Chronic TNF in the Aging Microenvironment Exacerbates TET2-loss-of-Function Myeloid Expansion

Introduction: Age-associated TET2 somatic mutations impart an intrinsic hematopoietic stem cell (HSC) advantage and contribute to the phenomenon of clonal hematopoiesis of indeterminate potential (CHIP). Individuals with TET2-mutantCHIP have a higher risk of developing myeloid neoplasms and other age-related conditions, including heart, lung, liver, kidney and infectious disease, and have increased risk of all-cause mortality. Despite its role in unhealthy aging, the extrinsic mechanisms driving TET2-mutant CHIP clonal expansion remain unclear. We previously showed an environment containing TNF favours TET2-mutant HSC expansion in vitro. We therefore postulated that age-related increases in TNF also provide an advantage to HSC and progeny with TET2-mutations in vivo.

Methods: All human and mouse studies met ethics approval. C57Bl/6 background wildtype (WT), TNF-α knockout ( TNF-/-), Tet2 hematopoietic knockout (Vav1-iCre-mediated; Tet2-/-) and floxed control mice ( Tet2f/f) mice were obtained from the Jackson Laboratory. Young (6-mo [n=9]) and old (18-22 mo [n=9]) WT mice, and old TNF-/- mice (n=7) were subjected to nonirradiative myeloablation with busulfan, and retro-orbital injections of 8x10 ⁶ cells/ml T-cell-depleted bone marrow (BM), equally harvested from 4-mo-old WT CD45.1 (n=5) and Tet2-/- CD45.2 donor mice (n=5). Flow cytometry was used to confirm and monitor engraftment. Mice were harvested 8 weeks post-transplant for analysis of HSC, progenitor, monocyte and neutrophil populations, and cytokine analyses. Consenting human research participants diagnosed with rheumatoid arthritis (RA) were recruited from the Greater Hamilton Area (Ontario, Canada) from 2016-2018. Blood draws occurred prior to any immunomodulatory treatment (baseline), and at 3- and 6-months following treatment with Adalimumab (Humira®), an anti-TNF agent. CHIP status was determined with our successful 48-gene, targeted, Ion-Torrent based sequencing approach to isolated genomic DNA from PBMC. Confirmation of calls and increased sensitivity to detect clones with VAF < 0.02 employed our established single molecule molecular inversion probes (smMIP)-targeted genomic capture technique and high-depth (47,500X avg. coverage) paired-end sequencing, employing high-stringency filters for error-suppression. Statistical analyses were performed in GraphPad Prism V9.2 or R 4.1.2.

Results: 1. Mixed BM chimeric mice of WT and TNF ^-/- genotypes reconstituted with WT CD45.1 ⁺ and Tet2 ^-/- CD45.2 ⁺ HSC showed that age-associated increases in TNF significantly increased by 2- to 2.5-fold the proportions of HSC in old recipient mice, with myeloid lineage skewing (2- to 2.5-fold more granulocyte-monocyte, monocyte-dendritic and common monocyte progenitor cells in BM, and dramatic expansion of Tet2-/- monocytes and neutrophils in blood). This aberrant myelomonocytic advantage was mitigated in old TNF ^-/- recipient mice, suggesting that TNF signalling in the BM is essential for Tet2-mutant myeloid expansion. 2. Age-associated TNF predisposed Tet2^-/- HSC to the development of Ly6C ^high inflammatory monocytes, including increased intracellular and serum TNF levels, further exacerbating an inflammatory environment in favor of Tet2-mutantexpansion. 3. Examination of human RA patients (n=7; avg. age 57y) with serial PBMC sampling revealed one patient with CHIP driven by CBL and PPM1D variants, with a 32% reduction in clone size between 3 and 6 months of anti-TNF therapy (adalimumab). Two additional variants were detected at baseline in this RA patient, TP53 R196* (VAF = 1.34%) and ASXL1 P689L (VAF = 3.13%), but these were not detected at 3- and 6-months post-treatment even with high-depth, error suppression sequencing. Two additional patients had detectible low-level CH clones prior to anti-TNF treatment - one with a STAG2 R1033Q variant at baseline (VAF = 1.65%) and the other with a TET2 baseline E798K variant (VAF = 1.68%) - but not detected during anti-TNF treatment.

Conclusions: We present novel findings that TNF in the aging BM environment has a causal role in driving TET2-mutant CHIP in vivo, and the first examination of the impact of TNF blockade on CH dynamics in humans, suggesting promise for reducing clonal burden. Larger, prospective studies are required to confirm these findings and to determine if inflammatory cytokine blockade may improve CHIP-comorbid conditions and myeloid cancer risk.