Project 1: Mechanism of autophagy

Mechanism of autophagosome biogenesis: During autophagy, a double membrane autophagosome engulfs cargo such as damaged organelles, cytosol, and microorganisms and then fuses with lysosome where its internal components are degraded by acid hydrolases. The formation of autophagosome is driven by the combined effort of lipids and proteins. Phosphatidylinositol 3-phosphate (PtdIns(3)P) has been shown to be extremely important for
autophagosome biogenesis. PtdIns(3)P production is characterized by the class III phosphatidylinositol 3-kinase (PI3KC3) that is evolutionary
conserved from yeast to humans. We are the first group to purify and identify the human PI3KC3 complex, which contains seven components hVps34,
p150/Vps30, Beclin 1/Atg6, UVRAG, Rubicon, Barkor/Atg14(L), and p40. Three of them, Barkor, Rubicon and p40, have not been previously linked to
autophagy. Seven subunits form two mutually exclusive subcomplexes that are targeted to either autophagosome or endosome. We report that the
Beclin 1-hVps34-p150 complex are targeted to autophagosome by Barkor/Atg14(L). Barkor positively regulates the PI3KC3 activity on the
autophagosome membrane, through which it determines the specificity of PI3KC3 in the autophagic pathway. Moreover, we found that Barkor interacts
with the autophagic membrane through its C-terminal 80 amino acids (named the BATS domain), and senses membrane curvature via an ALPS motif within the BATS domain. Hence, we propose that Barkor plays a dual role in autophagosome biogenesis by activating PI3KC3 on membrane and facilitating membrane curvature development, which are crucial for autophagosome biogenesis. The function of p40 is under investigation now. In the future study, we will focus on dissecting the signaling pathway regulating the PI3KC3 complex in autophgay, and we also aim to reconstitute the PI3KC3 lipid kinase activity and membrane deformation activity in vitro.

Selective degradation by autophagy: Autophagy degradation could be both non-selective and selective. We aim to identify selective autophagic
substrates and cargo adaptors. For this purpose, we purified the cellular complex of LC3. In addition to p62 and FYCO-1, two well-known adaptors, we have also identified Keap1. Keap1 is crucial for oxidative stress response. Keap1 is a cargo adaptor but not likely a substrate. Keap1
interacts with p62 and LC3 and colocalizes with ubiquitin aggregates in a stress-inducible manner. Genetic ablation of Keap1 leads to the
accumulation of ubiquitin aggregates, increased cytotoxicity of misfolded protein aggregates, and defective activation of autophagy. More
interestingly, in the LC3 complex, in addition to p62, Keap1 and FYCO-1, we have also identified another set of interacting proteins, the
biological significance of these interaction proteins are being studied in the lab.

Autophagosome maturation by the endosomal PI3KC3 complex: The endosome-localized PI3KC3 subcomplex is important for autophagosome and
endosome maturation. We found that two newly identified PI3KC3 subunits, Rubicon and UVRAG, are both localized to early endosomes. Rubicon
functions as a negative regulator since it antagonizes UVRAG by sequestering it in the complex form. Interestingly, this sequestration is
released by Rab7. GTP-bound Rab7 preferentially interacts with Rubicon and frees UVRAG. In addition, our mapping analysis reveals that different
regions of Rubicon, mediating the interaction with hVps34 and Rab7 respectively, contribute to autophagy and endocytosis differentially.

Project 2: Function of Mule in DNA damage and HDACi induced apoptosis

Apoptosis-inducing DNA damaging agents has been widely used as anti-cancer reagents. New emerging next-generation cancer drugs like histone
deacetylase inhibitors (HDACi) also elicit apoptosis. The apoptotic mechanisms underlying these drugs are not well understood. My lab is
working on one such critical mediator Mule in these drugs induced apoptosis. Mule is named as Mcl-1 ubiquitin ligase E3. Mule also targets
multiple substrates in the DNA damage response. In this study, we generated and characterized Mule knockout MEFs through collaboration with
Dr. Zhenyue Hao and Dr. Tak W Mak at University of Toronto. We found that Mule is crucial for DNA damage and HDACi induced apoptosis, which is due to the elevated HDAC2 activity in Mule null cells. Mule specifically targets HDAC2 for ubiquitination and degradation. The accumulation of
HDAC2 in Mule null cells leads to severe defects in p53 acetylation, phosphorylation, accumulation, transcriptional activation and apoptosis
upon stress. These phenotypes could be partially reversed by administration of HDACi, and fully rescued by lowering elevated HDAC2
protein levels in Mule null cells. This study identifies the Mule-HDAC2 pathway as key mediators in the apoptotic response to HDACi and DNA
damage, and this knowledge might be important for the HDACi application in cancer treatment.

The molecular mechanism underlying the crosstalk among apoptosis, autophagy and necrosis are also being extensively studied in the lab.