Genetic engineering oleaginous yeasts for improved lipid production

Oleaginous yeasts are promising organisms for the production of lipid-based chemicals and fuels from simple sugars. We chose oleaginous yeast Rhodosporidium toruloides for the production of lipid-based products, because this oleaginous yeast natively produces lipids at high titers and can grow on glucose and xylose. However genetic tool for this host is very limited. We successfully transformed R. toruloides using Agrobacterium tumefaciens mediated transformation. We engineered two R. toruloides strains for increased lipid production by over-expressing the native acetyl-CoA carboxylase and diacylglycerol acyltransferase genes. Our best strain RT880-AD, derived from IFO0880, was able to produce 16.4±1.1 g/L lipid from 70 g/L glucose and 9.5±1.3 g/L lipid from 70 g/L xylose in shake-flask experiments. Next, in order to further improve lipid production in R. toruloides, we explored more genetic engineering routes. We found over-expressing malic enzyme and stearoyl CoA desaturase could improve lipid production 23% and 28 % in R. toruloides IFO0880 strain, but over-expression of glycerol-3-P dehydrogenase and pyruvate carboxylase failed to improve lipid titers. Deletion of peroxisomal structure gene PEX10 decreased overall biomass growth and lipid titer as well. Over-expressing stearoyl CoA desaturase improved lipid production further 13% in shake-flask cultures from strain RT880-AD, where ACC1 and DGA1 were also over-expressed. Malic enzyme over-expression, on the contrary, decreased lipid production from RT880-AD strain. The triple over-expression strain RT-ADS (ACC1, DGA1 and SCD) achieved 89.4 g/L lipid in fed-batch bioreactor, representing 396% increase from IFO0880 negative control strain and 42% increase from RT880-AD. The production rate for RT880-ADS was 5.0 times the rate in RT880-N and 1.4 times the rate in RT880-AD. These work represents one of the first metabolic engineering in R. toruloides and establishes this yeast as a new platform for production of lipid-derived products. On the other hand, we also explored another oleaginous yeast Yarrowia lipolytica, which previously has received more attention in the literature. It is both a citric acid producing and oleaginous yeast. During a medium optimization process, we found pH influences citric acid and lipid production in Y. lipolytica, and its effects on the citric acid and lipid were opposite. We then used pH as the tuning parameter to investigate how carbon flux distributes into these two different pathways. From RNASeq data, we didn’t find significant overall elevated expression in lipid synthesis pathways at low pH when lipid production was favored. We proposed a model that transport limits citric acid secretion in acidic pH, because the reverse gradient for proton is too high between extracellular pH and intracellular (neutral) pH in more acidic medium to pump out proton