Inflammatory Crosstalk Drives Clonal Hematopoiesis

The inflammatory response is a protective host defense mechanism that is orchestrated by effector cells that are recruited to sites of tissue damage or infection. It has been recognized for many years that inflammation, often in association with aging, can also contribute to disease pathogenesis in many different contexts — so-called “inflamm-aging.” Numerous recent studies have highlighted the role of inflammation in promoting a fitness advantage of hematopoietic stem/progenitor cell (HSPC) clones carrying mutations of certain genes (e.g., DNMT3A, ASXL1, TET2, TP53), to promote the development of clonal hematopoiesis (CH). Because CH is also associated with development of inflammation-associated conditions including blood cancer, venous thromboembolism, and cardiovascular disease, understanding how inflammation contributes to enhanced HSPC fitness is crucial to developing rational strategies to intervene and reduce the risk of disease complications.

To understand the interplay between inflammation and CH, investigators have turned to model systems. TET2 and JAK2 mutation CH mouse models show increased inflammation in myeloid cells, contributing to atherosclerosis through heightened IL-1β/IL-6 signaling; inhibition of NLRP3 or AIM2 inflammasome ameliorates the atherosclerosis.^1,2  More recently, inflammation in association with chronic infection was shown to promote the development of DNMT3 mutation–associated CH via interferon-γ signaling.³  In this study, the authors used a bone marrow transplantation mouse model to generate chimeric mice carrying Dnmt3a knockout and wild-type hematopoietic cells in competition. Chronic infection promoted the fitness advantage of Dnmt3a knockout HSPC — an effect that could be recapitulated with recombinant interferon-γ. The authors of this study propose that DNA methylation silences differentiation programs in Dnmt3a knockout HSPC that might otherwise lead to exhaustion of wild-type HSPCs after excessive stimulation with interferon-γ. Although these model systems are innovative, they do not model true CH because the mutations do not originate in single hematopoietic stem cell clones. Consequently, the crucial process through which an HSPC clone exerts a substantial fitness advantage to become clonally dominant remains poorly understood.

To tackle this question, Dr. Serine Avagyan and colleagues⁴  from Dana-Farber/Boston Children's developed a system to color-label individual HSPC clones in zebrafish, allowing them to be tracked over time in their native microenvironment without perturbations such as transplantation. By combining this approach with a runx1 mutation, which generates fewer HSPCs during development and mosaic mutagenesis with CRISPR-Cas9, the authors developed an elegant system they termed “TWISTR” (tissue editing with inducible stem cell tagging via recombination). By targeting CRISPR-Cas9 guides to orthologues of CH-associated genes in humans, they were then able to track individual clones, carrying CH-associated mutations over time. CH-mutant zebrafish frequently showed a dominant clone, identified with a single color. By comparing the distribution of mutations in nonhematopoietic tissues, they found that frameshift mutations in asxl1, dnmt8 (DNMT3A orthologue), tet2, and tp53 were more likely to be found in dominant clones.

The beauty of the TWISTR method is that it then allowed the researchers to prospectively purify live cells from different clones for analysis. Single-cell RNA sequencing showed overexpression of inflammation-associated gene expression, with overexpression of inflammatory cytokines from CH-associated macrophages and neutrophils. As inflammation is a cardinal regulator of HSPC function, with many effects on HSPC fate and function⁵  including proliferation-induced DNA-damage and depletion of HSPCs,⁶  the authors next reasoned that mutant HSPCs might evade inflammation-associated depletion. They determined that CH-associated HSPCs overexpressed suppressors of inflammation such as nr4a1 and socs3a. A genetic rescue experiment showed that homozygous deletion of nr4a1 in asxl1-mutant HSPCs reversed the fitness advantage typically associated with frameshift asxl1 mutation, whereas nr4a1 deletion had no effect on normal HSPCs.

Taken together, these results provide evidence that CH-associated mutations lead to generation of an inflammatory microenvironment, generated by mature myeloid cells, which cell-extrinsically promote a fitness advantage of HSPCs from the same clone whilst suppressing normal HSPCs. This pathogenic cellular crosstalk adds to accumulating evidence that certain CH-associated mutations lead to dysregulation of specific inflammatory cytokines such as interleukin-6 with TET2,⁷  contributing both to HSPC fitness advantage and vascular complications of CH. Given the complex inflammatory milieux in older individuals with CH, it seems unlikely that CH in humans will be mediated by a single inflammatory axis. Furthermore, other understudied cell types such as platelets might also be implicated, in addition to myeloid cells. Understanding the cellular and molecular framework through which CH-associated mutations exert clonal dominance through these elegant model systems will pave the way to design rational multimodal anti-inflammatory strategies that might reverse the fitness advantage and disease-complications associated with inflamm-aging.