Defining and Characterizing Cell Signal Transductions in the Sestrin2 Pathway

Abstract

Sestrins are highly conserved, stress-inducible proteins that protect against cellular damage and the progression of age-associated and diabetic pathologies. These include fat accumulation, insulin resistance, muscle degeneration, cardiac dysfunction, mitochondrial pathologies, and tumorigenesis. Sestrin has two established anti-aging functions: reducing oxidative stress and regulating the mechanistic target of rapamycin (mTOR) signaling network. This thesis focuses on the regulation of mTOR signaling. First, I defined a new pathway in the Sestrin2 signaling network. Sestrin2 strongly activates AKT, a major metabolic regulator downstream of the insulin signaling cascade. I found that the mechanism for Sestrin2-induced AKT activation occurred through two large protein complexes: GAP Activity TOwards Rags 2 (GATOR2), a pentameric protein complex that binds to Sestrin2, and mTOR Complex 2 (mTORC2), a kinase upstream of AKT. GATOR2 bridged Sestrin2 and mTORC2 and was functionally required for Sestrin2-induced AKT activation. In addition, Sestrin2 bound to AKT’s pleckstrin homology (PH) domain and induced AKT translocation to the plasma membrane. Although Sestrin2 and GATOR2 were known to inhibit mTORC1, this work defined a new mechanism for Sestrin2 and GATOR2 in activating mTORC2. Next, I characterized other components in the Sestrin2 signaling network by generating transgenic mice with liver-specific knockouts. mTORC1 has two signaling arms required for full activation: nutrient sensing and growth factors. Nutrients input to the GATOR1 complex, which consists of Nprl2, Nprl3, and DEPDC5, while growth factors input to the TSC complex. First, I generated a DEPDC5 knockout mouse to constitutively activate the nutrient sensing arm of mTORC1. I found that liver-specific DEPDC5 deletion (Depdc5Δhep) showed enlarged zone 3 hepatocytes, increased sensitivity to acetaminophen toxicity, and decreased lipid accumulation in the liver. Next, I generated Depdc5Δhep/Tsc1Δhep double knockout (DKO) mice to concurrently activate both the growth factor and nutrient sensing arms of mTORC1 in the liver. The DKO mice showed severe phenotypes by two months with significantly lower body weights and more liver injury. Interestingly, the severe pathological phenotypes were all reversed after only 10 days of intraperitoneal injections of either rapamycin, an mTORC1 inhibitor, or Tempol, an oxidative stress reducer. Surprisingly, tauroursodeoxycholic acid (TUDCA), an ER stress reliever, was lethal to the DKO mice, suggesting a potential feedback mechanism for ER stress in the context of hyperactive mTORC1. In addition, transcriptomic analysis of WT, Depdc5Δhep, Tsc1Δhep, and DKO livers showed that DKO livers had distinct transcriptomic profiles with selective upregulation of oxidative stress genes. Taken together with the Tempol experiment, this suggests that hyperactive hepatic mTORC1 induces liver injury through oxidative stress signaling. Finally, systemic characterization without any interventions showed that 2-month-old DKO mice had insulin resistance, and 5-month-old DKO mice exhibited spontaneous hepatocellular carcinoma (HCC). Overall, this work developed and characterized a new mouse model of hyperactive mTOR that could potentially be utilized for future studies as a 2-month nonalcoholic steatohepatitis (NASH) model or 5-month HCC model. In summary, the work from this thesis has furthered our understanding of Sestrin2 signaling. First, we defined a novel mechanism where Sestrin2 and GATOR2 activate the mTORC2 pathway. Secondly, we developed and characterized a genetic mouse model with a hyperactive mTORC1 liver that spontaneously develops NASH by two months and HCC by five months. This work provides a mechanistic understanding of how Sestrin is able to protect against diabetic pathological progression and provide a new tool for future studies of liver disease.