TET2 mutations drive cell-autonomous type I interferon production and selective advantage through TRIM4 silencing Free

TET2 mutations cause clonal expansion of hematopoietic stem and progenitor cells (HSPCs) and chimeric antigen receptor (CAR)-expressing T cells, driving clonal hematopoiesis (CH) and CAR+ T cell lymphomas, respectively. TET2-mutated CH has been linked to hyperinflammation, but the mechanisms remain poorly understood.

We developed human isogenic induced pluripotent stem cell (iPSC)-based models of the 3 main CH mutations – DNMT3A R882H, TET2 haploinsufficiency and ASXL1 truncations –and performed integrated transcriptome, chromatin accessibility and DNA methylation (whole genome bisulfite sequencing) analyses in CD34+ HSPCs. Expectedly, DNMT3A- and TET2-mutant HSPCs showed a global decrease and increase, respectively, in DNA methylation, compared to isogenic wild-type (WT) controls. Single-cell RNA-seq analyses and immunoblotting experiments, unexpectedly, showed cell-autonomous activation of the type I interferon (IFN) induction pathway and secretion of bioactive type I IFN by TET2-mutated HSPCs. Furthermore, genomic and functional experiments in iPSC- and cord blood (CB)- HSPCs with TET2 shRNA knockdown or CRISPR/Cas9-mediated knockout demonstrated that, while type I IFN suppresses the expansion of WT CD34+/CD38- cells in a dose-dependent manner, TET2-mutant HSPCs downregulate their downstream IFN response and are thus protected. Importantly, while TET2-deficient CD34+/CD38- cells outgrew co-cultured WT cells, blockade of type I IFN signaling by a IFN alpha and beta receptor subunit 1 (IFNAR1) antibody mitigated their selective advantage.

To understand the mechanism, we searched for hypermethylated genes with concomitant loss of chromatin accessibility and gene expression specifically in TET2-mutant HSPCs. These analyses, followed by knockdown and knockout (KO) validation experiments, identified silencing of the E3 ubiquitin ligase tripartite motif containing 4 (TRIM4) gene as a putative effector of TET2 deficiency. Corroborating findings from scATAC-seq and TARGET-seq+ datasets of HSPCs from individuals with TET2-mutated CH, clonal cytopenia of undetermined significance (CCUS) and healthy donors, revealed decreased promoter accessibility and expression of TRIM4 in primary TET2-mutant HSPCs. Importantly, engineering TRIM4 deficiency by CRISPR in iPSC- and CB- HSPCs phenocopied TET2 deficiency phenotypes, including monocytic bias and cell-autonomous type I IFN induction. In addition, TRIM4-deficient iPSC- and CB- HSPCs outcompeted WT HSPCs in liquid culture and serial replating assays in semisolid media, recapitulating the clonal advantage of TET2-deficient HSPCs. Furthermore, ectopic TRIM4 re-expression in TET2-deficient HSPCs, rescued their clonal expansion both in vitro and in competitive repopulation assays in NSGS mice.

While TRIM proteins have known pleiotropic roles in immune signaling, TRIM4 is not well-characterized and lacks a mouse homolog. To identify substrates of TRIM4 we first performed integrated quantitative proteomics and ubiquitinome profiling in TET2+/-, TRIM4 KO, and isogenic WT iPSC-HSPCs. These analyses nominated 190 proteins with decreased ubiquitination and concomitant increased abundance by both TET2 and TRIM4 deficiency. Knockout of key components of the 3 main innate sensing pathways – TRIF/MYD88 (Toll-like receptor signaling pathway), MAVS (cytosolic RNA sensing) and STING (cytosolic DNA sensing) – identified MAVS as the dominant pathway. DExH-box helicase 58(DHX58), one of the 190 proteins identified by our proteomics analyses and a positive regulator of the MAVS pathway, was confirmed by knockout experiments to mediate the aberrant IFN production by TET2-deficient HSPCs. These results support a model whereby TET2 deficiency and consequent TRIM4 silencing activate type I IFN production through decrease in proteasome-mediated degradation of DHX58. Finally, we show that IFNAR1 blockade by the FDA-approved monoclonal antibody anifrolumab alleviates the clonal advantage of TET2-deficient T cells expressing a CD19-targeted CAR.

In summary, we report secretion of type I IFN by TET2-mutant HSPCs as the cause of a pro-inflammatory local microenvironment that favors the expansion of mutant cells; we discover TRIM4 silencing as a critical effector of TET2 deficiency; we characterize TRIM4 as a novel negative regulator of type I IFN induction through degradation of DHX58; and we propose a new pharmacologic approach to mitigate the adverse clinical consequences of TET2 mutations.