Age-Related Changes in Hematopoietic Stem Cell Proteostasis Promote the Emergence of Clonal Hematopoiesis and Leukemia in Older Adults

Hematopoietic stem cells (HSCs) regenerate blood and immune cells throughout life but their function declines with age. Aging HSCs exhibit reduced self-renewal, decreased regenerative capacity, and myeloid-biased differentiation that contribute to increased incidence of immune dysfunction, clonal hematopoiesis and leukemia in older adults. Mitigating age-related declines in HSC function could thus have a profound impact on extending healthspan. Efforts to enhance aged HSC function, however, have had limited success underscoring our incomplete understanding of the underlying causes of age-related stem cell dysfunction.

Loss of protein homeostasis (proteostasis) is a hallmark of aging, but whether it contributes to age-related HSC dysfunction is largely unknown. RNA-sequencing studies revealed that many of the top upregulated gene sets in old HSCs are related to proteostasis, suggesting that proteostasis may be challenged in aging stem cells. In surprising contrast, however, we found that old HSCs did not exhibit substantial changes in protein synthesis or accumulate misfolded or unfolded proteins. Given that proteostasis has emerged as fundamentally important for HSC self-renewal, these opposing results raised the possibility that age-related transcriptional rewiring of HSCs may be driven by a selective pressure to preserve proteostasis. Indeed, we determined that aging HSCs activate Hsf1, a key proteostasis sensor and master transcription factor of the heat shock response, whose activation was associated with elevated expression of multiple heat shock proteins in old HSCs. Conditional deletion of Hsf1 had minimal effects on young HSCs but resulted in significant accumulation of misfolded and unfolded protein and decreased self-renewal activity in old HSCs in vivo. These data indicate that HSCs activate Hsf1 to preserve proteostasis and fitness during aging.

Although Hsf1 activation helps preserve HSC function during aging, Hsf1 is also activated in many cancers, including acute myeloid leukemia (AML), where it confers stress resistance and disease progression. Hsf1 activation has always been proposed to occur after transformation, but whether pre-existing Hsf1 activation promotes cancer initiation is unknown. We tested if age-related Hsf1 activation in HSCs promoted the emergence of clonal hematopoiesis and leukemia in older adults. We developed a novel model of age-related clonal hematopoiesis by introducing Dnmt3a^R878H mutations in rare (~1%) young adult HSCs that enabled us to track clonal expansion throughout life. Dnmt3a^R878H HSCs expanded by ~20-fold and dominated the HSC pool by 17 months of age. Remarkably, conditional deletion of Hsf1 severely attenuated the age-related clonal expansion of Dnmt3a^R878H HSCs by 81%. These data indicate that age-related Hsf1 activation promotes the emergence of clonal hematopoiesis. Furthermore, we introduced a secondary transforming Nras^G12D mutation in Dnmt3a^R878H HSCs that led to splenomegaly, an aberrant expansion of pre-leukemic multipotent progenitors, and ultimately a fatal age-associated myeloid neoplasm. Strikingly, preventing age-related Hsf1 activation via conditional genetic deletion significantly extended overall survival and decreased penetrance of fatal neoplasms from 69% to 27% after 1 year.

These studies reveal that a selective pressure to preserve proteostasis results in transcriptional reprogramming of aging HSCs that helps preserve stem cell fitness at the cost of increasing the risk of clonal hematopoiesis and cancer. These new connections between molecular proteostasis and physiological changes in aging HSCs could uncover new therapeutic opportunities to prevent clonal hematopoiesis and leukemia by targeting underlying age-related changes in stem cell proteostasis.