Our lab has been interested in understanding mechanisms controlling distinct kinds of autophagy, in particular, macroautophagy and microautophagy, and how changes in their function lead to age-related diseases. Over the last several years, our focus has been on the following aspects: (a) autophagy and its regulation of hepatic and systemic lipid metabolism, (b) roles of autophagy in the CNS and regulation of systemic aging, (c) relationship between timed-feeding and the circadian clock on autophagy, (d) mechanisms of MASLD development, and (e) relationship between lipids, mTOR and regulation of autophagy.

Macroautophagy or autophagy maintains quality control by degrading unwanted cytoplasmic contents in acidic lysosomes. Autophagy occurs at basal levels in all cells and is induced with starvation and stress. Autophagy requires ~35 ATG proteins to orchestrate the formation of autophagosomes labelled with LC3, which sequester cargo, e.g., mitochondria, ER, lipid droplets, protein aggregates or even unwanted soluble proteins, and then fuse with lysosomes to deliver its contents. It is well-characterized that defects in autophagy lead to almost every known age-related chronic disease, e.g., diabetes, neurodegeneration, metabolic liver disease or sarcopenia. Hence our goal is to find ways to stimulate autophagy to extend healthspan.

Autophagy and hepatic lipid metabolism. We demonstrated a role for autophagy in the mobilization and degradation of intracellular lipid stores by a process we termed Lipophagy, thus mapping autophagy to metabolic regulation. Using live cell video-microscopy we found that lipid droplets are engulfed by autophagosomes or broken down piece meal. The sequestered lipid is delivered to lysosomes where lysosomal lipases generate free fatty acids that are utilized for energy. Our discovery of lipophagy in Nature, 2009 has since been cited in ~4,300 articles showing the existence of lipophagy in diverse cell types such as macrophages, neurons, foam cells, adipocytes and myocytes, to mention a few.

Discovery of the lipophagy receptor. Our proteomics of livers of fasted mice showed enrichment of phosphorylated VPS4A and VPS4B, which are AAA+ ATPases involved in the ESCRT pathway/microautophagy. While initially looking at microlipophagy, all our results indicated that phosphorylated VPS4A physically tethers lipid droplets to LC3. Indeed, using structural biology, super-resolution imaging, and biochemistry, we show that the N-terminus MIT domain of VPS4A binds to lipid droplets, and it uses at least two of its several LC3-interacting region domain (LIR) motifs to bind to LC3/autophagosomes. Mutating key residues on VPS4A to inactivate its MIT or LIR domains or knocking-down VPS4A per se blocks LD ingestion by autophagosomes, and marked lipid accumulation/hepatic steatosis. Furthermore, humans with obesity and age-associated steatotic liver disease show significantly reduced VPS4A levels in livers. The discovery of VPS4A as the selective receptor for lipophagy provides us with a molecular target to stimulate lipophagy, and has been published in Molecular Cell, 2024.

We have shown the Integrative role of hypothalamic autophagy in the regulation of feeding and peripheral metabolism. We have shown that fasting-induced increased availability of circulating free fatty acids leads to activation of lipophagy in hypothalamic neurons. This leads to generation of neuron-intrinsic free fatty acids in a controlled manner that drives orexigenic AgRP expression and stimulate feeding (first model on the right). As a result, mice knocked-out for the autophagy gene Atg7 in AgRP neurons ate less and remained lean. Please see paper in Cell Metabolism, 2011.

By contrast, blocking autophagy in anorexigenic POMC neurons resulted in lower levels of neuronal alpha-MSH, hyperphagia, and obesity, suggesting a role for autophagy in the processing of POMC to alpha-MSH (second model on the right); EMBO Reports, 2012. We also showed that POMCergic autophagy declines with age and correlates with age-associated obesity and insulin resistance.

On similar lines, we have also shown that Autophagy in POMC neurons is required to stimulate autophagy and lipophagy in peripheral tissues, liver and brown fat. We find that exposure to cold leads to induction of autophagy in both hypothalamic POMC neurons and peripheral tissues. This induction of POMCergic autophagy is needed for autophagy/lipophagy activation in liver and adipose tissues. Indeed, knocking-out the core autophagy gene Atg7 in POMC neurons led to the blockage of cold-induced lipophagy and accumulation of neutral lipid in liver and fat. In contrast, stimulating autophagy in POMC neurons through hypothalamic injections of rapamycin led to stimulation of autophagy and rapid depletion of lipid in peripheral tissues, see Cell Metabolism, 2016. This study reveals a CNS-to-peripheral autophagy axis that regulates systemic lipid utilization.

Our lab is interested in relationship between Autophagy and the Circadian Clock. We have demonstrated that autophagy degrades the clock repressor Cry1 to maintain glucose homeostasis, published in Cell Metabolism, 2018. We have found that autophagy in liver is stimulated between 3pm and 7pM, which is the time-frame when mice typically eat the least. During this time, the autophagy protein LC3 interacts with CRY1 via LC3-interacting region (LIR) motifs on CRY1. This results in the sequestration and degradation of CRY1. Under physiological states autophagy only degrades CRY1 and not Bmal1, Clock or Rev-Erb proteins. Given that CRY1 is a robust suppressor of gluconeogenesis, autophagic degradation of CRY1 allows gluconeogenesis to occur at maximal rates. On the basis of our studies, we are interested in developing drugs to prevent diabetes by interfering with LC3-CRY1 interaction, which will cause the stabilization of CRY1 such that gluconeogenesis is dampened and blood glucose levels are lowered.

Our laboratory is interested in understanding how autophagy decreases with age, and whether restoration of autophagy prevents the metabolic syndrome of aging. To this end, we have established an Isocaloric twice-a-day (ITAD) feeding intervention in mice that stimulates autophagy and prevents the metabolic syndrome of aging. Major benefits of ITAD feeding are improved glucose tolerance, lower lipids levels, and increased muscle mass. Using different mice that lacked autophagy in POMC neurons or liver or fat tissue or myogenic progenitors, we have found that autophagy in different tissues contributes to distinct benefits of ITAD feeding, published in Cell Metabolism, 2017.

Following up on this work to understand the mechanism of intermittent fasting and metabolic flexibility, we have now shown that feeding twice-a-night (TAN) generates food-driven surges of circulating insulin and leptin. Insulin and leptin surges cause cellular, functional, and metabolic remodeling of subcutaneous adipose tissue (sWAT), resulting in increased energy expenditure. Our single-cell RNA-sequencing analyses and flow cytometry show a role for insulin and leptin surges in innate lymphoid type 2 (ILC2) cell recruitment and sWAT browning. We also show that adipose tissue innervation is required for the adipose remodeling occurring with TAN feeding. Consistently, recreating insulin and leptin oscillations via once-a-day timed co-injections is sufficient to favorably remodel innervated sWAT. Hence, reorganization of nutrient-sensitive pathways by feeding only in distinct windows remodels sWAT and drives the metabolic benefits of intermittent fasting, published in Cell Reports, 2023.

Our lab has been interested in understanding the roles of mTOR signaling in relation to Autophagy, Lipid Metabolism and the Biology of Aging. We have recently shown in Nature Cell Biology, 2023 that fasting or lipid availability stimulates mTORC2, which phosphorylates its downstream target NDRG1 at serine 336. This drives mitochondrial fission and respiratory sufficiency. Using time-lapse imaging we show that NDRG1, but not the phosphorylation-deficient NDRG1 Ser336Ala mutant, engages with mitochondria to facilitate fission in cells, including those lacking the fission protein DRP1. Using proteomics, we identify that mTORC2-phosphorylated NDRG1 cooperates with small GTPase CDC42 and its effectors and regulators to orchestrate fission. In sum, while during nutrient surplus, mTOR complexes perform anabolic roles; however, paradoxical reactivation of mTORC2 during fasting unexpectedly drives mitochondrial fission and respiration. Please see Figure on the right (generated by Biorender).

The Laboratory of Autophagy and Integrative Aging