Abstract
It is well established that aging is accompanied by chronic low-grade inflammation that correlates well with
mortality and morbidity, but the association between aging and inflammation is not well understood. Similarly,
changes in DNA cytosine methylation occur reproducibly with age, suggesting that DNA methylation contributes
to the aging process, but the biology remains unclear. Certain families of transposable elements (TEs) are
repressed in somatic cells via DNA methylation; moreover, TEs are activated in aging mice and in cultured
human cells undergoing replicative senescence, and increased TE expression leads to DNA damage, mutations
and inflammation. These consequences have all been observed in cells from aged individuals, but the underlying
mechanisms are obscure. Understanding the mechanistic basis for the association of aging with inflammation
and changes in DNA methylation will be essential to design rational interventions for age-associated disorders.
Here we tackle the question of how changes in DNA methylation relate to inflammation during aging. DNA
methylation is regulated by DNA methyltransferases (DNMTs) and TET methylcytosine oxidases, which control
DNA methylation and demethylation respectively. DNMT deficiency predictably results in decreased DNA
methylation, but TET deficiency is paradoxically also associated with a striking loss of DNA methylation in hetero-
chromatin. Moreover, TET and/or DNMT-deficient cells show increased TE expression and cell-intrinsic
inflammation, similar to that observed in cells from aged humans and senescent cells in culture. We will test the
hypothesis that aging involves a progressive decrease in TET and/or DNMT activity with age, leading to selective
losses of DNA methylation in heterochromatin where repressed TEs reside, and hence in increased TE
expression and cell-intrinsic (“sterile”) inflammation. In Aim 1, we will define the DNA modification (5mC, 5hmC)
status of transposable elements in immune cells (CD4+ T lymphocytes and monocytes) of young, middle-aged
and old healthy subjects from the BLSA and GESTALT cohorts at NIA, and relate them to the observed increase
in expression of TEs as a function of age. These studies probe the hypothesis that decline of TET and/or DNMT
activity with age dysregulates TE expression. In Aim 2, we will correlate increased TE expression in immune
cells from the BLSA and GESTALT cohorts with signatures of inflammation and serum levels of pro-inflammatory
cytokines during aging. In Aim 3, we will perform complementary studies in genetically tractable mouse models
of clonal hematopoiesis, a malady of both aging and inflammation. Using mice deficient in Tet2 and/or Dnmt3a,
we will define the development of heterochromatic DNA hypomethylation and inflammation in immune cells of
these mice with age, and determine the relation to increased TE expression. Our studies will test the novel
hypothesis of a link between TET/DNMT deficiency, loss of heterochromatic DNA methylation and increased TE
expression with age, and deepen our understanding of the elusive mechanisms that connect heterochromatin
dysfunction with inflammation, aging and oncogenesis.
