UC Berkeley

Inflammation is an evolutionarily conserved host defense response during infection or injury that seeks to remove the causal agent that led to its initiation, repair the damaged tissue(s) and restore homeostasis. Thus, transient inflammation in response to an adequate threat with a quick return to a basal resting state is beneficial. However, when inflammation becomes inappropriately increased or prolonged it can have severe pathophysiological consequences.

During aging, the immune system shifts to a proinflammatory state characterized by low-grade, chronic, sterile inflammation that has been termed ‘inflammaging’. The proinflammatory state largely results from chronic activation of the innate immune system and includes elevated circulating levels of inflammatory mediators including the immune cell signaling molecules (cytokines) Interleukin (IL)-1, IL-6, IL-8, IL-13, IL-18, tumor necrosis factor (TNFα) and antivirals (the type I interferons (IFN-I). Numerous factors are thought to contribute to inflammaging and amongst potential mechanisms high fat feeding/obesity and associated increases in gut permeability to bacterial endotoxins, as well as, cellular senescence have emerged as key contributors. Importantly, inflammaging is tightly correlated to global indicators of poor health status, multimorbidity, impairment in day-to-day living activities and is thought to underlie or accelerate most age-dependent chronic diseases (e.g. cardiovascular disease, diabetes, cancer, as well as, neurodegenerative conditions like Parkinson’s disease (PD) and Alzheimer’s disease (AD)).

While mechanisms driving peripheral inflammaging are beginning to be understood, the etiology of neuro-inflammaging is a crucially unresolved issue. An important source of inflammation in the brain are the non-neuronal cell populations (glia) which provide structural, trophic, and other physiological support for neurons, and especially, the activity of microglia–the brain’s own resident innate immune cells. Microglia are essential for brain development, maintenance and protection throughout the life of an organism. As innate immune cells, however, they can also mount a full inflammatory response to infection or environmental challenge to restore brain health.

Environmental factors, for example, changes in peripheral levels of fatty acids, bacterial endotoxins or proinflammatory mediators resulting from high fat feeding or gut permeabilization can cause microglia to undergo changes that signal activation. Most importantly, microglia lose their homeostatic function during aging, becoming less neuroprotective and increasingly neurotoxic. Microglia-mediated inflammation, for example, is strongly linked to age-induced cognitive impairment, is a common hallmark of both PD and AD, and is believed to be mechanistically important in driving pathogenesis. Thus, there is great interest in discovering factors that regulate age-related changes in microglial inflammatory function.

Although inflammation is a complex and multicomponent response, a key point of its control occurs at the level of gene transcription and involves several classes of transcription factors, transcriptional co-regulators and chromatin modifications. A recent 2017 study by Soreq et al., identified a relatively unknown gene, PHD finger protein 15 (PHF15) as one of the top 25 differentially expressed genes in microglia during non-pathological aging in humans, with PHF15 levels increasing with age. Sequence and structural similarity to other members of the PHF family suggest that PHF15 might function as a putative chromatin-mediated gene regulator.

I first sought to determine whether PHF15 could repress inflammatory function in microglia. If so, I wanted to investigate whether factors known to be causal in inflammaging (e.g. high fat feeding/obesity or cellular senescence) lead to age-dependent cognitive impairment via modulation of microglial PHF15. A major hallmark of senescent cells is the secretion of inflammatory mediators including various cytokines (lL-6, IL-1β, IL-8), chemoattractant cytokines (chemokines; for example C-X-C motif chemokine 10 (CXCL10), C-C motif chemokine ligand (CCL-) 5 (CCL5) and CCL20), antivirals (IFN-I), growth factors and extracellular matrix proteases termed the senescence-associated secretory phenotype (SASP). Secretion of the SASP is partially controlled by the Cyclic GMP-AMP (cGAMP) synthase (cGAS)-Stimulator of interferon genes (STING) (cGAS-STING) cytosolic DNA sensing pathway at the molecular level. Thus, peripheral changes induced by high fat feeding/obesity or the SASP could lead to increased neuroinflammation via inhibition of microglial PHF15.

I show that Phf15 significantly represses proinflammatory gene expression in mouse microglia, modulating both the magnitude and duration of the inflammatory response. Importantly, Phf15 regulates both basal expression and signal-dependent upregulation of proinflammatory genes—which constitute different phases of the transcriptional inflammatory response and are controlled by distinct molecular mechanisms. Global transcriptional changes after Phf15 knockout in a microglial cell line further revealed that Phf15 may specifically regulate the antiviral response, as well as, proinflammatory factor production and secretion. Interestingly, loss of Phf15 resulted in increased IFN-I-dependent and inflammatory gene expression profiles that closely mimic transcriptional changes in aged microglia. Together, my data indicate that Phf15 is an important novel repressor of microglial inflammatory function which might counteract age-induced inflammation in the healthy, aging brain.

Interestingly, I found decreased expression levels of Il-6, a key proinflammatory cytokine expressed by senescent cells in the hippocampus–an area which mediates various memory-related process–of aged (27-month old) STING-deficient mice. However, this decrease did not translate to improved working memory or differences in Phf15 mRNA expression in the brain, suggesting that expression of SASP-related inflammatory factors by the cGAS-STING pathway does not proceed via inhibition of Phf15. Similarly, prolonged treatment with a high fat diet/obesity did not affect working memory or levels of Phf15 in the mouse brain, suggesting that brain inflammation resulting from a HFD or obesity is likewise not a result of Phf15 downregulation.

Overall, my results suggest that Phf15 could function as an immune regulatory checkpoint, restraining the transition from a homeostatic phenotype towards the chronic, proinflammatory, IFN-I responsive state seen in microglia in the aged brain. Further understanding of its exact mechanism of action could lend insight into possible future therapeutic intervention.