Stephan George University of Georgia
Faculty Sponsor(s): Christopher Doering Emory University, Trent Spencer Emory University, Harrison Brown Emory UniversityA growing body of evidence suggests that intracellular pools of rare tRNA isoacceptors are a rate-limiting step in protein biosynthesis, especially in cases of exogenous gene expression. To alleviate this restriction, codon optimization adjusts the codon usage of exogenous DNA to match genes expressed in the host cell’s genome. Recent evidence suggests codon usage of highly expressed genes in a cell can predict steady-state tRNA levels. Therefore, we predicted that codon optimization strategies can be tailored to specific tissues by matching codon usage frequencies with highly expressed genes in a given tissue. By interrogating the codon usage of highly expressed genes in human liver and myeloid cells, we developed liver (LCO) and myeloid (MCO) optimization strategies. LCO and MCO optimization strategies were applied to variants of human factor VIII (FVIII), a coagulation factor in the intrinsic coagulation pathway commonly synthesized in the liver. These optimization strategies were tested against traditional optimization strategies using the human genome (HCO) and a non-optimized control (NoCo). In the human hepatoma-derived HepG2 cell line, LCO FVIII expressed 2-3 fold higher than MCO or NoCo designs. In the non-hepatocyte derived BHK line, LCO and MCO expressed approximately 2-fold lower than NoCo fVIII. Drawing from these studies, we developed a liver-directed adeno-associated viral system expressing LCO and HCO FVIII, which was delivered to a mouse model of hemophilia A. LCO expressed 6-fold higher than HCO FVIII, supporting tissue-specific codon optimization as a platform to improve protein expression in both in vitro and in vivo expression systems.
When & Where
Gallahue Hall 101