controls. DKO BMDMs had reduced cytokine mRNA expression and significantly less TNF-α, IL-1β, IL-12p40, and IL-18 protein compared to single KO or WT controls (p<0.01, n=5). Similar results were observed in LPS-stimulated Tet2 KO and WT BMDMs treated with PLX51107. Further, genetic and pharmacologic Brd4 inhibition reduced pSTAT3, NLRP3, and phosphorylated p65 protein in LPS-stimulated Tet2 KO BMDMs. These results suggest Brd4 is required for the hyperinflammation characteristic of Tet2 CH, which promotes altered HSC function and ultimately malignant progression.

To study the effect of BETi on inflammation in human CH, PBMCs obtained from healthy donors or CCUS patients (n=4 each) at OSU’s leukemia predisposition clinic were treated with LPS ± PLX51107. CCUS PBMCs expressed increased IL-1β, IL-6, IL-12p40 and TNF-α measured by RT-qPCR (p<0.05) and multiplex cytokine analysis (p<0.01). Interestingly, PLX51107 reduced LPS-induced cytokines in both groups, but CCUS samples were markedly more sensitive to BET inhibition. Similarly, LPS-stimulated cytokines were more strikingly decreased in BETi treated murine Tet2 KO BMDMs, implying inflammation in CH may be particularly dependent upon Brd4.

In summary, Brd4 inhibition reduces self-renewal, inflammation, and myeloid expansion in murine Tet2 CH. Further, BETi treatment reduces excess cytokine production in response to LPS in human CCUS. Collectively, these results provide strong rationale for targeting Brd4 in CH at high risk of progression.